New insights into ergot alkaloid biosynthesis in Claviceps purpurea: An agroclavine synthase EasG catalyses, via a non-enzymatic adduct with reduced glutathione, the conversion of chanoclavine-I aldehyde to agroclavine

Article abstract of DOI:10.1039/c0ob01215g

Ergot alkaloids are indole derivatives with diverse structures and biological activities. They are produced by a wide range of fungi with Claviceps purpurea as the most important producer for medical use. Chanoclavine-I aldehyde is proposed as a branch point via festuclavine or pyroclavine to clavine-type alkaloids in Trichocomaceae and via agroclavine to ergoamides and ergopeptines in Clavicipitaceae. Here we report the conversion of chanoclavine-I aldehyde to agroclavine by EasG from Claviceps purpurea, a homologue of the festuclavine synthase FgaFS in Aspergillus fumigatus, in the presence of reduced glutathione and NADPH. EasG comprises 290 amino acids with a molecular mass of about 31.9 kDa. The soluble monomeric His₆-EasG was purified after overproduction in E. coli by affinity chromatography and used for enzyme assays. The structure of agroclavine was unequivocally elucidated by NMR and MS analyses.

Full text of DOI:10.1039/c0ob01215g

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 4328
www.rsc.org/obc
PAPER
New insights into ergot alkaloid biosynthesis in Claviceps purpurea: An
agroclavine synthase EasG catalyses, via a non-enzymatic adduct with
reduced glutathione, the conversion of chanoclavine-I aldehyde to agroclavine†
Marco Matuschek,a Christiane Wallwey,a Xiulan Xieb and Shu-Ming Li*a
Received 20th December 2010, Accepted 15th March 2011
DOI: 10.1039/c0ob01215g
Ergot alkaloids are indole derivatives with diverse structures and biological activities. They are
produced by a wide range of fungi with Claviceps purpurea as the most important producer for medical
use. Chanoclavine-I aldehyde is proposed as a branch point via festuclavine or pyroclavine to
clavine-type alkaloids in Trichocomaceae and via agroclavine to ergoamides and ergopeptines in
Clavicipitaceae. Here we report the conversion of chanoclavine-I aldehyde to agroclavine by EasG from
Claviceps purpurea, a homologue of the festuclavine synthase FgaFS in Aspergillus fumigatus, in the
presence of reduced glutathione and NADPH. EasG comprises 290 amino acids with a molecular mass
of about 31.9 kDa. The soluble monomeric His6-EasG was purified after overproduction in E. coli by
affinity chromatography and used for enzyme assays. The structure of agroclavine was unequivocally
elucidated by NMR and MS analyses.
46 to 66% on the amino acid level were found in both gene clusters,
which are assumed to be involved in the early steps towards the
ergoline ring (Fig. 2).1,11
Introduction
Ergot alkaloids are a complex family of indole derivatives with
diverse structures and biological activities.1 They are toxins
produced by fungi of the families Clavicipitaceae, e.g. Claviceps
and Neotyphodium, and Trichocomaceae including Aspergillus
and Penicillium.2,3 On the other hand, some of these compounds
and their semi-synthetic derivatives are important drugs, which are
used in modern medicine, e.g. ergotamine and dihydroergotamine
for the treatment of migraines4,5 and dihydroergotoxine for
dementia and age-related cognitive impairment.6,7
A characteristic structural feature of most ergot alkaloids
is the tetracyclic ergoline ring system.1,2,8 Fungi of the family
Trichocomaceae produce mainly clavine-type alkaloids consisting
of the ergoline system, whereas amides and peptides of lysergic
acid, also termed ergoamides and ergopeptines respectively, are
mainly found in fungi of the family Clavicipitaceae.1
The biosynthesis of the common ergoline ring begins with
the prenylation of L-tryptophan by
a dimethylallyltrypto-
phan synthase, e.g. FgaPT2, resulting in the formation of 4-
dimethylallyltryptophan (4-DMAT),11 which is then converted to
N-methyl-4-dimethylallyltryptophan (4-DMA-L-abrine) by an N-
methyltransferase, e.g. FgaMT.12 Lorenz et al.13 have shown that
the product of ccsA in C. purpurea (formerly known as cpox1 or
easE, a homologous gene of fgaOx1 in A. fumigatus) is involved
in the conversion of 4-DMA-L-abrine to chanoclavine-I (Fig. 2).
An alcohol dehydrogenase, e.g. FgaDH, catalyses the conversion
of chanoclavine-I to chanoclavine-I aldehyde (1).14
It has been demonstrated that 1 is the branch point for
the biosynthetic pathways of ergot alkaloids in Claviceps via
agroclavine (2) and in Aspergillus via festuclavine (3).15 2 and 3
were isolated from C. purpurea and A. fumigatus, respectively.16–18
2 differs from 3 in the presence of a double bond between C8
and C9. For the formation of 2 from 1, one reduction and an
isomerisation step would be necessary. In contrast, for the conver-
sion of 1 to 3, two reduction steps and possibly an isomerisation
process are required. It has been shown that 1 is converted to
3 by FgaFS in the presence of the old yellow enzyme FgaOx3
in A. fumigatus.19 3 serves then as a precursor of fumigaclavines
A, B and C.1 In C. purpurea, it is expected that 1 is converted
to 2, which is then converted via elymoclavine to lysergic acid,
the acidic component of ergoamides and ergopeptines (Fig. 2).20
During the preparation process of this manuscript, Cheng et al.21
reported the conversion of 1 to 2 by FgaFS from A. fumigatus in
The biosynthetic gene cluster for ergoamides and ergopeptines
in Claviceps purpurea has been proposed to contain 14 genes,
whereas the gene cluster of the clavine-type alkaloid fumigaclavine
C in Aspergillus fumigatus likely comprises 11 genes (Fig. 1).9–11
Seven orthologous/homologous genes with sequence identities of
aPhilipps-Universita¨t Marburg, Institut fu¨r Pharmazeutische Biologie
und Biotechnologie, Deutschhausstrasse 17A, D-35037, Marburg, Ger-
many. E-mail: shuming.li@Staff.uni-Marburg.de; Fax: 0049-6421-2825365;
Tel: 0049-6421-2822461
bPhilipps-Universita¨t Marburg, Fachbereich Chemie, Hans-Meerwein-
Strasse, 35032, Marburg, Germany
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c0ob01215g
4328 | Org. Biomol. Chem., 2011, 9, 4328–4335
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 1 Biosynthetic gene clusters of ergot alkaloids from Claviceps purpurea and of fumigaclavine C from Aspergillus fumigatus Af293. Homologous
genes are indicated as black arrows.
Fig. 2 Putative biosynthetic pathways of ergot alkaloids in Aspergillus fumigatus and Claviceps purpurea.
the presence of EasA from Neotyphodium lolii, a homologue of
FgaOx3 from A. fumigatus. These authors mentioned that EasA
from C. purpurea was catalytically inactive after heterologous
expression in E. coli due to a lack of the expected flavin
cofactor.21
NMR and MS analyses including HSQC, HMBC and ROESY
spectra.
Results and discussion
Here we report that EasA is not necessary for the conversion of
1 to 2, at least in our in vitro experiments. EasG of C. purpurea,
a homologue of FgaFS from A. fumigatus, is sufficient for the
formation of 2 from 1 in the presence of NADPH, via a non-
enzymatic adduct with reduced glutathione (GSH). The chemical
structure of the enzyme product 2 was unequivocally proven by
Sequence analysis and cloning of easG and easA
The deduced gene product of easG from C. purpurea (AY836771
in GenBank) comprises 290 amino acids and has a calculated
molecular mass of 31.9 kDa. In the case of easA (AJ703809 in
GenBank), the deduced gene product consists of 369 amino acids
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4328–4335 | 4329
©

View Article Online
and has a calculated molecular mass of 41.5 kDa. By using the pro-
gram BLAST 2 SEQUENCES (http://blast.ncbi.nlm.nih.gov),
EasG and EasA from C. purpurea share sequence identities of
47% and 58% with their homologues from A. fumigatus, i.e.
FgaFS (CAG28312.1) and FgaOx3 (AAW57089.1), respectively.
A structurally informed sequence alignment for EasG and FgaFS
as well as for EasA and FgaOx3 is provided as Fig. S1 (ESI†).
For the overexpression in E. coli, the sequences of easG
and easA from C. purpurea had been synthesised chemically
by MWG Biotech AG and cloned into the NcoI and BamHI
sites of the expression vector pQE60, resulting in the expression
vectors pMM044 containing easG and pMM046 containing easA,
respectively. The sequences in both vectors were also confirmed by
sequencing.
we speculated that the conversion of 1 to 2 in C. purpurea would
be catalysed by their homologues EasG and EasA. Therefore, 1
was incubated with the purified EasG and EasA in the presence
of NADH and FMN, which were successfully used in the enzyme
assays for the conversion of 1 to 3.19 1 is commercially not available
and was therefore prepared enzymatically from chanoclavine-I
with FgaDH according to Wallwey et al.14 (See experimental
section). Surprisingly, no product peak was observed in the
HPLC chromatogram of the incubation mixture of 1 with EasG
and EasA (Fig. 4B). Replacing NADH by NADPH did not
improve the result (Fig. 4C). One explanation for this unexpected
result could be that the recombinant proteins EasG and/or
EasA were catalytically inactive, as speculated by Cheng et al.22
However, this would be in contrast to the successful conversion
of 1 by FgaFS and FgaOx3.19 Another reason could be that
the reaction conditions are not correct, e.g. a required cofactor
is missing. Therefore, different cofactors were tested for the
incubation mixtures of 1 with both EasG and EasA. The HPLC
chromatogram of 1 with EasG and EasA, in the presence of 1 mM
reduced GSH, showed a minor peak with the same retention time
as that of 2 (Fig. 4D). A low conversion of 6% was observed under
this condition. Interestingly, the incubation mixture without EasA
also showed a similar profile in the HPLC chromatogram (Fig. 4E).
However, the consumption of 1 was much higher than the product
peak in Fig. 4D and 4E. This discrepancy is caused by the chemical
reaction of GSH with 1, which is clearly observed in the absence of
the enzymes (see below and Fig. S1, ESI†). Addition of NADPH
to the reaction mixture of EasG and GSH enhanced the product
formation significantly (Fig. 4F). A conversion rate of 59% was
detected in the reaction mixture. However, incubation of 1 with
EasG and NADPH, lacking GSH, did not result in the formation
of any product (Fig. 4G). Product formation was only detected in
assays with active (Fig. 4F), but not with heat-inactivated EasG
(Fig. 4H). In the presence of 1 mM GSH, product formation
showed a linear dependence on the protein amount up to 0.1 mg
(Fig. S2, ESI†) and on the incubation time up to 10 min (Fig. S3,
ESI†). This proved that EasA was not necessary for the conversion
of 1 to (very likely) 2. EasG was sufficient for this conversion in
the presence of GSH and NADPH. The low conversion without
NADPH (Fig. 4E) can be explained by traces of NADPH in
the purified EasG. A conversion of 2 to 3 was not observed in
incubation mixtures with EasG and EasA under various reaction
conditions (data not shown), indicating that agroclavine is not
likely to be an intermediate to festuclavine.
Overproduction and purification of His6-EasG and His6-EasA
For the expression of easG and easA, E. coli cells harbouring
pMM044 and pMM046 were cultivated at 26 ◦C and 22 ◦C
respectively, and were induced by addition of IPTG to a final
concentration of 0.1 mM (see Experimental Section). The His6-
tagged EasG and EasA proteins were purified on Ni-NTA agarose.
With pMM044, a major protein band with a migration above the
30 kDa size marker was detected (Fig. 3), as expected for the His6-
EasG with a calculated mass of 33.3 kDa. For EasA, a major
protein band with a migration below the 45 kDa size marker was
observed (Fig. 3), as expected for the His6-EasA with a calculated
mass of 41.5 kDa. The protein yield was 3.4 mg per litre of culture
for purified His6-EasG and 0.5 mg per litre of culture for purified
His6-EasA.
To determine the structure, 1.5 mg of 2 were isolated on HPLC
after incubation of 1 with EasG in the presence of GSH and
NADPH, and subjected to NMR and MS analyses. The 1H-NMR
spectrum of 2 in its protonated form corresponded perfectly to that
of an authentic agroclavine sample after treatment with HPLC
solvents containing trifluoroacetic acid (Fig. S13 and S14, ESI†).
The identity of the agroclavine standard had been confirmed by
Fig. 3 Analysis of the overproduction and purification of His6-EasA
(lanes 1-4) and His6-EasG (lanes 6-9). The proteins were separated on a
15% SDS-polyacrylamide gel and stained with Coomassie Brilliant Blue
R-250. Lane 1: Total proteins before induction; 2: Total proteins after
induction; 3: Soluble proteins after induction; 4: Purified His6-EasA; 5:
Molecular weight standard; 6: Total proteins before induction; 7: Total
proteins after induction; 8: Soluble proteins after induction; 9: Purified
His6-EasG.
1
comparison of the H-NMR and 13C-NMR data of its free base
form (Table S1, Fig. S7–S11, ESI†) with the data published by
Flieger et al.23,24 and Ninomiya et al.25 Assignments of 1H and 13
C
signals of the authentic agroclavine in its deprotonated form have
been confirmed by DQF-COSY, HSQC, HMBC and ROESY and
are given in Table S1 and the spectra are supplied as Fig. S9–12
(ESI†). These data proved unambiguously that 2 was agroclavine
with a chemical formula of C16H18N2, which was also confirmed
Enzyme activity and identification of the enzyme product
In a previous study, we have demonstrated the conversion of 1
to 3 by the festuclavine synthase FgaFS in the presence of the
old yellow enzyme FgaOx3, both from A. fumigatus.19 Initially,
4330 | Org. Biomol. Chem., 2011, 9, 4328–4335
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 4 HPLC chromatograms of incubation mixtures under different conditions. The reaction mixtures contained 1.5 mM of the cofactors FMN, NADH
or NADPH, 1 mM of GSH, 2-mercaptoethanol or L-ascorbic acid, 1 mM chanoclavine-I aldehyde and 2.5 mg of the recombinant enzymes and were
incubated at 30 ◦C for 2 h. The substances were detected with a Photo Diode Array detector and illustrated for absorption at 282 nm. (*: heat-inactivated
EasG by boiling for 60 min). 1: chanoclavine-I aldehyde; 2: agroclavine.
by detection of the [M+1]+ ion at m/z = 238.8 in the ESI-MS
spectrum.
more information about its involvement in the biosynthesis of
agroclavine.
In summary, EasG from C. purpurea, together with GSH,
catalyses the conversion of 1 to 2 in the presence of NADPH
in vitro and therefore functions as an agroclavine synthase. These
results also raised the question about the role of EasA in vivo in
the biosynthesis of ergot alkaloids in C. purpurea. Inactivation
of easA or its orthologues in a Claviceps strain would provide
NADPH and GSH can be replaced by NADH and other
thiol compounds, respectively. An incubation mixture of 1 and
GSH, with NADH, showed a comparable conversion rate of
54% to that with NADPH (Fig. 4I). Even higher conversion
rates of 99 and 67% were obtained in the incubation mixtures
with 2-mercaptoethanol (Fig. 4J) and dithiothreitol (data not
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4328–4335 | 4331
©

View Article Online
shown), respectively. GSH, 2-mercaptoethanol and dithiothreitol
are reducing agents. GSH and its oxidised form, GSSG, play
an important role as redox systems in living organisms.26–28 On
the other hand, GSH also serves as a thiolation agent, e.g. in
the detoxification of formaldehyde.29 No product formation was
detected in the incubation mixture with L-ascorbic acid instead
of GSH (Fig. 4K), indicating that GSH, 2-mercaptoethanol and
DTT act very likely as thiolation agents in the conversion of 1.
The results obtained from this study promoted us to re-
investigate the enzyme reaction catalysed by FgaFS and FgaOx3
from A. fumigatus in the presence and absence of GSH. When
we incubated FgaFS with NADPH and GSH in the absence of
FgaOx3, formation of product 2 with a conversion rate of 60%
was observed (Fig. 4L). ESI-MS analysis of the isolated product
from this incubation mixture (Fig. 4L) confirmed the presence of
2 and the absence of 3 by detection of an ion at m/z for [M+1] of
2. This indicates that, together with GSH, FgaFS also catalyses
the conversion of 1 to 2 in the absence of FgaOx3. However, the
production of 2 by A. fumigatus has not been reported in the
literature.
To further investigate these contradictory results, we carried
out additional incubations. An incubation of 1 with FgaFS and
FgaOx3, in the presence of NADPH and FMN, resulted in
the formation of 3 (Fig. 5B), as reported previously.19 Addition
of 1 mM GSH to the reaction mixture did not change the
result (Fig. 5C). The formation of a small amount of 2 was
observed, when 25 mM GSH was added to the reaction mixture
(Fig. 5D). The presence of 2 and 3 in the reaction mixture was
confirmed by detection of their [M+1]+ ions at m/z = 238.9 and
240.9, respectively. These results clearly demonstrated that in the
presence of FgaOx3 from A. fumigatus, the reaction of 1 with an
involvement of GSH was not preferred and 3 was the main enzyme
product. This explains why 3 instead of 2 has been identified in
A. fumigatus. In C. purpurea, it is plausible that EasG converts 1
to 2 with the help of cellular GSH, which is usually found in a
concentration range of 0.1–10 mM in living cells.27
Incubation of 1 with EasA from C. purpurea and FgaFS from
A. fumigatus, in the presence of NADH and FMN, resulted in
the formation of neither 3 nor any other products (Fig. 5E).
This indicates that EasA was catalytically inactive or the role of
FgaOx3 in the conversion of 1 to 3 could not be replaced by EasA.
Incubation of 1 with EasA and FgaFS in the presence of additional
GSH resulted in the formation of 2 (Fig. 5F), being identical to
the results with FgaFS, GSH and NADPH (see also Fig. 4L).
Non-enzymatic reaction of GSH with chanoclavine-I aldehyde and
proposed mechanism of the EasG reaction
In the enzyme assays with low yields for 2 (Fig. 4D and 4E),
consumption of 1 was much higher than the product formation,
indicating the involvement of a non-enzymatic reaction between 1
and GSH. To investigate this reaction, we incubated 1 only with
GSH at 30 ◦C. Dependencies of the decreasing of 1 on GSH
concentrations up to 10 mM and on time with GSH at 5 mM
and 1 at 1 mM are given in Fig. S4 and S5 (ESI†). About 22, 60
and 80% of 1 were consumed in the incubation mixtures with 1, 3
and 5 mM GSH for 80 min, respectively. With 5 mM GSH, 40%
of 1 was consumed after 15 min. This demonstrated clearly that
chemical reactions between 1 and GSH had taken place. It could
Fig. 5 HPLC chromatograms of the product formation with different
enzyme and cofactor combinations. The reaction mixtures contained
5 mM of the cofactors FMN and NADH, 1 mM or 25 mM of GSH,
1 mM chanoclavine-I aldehyde and 5 mg of the recombinant enzymes and
were incubated at 30 ◦C for 2 h. The substances were detected with a
Photo Diode Array detector and illustrated for absorption at 282 nm. 1:
chanoclavine-I aldehyde; 2: agroclavine; 3: festuclavine.
be speculated that GSH attacks C9 of 1, as observed for thiols with
other a,b-unsaturated aldehydes,30,31 resulting in the formation of
the enolate of a thiolated saturated aldehyde 4a (Fig. 6). Different
4332 | Org. Biomol. Chem., 2011, 9, 4328–4335
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 6 Proposed mechanism of the conversion of chanoclavine-I aldehyde (1) to agroclavine (2) by non-enzymatic isomerisation with the help of GSH
and reduction of an iminium by EasG.
forms of 4a, e.g. 4b and 4c, can be expected after protonation
and single bond rotation. Elimination of the thiol group from
4b would lead to the formation of isochanoclavine-I aldehyde,
5, which would undergo spontaneous cyclisation to the iminium,
6, by elimination of one molecule of water. Reduction of 6 by
EasG in the presence of NADPH would result in the formation of
agroclavine, 2 (Fig. 6). According to this hypothesis, the reaction
catalysed by EasG from C. purpurea is very similar to that by its
homologue FgaFS from A. fumigatus, i.e. a reduction reaction on
a double bond of an iminium ion.19
Attempts to identify reaction products of 1 with GSH, such
as 4a, 4b or 4c, failed. No noticeable peaks were observed in the
HPLC chromatograms of 1 with GSH, GSH and NADPH or
GSH with NADPH and EasG (Fig. S6, ESI†). As demonstrated
above, 2 was detected in the last case. In the reaction mixtures of
1 with GSH in the presence and absence of NADPH, insoluble
black precipitates were observed. Different solvents, e.g. acetone,
methanol or DMSO, were tested to dissolve the precipitate for MS
and NMR analysis, but without any success.
was unstable at room temperature and the peak at 8.1 min was
detected as the dominant one in the isolated sample (Fig. S7B,
ESI†).
Conclusions
In this study, we demonstrated clearly that the conversion of
chanoclavine-I aldehyde to agroclavine was catalysed by EasG
of C. purpurea in the presence of NADPH, via a non-enzymatic
adduct with reduced glutathione. Its homologue FgaFS of
Aspergillus fumigatus catalyses this conversion under the same
conditions as well. However, in the presence of FgaOx3, FgaFS
converted chanoclavine-I aldehyde to festuclavine, even in the
presence of 1 mM GSH. It can therefore be concluded that
chanoclavine-I aldehyde could be converted to agroclavine in C.
purpurea by EasG after double bond isomerisation by GSH and
to festuclavine in A. fumigatus by the two enzymes FgaFS and
FgaOx3 together.
We therefore decided to investigate the reactions of 1 with 2-
mercaptoethanol and dithiothreitol, under the same conditions as
for GSH (Fig. S6, ESI†). In the chromatograms of the incubation
mixtures of 1 with 2-mercaptoethanol and DTT without EasG,
an additional peak was detected at 6.7 and 8.1 min, respectively
(Fig. S6A and S6B, ESI†). These peaks were also found in the
incubation mixtures containing NADPH and EasG (Fig. S6C,
ESI†). Both peaks were isolated by HPLC and incubated with
EasG in the presence of NADPH. None of them were converted
by EasG (Fig. S7, ESI†). MS analysis provided no indication for
the presence of intermediates mentioned in Fig. 6.
Experimental Section
Computer-assisted sequence analysis
Sequence similarities were obtained by alignments of amino acid
sequences using the BLAST program “BLAST 2 SEQUENCES”
(www.ncbi.nlm.nih.gov). A codon optimisation was done for the
expression of the genes easG and easA in E. coli XL1 Blue MRF¢
cells using the program GENEius (www.genius.de).
Chemicals
Detailed inspection of the HPLC chromatogram of the incuba-
tion mixture of 1 with DTT revealed the presence of an additional
minor peak at 6.5 min (Fig. S6 and S7, ESI†). This peak was
also isolated on HPLC and subjected to MS analysis. ESI-MS
showed an ion at m/z = 409.1, corresponding to the [M+1]+ of
4c. More importantly, incubation of the isolated product with
EasG, in the presence of NADPH, resulted in the formation of a
product with the same retention time as that of 2 (Fig. S7A, ESI†).
Due to the low amount, a structure elucidation by NMR was not
possible in the present study. Furthermore, the isolated substance
The cofactors were obtained from Sigma–Aldrich (Munich,
Germany) and Carl Roth (Karlsruhe, Germany). Chanoclavine-I
and authentic agroclavine were kindly provided by Prof. Leistner
(Bonn, Germany).
Bacterial strains, plasmids and cultural conditions
The pQE60 vector was obtained from Qiagen (Hilden, Germany.
Escherichia coli XL1 Blue MRF¢ (Stratagene, Amsterdam, The
Netherlands) was used for cloning and expression experiments
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4328–4335 | 4333
©

View Article Online
and grown in liquid or on solid Luria–Bertani or Terrific–Broth
200 column (GE Health Care, Freiburg, Germany), with 50 mM
Tris-HCl buffer (pH 7.5) containing 150 mM NaCl as eluent.
The column was calibrated with blue dextran 2000 (2000 kDa),
ferritin (440 kDa), aldolase (158 kDa), conalbumin (75 kDa),
carbonic anhydrase (29 kDa) and ribonuclease A (13.7 kDa)
(GE Healthcare, Freiburg, Germany). The molecular mass of the
recombinant His6-EasG was determined to be 32.6 kDa. This
suggests that the native EasG is a monomer. For the recombinant
His6-EasA, the molecular mass was determined to be 114 kDa.
This suggests that the native EasA is likely a trimer.
◦
◦
medium with 1.5% (w/v) agar at 37 C or 22 C.32 Carbenicillin
(50 mg ml-1) was used for selection of recombinant E. coli strains.
Chemical synthesis, DNA isolation and cloning
Standard procedures for DNA isolation and manipulation were
performed as described.32 The genes easG and easA were syn-
thesized chemically and cloned into pCR2.1 and pBluescript II
SK (+) (MWG Biotech AG, Ebersberg, Germany), resulting in
plasmid pMM043 and pMM045, respectively. For cloning in the
expression vector pQE60, a NcoI restriction site was added to the
start codon. The original stop codon was removed by insertion of
a BamHI restriction site.
Enzyme assays with His6-EasG and His6-EasA
All of the enzyme assays contained 50 mM Tris-HCl, pH 7.5 and
1.3–3.0% (v/v) of glycerol. The reaction mixtures were incubated
at 30 ◦C for 2 h and were then extracted twice with 2 volumes
of ethyl acetate after adjusting to pH 9.0 with 1 M NaOH. The
combined organic phase was evaporated to dryness and dissolved
in 100 ml methanol. The enzyme products were analysed on a
HPLC system described below.
To create the expression vector pMM044 for easG and pMM046
for easA, the plasmids pMM043 and pMM045 were digested with
NcoI and BamHI. The resulted NcoI–BamHI fragments of 872
bp and 1109 bp, respectively, were isolated from the agarose gel
and ligated into pQE60, which had also been digested with NcoI
and BamHI, previously.
The reaction mixture for isolation of 1 (120 ml) contained
chanoclavine-I (1 mM), 12 mg (0.86 mM) of purified FgaDH and
5 mM NAD. The reaction mixture was incubated for 4 h at 30 ◦C.
The reaction mixture for isolation of the enzyme product, 2, for
structural elucidation (30 ml) contained chanoclavine-I (1 mM),
3 mg (0.86 mM) of purified FgaDH, 3 mg (3 mM) of purified EasG,
5 mM NAD, 5mM NADPH and 10 mM GSH. After incubation
for 15 h, the pH value of the reaction mixture was adjusted to
9 with aqueous ammonium hydroxide and extracted twice with
ethyl acetate. The combined organic phase was evaporated on a
rotation evaporator at 30 ◦C to dryness. The residue was dissolved
in 350 ml methanol and centrifuged at 14 000 ¥ g for 30 min at 4 ◦C
before injection onto HPLC for isolation of 2.
Overproduction and purification of His6-EasG and His6-EasA
For expression of the genes easG and easA, E. coli XL1 Blue
MRF¢ cells harbouring the plasmids pMM044 and pMM046 were
cultivated in 2000 ml Erlenmeyer flasks containing liquid Terrific–
Broth medium (1000 ml) supplemented with carbenicillin (50 mg
ml-1) and grown at 37 ◦C to an absorption at 600 nm of 0.6. For
induction, isopropyl thiogalactoside (IPTG) was added to a final
concentration of 0.1 mM and the cells containing the plasmid
pMM044 were cultivated for 15 h at 26 ◦C before harvest. The
cells containing the plasmid pMM046 were cultivated for 15 h at
22 ◦C before harvest.
The bacterial cultures were centrifuged and the pellets were
resuspended in lysis buffer (10 mM imidazole, 50 mM NaH2PO4,
300 mM NaCl, pH 8.0) at 1.4 ml per gram wet weight. After
addition of 1 mg ml-1 lysozyme, 10 mg ml-1 RNaseI, 5 mg ml-1
DNaseI and incubation on ice for 30 min, the cells were sonicated
6 times for 10 s each at 200 W. To separate the cellular debris
from the soluble proteins, the lysate was centrifuged at 14 000 ¥
g for 30 min at 4 ◦C. One-step purification of the recombinant
His6-tagged fusion proteins by affinity chromatography with Ni-
NTA agarose resin (Qiagen, Hilden, Germany) was carried out
according to the manufacturer’s instructions. The proteins were
eluted with 250 mM imidazole in 50 mM NaH2PO4, 300 mM
NaCl, pH 8.0. In order to change the buffer, the protein fractions
were passed through a PD-10 column (GE Healthcare, Freiburg,
Germany), which had been equilibrated with 50 mM Tris-HCl,
15% (v/v) of glycerol, pH 7.5, previously. The proteins were eluted
with the same buffer and stored frozen at -80 ◦C for enzyme assays.
HPLC analysis and isolation of agroclavine
Reaction mixtures were analysed on an Agilent HPLC Series 1200
by using a Multospher 120 RP18 column (4 ¥ 250 mm, 5 mm,
Agilent) at a flow rate of 1 ml min-1. Water (solvent A) and
acetonitrile (solvent B), each containing 0.5% (v/v) trifluoroacetic
acid, were used as solvents. The substances were detected with
a Photo Diode Array detector. The assays were analysed with a
gradient from 38 to 54% B over 10 min. After washing with 100%
solvent B for 5 min, the column was equilibrated with 38% solvent
B for 5 min. For better separation of 2 and 3 (Fig. 5), the substances
were eluted with 21% B for 38 min. After washing with 100%
solvent B for 5 min, the column was equilibrated with 21% solvent
B for 5 min. For isolation of the enzyme product 2, a gradient
from 30 to 37% B over 28 min was used. After washing with 100%
solvent B for 5 min, the column was equilibrated with 30% solvent
B for 5 min. The collected fractions containing the enzyme product
after HPLC separation were evaporated to dryness and subjected
to NMR and MS analyses.
Protein analysis
Proteins were analysed by SDS-PAGE according to the method of
Laemmli33 and stained with Coomassie Brilliant Blue R-250.
Non-enzymatic products of 1 with thiols were analysed, after
incubation at 30 ◦C for 5 or 16 h, on the same equipment and
with the same solvents aforementioned. After running with 10%
B for 14 min, a gradient from 10 to 74% B over 17 min was used.
The column was then washed with 100% solvent B for 5 min, and
equilibrated with 10% solvent B for 5 min.
Determination of the molecular mass of active His6-EasG and
His6-EasA
The molecular mass of the recombinant His6-EasG and His6-EasA
was determined by gel filtration on a HiLoad 16/60 Superdex
4334 | Org. Biomol. Chem., 2011, 9, 4328–4335
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
NMR experiments
References
1 C. Wallwey and S.-M. Li, Nat. Prod. Rep., 2011, 28, 496–510.
2 C. L. Schardl, D. G. Panaccione and P. Tudzynski, Alkaloids: Chem.
Biol., 2006, 63, 45–86.
3 D. Gro¨ger and H. G. Floss, Alkaloids: Chem. Biol., 1998, 50, 171–218.
4 C. E. Robertson, D. F. Black and J. W. Swanson, Semin. Neurol., 2010,
30, 201–211.
5 B. V. Lovell and M. J. Marmura, Curr. Opin. Neurol., 2010, 23, 254–258.
6 P. L. Schiff, Am. J. Pharma. Edu., 2006, 70, 1–10.
7 I. Setnikar et al., Arzneimittelforschung., 2001, 51, 2–6.
8 M. Flieger, M. Wurst and R. Shelby, Folia Microbiol., 1997, 42, 3–30.
9 T. Correia et al., Chem. Biol., 2003, 10, 1281–1292.
10 T. Haarmann et al., Phytochemistry, 2005, 66, 1312–1320.
11 I. A. Unso¨ld and S.-M. Li, Microbiology, 2005, 151, 1499–1505.
12 O. Rigbers and S.-M. Li, J. Biol. Chem., 2008, 283, 26859–26868.
13 N. Lorenz et al., Appl. Environ. Microbiol., 2010, 76, 1822–1830.
14 C. Wallwey, M. Matuschek and S.-M. Li, Arch. Microbiol., 2010, 192,
127–134.
15 C. M. Coyle et al., Appl. Environ. Microbiol., 2010, 76, 3898–3903.
16 S. Pazˇoutova´ et al., J. Nat. Prod., 1981, 44, 225–235.
17 B. Schumann et al., Planta Med., 1982, 45, 11–14.
Spectra were recorded at room temperature on a Bruker Avance
600 MHz spectrometer equipped with an inverse probe with z-
gradient. The DQF-COSY, HSQC, and HMBC spectra were
recorded with standard methods.34 ROESY experiment was per-
formed in phase-sensitive mode.35
For all two-dimensional spectra, 16 transients were used. For
ROESY spectra, a mixing time of 300 ms and a relaxation delay
of 3 s were used. 1H spectra were acquired with 65 536 data
points, while 2D spectra were collected using 4096 points in the
F2 dimension and 512 increments in the F1 dimension. Chemical
shifts were referenced to CD3OD. All spectra were processed with
Bruker TOPSPIN 2.1.
ESI-MS of the enzyme product
The positive electrospray ionization (ESI) mass spectrometry was
carried out with an AutoSpec instrument (Micromass Co. UK
Ltd).
18 V. Kren et al., FEMS Microbiol. Lett., 1986, 37, 31–34.
19 C. Wallwey et al., Org. Biomol. Chem., 2010, 8, 3500–3508.
20 T. Haarmann et al., ChemBioChem, 2006, 7, 645–652.
21 J. Z. Cheng et al., J. Am. Chem. Soc., 2010, 132, 12835–12837.
22 J. Z. Cheng et al., J. Am. Chem. Soc., 2010, 132, 1776–1777.
23 M. Flieger et al., J. Nat. Prod., 1993, 56, 810–814.
Nucleotide sequence accession numbers
The coding sequences of easG and easA from C. purpurea are
available at GenBank under the accession numbers AY836771 and
AJ703809.
24 M. Flieger et al., J. Nat. Prod., 1991, 54, 390–395.
25 I. Ninomiya et al., Chem. Pharm. Bull., 1991, 39, 23–30.
26 M. B. Toledano et al., FEBS Lett., 2007, 581, 3598–3607.
27 A. Meister, J Biol. Chem, 1988, 263, 17205–17208.
28 I. Sato et al., J. Biol. Chem., 2009, 284, 8042–8053.
29 R. J. Hopkinson et al., Org. Biomol. Chem., 2010, 8, 4915–4920.
30 I. V. Koval, Russ. J. Org. Chem., 2007, 43, 319–346.
31 H. Esterbauer, A. Ertl and N. Scholz, Tetrahedron, 1976, 32, 285–289.
32 J. Sambrook and D. W. Russell, Molecular Cloning: a Laboratory
Manual, Cold Spring Harbor Laboratory Press, New York, 2001.
33 U. K. Laemmli, Nature, 1970, 227, 680–685.
Acknowledgements
This work was supported by a grant from the Deutsche
Forschungsgemeinschaft (Li844/3-1 to S.-M. Li). We thank Prof.
E. Leistner for providing chanoclavine-I and agroclavine. Xie
acknowledges the Deutsche Forschungsgemeinschaft for funding
the Bruker AVANCE 600 spectrometer.
34 S. Berger and S. Braun, 200 and More NMR Experiments. A Practical
Course, Weiley-VCH, Weinheim, Germany, 2004.
35 D. Marion et al., J. Magn. Reson., 1989, 85, 393–399.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4328–4335 | 4335
©