Substrate promiscuity of the cyclic dipeptide prenyltransferases from Aspergillus fumigatus

Article abstract of DOI:10.1021/np800501m

This study reports that a series of tryptophan derivatives with modifications on the side chain or at the indole ring were accepted by two cyclic dipeptide prenyltransferases, CdpNPT and FtmPT1, and converted to prenylated derivatives. The structures of t

Full text of DOI:10.1021/np800501m

44
J. Nat. Prod. 2009, 72, 44–52
Substrate Promiscuity of the Cyclic Dipeptide Prenyltransferases from Aspergillus fumigatus§
Huixi Zou,†,‡ Xiaodong Zheng,‡ and Shu-Ming Li*,†
Institut fu¨r Pharmazeutische Biologie, Philipps-UniVersita¨t Marburg, Deutschhausstrasse 17A, D-35037 Marburg, Germany, and Department of
Food Science and Nutrition, Zhejiang UniVersity, Hangzhou 310029, Zhejiang, People’s Republic of China
ReceiVed August 13, 2008
This study reports that a series of tryptophan derivatives with modifications on the side chain or at the indole ring were
accepted by two cyclic dipeptide prenyltransferases, CdpNPT and FtmPT1, and converted to prenylated derivatives.
The structures of the enzymatic products were elucidated by NMR and MS analyses. In comparison to cyclic dipeptides,
which were reversely prenylated by CdpNPT at N-1 and in a regular manner by FtmPT1 at C-2, respectively, tryptophan
and its simple derivatives were prenylated reversely by both enzymes at N-1. These results demonstrated the substrate
promiscuity of both enzymes.
Recently, seven putative indole prenyltransferase genes were
identified in the genome sequence of Aspergillus fumigatus.1 Six
of these prenyltransferases have been characterized biochemically
and shown to catalyze transfer reactions of a prenyl moiety onto
different positions of the indole rings in the biosynthesis of diverse
secondary metabolites.2-7 For example, the two dimethylallyltryp-
tophan synthases FgaPT2 and 7-DMATS catalyze the prenylation
of L-tryptophan at C-4 and C-7 of the indole ring, respectively.3,6
CdpNPT and FtmPT1 were shown to prenylate tryptophan-
containing cyclic dipeptides with a diketopiperazine ring at N-1
and C-2, respectively (Scheme 1).2,4 All of these enzymes accepted
only dimethylallyl diphosphate (DMAPP) as the prenyl donor. On
the other hand, these enzymes showed broad substrate specificities
toward its aromatic substrates. In previous studies, it was shown
that FgaPT2 and 7-DMATS also accepted a series of simple indole
derivatives, affording conversion to C-4- and C-7-prenylated
derivatives,8,9 respectively. Correspondingly, CdpNPT and FtmPT1
accepted all of the tested tryptophan-containing cyclic dipeptides
and prenylated them at N-1 and C-2, respectively.2,4,10 As expected,
tryptophan was not a substrate for the cyclic dipeptide prenyltrans-
ferases CdpNPT and FtmPT1, and cyclic dipeptides were not
accepted by the dimethylallyltryptophan synthase FgaPT2.4,8,10
Interestingly, 7-DMATS accepted both tryptophan-containing linear
and cyclic dipeptides as substrates.6 The following questions had
arisen from these results: Is the broad substrate specificity of
7-DMATS just an exception among the indole prenyltransferases?
Were the assay conditions for CdpNPT and FtmPT1 with tryptophan
not sensitive enough to see any product formation? Very recently,
aminopeptidase activity of CdpNPT, FtmPT1, 7-DMATS, and
FgaPT1 was detected, indicating their common evolutionary
relationship.11 Therefore, these enzymes might share many features
including substrate promiscuity. These results prompted the rein-
vestigation of the acceptance of tryptophan and its simple deriva-
tives by CdpNPT and FtmPT1.
CdpNPT in 100 µL at 37 °C for 120 min, no enzymatic product
could be detected in the incubation mixture (Figure 1A). Under
the same conditions, the tryptophan-containing cyclic dipeptide
cyclo-D-Trp-L-Tyr was accepted, confirming previous results (Figure
1B). As expected, the conversion rate of cyclo-D-Trp-L-Tyr
increased with increasing amounts of CdpNPT (Figures 1D, 1F,
and 1H). By increasing protein amount, product formation was also
observed with L-tryptophan. As shown in Figure 1C, one product
peak at 13.9 min was detected in the reaction mixture containing
0.28 µM CdpNPT and even two product peaks at 13.9 and 16.5
min with 0.54 or 1.62 µM CdpNPT (Figures 1E and 1G). Both
peaks were absent in the incubation mixture with 1.62 µM heat-
inactivated CdpNPT (Figure 1I), indicating that these peaks were
very likely enzymatic products. With CdpNPT at a concentration
of 1.62 µM in a 100 µL assay, a conversion rate of 18.1% could
be achieved for L-tryptophan after incubation at 37 °C for 2 h
(Figure 1G). This result proved clearly that L-tryptophan was also
accepted by CdpNPT, indicating the broad substrate specificity of
this enzyme.
Similar results were also observed with the second cyclic
dipeptide prenyltransferase FtmPT1. Incubation of L-tryptophan and
DMAPP with FtmPT1 under similar conditions to those described
in a previous study,4 e.g., incubation with 0.11 µM monomeric
FtmPT1 for 2 h, showed no enzymatic conversion (Figure 2A),
while product formation at 13.5 min was clearly observed with
cyclo-L-Trp-L-Pro (brevianamide F) as substrate, the natural
substrate of FtmPT14 (Figure 2B). The product of incubation with
brevianamide F was confirmed by preparative isolation and NMR
analysis as tryprostatin B, which carries a regular prenyl moiety at
C-2 of the indole ring (Scheme 1 and Figure 2J).4 Brevianamide F
was almost completely converted by 0.28 µM or more FtmPT1 in
a 100 µL assay (Figures 2D, 2F, and 2H). Under these conditions,
a product peak at 13.9 min could also be detected with L-tryptophan
as substrate (Figures 2C, 2E, and 2G). Interestingly, the enzymatic
product of L-tryptophan with FtmPT1 had the same retention time
as the first enzymatic product of the incubation with CdpNPT, i.e.,
13.9 min (Figures 1G and 2G).
Results and Discussion
Tryptophan Acceptance Was Detected at Higher Enzyme
Concentration by Both Cyclic Dipeptide N-Prenyltransferase
(CdpNPT) and Brevianamide F Prenyltransferase (FtmPT1).
Under conditions described previously,2 e.g., incubation of 1 mM
tryptophan, 2 mM DMAPP, and 0.11 µM recombinant homodimeric
With 1.62 µM CdpNPT or 1.12 µM FtmPT1 in a 100 µL enzyme
assay, no conversion of tyrosine was detected in the presence of
DMAPP (data not shown).
Simple Tryptophan Derivatives Were Also Substrates of
CdpNPT and FtmPT1. After confirmation of the enzymatic
products of L-tryptophan with CdpNPT and FtmPT1 as prenylated
tryptophan by detection of the [M - 1]- ion at m/z 271.2 and [M
+ 1]+ ion at m/z 273.1 with LC-MS (Supporting Information, Table
S1), incubations of CdpNPT and FtmPT1 with simple indole
derivatives were carried out. These substances were accepted by
dimethylallyltryptophan synthases FgaPT2 and/or 7-DMATS as
§ Dedicated to Prof. Dr. Richard D. Hutchison on the occasion of his
65th birthday.
* To whom correspondence should be addressed. Tel: 0049-6421-2822461.
Fax: 0049-6421-2826678. E-mail: shuming.li@staff.uni-marburg.de.
† Philipps-Universita¨t Marburg.
‡ Zhejiang University.
10.1021/np800501m CCC: $40.75
2009 American Chemical Society and American Society of Pharmacognosy
Published on Web 12/29/2008

Cyclic Dipeptide Prenyltransferases from Aspergillus
Journal of Natural Products, 2009, Vol. 72, No. 1 45
Figure 1. HPLC chromatograms of incubation mixtures of L-Trp and cyclo-D-Trp-L-Tyr with different amounts of recombinant CdpNPT.

46 Journal of Natural Products, 2009, Vol. 72, No. 1
Zou et al.
Scheme 1. Prenyltransfer Reactions Catalyzed by Different Prenyltransferases from Aspergillus fumigatus: Conversion of L-Trp to
4-Dimethylallyltryptophan (4-DMAT) by FgaPT2 (A) and to 7-Dimethylallyltryptophan (7-DMAT) by 7-DMATS (B); Conversion of
Brevianamide F by FtmPT1 to Tryprostatin B (C) and to N-1-3′-DMA-cyclo-L-Trp-L-Pro by CdpNPT (D)
Table 1. Relative Substrate Specificity of CdpNPT and FtmPT1 toward Indole Derivativesa
CdpNPT (1.62 µM)
FtmPT1 (1.68 µM)
2 h
conversion
16 h
2 h
conversion
16 h
rel yield
[%]
conversion
rel yield
[%]
rel yield
[%]
conversion
rel yield
[%]
substrate
rate [%]
rate [%]
rate [%]
rate [%]
L-tryptophan
indole-3-acetic acid
L-ꢀ-homotryptophan
tryptamin
12.05
e0.21
e0.41
e0.40
e0.10
15.14
3.01
e0.13
8.75
e0.18
e0.11
6.1
100
38.24
17.33
e0.41
e0.40
9.94
43.4
9.38
e0.13
23.48
7.71
100
7.51
e0.21
e0.41
e0.40
e0.10
e0.28
e0.17
e0.13
1.20
e0.18
11.08
5.74
e0.10
e0.14
e0.09
4.14
100
37.84
e0.21
e0.41
e0.40
0.29
11.12
0.38
e0.13
7.33
100
e1.7
e3.4
e3.3
e0.8
125.6
25.0
e1.1
72.6
e1.5
e0.9
50.6
45.3
e1.1
e1.0
26.0
113.5
24.5
e0.3
61.4
20.2
13.5
37.9
e0.3
e0.4
1.9
e2.8
e5.5
e5.3
e1.3
e3.7
e2.3
e1.7
16.0
e2.4
147.5
76.4
e1.3
e1.9
e1.2
55.1
e0.6
e1.1
e1.1
0.8
29.4
1.0
e0.3
19.4
10.3
96.0
64.1
e0.3
e0.4
4.3
N-acetyl-DL-tryptophan
L-abrine (NR-methyl-L-tryptophan)
R-methyl-DL-tryptophan
1-methyl-DL-tryptophan
4-methyl-DL-tryptophan
5-methyl-DL-tryptophan
6-methyl-DL-tryptophan
7-methyl-DL-tryptophan
5-hydroxy-L-tryptophan
5-bromo-DL-tryptophan
5-fluoro-L-tryptophan
6-fluoro-DL-tryptophan
3.89
5.16
36.33
24.26
e0.10
e0.14
1.64
14.48
e0.10
e0.14
0.72
e0.10
e0.14
e0.09
2.13
e0.8
e1.2
e0.7
17.7
5.29
13.8
21
55.5
a The reaction mixtures with 1.62 µM CdpNPT or 1.68 µM FtmPT1 were incubated for 2 and 16 h. The relative yield or relative activity was defined
as the ratio of the conversion rate of a substrate to that of L-tryptophan.
Scheme 2. Proposed Relationship between N-1- and C-2-Prenylated Indole Derivatives
substrates.8,9 In total, 16 substances (Table 1) were tested with both
enzymes in the presence of DMAPP. Product formation could be
detected for six and five of these substrates after incubation with
CdpNPT and FtmPT1 for 2 h at 37 °C, respectively (Table 1). By
expanding the incubation time to 16 h, 11 and 10 substances were
accepted by CdpNPT and FtmPT1, respectively (Table 1). In
summary, CdpNPT and FtmPT1 showed similar substrate specifici-
ties toward simple tryptophan derivatives, although the relative
activities differed from each other. With the exception of 1-methyl-,
5-hydroxy-, and 5-bromo-DL-tryptophan, which were not accepted
by any of the enzymes, tryptophan derivatives with modifications
at the indole ring were in general better accepted than those with
modifications at the side chain of tryptophan. Indole-3-acetic acid
and N-acetyl-DL-tryptophan were not or almost not accepted by
FtmPT1, but well accepted by CdpNPT, with a relative activity of
45% and 26% in comparison to that of L-tryptophan. L-Tryptophan
and L-abrine were found to be the best simple tryptophan derivatives
for CdpNPT as substrates, while L-tryptophan and 6-methyl-DL-
tryptophan were the best simple tryptophan derivatives for the
FtmPT1 reaction (Table 1). Interestingly, L-ꢀ-homotryptophan,
which was well accepted by FgaPT2 and 7-DMATS,8,9 was not a
substrate for CdpNPT and FtmPT1. As examples shown in Figure
3, two enzymatic products were detected with all of the accepted
substrates in the reaction mixtures with CdpNPT and one with

Cyclic Dipeptide Prenyltransferases from Aspergillus
Journal of Natural Products, 2009, Vol. 72, No. 1 47
Figure 2. HPLC chromatograms of incubation mixtures of L-Trp and brevianamide F with different amounts of recombinant FtmPT1.
FtmPT1, respectively. The products in the reaction mixtures of
FtmPT1 had the same retention time as those of the first product
in the reaction mixtures of the respective substrate with CdpNPT
(Figure 3).
N-1-Prenylated Derivatives Are Products of CdpNPT and
FtmPT1 Reactions, When Simple Tryptophan Derivatives
Were Used As Substrates. For structural elucidation, the
enzymatic products of L-tryptophan, L-abrine, and 4- and 7-methyl-
DL-tryptophan were isolated from the incubation mixtures of
CdpNPT on a preparative scale and subjected to NMR and MS
analyses. As expected, CdpNPT catalyzes the transfer of a reverse
prenyl moiety [3′-(3′,3′-dimethylallyl), 3′-DMA] onto N-1 of the
tryptophan-containing cyclic dipeptides.2 The enzymatic product
peaks at 13.9, 14.9, 16.0, and 16.2 min (Figure 3) indeed carry
reverse prenyl moieties at N-1 of the indole ring (Figure 4). This
was unequivocally proven by detection of the prenyl signals of N-1-

48 Journal of Natural Products, 2009, Vol. 72, No. 1
Zou et al.
Figure 3. HPLC chromatograms of incubation mixtures of selected tryptophan derivatives with recombinant CdpNPT or FtmPT1. The
reaction mixtures contained 1.62 µM CdpNPT or 1.68 µM FtmPT1 and were incubated at 37 °C for 16 h. S: substrate. P, P1, and P2:
products.
3′-DMAT, 4-methyl-N-1-3′-DMAT, and 7-methyl-N-1-3′-DMAT
in D2O at δ 5.99-6.00 (H-2′), 5.10-5.17 (H-1′), 5.09-5.12 (H-
1′), 1.08-1.10 (3H-4′), and 0.99-1.00 (3H-5′) ppm or of N-1-3′-
DMA-L-abrine in DCl at 6.19 (H-2′), 5.34 (H-1′), 5.27 (H-1′), 1.24
(3H-4′), and 1.18 (3H-5′) ppm, respectively (Table 2). The upfield
shifting of the signal for H-2 to 5.2-5.5 ppm (Table 2) confirmed
the prenylation at N-1.2,12 The structures of these products were
also confirmed by MS analysis (Table S1). Due to low solubility
in D2O and low stability in DCl, no interpretable NMR spectra
could be obtained for the second peak of the reaction mixtures.
Therefore, the structures of these compounds could not be
elucidated. However, from NMR spectra of the products isolated
from the incubation mixtures of 4- and 7-methyl-DL-tryptophan, it
seems that these peaks contain prenylated derivatives (data not
shown).
By analogy to CdpNPT, the enzymatic products of L-tryptophan,
L-abrine, and 4- and 7-methyl-DL-tryptophan with retention times
at 13.9, 14.9, 16.0, and 16.2 min, respectively, were also isolated
from the reaction mixtures of FtmPT1. In addition, the enzymatic
product of 6-methyl-DL-tryptophan was also isolated from the
incubation mixture of FtmPT1. The isolated products were subjected
to NMR and MS analyses. Inspection of the 1H NMR spectra
surprisingly revealed that the enzymatic products isolated from
incubation mixtures of FtmPT1 are identical to those identified from

Cyclic Dipeptide Prenyltransferases from Aspergillus
Journal of Natural Products, 2009, Vol. 72, No. 1 49
the first reaction on the second one, the behavior of N-1-3′-
DMAT and simple derivatives in the absence of the substrate
was investigated. For this purpose, 4-methyl-DL-tryptophan was
incubated with CdpNPT in the presence of DMAPP (Figure 5A
) Figure 3E). The two product peaks, i.e., 4-methyl-N-1-3′-
DMAT and the unidentified products of the second peak with
retention time of 16.5 min, were then isolated on HPLC and
incubated again with FtmPT1 or CdpNPT in the presence or
absence of DMAPP at pH 7.5 for 16 h (Figure 5). As shown in
Figures 5B-D, no additional peaks were detected after the
incubation with FtmPT1, demonstrating that 4-methyl-N-1-3′-
DMAT was chemically stable and not converted by FtmPT1
under the tested condition. This is also true for the second
product peak (Figures 5G-I). This proved that the final product
of FtmPT1 reaction of 4-methyl-DL-tryptophan was indeed
4-methyl-N-1-3′-DMAT, i.e., reverse prenylation at N-1 (Figure
2I) rather than regular prenylation at C-2. Similar results are
expected for tryptophan and other simple derivatives.
CdpNPT Catalyzes Prenylations at Different Positions of
Tryptophan and Its Simple Derivatives. Enzymatic products
of cyclic dipeptides with CdpNPT were found to be unstable
under acidic conditions and underwent rearrangement to different
products, e.g., to regular N-1-prenylated derivatives.2 Even
during the enzymatic incubation at pH 7.5 and 37 °C, two or
more products could be detected when cyclo-L-Trp-L-Pro and
cyclo-D-Trp-L-Pro were used as substrates.2 This could also be
true for the CdpNPT products of tryptophan and its derivatives.
It could be speculated that rearrangement of N-1-3′-DMAT
derivatives had already taken place during the enzymatic
incubation to the unknown products of the second peak (Figure
3). In comparison to those of CdpNPT, the incubation mixtures
of L-tryptophan, L-abrine, and 4- and 7-methyl-DL-tryptophan
as well as 6-fluoro-DL-tryptophan with FtmPT1 showed only one
product peak each. Therefore, the second peak in the incubation
mixtures of CdpNPT was an enzymatic product rather than a
result of a chemical rearrangement of N-1-3′-DMAT derivatives,
which would also take place during incubation with FtmPT1,
because the assay conditions, e.g., pH value, incubation tem-
perature, and time, and substrate and buffer components were
identical for both enzyme reactions. As shown in Figure 1, both
Figure 4. Enzymatic products of tryptophan derivatives after
incubation with CdpNPT or FtmPT1.
CdpNPT assays, i.e., reversely N-1-prenylated tryptophan and
derivatives (Figures 1J and 2I). This is in contrast to the regular
prenylation at C-2 (Scheme 1 and Figure 2J), when tryptophan-
containing cyclic dipeptides were used as substrates for FtmPT1
reactions.4
Two questions arise from the results obtained from incubation
mixtures of tryptophan and its simple derivatives with CdpNPT
and FtmPT1. Are the FtmPT1 products identified in this study
probable precursors of C-2-prenylated derivatives? Furthermore,
what is the relationship of the two product peaks in the incubation
mixtures with CdpNPT?
4-Methyl-N-1-3′-DMAT Was Stable in the FtmPT1 Assay.
It could be speculated that FtmPT1 catalyzes the prenylation of
L-tryptophan and its simple derivatives as well as of tryptophan-
containing cyclic dipeptides in the same manner, i.e., prenylation
at N-1 followed by immigration of the prenyl moiety from N-1
to C-2 (Scheme 2), as reported for 3-alkyl-1-allylindoles.13 In
the case of cyclic dipeptides, the products of the first reaction
would be immediately converted, so that only C-2-prenylated
derivatives could be detected. In the case of tryptophan or its
simple derivatives, the second reaction would be much slower
than the first one, so that N-1-3′-DMAT or its derivatives were
detected as described above. Incubation mixtures of L-tryptophan
and its simple derivatives with FtmPT1 for 16 h showed only
one product peak each (Figure 3). To exclude the influence of
Table 2. 1H NMR Data of Isolated Enzymatic Productsa
N-1-3′-DMAT
4-methyl-N-1-3′-DMAT
6-methyl-N-1-3′-DMAT
7-methyl-N-1-3′-DMAT
N-1-(3′-DMA)-L-abrine
proton
in D2O
in D2O
in D2O
in D2O
in DCl (0.1 M)
H-2
H-4
H-5
H-6
H-7
H-10
5.46, s
7.39, d, 7.6
5.24, s
5.24, s
5.46, s
7.25, d, 7.6
6.91, t, 7.6
7.25, d, 7.1
CH3: 2.17, s
2.58, dd, 13.4, 6.5
2.46, t, 11.5
3.63, dd, 11.5, 6.5
5.66, s
CH3: 2.44, s
6.76, d, 7.3
7.15, t, 7.7
6.62, d, 7.6
2.91. dd, 13.4, 6.3
2.52, dd, 13.4, 8.2
3.68, t, 8.2
7.24, d, 7.9
6.75, d, 7.6
CH3: 2.28, s
6.62, s
2.42, dd, 13.0, 6.0
2.28, m
7.48, d, 7.6
7.03, t, 7.6
7.39, t, 7.6
6.97, d, 7.9
2.84, m
6.95, td, 7.6, 1.0
7.24, td, 7.7, 1.2
6.80, d, 7.6
2.60, dd, 13.1, 6.6
2.46, dd, 13.1, 11.4
3.68, dd, 11.7, 6.3
2.67, t, 10.7
3.77, m
H-11
H-12
3.68, m
N-CH3: 2.73, s
5.34, d, 10.4
5.27, d, 16.7
6.19, dd, 16.7, 10.4
CH3: 1.24, s
CH3: 1.18, s
H-1′
5.17, dd, 10.7, 1.3
5.12, dd, 17.5, 1.3
6.00, dd, 17.5, 10.7
CH3: 1.10, s
5.10, d, 11.2
5.09, d, 17.2
5.99, dd, 17.2, 11.2
CH3: 1.08, s
CH3: 1.00, s
5.13, d, 10.7
5.08, d, 17.6
6.01, dd, 17.6, 10.7
CH3: 1.07, s
CH3: 0.99, s
5.16, dd, 10.8, 1.3
5.12, dd, 17.3, 1.3
5.99, dd, 17.3, 10.8
CH3: 1.10, s
H-2′
H-4′
H-5′
CH3: 0.99, s
CH3: 0.99, s
a The chemical shifts (δ) are given in ppm and coupling constants in Hz.
Table 3. Km Values and Turnover Numbers of CdpNPT and FtmPT1
CdpNPT
FtmPT1
kcat/Km
ratio
kcat/Km
kcat/Km
ratio
kcat/Km
substrate
Km [mM]
kcat [s-1
]
[mM-1 s-1
0.006
]
Km [mM]
kcat [s-1
]
[mM-1 s-1
]
L-tryptophan
0.83
0.005
1
0.82
0.055
0.12
5.57
0.15
101.3
1
675
cyclo-L-Trp-L-Pro
cyclo-D-Trp-L-Tyr
0.128
0.46
3.59
598

50 Journal of Natural Products, 2009, Vol. 72, No. 1
Zou et al.
Figure 5. HPLC chromatograms of incubation mixtures of isolated products of 4-methyl-DL-tryptophan with CdpNPT or FtmPT1. The
reaction mixtures contained 1.62 µM CdpNPT or 1.68 µM FtmPT1 and were incubated at 37 °C for 16 h.
product peaks in the incubation mixtures of tryptophan were
detected only in the presence of active, but not of heat-
inactivated, CdpNPT.
By comparison of the results obtained from CdpNPT and
FtmPT1, it could be speculated that the unidentified compounds of
the second peak were enzymatic products of CdpNPT with
L-tryptophan or derivatives. Three possibilities could be envisaged:
(a) Both products were formed directly from L-tryptophan or its
derivatives; (b) N-1-3′-DMAT derivatives were formed from
L-tryptophan or its derivatives and converted to the products of the
second peaks; (c) N-1-3′-DMAT derivatives were formed from the
products of the second peak.

Cyclic Dipeptide Prenyltransferases from Aspergillus
Journal of Natural Products, 2009, Vol. 72, No. 1 51
Figure 6. Determination of kinetic parameters of L-tryptophan for CdpNPT (A) and FtmPT1 (B).
To prove the possibility (b), 4-methyl-N-1-3′-DMAT was
incubated with CdpNPT in the absence or presence of DMAPP for
16 h. No additional peak was detected after incubation (Figures
5B, 5E, and 5F), proving that possibility (b) could be excluded.
Similar results were obtained after incubation of the isolated
compounds of the second peak with CdpNPT in the absence or
presence of DMAPP [possibility (c)] (Figures 5G, 5J, and 5K). This
indicated that the two product peaks of CdpNPT reactions are very
probably independently formed from tryptophan and its simple
derivatives [possibility (a)]. However, it cannot be excluded that
the conversion between the two compounds occurs only during the
enzymatic reactions, but not after releasing of the products from
the enzyme.
Kinetic Parameters of Selected Compounds. For comparison
of the behavior of CdpNPT and FtmPT1 toward tryptophan and
cyclic dipeptides, kinetic parameters of CdpNPT were determined
for L-tryptophan and cyclo-D-Trp-L-Tyr. Kinetic parameters of
FtmPT1 were also obtained for L-tryptophan and compared with
those of cyclo-L-Trp-L-Pro (brevianamide F), the natural substrate
of FtmPT1.4 The reactions of CdpNPT and FtmPT1 apparently
followed Michaelis-Menten kinetics by using these substrates in
the presence of DMAPP. Michaelis constant Km, turnover number
kcat, and enzymatic rate constant kcat and Km-1 were determined by
Lineweaver-Burk and Hanes-Woolf analyses14 and are given in
Table 3. Km values of CdpNPT and FtmPT1 were determined for
L-tryptophan at 0.83 and 0.82, respectively, and are much higher
than those for cyclic dipeptides (Table 3). The ratio of enzymatic
rate constants kcat and Km-1 of CdpNPT for L-tryptophan and cyclo-
D-Trp-L-Tyr was found to be 1:598. In the case of FtmPT1, it was
found to be 1:675 for L-tryptophan and cyclo-L-Trp-L-Pro (Table
3) (Figure 6). These data proved the high efficiency of the
prenyltransferases CdpNPT and FtmPT1 toward cyclic dipeptides.
Regarding the low enzymatic rate constants of CdpNPT and
FtmPT1 toward tryptophan, no plausible biological relevance could
be predicted for their substrate promiscuity in ViVo.
In this study, evidence that the two cyclic dipeptide prenyltrans-
ferases CdpNPT and FtmPT1 from Aspergillus fumigatus accepted
also tryptophan and its simple derivatives was provided and,
therefore, demonstrated that the broad substrate specificity observed
for 7-DMATS9 is also applicable for other indole prenyltransferases,
at least for the enzymes described in this study. This feature makes
this new enzyme group especially useful as tools for production of

52 Journal of Natural Products, 2009, Vol. 72, No. 1
Zou et al.
prenylated compounds by chemoenzymatic synthesis.8,9 Interest-
ingly, the FtmPT1 products of tryptophan and simple derivatives
differ clearly from those of tryptophan-containing cyclic dipeptides
by different prenylation positions and prenylation patterns, i.e.,
reverse N-1-prenylation for the former and regular C-2-prenylation
for the last case (Figures 2I and 2J). These results demonstrated
the substrate promiscuity of this enzyme group, in addition to their
catalytic promiscuity.11
Spectroscopic Analysis. The isolated products (50-200 µg) were
1
1
analyzed by H NMR spectroscopy, H-1H-COSY, and positive and
negative electrospray ionization (ESI) mass spectrometry with a
ThermoFinnigan TSQ Quantum. The mass spectrometer was coupled
with an Agilent HPLC series 1100 equipped with a RP18-column (10
× 250 mm, 5 µm). For separation, the column was run with 50% solvent
B (MeOH) in solvent A (H2O), each containing 0.1% HCO2H for 5
min, followed by a gradient from 10% to 100% B over 30 min. After
washing with 100% B, the column was equilibrated with 10% B for
10 min. The flow rate was 0.2 mL min-1
.
NMR spectra were recorded on an Avance DRX 500 spectrometer
(Bruker). The solvent signal of D2O at 4.81 ppm was used as reference.
NMR and MS data are given in Tables 2 and S1, respectively.
Experimental Section
Chemicals. Trisammonium salt of DMAPP was prepared by analogy
with the synthesis of trisammonium geranyl diphosphate reported by
Woodside.15
OverproductionandPurificationofHis6-CdpNPTandHis6-FtmPT1.
Overproduction and purification of His6-CdpNPT and His6-FtmPT1
were described previously.4,10
Acknowledgment. This work was supported by a grant from the
Deutsche Forschungsgemeinschaft (to S.-M.L.). H.Z. was a recipient
of a fellowship from China Scholarship Council.
Enzyme Assays. The enzyme assays contained 50 mM Tris-HCl
(pH 7.5) and 10 mM CaCl2, differing from each other by incubation
volumes, amounts of FtmPT1 or CdpNPT, and incubation times. The
reaction mixtures of the standard assay for determination of the substrate
specificity (100 µL) contained 1 mM tryptophan or derivatives, 2 mM
DMAPP, and 1.68 µM purified His6-FtmPT1 or 1.62 µM His6-CdpNPT.
After incubation for 2 h at 37 °C, the reaction was stopped by addition
of 100 µL of MeOH. After removal of the protein by centrifugation
(15000g, 10 min, 4 °C), the enzymatic products were analyzed on an
HPLC system described below. For quantification, two independent
assays were carried out routinely. For determination of the kinetic
parameters, DMAPP at a final concentration of 2 mM and L-tryptophan
of up to 5 mM were used. The assays for isolation of the enzymatic
products for structural elucidation (5 mL) used 1 mM tryptophan or
derivatives, 2 mM DMAPP, and 1.68 µM purified His6-FtmPT1 or 1.62
µM His6-CdpNPT and were incubated for 16 h. The reaction mixtures
were concentrated on a rotation evaporator at 30 °C to a volume of
500 µL before injection.
Supporting Information Available: This material is available free
of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) Nierman, W. C.; Pain, A.; Anderson, M. J.; Wortman, J. R.; Kim,
H. S.; Arroyo, J.; Berriman, M.; Abe, K.; Archer, D. B.; Bermejo, C.;
et al. Nature 2005, 438, 1151–1156.
(2) Ruan, H.-L.; Yin, W.-B.; Wu, J.-Z.; Li, S.-M. ChemBioChem 2008,
9, 1044–1047.
(3) Unso¨ld, I. A.; Li, S.-M. Microbiology 2005, 151, 1499–1505.
(4) Grundmann, A.; Li, S.-M. Microbiology 2005, 151, 2199–2207.
(5) Unso¨ld, I. A.; Li, S.-M. ChemBioChem 2006, 7, 158–164.
(6) Kremer, A.; Westrich, L.; Li, S.-M. Microbiology 2007, 153, 3409–
3416.
(7) Grundmann, A.; Kuznetsova, T.; Afiyatullov, S. S.; Li, S.-M. Chem-
biochem 2008, 9, 2059–2063.
(8) Steffan, N.; Unso¨ld, I. A.; Li, S.-M. ChemBioChem 2007, 8, 1298–
1307.
(9) Kremer, A.; Li, S.-M. Appl. Microbiol. Biotechnol. 2008, 79, 951–
961.
HPLC Analysis and Determination of the Conversion Rate.
Reaction mixtures of CdpNPT and FtmPT1 were analyzed on an Agilent
HPLC Series 1100 by using a LiChrospher RP 18-5 column (125 × 4
mm, 5 µm, Agilent) at a flow rate of 1 mL min-1. The substances
were detected with a photodiode array detector. H2O (solvent A) and
MeOH (solvent B) were used as solvents. For separation of L-tyrptophan
and its simple derivatives, the samples were eluated with 30% B for 2
min, followed by a gradient from 30% to 70% B in 13 min. After
washing with 100% solvent B for 5 min, the column was equilibrated
with 30% solvent B for 5 min. For cyclic dipeptides, a gradient 50%
to 80% was used instead of 30% to 70%. To determine the stability of
the enzymatic products (Figures 5Band 5I), a gradient of 40% to 75%
was used instead.
(10) Yin, W.-B.; Ruan, H.-L.; Westrich, L.; Grundmann, A.; Li, S.-M.
ChemBioChem 2007, 8, 1154–1161.
(11) Kremer, A.; Li, S.-M. Chem. Biol. 2008, 15, 729–738.
(12) Baran, P. S.; Guerrero, C. A.; Corey, E. J. J. Am. Chem. Soc. 2003,
125, 5628–5629.
(13) Casnati, G.; Marchelli, R.; Pochini, A. J. Chem. Soc., Perkin Trans.
1 1974, 75, 4–757.
(14) Horton, H. R; Moran, L. A.; Scrimgeour, K. G.; Perry, M. D.; Rawn,
J. D. Principles of Biochemistry; Pearson Education, Inc.: London
2006; p 141.
(15) Woodside, A. B.; Huang, Z.; Poulter, C. D. Org. Synth. 1988, 66,
211–215.
NP800501M