Mapping the sirodesmin PL biosynthetic pathway- A remarkable intrinsic steric deuterium isotope effect on a 1H NMR chemical shift determines β-proton exchange in tyrosine

Article abstract of DOI:10.1139/V09-019

Sirodesmin PL is both an antibiotic and a phytotoxin produced by a fungal plant pathogen (Leptosphaeria maculans,asexual stage Phoma lingam) that causes blackleg disease on crucifers. To determine potential biosynthetic precursors of sirodesmin PL, deuterated compounds were synthesized and incubated with cultures of L. maculans. Incorporations of deuterium into sirodesmin PL (7) and its precursor phomamide (4) were determined using ¹H and ¹³C NMR spectroscopy, LCHRMS-ESI and HRMS-EI spectrometry. Spectroscopic analyses established that [3,3-²H₂]L- tyrosine (1a), [3,3-²H₂]O-prenyl-L-tyrosine (9a), [3,3,5',5',5'-²H₅]O-prenyl-L-tyrosine (9b), and [5,5- ²H₂]phomamide (4a) were incorporated efficiently into sirodesmin PL (7). Interestingly, an unexpected "twist" revealed that one of the β-deuteria (pro-R) of [3,3-²H₂]L-tyrosine (1a) was exchanged stereospecifically before incorporation into sirodesmin PL (7). As well, our studies revealed that O-prenyl-Ltyrosine is likely to be the first committed precursor en route to sirodesmin PL (7).

Full text of DOI:10.1139/V09-019

556
Mapping the sirodesmin PL biosynthetic pathway —
A remarkable intrinsic steric deuterium isotope
1
effect on a H NMR chemical shift determines
b-proton exchange in tyrosine
M. Soledade C. Pedras and Yang Yu
Abstract: Sirodesmin PL is both an antibiotic and a phytotoxin produced by a fungal plant pathogen (Leptosphaeria maculans,
asexual stage Phoma lingam) that causes blackleg disease on crucifers. To determine potential biosynthetic precursors of si-
rodesmin PL, deuterated compounds were synthesized and incubated with cultures of L. maculans. Incorporations of deute-
rium into sirodesmin PL (7) and its precursor phomamide (4) were determined using 1H and 13C NMR spectroscopy, LC-
HRMS-ESI and HRMS-EI spectrometry. Spectroscopic analyses established that [3,3-2H2]L-tyrosine (1a), [3,3-2H2]O-prenyl-
L-tyrosine (9a), [3,3,5’,5’,5’-2H5]O-prenyl-L-tyrosine (9b), and [5,5-2H2]phomamide (4a) were incorporated efficiently into si-
rodesmin PL (7). Interestingly, an unexpected ‘‘twist’’ revealed that one of the b-deuteria (pro-R) of [3,3-2H2]L-tyrosine (1a)
was exchanged stereospecifically before incorporation into sirodesmin PL (7). As well, our studies revealed that O-prenyl-L-
tyrosine is likely to be the first committed precursor en route to sirodesmin PL (7).
Key words: biosynthesis, isotope effect, deuterium incorporation, Leptosphaeria maculans, phytotoxin, sirodesminbiosynth-
`
´
ese, effet isotopique, incorporation de deuterium, , Leptosphaeria maculans, phytotoxine, sirodesmine.
´
´
`
`
Resume : La sirodesmine PL est a la fois un antibiotique et une phytotoxine produite par un champignon pathogene (Leptos-
`
phaeria maculans, stade asexuel du Phoma lingam) qui provoque la maladie de la jambe noire chez les cruciferes. Dans le but
´
´
´
´
`
de determiner la nature des precurseurs biosynthetiques potentiels de la sirodesmine PL, on a effectue la synthese de produits
´ ´ ´ ´
´
´
deuteries qui ont ete incubes avec des cultures de L. maculans. L’incorporation du deuterium dans la sirodesmine PL (7) et
dans son precurseur, la phomamide (4), a ete determine en faisant appel a la RMN du H et du 13C ainsi que par chromatogra-
1
´
´ ´
´
´
`
´
`
´
`
´
´
´
phie liquide liee a la spectrometrie de masse a haute resolution avec ionisation par impact electronebulisation (CL-SMHR-
´
`
´
´
IEN) et par spectrometrie de masse a haute resolution avec ionisation par impact electronique (SMHR-IE). Les analyses spec-
2
2
´
´
troscopiques ont permis de determiner que la [3,3- H2]L-tyrosine (1a), la [3,3- H2]O-prenyl-l-tyrosine (9a), la [3,3,5’,5’,5’-
2
2
´
H5]O-prenyl-l-tyrosine (9b) et la [5,5- H2]phomamide (4a) s’incorporent d’une fac¸on efficace dans la sirodesmine PL (7).
2
´
` ´ ´
´ `
´
Un fait interessant a ete note, a savoir celui qu’un des atomes de deuterium en b, le pro-R, de la [3,3- H2]L-tyrosine (1a)
´
´ ´
´
´
´ ´ ´
s’echange d’une fac¸on stereospecifique avant l’incorporation dans la sirodesmine PL (7). Des plus, nos etudes ont revele que
´
´
´ ´
la O-prenyl-l-tyrosine est probablement le premier precurseur dedie en route vers la sirodesmine PL (7).
´
[Traduit par la Redaction]
dioxopiperazines (ETPs) produced by two unrelated fungal
species, Sirodesmium diversum (sirodesmins A–G)3 and Lep-
tosphaeria maculans (asexual stage Phoma lingam, sirodes-
min PL, deacetylsirodesmin PL,4 and sirodesmins H,5 J, and
K6). Sirodesmin PL (7) is both an antibiotic and a phytotoxin,
i.e., toxic to plant cells. Interest in ETPs has increased over
the past years because of their biological activity7 and poten-
tial ecological roles. The current focus on sirodesmin PL (7)
is motivated by its potential involvement in the interaction of
the plant pathogen L. maculans with its host plants.8,9
The primary biosynthetic precursors of sirodesmin PL
(7) together with its biosynthetic pathway were proposed
based on results of experiments using radiolabeled (3H and
14C) and stable (13C) isotopes,10 and later on confirmed by
Bu’Lock and Clough.11 The proposed pathway involved di-
oxopiperazine formation between L-tyrosine (1) and L-serine
(2), isoprenylation with dimethylallylpyrophosphate (DMAPP)
to form phomamide (4), followed by a number of oxidative
rearrangements. As well, the isolation of a toxin structur-
Introduction
Mapping biosynthetic pathways used by living organisms to
assemble complex libraries of natural products is generally
accomplished using isotopically labeled precursors. The
choice of isotope depends on the chemical structure of the nat-
ural product and hypothetical precursors. Currently, the use of
stable isotopes is common due to the availability of highly
sensitive analytical tools, while a few decades ago only radio-
active isotopes were thinkable.1 Nonetheless, great challenges
remain, particularly for unusual and complex pathways as-
sembled by nonribosomal peptide synthases, such as those
producing sirodesmins.2 Sirodesmins are epipolythio-2,5-
Received 12 November 2008. Accepted 24 December 2008.
Published on the NRC Research Press Web site at
canjchem.nrc.ca on 17 March 2009.
M.S.C. Pedras1 and Y. Yu. Department of Chemistry,
University of Saskatchewan, 110 Science Place, Saskatoon, SK
S7N 5C9, Canada.
ally related to sirodesmin PL (7), phomalirazine (6), led us
1Corresponding author (e-mail: s.pedras@usask.ca).
´ ´
12
to propose a modification of the Curtis–Ferezou pathway.
Can. J. Chem. 87: 556–562 (2009)
doi:10.1139/V09-019
Published by NRC Research Press

Pedras and Yu
557
Scheme 2. Hypothetical intermediate 8 for introduction of sulfur
Scheme 1. Biosynthetic pathway of sirodesmin PL (7): precursors
L-tyrosine (1), L-serine (2), and DMAPP (dimethylallylpyropho-
sphate) and potential intermediates 3–6.
into sirodesmin PL (7).
Incorporation experiments using deuterated compounds
[3,3-2H2]L-tyrosine (1a), [3,3-2H2]O-prenyl-L-tyrosine (9a),
[3,3,5’,5’,5’-2H5]O-prenyl-L-tyrosine (9b), [2,3,3-2H3]L-serine
(2a), [5,5-2H2]phomamide (4a), and [5,5-2H2]cyclo-L-Tyr-L-
Ser (3a) were carried out using a wild type isolate of
L. maculans. After incubation of the fungal cultures with
deuterated compounds, cultures were extracted and phoma-
mide (4) and sirodesmin PL (7) were purified as described
in the experimental. Similar conditions were used for control
experiments in which the corresponding natural abundance
compounds were administered to fungal cultures. Incorpora-
The recent cloning of a gene cluster involved in the bio-
synthesis of sirodesmin PL (7) suggested a more detailed
pathway that took into account the ascribed functions of
the cloned genes.8 A simplified pathway showing relevant
potential intermediates is depicted in Scheme 1.
1
tion of deuterium, or lack of it, was determined using H
and 13C NMR spectroscopy, LC-HRMS-ESI and HRMS-EI
spectrometry.
First analysis of HRMS data established that (i) [3,3-2H2]L-
tyrosine (1a), [3,3-2H2]O-prenyl-L-tyrosine (9a), [3,3,5’,5’,5’-
2H5]O-prenyl-L-tyrosine (9b), and [5,5-2H2]phomamide (4a)
were incorporated efficiently (>10%) into sirodesmin PL (7);
(ii) [3,3-2H2]L-tyrosine (1a), [3,3-2H2]O-prenyl-L-tyrosine
(9a), and [3,3,5’,5’,5’-2H5]O-prenyl-L-tyrosine (9b) were in-
corporated efficiently (>20%) into phomamide (4); (iii) [5,5-
2H2]cyclo-L-Tyr-L-Ser (3a) and [2,3,3-2H3]L-serine (2a) were
incorporated poorly (<5%) into either phomamide (4) or siro-
desmin PL (7) (Table 1). Detailed analyses of spectroscopic
data obtained for phomamide (4) and sirodesmin PL (7) iso-
lated from fungal cultures incubated with deuterated
compounds (summarized in Table 1 and Fig. 1) provided im-
portant additional information, as described below.
The HRMS-EI data of purified phomamide (4) isolated
from cultures incubated with [3,3-2H2]L-tyrosine (1a) indi-
cated that its [M + 1]+ and [M + 2]+ were about 20% and
12% higher, respectively, than those of control samples, sug-
gesting that 20% of phomamide (4) was monodeuterated
(equation for calculation is shown in Table 1) and 12% was
dideuterated, i.e., 32% of phomamide (4) resulted from in-
corporation of [3,3-2H2]L-tyrosine (1a) (Table 1, entry 1).
The 1H NMR spectrum (provided in the Supplementary
data) of phomamide (4) showed that the protons H-5a (dH
3.30, dd, J = 14.0, 3.5 Hz) and H-5b (dH 2.99, J = 14.0,
8.9 Hz) integrated for ca. 0.7 and 0.9, respectively, whereas
control samples showed the expected integration of ca. 1 for
H-5a and H-5b (H-5a (dH 3.30) / H-5b (dH 2.99) = 1.0). The
These recent developments reawakened our interest in the
sirodesmin PL (7) biosynthetic pathway. Specific goals to
uncover the structure(s) of intermediate(s) immediately pre-
ceding insertion of the disulfide bridge and to determine the
stage at which prenylation occurred directed our experimen-
tal design. Based on the structures of metabolites recently
isolated,13 we hypothesized that a compound containing a
bis-ylidine 2,5-dioxopiperazine moiety (8) might be a plausi-
ble substrate for enzymatic introduction of the disulfide
bridge into sirodesmin PL (7). If 8 was a precursor, then
incorporation of dideuterated [3,3-2H2]L-tyrosine (1a) or
[5,5-2H2]phomamide (4a) would lead to the incorporation of
only one deuterium into sirodesmin PL (7, 2H–C-5, Scheme 2).
Alternatively, the presence of two deuteria in sirodesmin PL
would eliminate 8 from this biosynthetic map (Scheme 2). To-
ward this end, we describe work to establish biosynthetic pre-
cursors of phomamide (4) and sirodesmin PL (7).
Results and discussion
The deuterated compounds [3,3-2H2]O-prenyl-L-tyrosine
(9a),14 [3,3,5’,5’,5’-2H5]O-prenyl-L-tyrosine (9b), [5,5-
2H2]cyclo-L-Tyr-L-Ser (3a), and [5,5-2H2]phomamide (4a)15
were prepared from [3,3-2H2]L-tyrosine (1a) following mod-
ifications of previously reported methods, as described in the
Supplementary data. HRMS-EI analyses indicated that the
percentage of deuterium in the synthetic deuterated com-
pounds was higher than 98%.
Published by NRC Research Press

558
Can. J. Chem. Vol. 87, 2009
Table 1. Percentages of incorporation of deuterated compounds into phomamide (4) and sirodesmin PL (7) in cultures of Leptosphaeria
maculans.
Deuterated
Entry compound
2
2
Phomamide (4) (% of H, analytical method)
Sirodesmin PL (7) (% of H, analytical method)
32 (HRMS-EI: monodeuterated = 20;
32 (HRMS-EI: monodeuterated = 19; dideuterated =
13)a; 30 (1H NMR, H-7)b
1
dideuterated = 12)a; 30 (1H NMR, H2-5a,5b)
41 (HRMS-EI: monodeuterated = 24;
35 (HRMS-EI: monodeuterated = 21; dideuterated =
15)a; 32 (1H NMR, H-7)b
2
3
dideuterated = 17)a; 39 (1H NMR, H2-5a,5b)
3 (HRMS-EI)a
4 (HRMS-EI)a
30 (HRMS-EI: monodeuterated = 4; dideuterated =
26)a; 27 (1H NMR, H-7)b
4
—
3 (HRMS-EI: monodeuterated = 1;
dideuterated = 2)a
5
6
2 (HRMS-EI: monodeuterated = 0; dideuterated = 2)a
23 (HRMS-ESI: tetradeuterated = 16;
13 (HRMS-ESI: tetradeuterated = 9; pentadeuterated =
4)c; 23 (1H NMR, H-7)b
pentadeuterated = 7)c; 26 (1H NMR, H2-5a,5b)
aThe percentage of incorporation (I) was determined by HRMS-EI and was calculated using the formula I = {([M + n]+ – [M + n]C+tl) / ([M]+ + [M + 1]+ + [M +
2]+)} Â 100; n = 1, 2; [M + n]+ = intensity of deuterated molecular ion; [M + n]C+tl = intensity of molecular ion of control samples (natural abundance).
bResolution of signals was enhanced through a Lorentz–Gaussian lineshape transformation (gb = 0.3, lb = –1).
cThe percentage of incorporation (I) was determined by HRMS-ESI (positive mode) and was calculated using the formula I = {([M + 1 + n]+ – [M + 1 + n]+Ctl
)
/ ([M + 1]+ + [M + 2]+ + [M + 5]+ + [M + 6]+)} Â 100; n = 4, 5; [M + n]+ = intensity of deuterated molecular ion; [M + n]+Ctl = intensity of molecular ion of
control samples (natural abundance).
percentage of deuterium at the C-5 of phomamide resulting
from feeding [3,3-2H2]L-tyrosine (1a) was calculated to be
30% for H-5a (dH 3.30) and 11% for H-5b (dH 2.99). These
1H NMR data indicated that about 11% of phomamide (4)
was dideuterated, whereas the remaining 19% of phomamide
was monodeuterated (Table 1, entry 1), in complete agree-
ment with the HRMS data. Similar results were obtained
when cultures were incubated with [3,3-2H2]O-prenyl-L-tyro-
sine (9a, Table 1, entry 2).
for ca. 2 in natural abundance samples of 7 and for <2 (1.4–
1.6) in samples resulting from incorporation of [3,3-2H2]L-
tyrosine (1a). Unexpectedly, a closer inspection of the signal
1
due to H-7 (dH 5.56) in the H NMR spectrum of sirodesmin
PL (7) isolated from feeding experiments using [3,3-2H2]L-
tyrosine (1a) showed two partially overlapping peaks attrib-
utable to H-7 (dH 5.56) that were not apparent in nondeuter-
ated samples (i.e., natural abundance samples of 7).
Integration of these signals (dH 5.56) using resolution en-
hancement (Lorentz–Gaussian lineshape transformations and
baseline corrections) showed resonances at dH 5.556 and dH
5.550 in a 2:1 ratio (Figs. 1a and 1b). The presence of both
peaks suggested that replacement of H2-5 of sirodesmin PL
(7) with deuterium affected the chemical shift of H-7, i.e.,
the new peak at dH 5.550 was caused by the isotope shift ef-
Similarly, the HRMS-EI spectroscopic data of purified si-
rodesmin PL (7) isolated from cultures fed with [3,3-2H2]L-
tyrosine (1a) provided percentages of deuterium incorpora-
tion. Because the [M]+ of sirodesmin PL (7) observed in the
HRMS-EI spectrum was very weak, the base peak [M – S2]+
was used to calculate the percentage of deuterium incorpora-
tion, as follows. The intensity of the [M – S2 + 1]+ of siro-
desmin PL was ca. 19% higher than that of control samples,
whereas that of [M – S2 + 2]+ was ca. 13% higher than that
of control samples. These results indicated that both mono-
deuterated (19%) and dideuterated (13%) sirodesmin PL (7)
were biosynthesized from 1a (Table 1, entry 1). Therefore,
the total amount of deuterated 7 (mono plus di) calculated
from the HRMS-EI data was ca. 32%.
2
fect of H2-5 on H-7. The integrated area of this peak (dH
5.550) in conjunction with the HRMS-EI data (Table 1, en-
try 1) indicated that it accounted for the total amount of
monodeuterated and dideuterated 7 (ca. 33%). Nonetheless,
detection of a deuterium isotope shift effect over four cova-
lent bonds was not expected.1,16 To further understand this
result, the proton decoupled 13C NMR spectrum of this deu-
terated sample of sirodesmin PL (7) was obtained and ana-
lyzed. The signal of C-5 (single plus multiplet) had low
intensity due to the presence of deuterium; as well, C-4 (dC
75.46, 75.40, 75.32) and C-6 (dC 82.44, 82.38, 82.34,
Figs. 1c and 1d) displayed the expected two additional peaks
1
The H NMR spectrum of sirodesmin PL (7) was less in-
formative (provided in the Supplementary data) than that of
phomamide (4) because the signals due to protons H2-5 (dH
3.27) appeared as a broad AB quartet; this signal integrated
Published by NRC Research Press

Pedras and Yu
559
1
Fig. 1. Sections of NMR spectra of sirodesmin PL: (a) H NMR,
ca. 33% of H-7 and this amount corresponded to the total
1
H-7 signal of natural abundance 7; (b) H NMR, H-7 signal (peaks
amount of deuterated 7 determined by HRMS-EI (19:13 ra-
1
1 at d 5.556 and 2 at d 5.550) of 7 due to partial deuteration at C-5;
(c) 13C NMR, peak 3: C-6 signal of natural abundance 7; (d) 13C
NMR, peak 3: C-6 signal of 7, peak 4: C-6 signal of 7b, and peak
5: C-6 signal of 7a due to partial deuteration at C-5 (full H NMR
spectra of isolated deuterated phomamides and sirodesmins PL are
provided in the Supplementary data).
tio), the 33% accounted by H NMR must necessarily repre-
sent a mixture of dideuterated (7a) and monodeuterated (7b)
compound in a 19:13 ratio, respectively. This unexpected
isotope shift effect (higher field signal of H-7, dH 5.550)
could be explained by a through space intrinsic deuterium
isotope effect on H-7.
1
Isotope effects on proton and (or) carbon chemical shifts
in compounds where deuterium and the observed proton(s)/
carbon(s) were separated by four or more covalent bonds but
where the nuclei were close together in space, termed ‘‘in-
trinsic steric isotope effects’’, have been reported and char-
acterized.17 The dideuterated sirodesmin molecules had both
pro-R and pro-S deuteria at C-5 (13%), whereas the mono-
deuterated molecules (19%) in principle could have either
the pro-R or the pro-S deuterium (corresponding to the pro-
R or the pro-S hydrogens) or a mixture of pro-R and pro-S
(e.g., a 1:1 ratio). However, considering the nature of intrin-
sic steric deuterium isotope effects, it would be expected
that only the deuterated molecules having the deuterium
˚
close to H-7 (ca. <3 A) would display this effect. Molecular
mechanics calculations18 showed that structure 7b has the
˚
pro-S hydrogen at C-5 in closer proximity of H-7 (2.47 A)
˚
than the pro-R hydrogen (3.59 A). Therefore, these intera-
tomic distances are in close agreement with previous re-
ports17 on intrinsic steric deuterium isotope effects on
proton(s).
From the overall analysis it was concluded that only the
monodeuterated structure 7b was formed (but not 7c), in ad-
dition to the dideuterated 7a. Finally, extrapolation of this
conclusion to the precursor [3,3-2H2]L-tyrosine (1a) indi-
due to the b-isotope shift.1,16 Importantly, only one 13C
NMR peak was observed for C-7 (dC 79.2), indicating that
the deuterium at C-5 caused no isotope shift effect on C-7,
as expected due to the additional bond separation. Similar
results were obtained when cultures were incubated with
[3,3-2H2]O-prenyl-L-tyrosine (9a) (Table 1, entry 2). Fur-
thermore, feeding [2,3,3-2H3]L-serine (2a) and [5,5-2H2]cy-
clo-L-Tyr-L-Ser (3a) led to very low incorporation of
deuterium into sirodesmin PL (7) (Table 1, entries 3 and 5,
respectively).
2
cated that its HR was partially and stereospecifically ex-
changed before incorporation into phomamide (4), which in
turn was incorporated without further deuterium exchange
into sirodesmin PL. Furthermore, considering that feeding
either [3,3-2H2]L-tyrosine (1a) or [3,3-2H2]O-prenyl-L-tyro-
sine (9a) yielded monodeuterated products as well, it is sur-
mised that deuterium exchange occurred before formation of
phomamide, i.e., either in L-tyrosine or O-prenyl-L-tyrosine
or both (Scheme 3).
Comparable results in which dideuterated (11a) and
monodeuterated (11b) gliotoxin were obtained from feeding
experiments using [3,3-2H2]D,L-phenylalanine (8a) were re-
ported previously;19 however, stereospecific exchange of the
pro-S deuterium of deuterated phenylalanine (Scheme 4)
was established using stereospecifically monodeuterated
phenylalanines 10b and 10c.
The results described above and summarized in entries 1
and 2 of Table 1 demonstrated that 30%–35% of both
monodeuterated (7b) and dideuterated (7a) sirodesmin PL
(7) was obtained from incorporation experiments with di-
deuterated precursors 1a and 9a, whereas mostly dideuter-
ated 7a was obtained from incorporation of 4a (full 1H
NMR spectra of isolated deuterated phomamides and siro-
desmins PL are provided in the Supplementary data). Conse-
quently, these results indicated that deuterium exchange took
place before the final step of phomamide (4) biosynthesis
and thus the proposed bis-ylidine 2,5-dioxopiperazine (8) is
not a likely intermediate of the biosynthetic pathway.
Further data analysis and additional calculations clarified
the specific nature of this deuterium exchange, as follows.
The HRMS-EI data (Table 1, entry 1) showed that the total
amount of monodeuterated (19%) and dideuterated (13%)
molecules of 7 was 32% (19% + 13%, relative to natural
Finally, to clarify the stage at which prenylation occurred
(i.e., amino acid or dioxopiperazine), incorporation of
[3,3,5’,5’,5’-2H5]O-prenyl-L-tyrosine (9b) was investigated.
1
The H NMR spectra of both phomamide (4) and sirodesmin
PL (7) isolated from cultures incubated with [3,3,5’,5’,5’-
2H5]O-prenyl-L-tyrosine (9b) showed that integration of the
peak area of protons H-5a, H-5b, and H3-5’ of phomamide
(4) and H2-5 and H3-17 of sirodesmin PL (7) were lower
than those of control samples. That is, both pentadeuterated
and tetradeuterated phomamide (4c, 7%; 4d, 16%) and siro-
desmin PL (7d, 4%; 7e, 9%) were produced in cultures in-
cubated with pentadeuterated prenyl tyrosine 9b (Table 1,
entry 6). These amounts of deuterium incorporation and the
1
abundance compound), whereas the two H NMR peaks due
to H-7 at dH 5.556 and 5.550 (2:1, Figs. 1a and 1b) were due
to the total amount of 7, i.e., natural abundance, monodeu-
terated, and dideuterated. Since integration of both H-7
NMR peaks showed that the signal at dH 5.550 represented
Published by NRC Research Press

560
Can. J. Chem. Vol. 87, 2009
Scheme 3. Incorporation of [3,3-2H2]L-tyrosine (1a), [3,3-2H2]O-prenyl-L-tyrosine (9a), and [5,5-2H2]phomamide (4a) into sirodesmin PL
showing partial structures of dideuterated (7a) and monodeuterated (7b) sirodesmin PL; 7c is not formed; 7d and 7e result from incorpora-
tion of [3,3,5’,5’,5’-2H5]O-prenyl-L-tyrosine (9b). The curved arrow indicates an intrinsic steric deuterium isotope (ISDI) effect (ca. 3 Hz) of
1
2HS-5 on the H NMR chemical shift of H-7.
Scheme 4. Incorporation of [3,3-2H2]D,L-phenylalanine (10a) into
dideuterogliotoxin (11a) and monodeuterogliotoxin (11b).19
other primary pathways (e.g., synthesis of proteins and tryp-
tophan and (or) degradation to pyruvate and glycine). Pre-
vious results using radiolabeled serine and tyrosine reported
a rather low incorporation of both amino acids into phoma-
mide (4) and sirodesmin PL (7).10
Conclusion
In summary, this work demonstrated that both dideuter-
ated and monodeuterated sirodesmin PL (7a and 7b) and
phomamide (4a and 4b) were produced when the precursors
[3,3-2H2]L-tyrosine (1a) and [3,3-2H2]O-prenyl-L-tyrosine
(9a) were incubated with cultures of L. maculans. Impor-
tantly, a remarkable intrinsic steric deuterium isotope effect
detected on the NMR chemical shift of H-7 of monodeuter-
ated sirodesmin PL (7b) revealed that the pro-R deuterium
of sirodesmin PL exchanged stereospecifically. This stereo-
specific exchange occurred before incorporation of tyrosine
(1a) into phomamide (4), thus the proposed bis-ylidine 2,5-
dioxopiperazine (8) is not a likely intermediate of the siro-
desmin pathway. That is, a percentage of [3,3-2H2]L-tyrosine
(1a) is directly channelled into sirodesmin biosynthesis and
another fraction rechannelled after diverging into a path-
way(s) where a b-deuterium is exchanged stereospecifically.
Such a pathway(s) might involve transamination via pyri-
much lower incorporation of 3a (£2%) relative to com-
pounds 1a, 9a, and 4a indicated that 3 was not a likely in-
termediate in the sirodesmin PL (7) pathway. In addition,
the incorporation of specifically deuterated prenyl derivative
demonstrated clearly that CH3-17 is derived from the deuter-
ated methyl of substituted tyrosine 9b, which agrees with
previous data.10 Substantially lower incorporations of serine
than tyrosine may reflect a faster deployment of serine into
Published by NRC Research Press

Pedras and Yu
561
doxal phosphate with concomitant stereospecific exchange
of the pro-R b-deuterium with a proton.20 A fraction of the
resulting monodeuterated tyrosine ([3-2HS]L-tyrosine) could
eventually be redirected into the sirodesmin pathway. In ad-
dition, our results strongly suggested that prenylation oc-
curred before dioxopiperazine formation and thus O-prenyl-
L-tyrosine is a likely intermediate of phomamide (4) en
route to sirodesmin PL (7). Therefore, prenylation of tyro-
sine by the prenylase sirD appears to be the first committed
step in the biosynthetic pathway to sirodesmin PL (7).
As a final point, it is of interest to highlight that, in general,
demonstration of stereospecific isotope exchange observed in
biological pathways20,21 require feeding experiments with
stereospecifically labeled substrates, e.g., phenylalanine (10b
and 10c) ? gliotoxin (11 and 11b) (Scheme 4). By contrast,
in this work, the detection and substantiation of an intrinsic
steric deuterium isotope shift effect revealed clearly a stereo-
specific deuterium exchange (7b and 7e) in the biosynthetic
pathway to sirodesmin PL (7).
in sterile distilled water, 5.0 mmol/L) were added to cul-
tures. At the 5th day, the broth of each flask was extracted
with EtOAc (100 mL Â 3) and concentrated to dryness. The
residue was separated by preparative TLC (MeOH–CH2Cl2,
10:90) to give sirodesmin PL (7) (Rf = 0.75, ca. 90 mg/L)
and phomamide (4) (Rf = 0.25, ca. 10 mg/L).
Supplementary data
Supplementary data for this article are available on the
journal Web site (canjchem.nrc.ca) or may be purchased
from the Depository of Unpublished Data, Document Deliv-
ery, CISTI, National Research Council Canada, Ottawa, ON
K1A 0R6, Canada. DUD 3898. For more information on ob-
taining material refer to cisti-icist.nrc-cnrc.gc.ca/cms/
unpub_e.shtml.
Acknowledgments
Support for the authors’ work was obtained from the Nat-
ural Sciences and Engineering Research Council of Canada
(Discovery grant to MSCP), Canada Foundation for Innova-
tion, Canada Research Chairs (MSCP), and the University of
Saskatchewan (teaching assistantship to YY). We acknowl-
edge the technical assistance of K. Brown (NMR), G.
Schatte (NMR), and K. Thoms (MS) for data acquisition.
References
Materials and methods
(1) For a relevant review see: Schneider, B. Prog. Nucl. Magn.
Reson. Spectrosc. 2007, 51, 155. doi:10.1016/j.pnmrs.2007.
02.006.
(2) For a recent review on ETPs see: Gardiner, D. M.; Howlett,
B. J. Microbiology 2005, 151, 1021. doi:10.1099/mic.0.
27847-0. PMID:15817772.
(3) Curtis, P. J.; Greatbanks, D.; Hesp, B.; Forbes, C. A.; Freer,
A. A. J. Chem. Soc. Perkin Trans. 1977, 2, 180. doi:10.1039/
p19770000180.
(4) Quesneau-Thierry, A.; Boudart, G. Nouv. J. Chim. 1977, 1,
327.
General experimental procedures
All solvents were HPLC grade and used as such. Organic
extracts were dried with Na2SO4 and solvents removed
under reduced pressure in a rotary evaporator. Preparative
thin layer chromatography was carried out on silica gel
plates, Kieselgel 60 F254 (20 cm  20 cm  0.25 mm), com-
pounds were visualized under UV light. NMR spectra were
1
recorded on Bruker Avance 500 series spectrometers; for H
1
(500 MHz), d values were referenced as follows for H to
CHCl3 (7.27 ppm) and for 13C to CDCl3 (77.23 ppm). Mass
spectra (MS) were obtained on a mass spectrometer using a
solids probe.
´
(5) Pedras, M. S. C.; Abrams, S. R.; Seguin-Swartz, G. Tetrahe-
dron Lett. 1988, 29, 3471. doi:10.1016/0040-4039(88)85192-
X.
(6) Pedras, M. S. C.; Taylor, J. L.; Nakashima, T. T. J. Org.
Chem. 1993, 58, 4778. doi:10.1021/jo00070a002.
(7) Kupfahl, C.; Heinekamp, T.; Geginat, G.; Ruppert, T.; Haertl,
A.; Hof, H.; Brakhage, A. A. Mol. Microbiol. 2006, 62, 292.
doi:10.1111/j.1365-2958.2006.05373.x. PMID:16956378.
(8) Gardiner, D. M.; Cozijnsen, A. J.; Wilson, L. M.; Pedras, M.
S. C.; Howlett, B. J. Mol. Microbiol. 2004, 53, 1307. doi:10.
1111/j.1365-2958.2004.04215.x. PMID:15387811.
(9) Elliot, C. E.; Gardiner, D. M.; Thomas, G.; Cozijnsen, A.;
Van de Wouw, A.; Howlett, B. J. Molecular Plant Pathology
2007, 8, 791. doi:10.1111/j.1364-3703.2007.00433.x.
(10) Ferezou, J.-P.; Quesneau-Thierry, A.; Servy, C.; Zissmann ,
E.; Barbier, M. J. Chem. Soc. Perkin Trans. 1 1980, 1739.
doi:1.1039/P19800001739.
Synthesis of labeled precursors
The deuterated materials [3,3-2H2]O-prenyl-L-tyrosine
(9a), [3,3,5’,5’,5’-2H5]O-prenyl-L-tyrosine (9b), [5,5-2H2]cy-
clo-L-Tyr-L-Ser (3a), and [5,5-2H2]phomamide (4a) were
prepared following modifications of previously reported
methods, as described in the Supplementary data. All com-
pounds gave satisfactory spectroscopic data; in each case
the percentage of deuterated synthetic compound was ‡98%
(HRMS-EI).
Incorporation of labeled precursors
Fungal cultures of L. maculans isolate BJ 125 were ob-
tained from the IBCN collection, Agriculture and Agri-Food
Canada Research Station, Saskatoon, SK. Cultures were
handled as described previously.22 Isolate BJ 125 was grown
in 250 mL Erlenmeyer flasks containing 100 mL minimal
media inoculated with fungal spores at 1 Â 109 cells per
flask. The cultures were incubated on a shaker at 130 rpm,
at 24 2 8C for 3 d, then deuterated compounds (dissolved
(11) Bu’Lock, J. D.; Clough, L. E. Aust. J. Chem. 1992, 45, 39.
doi:10.1071/CH9920039.
´
(12) Pedras, M. S. C.; Abrams, S. R.; Seguin-Swartz, G.; Quail, J.
W.; Jia, Z. J. Am. Chem. Soc. 1989, 111, 1904. doi:10.1021/
ja00187a068.
(13) Pedras, M. S. C.; Yu, Y. Bioorg. Med. Chem. 2008, 16,
8063. doi:10.1016/j.bmc.2008.07.060. PMID:18701303.
Published by NRC Research Press

562
(14) Fraile, J. M.; Garcia, J. I.; Mayoral, J. A.; Royo, A. J. Tetra-
Can. J. Chem. Vol. 87, 2009
Anet, F. A. L.; Kronja, O. J. J. Am. Chem. Soc. 2007, 129,
10276. doi:10.1021/ja072375r. PMID:17655301.
(18) Computational calculations of the interatomic distances in
sirodesmin PL (7) were carried out using the Spartan 06
software package, method MMFF94.
(19) Bu’Lock, J. D.; Ryles, A. P.; Johns, N.; Kirby, G. W. J.
Chem. Soc. Chem. Comm. 1972, 100. doi:10.1039/
c39720000100.
(20) Sawada, S.; Kumagai, H.; Yamada, H.; Hill, R. K. J. Am.
Chem. Soc. 1975, 97, 4334. doi:10.1021/ja00848a033.
PMID:1141596.
(21) Hanson, K. R. Annu. Rev. Biochem. 1976, 45, 307. doi:10.
1146/annurev.bi.45.070176.001515. PMID:786152.
(22) Pedras, M. S. C.; Khan, A. Q. J. Agr. Food Chem. 1996, 44,
3403. doi:10.1021/jf960098u.
hedron Asymmetry 1996, 7, 2263. doi:10.1016/0957-
4166(96)00281-9.
´ ´
(15) (a) Ferezou, J. P.; Quesneau-Thierry, A.; Barbier, M.;
Kollmann, A.; Bousquet, J. F. J. Chem. Soc. Perkin Trans.
1980, 1, 113. doi:10.1039/p19800000113.; (b) Mancilla, T.;
Carrillo, L.; Zamudio-Rivera, L. S.; Beltran, H. I.; Farfan,
N. Org. Prep. Proced. Int. 2001, 33, 34; (c) Nitecki, D. E.;
Halpern, B.; Westley, J. W. J. Org. Chem. 1968, 33, 864.
doi:10.1021/jo01266a091.
(16) Jurlina, J. L.; Stothers, J. B. J. Am. Chem. Soc. 1982, 104,
4677. doi:10.1021/ja00381a030.
(17) (a) Anet, F. A. L.; Dekmezian, H. J. J. Am. Chem. Soc.
1979, 101, 5449 doi:10.1021/ja00512a073.; (b) Ernst, L.;
Eltamany, S.; Hopf, H. J. Am. Chem. Soc. 1982, 104, 299
doi:10.1021/ja00365a060.; (c) Saunders, M.; Wolfsberg, M.;
Published by NRC Research Press