Dissecting metabolic puzzles through isotope feeding: A novel amino acid in the biosynthetic pathway of the cruciferous phytoalexins rapalexin A and isocyalexin A

Article abstract of DOI:10.1039/c2ob27076e

Understanding defence pathways of plants is crucial to develop disease-resistant agronomic crops, an important element of sustainable agriculture. For this reason, natural plant defenses such as phytoalexins, involved in protecting plants against microbial pathogens, have enormous biotechnological appeal. Crucifers are economically important plants, with worldwide impact as oilseeds, vegetables of great dietetic value and even nutraceuticals. Notably, the intermediates involved in the biosynthetic pathways of unique cruciferous phytoalexins such as rapalexin A and isocyalexin A remain unknown. Toward this end, using numerous perdeuterated compounds, we have established the potential precursors of these unique phytoalexins and propose for the first time their detailed biosynthetic pathway. This pathway involves a variety of intermediates and a novel amino acid as the central piece of this complex puzzle. This work has set the stage for the discovery of enzymes and genes of the biosynthetic pathway of cruciferous phytoalexins of unique scaffolds.

Full text of DOI:10.1039/c2ob27076e

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Dissecting metabolic puzzles through isotope feeding:
a novel amino acid in the biosynthetic pathway of the
cruciferous phytoalexins rapalexin A and isocyalexin A†
Cite this: Org. Biomol. Chem., 2013, 11,
1149
M. Soledade C. Pedras* and Estifanos E. Yaya
Understanding defence pathways of plants is crucial to develop disease-resistant agronomic crops, an
important element of sustainable agriculture. For this reason, natural plant defenses such as phytoalex-
ins, involved in protecting plants against microbial pathogens, have enormous biotechnological appeal.
Crucifers are economically important plants, with worldwide impact as oilseeds, vegetables of great
dietetic value and even nutraceuticals. Notably, the intermediates involved in the biosynthetic pathways
of unique cruciferous phytoalexins such as rapalexin A and isocyalexin A remain unknown. Toward this
end, using numerous perdeuterated compounds, we have established the potential precursors of these
unique phytoalexins and propose for the first time their detailed biosynthetic pathway. This pathway
involves a variety of intermediates and a novel amino acid as the central piece of this complex puzzle.
This work has set the stage for the discovery of enzymes and genes of the biosynthetic pathway of cruci-
ferous phytoalexins of unique scaffolds.
Received 25th October 2012,
Accepted 14th December 2012
DOI: 10.1039/c2ob27076e
www.rsc.org/obc
families. Indeed, cruciferous defence-related metabolites seem
to combine structures concisely assembled, but containing a
Introduction
Plants, microbes and animals produce an astonishing array of relatively high number of functional groups, which is possible
secondary metabolites, better known as natural products.1,2 due to the presence of sulfur, nitrogen and oxygen hetero-
These products are biosynthesized from primary building atoms. Crucifers (plant family Brassicaceae) have worldwide
blocks such as amino acids, the precursors of alkaloids, impact due to all sorts of uses, from edible oils and vegetables
acetate/malonate, the precursors of polyketides, or shikimate, (cauliflower, broccoli, cabbage) to condiments (mustard,
the precursor of phenylpropanoids, flavonoids and others. wasabi), fodder, ornamentals, industrial oils and fuels.7
Such metabolic arrays are characteristic of individual groups
Since the discovery of the first cruciferous phytoalexins in
of organisms and are crucial to their overall fitness. For 1986,8 over 40 chemical structures have been solved.5 Some of
example, plants produce vital defensive metabolites commonly these metabolites have been found only in wild species, as for
known as phytoalexins and phytoanticipins that can be example wasalexins A (5) and B (6)9 in Thellungiella salsuginea
recruited into specific defence programs depending on the and camalexin (8)10 in Arabidopsis thaliana. For decades,
attacker. Phytoalexins are low molecular weight antimicrobial A. thaliana11 and more recently T. salsuginea12 have been devel-
metabolites biosynthesized de novo by plants in response to oped and exploited as model-systems for dissecting cellular
pathogen attack or other types of stress such as UV radiation,3 and metabolic processes. In fact, the availability of numerous
while phytoanticipins are constitutive metabolites.4 Remark- A. thaliana mutants has facilitated the isolation and cloning of
ably, most cruciferous phytoalexins,5 and some phytoantici- genes involved in the biosynthetic pathway of camalexin (8),
pins,6 contain nitrogen and sulfur atoms in addition to the but to date no genes or enzymes have been reported for any
carbon scaffold, a rare feature in phytoalexins of other plant other cruciferous phytoalexin. The structural diversity of cruci-
ferous phytoalexins points toward complex biosynthetic path-
ways and unique enzymes that have yet to be uncovered.
(S)-Tryptophan (Trp) is the primary building block of crucifer-
ous phytoalexins, including wasalexins and camalexins.5 Com-
parison of the currently known phytoalexin structures (45)
with that of Trp indicates only two carbon scaffold types, as
summarized in Fig. 1: scaffold A, encompassing the majority
of phytoalexins, which have indole plus a side-chain with one
Department of Chemistry, University of Saskatchewan, 110 Science Place, Saskatoon
SK S7N 5C9, Canada. E-mail: s.pedras@usask.ca; Fax: +(306)966-4730;
Tel: +(306)966-4772
†Electronic supplementary information (ESI) available: General experimental
details, Tables S1–S3 showing percentage of isotope incorporation, 1H and 13C
NMR spectra of new compounds shown in Table 1. See DOI:
10.1039/c2ob27076e
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1149

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 Cruciferous metabolites grouped according to carbon scaffolds. Scaffold A, indole plus side-chain C–N; scaffold B, indole plus side-chain N or O; scaffold C,
indole plus side-chain C–C–N.
carbon attached to a heteroatom (usually at C-3′; brassinin (1) recent discovery of isocyalexin A (13),20 apparently the only
and derivatives 2 and 3, cyclobrassinin (4), camalexin (8), etc., plant isocyanide reported to date, and the exceptional opportu-
ca. 37 metabolites), and scaffold B, representing four phyto- nity to discover a novel route leading to scaffold B, we have
alexins, which have indole directly attached to a heteroatom (at undertaken to dissect this challenging biosynthetic pathway.
C-3′; rapalexins A (10) and B (11), isocyalexin A (12) and isa- This investigation has revealed that a novel amino acid derived
lexin (13)). Furthermore, the cruciferous phytoanticipins gluco- from Trp is the central piece of this puzzle, setting the stage
brassicins 14–16 (indolyl-3′-methyl glucosinolates)13 are for biotechnological applications.
constitutive metabolites also derived from (S)-Trp,14,15 having
scaffold C, which contains indole plus a side-chain with two
carbons attached to nitrogen (at C-3′).
Results
While most of the biosynthetic intermediates involved in
the multistep biosynthesis of brassinins16–18 and wasalexins19
Hypothetical pathway of rapalexin A (10) and isocyalexin A (12)
and incorporation of (S)-[13C11,15N2]Trp
have been identified or deduced (Scheme 1), the biosynthetic
steps of phytoalexins containing scaffold B remain undeter-
mined. Although rapalexin A (10) has been shown to be a close
precursor of isalexin (13)20 and to incorporate perdeuterated
Trp,19 no other precursors have been revealed. In view of the
Inspired by the presence of an isothiocyanate group in rapa-
lexin A (10), we proposed that the hypothetical glucosinolate
glucorapassicin (21, 4′-methoxyindolyl-3′-glucosinolate, trivial
name derived from rapalexin and glucobrassicin) would be its
1150 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 1 Biosynthetic pathway from (S)-Trp to brassinins (two arrows indicate multiple steps; the compound in brackets is unstable, isolated only for quasi-
natural products18 with R = Ac or t-Boc).
Scheme 2 Proposed biosynthetic pathway of rapalexin A (10) (two arrows indicate multiple steps; the compound in brackets is expected to be unstable).
immediate precursor,21 by analogy to the brassinins pathway precursors and feeding of each labeled precursor to the par-
(Scheme 1) and to reactions catalysed by plant myrosinases ticular organism. The choice of isotope(s) depends on the
that use glucosinolates as substrates.13 Further biosynthetic chemical structures of metabolites and hypothetical precur-
considerations suggest that the precursor of glucorapassicin sors.22 The use of stable isotopes such as 2H, 13C, 15N, 34S
(21) would be 4′-methoxyindole-3′-carboxaldehyde oxime (20), instead of radioactive isotopes is possible and desirable due to
which in turn would derive from the corresponding amino the availability of highly sensitive analytical tools, namely
acid 4′-methoxyindolyl-3′-glycine (19). Although this idea is a HPLC-DAD-ESI-MS.17,19 Rutabaga roots (Brassica napus L. ssp.
greatly appealing and exciting possibility, it has not been rapifera) are a good choice to uncover the biosynthetic inter-
advanced or even proposed previously (Scheme 2), neither mediates leading to rapalexin A (10) and isocyalexin A (12),
have glucorapassicin (21) or amino acid 19 been reported in due to the straightforward nature of elicitation and feeding
crucifers or any other plants. Undoubtedly, an explanation for experiments.20
the conundrum of how amino acid 19 is formed in planta
To establish unambiguously the origin of the carbon and
must be found.
nitrogen atoms of both rapalexin A (10) and isocyalexin A (12),
Dissecting biosynthetic pathways involves retrobiosynthetic fully labeled (S)-[13C11,15N2]Trp was used. Feeding experiments
analysis of each structure to infer potential precursors of involved addition of aqueous solutions of labeled (S)-
metabolite, followed by synthesis of isotopically labeled [13C11,15N2]Trp to elicited slices of rutabaga roots, followed by
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1151

View Article Online
Paper
Organic & Biomolecular Chemistry
incubation and extraction of solutions and tissues. Control
experiments were conducted similarly, but using either non-
labeled, i.e. naturally occurring (S)-Trp, or the carrier solution
only. HPLC-DAD-ESI-MS analysis of each extract was carried
out; the identity of rutabaga metabolites corresponding to
each peak detected in HPLC-DAD-ESI-MS chromatograms was
established by direct comparison with authentic samples avail-
able in our libraries (UV spectra and ESI-MS data).
HPLC-ESI-MS analyses of extracts demonstrated that (S)-
[13C11,15N2]Trp was incorporated into various rutabaga meta-
bolites (Table S1, ESI†). To our absolute delight, the mass spec-
tral data (HPLC-ESI-MS) showed that both rapalexin A (10) and
isocyalexin A (12) incorporated nine carbons (eight indole
carbons plus one side-chain [13C9]) and both nitrogens (15N2)
of Trp (determined from the presence of [M − 1 + 11]− ions in
the mass spectrum of rapalexin A (10) and isocyalexin A (12),
Fig. 2). These results clearly indicated that the carbons of the
side-chain groups of both phytoalexins 10 and 12 resulted
Scheme 3 Incorporation of (S)-[13C11
,
15N2]Trp into rapalexin A (10), isocyalexin
A (12) and isalexin (13) in rutabaga roots (determined by HPLC-ESI-MS).
from (S)-Trp, suggesting that
a
rearrangement reaction [13C11,15N2]Trp. Further examination of the chromatograms of
occurred in the formation of both phytoalexins (Scheme 3; the extracts of fed tissues and analyses of mass spectra showed
nitrogen of the amino group of Trp became directly attached incorporation of carbons and nitrogens of (S)-[13C11,15N2]Trp
to C-3′ of indole). As well, isalexin (13) showed incorporation into the expected metabolites glucobrassicins 14–16, cyclobras-
of all carbons and one nitrogen of the indole ring of (S)- sinin (4), spirobrassinin (7) and rutalexin (9) (ESI Table S1†).
Fig. 2 ESI-MS spectra of: A, rapalexin A (10) [M − 1]− = 202.9, [M − 1 + 1]− = 203.9, [M − 1 + 2]− = 204.9, [M − 1 + 11]− = 213.9, and A-control; B, isocyalexin A
(12) [M − 1]− = 170.9, [M − 1 + 1]− = 171.9, [M − 1 + 11]− = 181.9, B-control, resulting from incorporation of (S)-[13C11 15N2]Trp showing ions corresponding to 13
, C
and 15N enriched and natural abundance (control).
1152 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
That is, this investigation confirmed previously established thioglucose tetraacetate and sulfonation using HSO3Cl. Depro-
biosynthetic relationships among the various phytoalexins bio- tection of tetraacetyl glucose with KOCH3 yielded N-t-Boc-glu-
synthesized in rutabaga roots and other agronomic Brassica corapassicin (28), which on standing in either aqueous or
species.5
organic solvents decomposed to various products over a period
of 12–24 hours. Attempts to deprotect N-t-Boc-glucorapassicin
(28) also led to extensive decomposition. For this reason, it
was investigated whether desulfoglucorapassicin (27) could be
prepared. Desulfoglucosinolates have been shown to be precur-
sors of glucosinolates (introduction of sulfate is the last bio-
synthetic step), hence this alternative compound seemed a
reasonable choice.15,25 Deprotection of the indole nitrogen fol-
lowed by methanolysis of N-t-Boc-desulfotetraacetyl gluco-
rapassicin using KOCH3 yielded desulfoglucorapassicin (27).
Fortunately, desulfoglucorapassicin (27) was stable in aqueous
solutions and in organic solvents, hence was judged satisfac-
tory to use in feeding experiments. Synthesis of the corres-
ponding hexadeuterated compound was carried out similarly,
using N-t-Boc-20a (N-t-Boc-4′-methoxyindole-3′-carboxaldehyde
oxime). The t-Boc deprotection step was carried out with deut-
erated TFA because the 4′-methoxy group caused exchange of
indole deuterons 2H-5′ and 2H-7′ with protons of TFA
(Scheme 5).
Synthetic planning and syntheses of perdeuterated
intermediates
Excited with the promising results obtained from incorpor-
ation of (S)-[13C11,15N2]Trp, we set out to synthesize 4′-methoxy-
indolyl-3′-glycine (19), 4′-methoxyindole-3′-carboxaldehyde
oxime (20) and glucorapassicin (21), and their perdeuterated
counterparts [2H3CO,5′,6′,7′-2H3]-4′-methoxyindolyl-3′-glycine
(19a), [2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-carboxaldehyde
oxime (20a), [2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-carbox-
aldehyde (24a) and [2H3CO,5′,6′,7′-2H3]glucorapassicin (21a). A
fairly straightforward retrosynthetic analysis suggested tetra-
deuterated indole as the starting material for very concise
syntheses of 19a and 21a via 24a (Scheme 4). However, a
concern remained regarding glucorapassicin (21),
a yet
unknown compound having somewhat unpredictable stability,
but since the indole nitrogen could be protected, the synthetic
challenge seemed solvable.
The synthesis of indolyl-3′-glycine (29) was previously
carried out by coupling indole (23) with oxalyl chloride fol-
lowed by hydrolysis, oximation and reduction.26,27 Following a
similar route and starting with tetradeuteroindole (23a),
[4′,5′,6′,7′-2H4]indolyl-3′-glycine (29a) was readily prepared.
Nonetheless, synthesis of the corresponding [2H3CO,5′,
6′,7′-2H3]-4′-methoxyindolyl-3′-glycine (19a) required a route
modification because the starting material, hexadeuterated
4-methoxyindole, was not readily accessible. First, it was antici-
pated that the α-keto acid or α-keto ester 25a might direct
methoxylation of the indole nucleus to C-4′. Although the reac-
tion did not work with the α-keto acid, the corresponding
α-keto ester 25a afforded the desired 4′-methoxyindolyl α-keto
acid upon standard 4-methoxylation conditions applicable to
indole derivatives.23 Next, oximation of the α-keto acid yielded
26a, which was reduced with NaBH4/NiCl2 to afford the
desired amino acid 19a (Scheme 5). In the first instance, syn-
thesis of both enantiomers of 19a was deemed unnecessary
because the racemic form would provide 50% of the desired
isomer. Additional perdeuterated compounds (Table 1, Fig. 3)
used in feeding experiments described below were synthesized
as reported in the Experimental section.
The synthesis of glucorapassicin (21) used N-t-Boc-4′-meth-
oxyindole-3′-carboxaldehyde oxime (N-t-Boc-20), prepared from
N-t-Boc-4′-methoxyindole-3′-carboxaldehyde, obtained from
indole (23), following a standard 4-methoxylation procedure.23
Chlorination of the resulting oxime N-t-Boc-20 using N-chloro-
succinimide (NCS)24 was followed by coupling with
Feeding experiments and determination of deuterium
incorporation
Feeding experiments were carried out with all deuterated com-
pounds shown in Table 1 and the corresponding natural abun-
dance compounds prepared using similar methods, to ensure
that biotransformations were carried out with comparable
materials. Rutabaga roots were elicited and incubated with
aqueous solutions of perdeuterated compounds and solutions
of the corresponding non-deuterated compounds, as described
in the Experimental section. The percentages of deuterium
incorporation into each metabolite were determined by
Scheme 4 Retrosynthetic analysis of [2H3CO,5’,6’,7’-2H3]-4’-methoxyindolyl-3’-
glycine (19a) and [2H3CO,5’,6’,7’-2H3]glucorapassicin (21a).
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1153

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 5 Synthesis of perdeuterated precursors 19a, 20a, 24a and 27a (compound 28 decomposed upon standing in solution).
Table 1 Perdeuterated compounds used in precursor feeding experiments
Compound number
or abbreviation
Deuterated compound (origin)
13
(S)-[U-
C
11
,U-15N2]Trp (commercial)
(S)-[13C11
14a
19a
,
15N2]Trp
[2,2,4′,5′,6′,7′-2H6]Glucobrassicin (synthetic)19
(R,S)-[2H3CO,5′,6′,7′-2H3]-4′-Methoxyindolyl-3′-
glycine (this work)
[2H3CO,5′,6′,7′-2H3]-4′-Methoxyindole-3′-
carboxaldehyde oxime (this work)
[2H3CO]-4′-Methoxyindole-3′-carboxaldehyde
(this work)
20a
24b
27a
[2H3CO,5′,6′,7′-2H3]Desulfoglucorapassicin
(this work)
(R,S)-[4′,5′,6′,7′-2H4]Indolyl-3′-glycine (this work)
[4′,5′,6′,7′-2H4]Indole desulfoglucosinolate
(this work)
29a
30a
Fig. 3 Chemical structures of perdeuterated compounds used in feeding
experiments.
[4′,5′,6′,7′-2H4]Indole-3′-carboxaldehyde oxime
(this work)
31a
[2H3CO]-4-Methoxyindole (this work)
[2H3CO]-4′-Methoxyindolyl-3′-acetonitrile
(this work)
32a
33a
Incorporations of perdeuterated 4′-methoxyindolyl-3′-glycine
(19a), 4′-methoxyindole-3′-carboxaldehyde oxime (20a) and
desulfoglucorapassicin (27a)
[4′,5′,6′,7′-2H4]Desulfoglucobrassicin (this work)
34a
Perdeuterated compounds 4′-methoxyindolyl-3′-glycine (19a),
4′-methoxyindole-3′-carboxaldehyde oxime (20a) and desulfo-
glucorapassicin (27a) were added to rutabaga root slices, roots
were incubated, extracted and the extracts analyzed by
analysis of data obtained by HPLC-ESI-MS (positive or negative
ion modes), as described in the Experimental section and
Tables S1–S3 (ESI†).
1154 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 4 ESI-MS spectra of: A, rapalexin A (10) [M − 1]− = 202.9, [M − 1 + 1]− = 203.9, [M − 1 + 2]− = 204.9, [M − 1 + 6]− = 208.9, A-control; B, isocyalexin A (12)
[M − 1]− = 170.9, [M − 1 + 1]− = 171.9, [M − 1 + 6]− = 176.9, B-control, resulting from incorporation of (R,S)-[2H3CO,5’,6’,7’-2H3]-4’-methoxyindolyl-3’-glycine (19a)
showing ions corresponding to deuterated and natural abundance (control).
HPLC-DAD-ESI-MS. From the amounts of deuterium present (Scheme 6). As expected, none of the metabolites with scaffold
in each phytoalexin, it was concluded that the amino acid 19a types A or C incorporated any of these perdeuterated meta-
and the oxime 20a were both incorporated into rapalexin A bolites (Table S2, ESI†). In addition, similarly conducted
(10a), isocyalexin A (12a) and isalexin (13a), while desulfoglu- feeding experiments using [2H3CO]-4-methoxyindole (32a),
corapassicin (27a) was incorporated into rapalexin A (10a) and [2H3CO]-4′-methoxyindole-3′-carboxaldehyde
(24b)
and
isalexin (13a), but not into isocyalexin A (12a) (Fig. 4–6, [2H3CO]-4′-methoxyindolyl-3′-acetonitrile (33a) followed by
Table S2, ESI†). Similarly conducted experiments using the HPLC-DAD-ESI-MS analyses of the tissues indicated that these
perdeuterated and non-methoxylated counterparts, i.e. indolyl- compounds were not incorporated into any of the metabolites
3′-glycine (29a), indole-3′-carboxaldehyde oxime (31a) and shown.
indole desulfoglucosinolate (30a), followed by HPLC-DAD-
ESI-MS analyses of extracts showed that these compounds
were not incorporated into rapalexin A (10a), isocyalexin A
Incorporations of perdeuterated glucobrassicin (14a) and
desulfoglucobrassicin (34a)
(12a), isalexin (13a) or any of the metabolites shown in
Table S2 (ESI†).
Previously, we have demonstrated that administration of hexa-
deuterated glucobrassicin (14a) led to incorporation of several
deuterium atoms into both wasalexins A (5a) and B (6a) and
into methoxyglucobrassicins 15a and 16a.19 In this work,
although only a maximum of three deuteria from hexadeuter-
ated glucobrassicin (14a) could be incorporated into rapalexin
A (10a), isocyalexin A (12a) and isalexin (13a), the availability
of hexadeuterated glucobrassicin (14a) was crucial to
These results indicate that 4′-methoxyindolyl-3′-glycine
(19a) is a precursor of phytoalexins with carbon scaffold type
B: rapalexin A (10a), isocyalexin A (12a) and isalexin (13a). Fur-
thermore, desulfoglucorapassicin (27a) is a precursor of rapa-
lexin A (10a) and isalexin (13a), but is not a precursor of
isocyalexin A (12a), that is 12a is formed upstream of 27a
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1155

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 5 ESI-MS spectra of: A, rapalexin A (10) [M − 1]− = 202.9, [M − 1 + 1]− = 203.9, [M − 1 + 2]− = 204.9, [M − 1 + 6]− = 208.9, A-control; B, isocyalexin A (12)
[M − 1]− = 170.9, [M − 1 + 1]− = 171.9, [M − 1 + 6]− = 176.9, B-control, resulting from incorporation of [2H3CO,5’,6’,7’-2H3]-4’-methoxyindole-3’-carboxaldehyde
oxime (20a) showing ions corresponding to deuterated and natural abundance (control).
Fig. 6 ESI-MS spectra of: A, rapalexin A (10) [M − 1]− = 202.9, [M − 1 + 1]− = 203.9, [M − 1 + 2]− = 204.9, [M − 1 + 6]− = 208.9, A-control, resulting from incorpor-
ation of [2H3CO,5’,6’,7’-2H3]desulfoglucorapassicin (27a) showing ions corresponding to deuterated and natural abundance (control); ESI-MS peaks corresponding
to isocyalexin A (12) indicated no incorporation of deuterium.
distinguish between two potential biosynthetic routes to phy- metabolic recycling of [4,5,6,7-2H4]indole (23a) to (S)-
toalexins of carbon scaffold A (brassinins, cyclobrassinins, [4′,5′,6′,7′-2H4]Trp to yield tetradeuterated metabolites, or
etc.): degradation of the side chain of 14a followed by direct incorporation of 14a to yield hexadeuterated
1156 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 6 Incorporation of perdeuterated compounds 19a, 20a and 27a into rapalexin A (10), isocyalexin A (12) and isalexin (13) in rutabaga roots (determined
by HPLC-ESI-MS).
Scheme 7 Incorporation of perdeuterated glucobrassicin (14a) into spirobrassinin (7a), cyclobrassinin (4a), rapalexin A (10a), isocyalexin A (12a) and isalexin (13a)
in rutabaga roots (determined by HPLC-ESI-MS).
metabolites. To establish if any of these alternative routes was glucobrassicin (14a) were present in rapalexin A (10a), isocya-
operating, [2,2,4′,5′,6′,7′-2H6]glucobrassicin (14a) was syn- lexin A (12a) and isalexin (13a), demonstrating that 14a was
thesized and added to prepared slices of rutabaga roots. After also incorporated into phytoalexins of scaffold type B. Further-
incubation and extractions, fractions were analyzed by more, considering the substantial percentages of deuterium
HPLC-DAD-ESI-MS and the percentages of deuterium content incorporation from glucobrassicin (14a) into both 1-methoxy-
were determined (Scheme 7, Table S3, ESI†). The presence of glucobrassicin (15a) and 4-methoxyglucobrassicin (16a), the
six deuteria in phytoalexins with carbon scaffold A (cyclobras- precursor relationship was once more confirmed.19 Similarly,
sinin and spirobrassinin) demonstrated that 14a was incorpor- [4′,5′,6′,7′-2H4]desulfoglucobrassicin (34a) was synthesized and
ated intact into these phytoalexins (Scheme 7 and Table S3, added to prepared slices of rutabaga roots, as reported above
ESI†).
Similarly,
three
deuteria
of
hexadeuterated for 14a. After incubation, extractions, and analyses using
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1157

View Article Online
Paper
Organic & Biomolecular Chemistry
HPLC-DAD-ESI-MS, the percentages of deuterium content were deprotection step failed) possibly due to its low chemical stabi-
determined (Table S3, ESI†). The presence of four deuteria in lity. However, this reasoning may not be applicable to the
phytoalexins of carbon scaffold A (cyclobrassinin (4a) and chemical stability of glucorapassicin (21) in plant cells, since
spirobrassinin (7a)) and three deuteria in rapalexin A (10a) and the conditions are quite different. Detailed steps of the trans-
isocyalexin A (12a) demonstrated that 34a was incorporated formation of Trp to glucobrassicins 14 and 16 are not shown
into phytoalexins of scaffold types A and B. Furthermore, 34a in Scheme 8, as the glucosinolate pathway has been the
was incorporated into both 1-methoxyglucobrassicin (15a) and subject of extensive studies reviewed recently.14,15,30 Nonethe-
4-methoxyglucobrassicin (16a) (Table S3, ESI†).
less, it is pertinent to stress that our data confirm that all
carbon (except for the methyl group of 16) and nitrogen atoms
of glucobrassicins 14–16 derive from (S)-Trp.
Although our proposed metabolic map of rapalexin A (10),
isocyalexin A (12) and isalexin (13) is complex (Scheme 8), con-
Discussion
The biosynthetic pathway to rapalexin A (10), isocyalexin A (12) sidering that most of the enzymes of the glucosinolate
and isalexin (13) starting from the primary building block pathway have been cloned,15 this knowledge should facilitate
(S)-Trp can be proposed as shown in Scheme 8. This pathway the isolation of the corresponding genes involved in the bio-
is based on incorporations of labeled Trp and perdeuterated synthesis of phytoalexins with scaffold B. Specially because
compounds shown in Table 1 and Fig. 3. The key step in the rapalexin
A is resistant to degradation by the blackleg
biosynthetic puzzle of phytoalexins with carbon scaffold B is fungus,31 one of the major pathogens of oilseed Brassica
the transformation of 4-methoxyglucobrassicin (16) to 4′-meth- species, this phytoalexin is likely to motivate great interest in
oxyindolyl-3′-glycine (19), for which we propose a Neber-type breeding programs aiming to generate plants resistant to
rearrangement via an azirine intermediate.28 By analogy to the blackleg disease.
base-catalysed chemical hydrolysis of benzylglucosinolate to
Overall, it appears that the important difference between
phenyl glycine,28 it is proposed that glucosinolate 15 the pathways of phytoalexins with scaffolds A or B lies in the
rearranges under enzymatic catalysis involving formation of stability of isothiocyanates 10 and 41; while isothiocyanate 10
the thioglucose azirine intermediate 35 followed by hydrolysis is chemically stable, 41 is not. As it stands, this difference
to the novel 4-methoxyindolyl containing amino acid 19. could directly influence the number/variety of phytoalexins
Transformation of 4′-methoxyindolyl-3′-glycine (19), the central having scaffolds A and B. Furthermore, the discovery of this
piece of this puzzle, to glucorapassicin (21) is proposed to new pathway derived from a new glucosinolate suggests that
follow a biosynthetic pathway similar to that of glucosino- other aromatic glucosinolates might be precursors of novel
lates.15 First, oxidative decarboxylation of 19 to oxime 20, fol- cruciferous phytoalexins. For example, benzyl glucosinolates
lowed by oxidation to the nitrile oxide 36 and introduction of might be precursors of yet unknown phytoalexins containing
sulfur likely via glutathione (GST) is proposed. Next, hydrolysis the benzyl scaffold and functional groups similar to those
of the glutamyl residue of GST followed by lyase mediated C–S found in scaffolds A and B. Hence, to evaluate the diversity
cleavage would lead to thiohydroximic acid 37. Glucosylation and evolution of phytoalexin structures, future work should
of 37 (likely the K salt) to yield desulfoglucorapassicin (27) fol- include additional plant families of the order Brassicales32
lowed by sulfation would yield glucorapassicin (21). Lossen- known to produce aromatic glucosinolates other than those
type rearrangement of 21 via 39 can lead to rapalexin A (10), containing indole. It would only be surprising if novel struc-
similar to the rearrangement of 40 to isothiocyanate 41, en tures were not discovered, as metabolic pathways of plants are
route to brassinin (1). Finally, oxidation of rapalexin A (10) can highly evolvable.33,34
lead to isalexin (13). The fact that perdeuterated compounds
Previously, it was shown that the callose defence response
14, 19a, 20a and 27a were incorporated into rapalexin A (10) in A. thaliana triggered by a microbial associated molecular
and isalexin (13) strongly supports this hypothesis. Further- pattern (Flg22, a synthetic 22-amino acid peptide) involved
more, since the non-methoxylated perdeuterated compounds GST, ascorbate, cadmium, and 4-methoxyglucobrassicin (16).
indolyl-3′-glycine (29a), indole desulfoglucosinolate (30a) and That is, callose deposition in A. thaliana, an innate immune
indole-3′-carboxaldehyde oxime (31a) were not incorporated, it response of plants to pathogens, requires the simultaneous
is likely that the enzymes downstream from 4′-methoxyindolyl- induction of three pathways, including 4-methoxylation (oxi-
3′-glycine (19) are selective. The biosynthetic pathway to isocya- dation followed by methylation) of glucobrassicin (14) and
lexin A (12) is interesting in its own right, rather different from hydrolysis of 4-methoxyglucobrassicin (16).35 Importantly,
the biosynthesis of the isocyanide functional group in other independent work confirmed these findings and established
organisms.29 Hence, the biosynthesis of isocyalexin A (12) is that only metabolite X derived from 4-methoxyglucobrassicin
proposed to involve sulfation of oxime 20 followed by a Beck- (16) was important for the restriction of fungal growth.36
mann-type rearrangement, because it incorporated 19a and Efforts to determine the structure of this metabolite X did not
20a, but did not incorporate desulfoglucorapassicin (27a) or lead to a concrete structure, but a reference was made to the
the non-methoxylated counterparts.
A pertinent question potential involvement of cruciferous phytoalexins. Considering
could be asked about the chemical stability of glucorapassicin that the steps required for callose deposition and formation
(21) since its synthesis could not be completed (the indole of metabolite X are also required for the biosynthesis of
1158 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 8 Biosynthetic pathway of phytoalexins rapalexin A (10), isocyalexin A (12) and isalexin (13) and biosynthetic relationships of glucobrassicin (14) with
brassinin (1) and cyclobrassinin (4).
4′-methoxyindolyl-3′-glycine (19), rapalexin A (10), isocyalexin A confirm this hypothesis as the phytoalexin profiles of rutabaga
(12) and isalexin (13), it is possible that one of these com- are more complex than those of A. thaliana.
pounds is either metabolite X or a precursor. Although we
As a final point, it is important to stress that breeding strat-
have previously detected rapalexin A in A. thaliana,9 we cannot egies to enhance natural resistance traits of plants to microbial
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1159

View Article Online
Paper
Organic & Biomolecular Chemistry
pathogens can be improved if detailed metabolic maps of extracts), H2O (with 0.2% HCO2H)–CH3CN (with 0.2% HCO2H)
defence pathways are available. Yet, among the 45 cruciferous from 90 : 10 to 50 : 50 in 25 min and a flow rate of 1.0 mL
phytoalexins reported thus far, only genes of the camalexin (8) min−1. Data acquisition was carried out in positive and nega-
pathway in the model plant A. thaliana have been cloned.5 To tive polarity modes in a single LC run, and data processing
facilitate similar discoveries within economically important carried out with Agilent Chemstation Software. Samples were
crops, we have investigated the potential precursors of the dissolved in CH3CN or CH3CN–H2O.19 Additional conditions
phytoalexins rapalexin A, isocyalexin A and isalexin using are as reported in ESI.†
numerous perdeuterated compounds, and proposed for the
The percentage of deuterium incorporation into each
first time their detailed biosynthetic pathway. We have shown metabolite was established from analysis of data obtained by
that the novel amino acid 4′-methoxyindolyl-3′-glycine is the HPLC-ESI-MS (positive or negative ion modes) using the
central piece of this puzzle, filling an important void in the expressions shown in footnotes to each table presented in ESI†
2
cruciferous phytoalexin biosynthetic conundrum. The stage is (e.g., % of H incorporation = {[M 1 + n]+/−/([M 1]+/− + [M
set to facilitate the discovery of the corresponding enzymes 1 + n]+/−)} × 100 (n = 3, 4, 6), where n is the number of deuter-
and genes of the biosynthetic pathway of cruciferous phytoa- ium atoms, and M is the peak intensity of the quasi-molecular
lexins of scaffolds A and B.
ion peak [M 1]+/−); the normalized peak intensities of deuter-
ated and corresponding non-deuterated compounds were iden-
tical; unless otherwise stated, the ions [M 1 + n]+/− were not
detected in natural abundance samples.
Experimental
General experimental
Compound synthesis and characterization
All solvents were HPLC grade and used as such, except for
those used in chemical syntheses, as noted. Unless otherwise
noted, materials were obtained from commercial suppliers
and used without further purification. Labeled (S)-Trp contain-
ing U-13C11, 97–99%; U-15N2, 97–99% was purchased from
Cambridge Isotope Laboratories, Inc., Andover, MA, USA. Ruta-
bagas were purchased from local markets throughout the year,
no substantial seasonal changes were observed.
All compounds gave satisfactory spectroscopic data; in each
case the percentage of deuterium in the synthetic compound
was ≥97%. All deuterated compounds have the additional
letter a (e.g. 19a), except for [2H3CO]-4′-methoxyindole-3′-car-
boxaldehyde (24b).
(R,S)-[2H3CO,5′,6′,7′-2H3]-4′-Methoxyindolyl-3′-glycine (19a)
Oxalyl chloride (52 μL, 0.60 mmol) was added to a solution of
[4,5,6,7-2H4]indole (60 mg, 0.50 mmol) in anhydrous diethyl
ether (2 mL) at 0 °C. The reaction mixture was stirred for 1 h at
0 °C, was filtered and the residue was washed with ice cold
ether and concentrated under reduced pressure to yield
crude [4′,5′,6′,7′-2H4]indole-3′-oxalylchloride quantitatively.
[4′,5′,6′,7′-2H4]Indole-3′-oxalylchloride (100 mg, 0.48 mmol)
was dissolved in MeOH (2 mL) and stirred at rt for 1 h. The
solvent was evaporated under reduced pressure to yield methyl
indolyl-3′-oxoacetate quantitatively. Thallium trifluoroacetate
(TTFA, 209 mg, 0.38 mmol) was added to a solution of methyl
indolyl-3′-oxoacetate (53 mg, 0.26 mmol) in TFA (2 mL) and
stirred for 15 h at 60 °C. The solvent was evaporated under
reduced pressure, DMF (2 mL), I2 (195 mg, 0.77 mmol) and
CuI (195 mg, 1.0 mmol) were added to the reaction mixture
and stirred for 1 h at 25 °C. NaOCD3 (prepared from Na,
400 mg, 17.4 mmol, in anhydrous CD3OD, 3 mL) was added to
Flash column chromatography (FCC): silica gel, grade 60,
230–400 μm. Organic extracts were dried over Na2SO4 and the
solvents were removed using a rotary evaporator.
NMR spectra were recorded on Bruker Avance 500 MHz
spectrometers. For 1H NMR (500 MHz) and 13C NMR
(125.8 MHz) spectra, the chemical shifts (δ) are reported in
parts per million (ppm) relative to TMS; spectra were cali-
brated using the solvent peak; spin coupling constants (J) are
reported to the nearest 0.5 Hz. Fourier transform infrared
(FT-IR) data were recorded on a spectrometer and spectra were
measured by the diffuse reflectance method on samples dis-
persed in KBr. MS [high resolution (HR), electron impact (EI)]
were obtained on a VG 70 SE mass spectrometer employing a
solids probe.
HPLC analyses and calculation of deuterium incorporation
HPLC-DAD-ESI-MS analysis was carried out with an Agilent the reaction mixture and refluxed at 110 °C for 1 h. The reac-
1100 series HPLC system equipped with an autosampler, tion mixture was acidified with HCl (1 M, pH 3), filtered
binary pump, degasser, and a diode array detector connected through celite and extracted with EtOAc. The organic layer was
directly to a mass detector (Agilent G2440A MSD-Trap-XCT ion dried and concentrated. The crude product was subjected
trap mass spectrometer) with an electrospray ionization (ESI) to
FCC
(CH2Cl2–MeOH–AcOH,
80 : 20 : 1)
to
yield
source. Chromatographic separations were carried out at room [2H3CO,5′,6′,7′-2H3]-4-methoxyindolyl-3′-oxo acid (0.090 mmol,
temperature using an Eclipse XDB-C-18 column (5 μm particle 22 mg, 37%). A solution of NH2OH·HCl (13 mg, 0.19 mmol)
size silica, 150 mm × 4.6 mm I.D.). The mobile phase con- and Na2CO3 (10 mg, 0.09 mmol) in water (1 mL) was added to
sisted of a linear gradient of: Method A (used with non-polar a solution of [2H3CO,5′,6′,7′-2H3]-4-methoxyindolyl-3′-oxo acid
extracts), in H2O (with 0.2% HCO2H)–CH3CN (with 0.2% (21 mg, 0.090 mmol) in EtOH (2 mL). The reaction mixture
HCO2H) from 75 : 25 to 25 : 75 in 25 min, to 0 : 100 in 5 min was stirred for 3 h at 70 °C, EtOH was removed under reduced
and a flow rate of 1.0 mL min−1; method B (used with polar pressure and the residue was washed with CH2Cl2. The residue
1160 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
was dissolved in water (10 mL), acidified with HCl (1 M, pH 3)
A solution of NH2OH·HCl (15 mg, 0.22 mmol) and Na2CO3
and extracted with EtOAc. The organic extract was dried and (13 mg, 0.12 mmol) in water (1 mL) was added to a solution of
concentrated to afford crude [2H3CO,5′,6′,7′-2H3]-4-methoxyin- [2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-carboxaldehyde (24a,
dolyl-3′-oximino acid quantitatively. NiCl2·6H2O (11 mg, 20 mg, 0.11 mmol) in MeOH (2 mL) at 60 °C. After 3 h, the
0.05 mmol) in EtOH (0.5 mL) was added to a solution of solvent was removed, the residue was dissolved in water,
[2H3CO,5′,6′,7′-2H3]-4-methoxyindolyl-3′-oximino acid (11 mg, extracted with EtOAc, the organic phase was dried, concen-
0.05 mmol) in EtOH (1 mL) followed by addition of NaBH4 trated and subjected to FCC (CH2Cl2–MeOH, 98 : 2) to yield
(5.2 mg, 0.12 mmol) over a 15 min period. The reaction [2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-carboxaldehyde oxime
mixture was stirred at rt for 48 h, acidified (HCl, 1 M) and the (20a) (17.6 mg, 0.090 mmol, 82%). The corresponding non-
solvent was removed. The residue was extracted with MeOH deuterated compound 20 was prepared similarly but using
and the extract was concentrated to yield [2H3CO,5′,6′,7′-2H3]-4- non-deuterated materials.
methoxyindolyl-3′-glycine (19a) quantitatively (21 mg,
0.090 mmol). The corresponding non-deuterated compound H2O) λmax (nm): 220, 240, 295. FTIR (KBr, cm−1) νmax: 3260,
19 was prepared similarly but using non-deuterated materials.
2932, 1620, 1379, 1267, 1100, 1026. 1H NMR (500 MHz CD3CN)
Compound 20. HPLC tR = 7.9 min, method A. UV (CH3CN–
Compound 19. HPLC tR = 6.8 min, method B. UV (CH3CN– δ 9.68 (1H, brs), 8.23 (1H, brs), 8.22 (1H, s), 8.21 (1H, d, J =
H2O) λmax (nm): 220, 265. FTIR (KBr, cm−1) νmax 3181, 2928, 3 Hz), 7.13 (1H, dd, J = 8, 8 Hz), 7.08 (1H, d, J = 8 Hz), 6.64
1624, 1504, 1466, 1362, 1246, 1088, 731. 1H NMR (500 MHz, (1H, d, J = 7.5 Hz), 3.94 (3H, s). 13C NMR δ (125.8 MHz,
CD3OD) δ 7.11 (1H, s), 6.97 (1H, dd, J = 8, 8 Hz), 6.91 (1H, d, J CD3CN) 155.5, 146.7, 142.2, 137.7, 131.1, 124.3, 108.0, 106.0,
= 8 Hz), 6.46 (1H, d, J = 7.5 Hz), 5.03 (1H, s), 3.83 (3H, s). 13C 102.0, 55.9. HR-EI-MS m/z 190.0741 (M+), calcd for C10H10N2O2
NMR (125.8 MHz, CD3OD) δ 173.9, 154.5, 139.9, 125.7, 124.2, 190.0742 (32%), 172.06 (100%), 157.04 (65%), 129.05 (72%).
117.4, 109.9, 106.6, 100.9, 55.9, 53.9. HR-ESI-MS m/z: [M + H − HPLC-ESI-MS m/z [M + H]+ 191.0 (100%), 174.0 (11%).
NH2]+ 204.0662, calcd for C11H10NO3 204.0665. HPLC-ESI-MS
m/z [M − H]− 219.2 (41%), 202.2 (24%), 176.2 (29%), 158.2 (CH3CN–H2O) λmax (nm): 220, 240, 295. 1H NMR (500 MHz
Compound 20a. HPLC-DAD, tR = 7.9 min, method A. UV
(100%).
CD3CN) δ 9.68 (1H, brs), 8.23 (1H, s), 8.21 (1H, s). HR-EI-MS
2
Compound 19a. HPLC tR = 6.8 min, method B. UV (CH3CN– m/z 196.1124 (M+), calcd for C10H4 H6N2O2 196.1119 (88%),
1
H2O) λmax (nm): 220, 265. H NMR (500 MHz, CD3OD) δ 7.43 178.10 (100%), 160.05 (62%), 132.06 (44%). HPLC-ESI-MS m/z:
(1H, s), 5.32 (1H, s). HR-ESI-MS m/z: 210.1033 [M − NH2]+, [M + H]+ 197.0 (100%), 180.0 (12%).
2
calcd for C11H4 H6NO3 210.1031. HPLC-ESI-MS m/z: [M − H]−
224.9 (34%), 207.9 (47%), 180.9 (38%), 162.9 (100).
[2H3CO]-4′-Methoxyindole-3′-carboxaldehyde (24b)
A solution of TTFA (280 mg, 0.51 mmol) in TFA (2 mL) was
[2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-carboxaldehyde oxime
added to indole-3′-carboxaldehyde (50 mg, 0.34 mmol).23 The
(20a) and [2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-
mixture was stirred for 3 h at 30 °C and the solvent was
removed under reduced pressure. Iodine (260 mg, 1.02 mmol),
carboxaldehyde (24a)
Freshly distilled POCl3 (96 μL, 1.02 mmol) was added dropwise CuI (260 mg, 1.36 mmol) and DMF (2 mL) were added to the
to a solution of [4,5,6,7-2H4]indole (83 mg, 0.69 mmol) in reaction mixture and the mixture was stirred for 1 h at 25 °C.
anhydrous DMF (1 mL) at 0 °C. The reaction mixture was NaOCD3 (prepared from Na 400 mg, 17.4 mmol in CD3OD,
stirred for 1 h at rt, basified with aq. NH3 (28%), diluted with 3 mL) was added and the reaction mixture was further stirred
water and extracted with EtOAc. The organic extract was dried at 110 °C for 1 h. The reaction mixture was cooled, neutralized
and concentrated to yield [4′,5′,6′,7′-2H4]indole-3′-carboxalde- (HCl, 1 M), filtered through celite, the filtrate was diluted with
hyde (98 mg, 0.66 mmol, 95%). A solution of TTFA (260 mg, water and extracted with EtOAc. The organic layer was dried,
0.49 mmol) in TFA (2 mL) was added to [4′,5′,6′,7′-2H4]indole- concentrated and the crude product was subjected to FCC
3′-carboxaldehyde (49 mg, 0.33 mmol), the mixture was stirred (EtOAc–Hex, 3 : 7) to yield [2H3CO]-4′-methoxyindole-3′-carbox-
overnight at 30 °C, and the solvent was removed under aldehyde (47 mg, 0.26 mmol, 78%). The corresponding non-
reduced pressure. DMF (2 mL), iodine (260 mg, 1.02 mmol) deuterated compound 24 was prepared similarly but using
and CuI (260 mg, 1.36 mmol) were added to the residue and non-deuterated materials.
the reaction mixture was stirred for 2 h at 25 °C. NaOCD3 (pre-
Compound 24. HPLC tR = 7.3 min, method A. UV (CH3CN–
pared from Na 400 mg, 17.4 mmol in CD3OD 3 mL) was added H2O) λmax (nm): 210, 250, 325. FTIR (KBr, cm−1) νmax: 3194,
to the reaction mixture and the mixture was further stirred at 1718, 1644, 1624, 1512, 1383, 1271, 1100, 781. 1H NMR
110 °C for 1 h. The reaction mixture was cooled to room temp- (500 MHz, CD3CN) δ 10.41 (1H, s), 10.2 (1H, brs), 7.90 (1H, s),
erature, neutralized (HCl, 1 M) and filtered through celite. The 7.19 (1H, dd, J = 8, 7.5 Hz), 7.14 (1H, d, J = 8 Hz), 6.75 (1H, d,
filtrate was diluted with water and extracted with EtOAc. The J = 7.5 Hz), 3.95 (3H, s). 13C NMR δ (125.8 MHz, CD3CN) 188.2,
organic layer was dried, concentrated and the crude product 155.4, 139.1, 129.8, 125.0, 119.9, 116.9, 106.6, 103.4, 56.1.
was subjected to FCC (EtOAc–Hex, 3 : 7) to afford HR-EI-MS m/z 175.0629 (M+), calcd for C10H9NO2 175.0633
[2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-carboxaldehyde (24a, (100%), 160.04 (28%), 144.04 (26%), 129.06 (43%). HPLC-
27 mg, 0.09 mmol, 45%).
ESI-MS m/z [M − H]− 173.9 (100%), 158.9 (98%).
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1161

View Article Online
Paper
Organic & Biomolecular Chemistry
Compound 24b. HPLC tR = 7.3 min, method A. UV (CH3CN– (1H, d, J = 10 Hz), 3.87 (3H, s), 3.45–3.23 (4H, m), 2.99 (1H, dd,
H2O) λmax (nm): 210, 250, 325. 1H NMR (500 MHz, CD3Cl) δ J = 9, 9 Hz), 2.24–2.21 (1H, m). 13C NMR (125.8 MHz, CD3OD) δ
10.51 (1H, s), 8.95 (1H, brs), 7.93 (1H, d, J = 3 Hz), 7.22 (1H, 155.3, 153.2, 139.0, 127.0, 124.3, 118.7, 107.7, 106.2, 102.1,
dd, J = 8, 8 Hz), 7.09 (1H, d, J = 8 Hz), 6.72 (1H, d, J = 8 Hz). 85.3, 81.6, 79.8, 73.7, 70.7, 61.9, 56.2. HR-EI-MS m/z 423.0632
2
HR-EI-MS m/z 178.0819 (M+), calcd for C10H6 H3NO2 178.0821 [M + K]+, calcd for C16H20N2O7SK 423.0622 (100%). HPLC-
(100%), 145.05 (21%), 131.07 (32%). HR-EI-MS m/z 175.0629 ESI-MS m/z [M + H]+ 385.0 (100%), 223.0 (38%).
(M+), calcd for C10H9NO2 175.0633 (100%), 160.04 (28%),
Compound 27a. HPLC tR = 9.3 min, method B. UV (CH3CN–
H2O) λmax (nm): 220, 270. H NMR (500 MHz, CD3OD) δ 7.28
1
144.0. HPLC-ESI-MS m/z: [M − H]− 176.9 (92%), 158.9 (100%).
(1H, s), 4.20 (1H, d, J = 10 Hz), 3.43–3.23 (4H, m), 2.98 (1H, dd,
J = 9, 9 Hz), 2.24–2.21 (1H, m). HR-MS-ESI m/z 429.1009 [M +
[2H3CO,5′,6′,7′-2H3]Desulfoglucorapassicin (27a)
(t-Boc)2O (23 mg, 0.19 mmol) and DMAP (2 mg, 0.02 mmol) K]+, calcd for C16H142H6N2O7SK 429.1004 (100%), 391.1440
were added to a solution of [2H3CO,5′,6′,7′-2H3]-4′-methoxy- [M + 1]+ (82%), 229.0 (17%), 201 (36%). HPLC-ESI-MS m/z
indole-3′-carboxaldehyde (40 mg, 0.22 mmol) in THF (3 mL) at [M + H]+ 391.0 (100%), 229.0 (38%).
0 °C. The reaction mixture was stirred at rt for 2 h, was neutral-
(R,S)-[4′,5′,6′,7′-2H4]Indolyl-3′-glycine (29a)
ized (HCl, 1 M), diluted with water and extracted with CH2Cl2.
The organic phase was dried and concentrated to yield crude Oxalyl chloride (52 μL, 0.6 mmol) was added to a solution of
N-t-Boc[2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-carboxaldehyde [4,5,6,7-2H4]indole (60 mg, 0.50 mmol) in anhydrous Et2O
(63 mg, 0.22 mmol). A solution of NH2OH·HCl (31 mg, (2 mL) at 0 °C, and the reaction mixture stirred for 1 h at 0 °C.
0.44 mmol) and Na2CO3 (24 mg, 0.22 mmol) in water (1 mL) The reaction mixture was filtered and the residue was washed
was added to a solution of t-Boc[2H3CO,5′,6′,7′-2H3]-4′-methoxy- with ice cold ether and concentrated under reduced pressure
indole-3′-carboxaldehyde (63 mg, 0.22 mmol) in MeOH (4 mL) to yield crude [4′,5′,6′,7′-2H4]indolyl-3′-oxalylchloride in quanti-
at 60 °C. After 3 h the solvent was removed, the residue was tative yield. NaOH (20%, w/v) was added to a suspension of
dissolved in water and extracted with CH2Cl2. The organic [4′,5′,6′,7′-2H4]indolyl-3′-oxalylchloride (105 mg, 0.50 mmol) in
phase was dried, and concentrated to yield crude N-t-Boc- THF to obtain pH 10. The reaction mixture was stirred at room
[2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′-carboxaldehyde oxime temperature for 30 min, acidified (HCl, 2 M) and further
(65 mg, 0.21 mmol, 95%). NCS (16 mg, 0.12 mmol) was added stirred for 30 min. The reaction mixture was diluted with
to a solution of N-t-Boc[2H3CO,5′,6′,7′-2H3]-4′-methoxyindole-3′- water, extracted with EtOAc, the organic layer was dried and
carboxaldehyde oxime (35 mg, 0.12 mmol) in CH2Cl2 (1 mL) concentrated to yield [4′,5′,6′,7′-2H4]indolyl-3′-oxo acid (72 mg,
and pyridine (100 μL) at 0 °C and the reaction mixture was 0.37 mmol, 75% over two steps). A solution of NH2OH·HCl
stirred for 30 min at rt. 1-Thio-β-D-glucopyranose-2,3,4,6-tetra- (108 mg, 1.55 mmol) and NaOAc (211 mg, 1.55 mmol) in water
acetate (26 mg, 0.060 mmol) in CH2Cl2 (200 μL) and Et3N (1 mL) was added to a solution of [4′,5′,6′,7′-2H4]indolyl-3′-oxo
(50 μL, 0.36 mmol) were added to the reaction mixture and the acid (50 mg, 0.26 mmol) in EtOH (2 mL). The reaction mixture
mixture was stirred for 3 h at rt. The reaction mixture was was stirred for 3 hours at 70 °C, EtOH was removed under
diluted with water, extracted with CH2Cl2, the organic layer reduced pressure and the residue was washed with CH2Cl2.
was dried, concentrated and the residue was subjected to FCC The residue was diluted with water (10 mL), acidified (HCl,
(EtOAc–Hex, 1 : 1) to yield t-Boc[2H3CO,5′,6′,7′-2H3]desulfoglu- 1 M) and extracted with EtOAc. The organic extract was dried
corapassicin tetraacetate (41 mg, 0.060 mmol, 52%). t-Boc and concentrated to yield crude [4′,5′,6′,7′-2H4]indolyl-3′-
[2H3CO,5′,6′,7′-2H3]desulfoglucorapassicin tetraacetate (41 mg, oximino acid, which was taken to the next step without further
0.06 mmol) was dissolved in CH2Cl2 (3.7 mL) and D-TFA purification. A solution of NiCl2·6H2O (61 mg, 0.26 mmol) in
(750 μL) and stirred at rt for 3 h. The reaction mixture was EtOH (1 mL) was added to a solution of [4′,5′,6′,7′-2H4]indole-
diluted with water and extracted with CH2Cl2. The organic 3′-oximino acid (54 mg, 0.26 mmol) in EtOH (3 mL), followed
layer was dried, concentrated and subjected to FCC (EtOAc– by addition of NaBH4 (29.6 mg, 0.78 mmol) over a 20 min
Hex, 1 : 1) to yield [2H3CO,5′,6′,7′-2H3]desulfoglucorapassicin period. The reaction mixture was stirred at rt for 48 h, was
tetraacetate (31 mg, 0.05, 92%). Freshly prepared KOCH3 acidified with HCl (1 M, pH 3), the solvent was removed under
(1 mM, 75 μL) was added to the solution of [2H3CO,5′,6′,7′-2H3] reduced pressure, the residue was extracted with MeOH and
desulfoglucorapassicin tetraacetate (31 mg, 0.06 mmol) in the extract was concentrated to yield [4′,5′,6′,7′-2H4]indolyl-3′-
anhydrous MeOH (1 mL) and stirred at rt for 30 min. The reac- glycine (29a) (50 mg, 0.26 mmol) quantitatively. The corre-
tion mixture was neutralized with acetic acid and the solvent was sponding non-deuterated compound 29 was prepared similarly
evaporated under reduced pressure to yield [2H3CO,5′,6′,7′-2H3]- but using non-deuterated materials.
desulfoglucorapassicin (27a) (33 mg, 0.08 mmol) quantitatively.
The corresponding non-deuterated compound 27 was prepared H2O) λmax (nm): 220, 275. FTIR (KBr, cm−1) νmax: 3389, 3000,
similarly but using non-deuterated materials.
1737, 1488, 1454, 1426, 1193, 1105, 753. 1H NMR (500 MHz,
Compound 29. HPLC tR = 6.9 min, method B. UV (CH3CN–
Compound 27. HPLC tR = 9.3 min, method B. UV (CH3CN– D2O) δ 7.68 (1H, d, J = 8 Hz), 7.59 (1H, s), 7.57 (1H, d, J =
H2O) λmax (nm): 220, 270. FTIR (KBr, cm−1) νmax: 3160, 1667, 8.5 Hz), 7.31 (1H, dd, J = 8, 8 Hz), 7.23 (1H, dd, J = 8, 8 Hz),
1
1405. H NMR (500 MHz, CD3OD) δ 7.32 (1H, s), 7.08 (1H, dd, 5.51 (1H, s). 13C NMR (125.8 MHz, D2O) δ 171.7, 136.2, 127.0,
J = 8, 8 Hz), 7.02 (1H, d, J = 8 Hz), 6.56 (1H, d, J = 7.5 Hz), 4.19 124.4, 122.7, 120.4, 117.9, 112.4, 105.0, 49.7. HR-ESI-MS m/z
1162 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
174.0556 [M − NH2]+, calcd for C10H8NO2 174.0555 (100%). C15H142H4N2O6SK 397.0768 (28%), 224.11 (100%). ESI-MS m/z:
HPLC-ESI-MS m/z [M − H]− 189.1 (100%), 146.2 (16%).
[M + 1]+ 197.1 (100%), 359.1 (96%).
Compound 29a. HPLC tR = 6.9 min, method B. UV (CH3CN–
1
[4′,5′,6′,7′-2H4]Indole-3′-carboxaldehyde oxime (31a)
H2O) λmax (nm): 220, 275. H NMR (500 MHz, CD3OD) δ 7.58
(1H, s), 5.50 (1H, s). HR-ESI-MS m/z 178.0796 [M − NH2]+, Freshly distilled POCl3 (96 μL, 1.02 mmol) was added dropwise
2
calcd for C10H4 H4NO2 178.0800 (100%). HPLC-ESI-MS m/z: to a solution of [4,5,6,7-2H4]indole (83 mg, 0.69 mmol) in
[M − H]− 193.1 (100%).
anhydrous DMF (1 mL) at 0 °C. The reaction mixture was
stirred for 1 h at room temperature, basified with NH3 (28%),
diluted with water and extracted with EtOAc. The organic
[4′,5′,6′,7′-2H4]Indole desulfoglucosinolate (30a)
To
a
solution of [4′,5′,6′,7′-2H4]indole-3′-carboxaldehyde extract was dried over Na2SO4 and concentrated to yield
(39 mg, 0.26 mmol) in EtOH (2 mL), a solution of Na2CO3 [4′,5′,6′,7′-2H4]indole-3′-carboxaldehyde (98 mg, 0.54 mmol,
(28 mg, 0.26 mmol) and NH2OH·HCl (36 mg, 0.52 mmol) in 95%). A solution of NH2OH·HCl (69.5 mg, 1.0 mmol) and
water (1 mL) was added and stirred at 70 °C for 3 h. EtOH was Na2CO3 (53 mg, 0.5 mmol) in water (1 mL) was added to a
removed, the reaction mixture was diluted with water and solution of [4′,5′,6′,7′-2H4]indole-3′-carboxaldehyde (75 mg,
extracted with EtOAc. The organic layer was dried and concen- 0.5 mmol) in EtOH (2 mL) at 60 °C. After 3 h the solvent was
trated under reduced pressure to yield the corresponding removed, the residue was dissolved in water and extracted with
[4′,5′,6′,7′-2H4]indole-3′-carboxaldoxime quantitatively. NCS EtOAc. The organic phase was dried, concentrated and sub-
(27 mg, 0.20 mmol) was added to a solution of [4′,5′,6′,7′-2H4]- jected to FCC (CH2Cl2–MeOH, 98 : 2) to yield [4′,5′,6′,7′-2H4]-
indole-3′-carboxaldoxime (28 mg, 0.17 mmol) in pyridine indole-3′-carboxaldoxime (31a) (76 mg, 0.46 mmol, 92%).
(100 μL) and CH2Cl2 (1 mL) at 0 °C. The reaction mixture was
Compound 31a. HPLC-DAD, tR = 6.3 min, method A. UV
1
stirred for 30 min at rt, 1-thio-β-D-glucopyranose-2,3,4,6-tetra- (CH3CN–H2O) λmax (nm): 230, 260. H NMR (500 MHz CD3CN)
acetate (37 mg, 0.10 mmol) in CH2Cl2 (0.2 mL) was added, and δ 9.76 (1H, brs), 8.25 (1H, d, J = 2.5 Hz), 7.73 (1H, s). HR-EI-MS
2
then Et3N (71 μL, 0.51 mmol) and the reaction mixture was m/z 164.0885 (M+), calcd for C9H4 H4N2O 164.0887 (20%),
stirred for 3 h at rt. The reaction mixture was diluted with 147.08 (100%), 119.07 (35%). HPLC-ESI-MS m/z [M + H]+ 165.1
water (5 mL), extracted with CH2Cl2, the organic layer was (100%), 148.1 (48%).
dried and concentrated. The residue was subjected to FCC
[2H3CO]-4-Methoxyindole (32a)
(EtOAc–Hex, 1 : 1) to yield the tetraacetyl derivative (20 mg,
0.04 mmol, 22% from oxime). KOCH3 (75 μL, 1 mM) was A solution of 3-nitrophenol (100 mg, 0.70 mmol) in THF
added to the solution of the tetraacetyl derivative (20 mg, (1 mL) was added to a suspension of NaH (24 mg, 1 mmol) in
0.04 mmol) in anhydrous MeOH (1 mL) to obtain pH 9. The THF (2 mL). The reaction mixture was stirred at room tempera-
2
reaction mixture was stirred for 30 min and neutralized with ture for 10 min, H3CI (90 μL, 1.4 mmol) was added and the
acetic acid. The solvent was evaporated, the residue was dis- reaction mixture was stirred at 65 °C for 5 hours. The reaction
solved in water and freeze dried to yield [4′,5′,6′,7′-2H4]indole mixture was diluted with water, extracted with CH2Cl2, the
desulfoglucosinolate (30a) in quantitative yield (22 mg, organic layer was dried and concentrated to yield crude
0.06 mmol). The corresponding non-deuterated compound 30 3-[2H3C-O]nitrobenzene (112 mg, 0.73 mmol) in quantitative
was prepared similarly but using non-deuterated materials. yield.
A
solution of 3-[2H3C-O]nitrobenzene (110 mg,
Compound 30 was previously synthesized to be used as 0.70 mmol) and 4-chlorophenyl acetonitrile (127 mg,
internal standards in seed meals, but only the 1H NMR data 0.75 mmol) in DMF (2 mL) was slowly added to a suspension
were reported.37
of t-BuOK (300 mg, 2.67 mmol) in DMF (1.5 mL) at −20 °C.
Compound 30. HPLC tR = 6.1 min, method B. UV (CH3CN– The reaction mixture was stirred at −20 °C for 2 h, quenched
H2O) λmax (nm): 270. FTIR (KBr, cm−1) νmax: 3270, 2932, 1565, with ice cold HCl (1 M) and extracted with EtOAc. The organic
1
1413, 1105, 1048, 747. H NMR (500 MHz, CD3OD) δ 7.83 (1H, extract was washed with NaOH (2 M), dried and concentrated.
d, J = 8 Hz), 7.70 (1H, s), 7.39 (1H, d, J = 8 Hz), 7.15 (1H, dd, J = The crude reaction mixture was subjected to FCC (EtOAc–Hex
7, 8 Hz), 7.08 (1H, dd, J = 7, 8 Hz), 4.52 (1H, d, J = 10 Hz), 3.66 1 : 3) to yield (6-[2H3C-O]-2-nitrophenyl)acetonitrile (37 mg,
(1H, dd, J = 12, 2 Hz), 3.54 (1H, dd, J = 12, 5.5 Hz), 3.29–3.25 0.19 mmol, 26%). Palladium on carbon (10% Pd/C, 10 mg) was
(2H, m), 3.13 (1H, dd, 9, 9), 2.74 (1H, m). 13C NMR δ added to the solution of (6-[2H3CO]-2-nitrophenyl)acetonitrile
(125.8 MHz, CD3OD) 149.3, 137.9, 128.7, 127.6, 123.4, 121.6, (50 mg, 0.25 mmol) in MeOH (2.5 mL) and AcOH (244 μL) and
121.3, 112.6, 110.4, 86.1, 82.0, 79.6, 74.4, 71.1, 62.5. the resulting reaction mixture was stirred under a H2 atmos-
HR-ESI-MS m/z 393.0572 [M + K]+, calcd for C15H19N2O6SK phere (balloon pressure) at rt. After 12 h, the catalyst was fil-
393.0517 (16%). HPLC-ESI-MS m/z [M + 1]+ 355.1 (96%), 193.1 tered off, and the filtrate was concentrated under reduced
(100%).
pressure. The residue was suspended in sat. NaHCO3 (5 mL)
Compound 30a. HPLC tR = 6.1 min, method B. UV (CH3CN– and extracted with CH2Cl2. The organic layer was washed with
1
H2O) λmax (nm): 270. H NMR δ (500 MHz, CD3OD) 7.71 (1H, water, dried and concentrated under reduced pressure to yield
s), 4.54 (1H, d, J = 10 Hz), 3.66 (1H, dd, J = 12, 2 Hz), 3.56 (1H, 4-[2H3C-O]indole (32a, 30.4 mg, 0.20 mmol, 81%). Compound
dd, J = 12, 5 Hz), 3.28–3.15 (2H, m), 3.15 (1H, dd, 9, 9), 2.77 32 was synthesized similarly but using non-deuterated
(1H, m). HR-ESI-MS m/z 397.0769 [M
+
K]+, calcd for materials.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1163

View Article Online
Paper
Organic & Biomolecular Chemistry
Compound 32. HPLC tR = 12.5 min, method A. UV (CH3CN– 171.05 (92%). HPLC-ESI-MS m/z: [M + H]+ 190.2 (100%), 163.2
H2O) λmax (nm): 220, 263. Other data as previously reported.38
Compound 32a. HPLC tR = 12.5 min, method A. UV
(48%), 150.3 (89%).
1
[4′,5′,6′,7′-2H4]Desulfoglucobrassicin (34a)
(CH3CN–H2O) λmax (nm): 220, 263. H NMR (500 MHz, CD3Cl)
δ 8.16 (1H, brs), 7.13 (1H, dd, J = 8, 7.5 Hz), 7.13 (1H, s), 7.03 NH4OAc (12.9 mg, 0.16 mmol) was added to a solution of
(1H, d, J = 8 Hz), 6.67 (1H, broad s), 6.53 (1H, d, J = 7.5 Hz). [4′,5′,6′,7′-2H4]indole-3′-carboxaldehyde
in
nitromethane
2
HR-EI-MS m/z 150.0872 (M+), calcd for C9H6 H3NO 150.0872 (1 mL) and refluxed at 105 °C for 90 min. The reaction mixture
(100%), 132.04 (73%), 104.04 (36%). HPLC-ESI-MS m/z: was diluted with water, extracted with CH2Cl2 and concen-
[M + H]+ 151.1 (100%), 133.0 (44%).
trated to yield crude [4,5,6,7-2H4]-3-(2′-nitrovinyl)indole, which
was taken to the next step without further purification. NaBH4
(38 mg, 0.96 mmol) was added to a solution of crude
[2H3CO]-4′-Methoxyindolyl-3′-acetonitrile (33a) (arvelexin)
A solution of TTFA (280 mg, 0.51 mmol) in TFA (2 mL) was [4,5,6,7-2H4]-3-(2′-nitrovinyl)indole (61.6 mg, 0.32 mmol) in
added to indole-3′-carboxaldehyde (50 mg, 0.34 mmol). The THF (2 mL) and MeOH (200 μL). The reaction mixture was
mixture was stirred for 3 h at 30 °C. Solvent was removed stirred for 3 h at rt and excess NaBH4 was destroyed by adding
under reduced pressure. Iodine (260 mg, 1.02 mmol), CuI water to the reaction mixture. The reaction mixture was
(260 mg, 1.36 mmol) and DMF (2 mL) were added and stirred extracted with CH2Cl2, the solvent was removed and the
for 1 h at 25 °C. NaOCD3 (prepared from Na 400 mg, 17.4 mg residue subjected to FCC (100% CH2Cl2) to yield [4,5,6,7-2H4]-
in CD3OD 3 mL) was added and the reaction mixture was 3-(2′-nitroethyl)indole (23 mg, 0.12 mmol, 38%). NaOCH3
further stirred at 110 °C for 1 h. The reaction mixture was (23 mg mL−1, 1 mL) was added to a solution of [4,5,6,7-2H4]-3-
cooled, neutralized (HCl, 1 M), filtered through celite and the (2′-nitroethyl)indole (23 mg, 0.12 mmol) in MeOH (1 mL), the
filtrate was diluted with water and extracted with EtOAc. The reaction mixture was stirred for 30 min at rt and the solvent
organic layer was dried, concentrated and the crude product was removed. The residue was suspended in DME (1 mL) at
was subjected to FCC (EtOAc–Hex, 3 : 7) to afford [2H3CO]-4′- −40 °C, thionyl chloride (35 μL, 0.29 mmol) in DME (0.5 mL)
methoxyindole-3′-carboxaldehyde (47 mg, 0.26 mmol, 79%). was added dropwise and the reaction mixture was stirred for
NH4OAc (14 mg, 0.18 mmol) was added to a solution 1 h at −40 °C. The solvent was removed, the residue was
of [2H3CO]-4′-methoxyindole-3′-carboxaldehyde (65 mg, diluted with water, extracted with CH2Cl2 and concentrated to
0.36 mmol) in nitromethane (1 mL) and refluxed at 105 °C for yield crude oximoyl chloride, which was taken to the next step
90 min. The reaction mixture was allowed to cool, diluted with without further purification. A solution of 1-thio-β-D-glucopyra-
water, extracted with CH2Cl2 and concentrated to yield crude nose tetraacetate (22 mg, 0.09 mmol) in anhydrous CH2Cl2
4-[2H3C-O]-3-nitrovinylindole, which was taken to the next step (1 mL) and Et3N (50 μL, 0.36 mmol) in Et2O (1 mL) were suc-
without further purification. NaBH4 (42 mg, 1.1 mmol) was cessively added with stirring to a solution of crude oximoyl
added to a solution of crude [2H3CO]-4-methoxy-3-nitrovinylin- chloride in anhydrous Et2O–CH2Cl2 (2 : 1, 3 mL) at rt. After
dole (79 mg, 0.36 mmol) in THF (3 mL) and MeOH (300 μL). 3 h, the reaction mixture was diluted with water, extracted with
The reaction mixture was stirred for 3 h at rt, after which CH2Cl2 and concentrated. The residue was subjected to FCC
excess NaBH4 was destroyed by adding ice-cold water to the (CH2Cl2–MeOH, 98 : 2) to yield [4′,5′,6′,7′-2H4]tetraacetyl desul-
reaction mixture. The reaction mixture was extracted with foglucobrassicin (22 mg, 0.040 mmol, 33%). Freshly prepared
CH2Cl2, concentrated and subjected to FCC (CH2Cl2, 100%) KOCH3 (1 mM, 75 μL) was added to
a solution of
to yield [2H3CO]-4-methoxy-3-(2′-nitroethyl)indole (27 mg, [4′,5′,6′,7′-2H4]tetraacetyl desulfoglucobrassicin (22 mg,
0.12 mmol, 34%). Et3N (324 μL, 2.3 mmol) and CS2 (72 μL, 0.04 mmol) in anhydrous MeOH (1 mL) at rt under an inert
1.2 mmol) were added to a solution of [2H3CO]-4-methoxy-3- atmosphere, and the mixture was stirred for 30 min. The reac-
(2′-nitroethyl)indole (26 mg, 0.12 mmol) in acetonitrile tion mixture was neutralized with acetic acid and concen-
(1.5 mL) and the reaction mixture was stirred at 40 °C for 20 h. trated. The residue was dissolved in water and freeze-dried to
The solvent was removed, the residue was diluted with water yield [4′,5′,6′,7′-2H4]desulfoglucobrassicin (23 mg, 0.08 mmol)
(10 mL) and extracted with CH2Cl2. The organic layer was quantitatively. The corresponding non-deuterated compound
dried, concentrated and the residue was subjected to FCC 34 was prepared similarly but using non-deuterated materials.
(CH2Cl2, 100% to CH2Cl2–MeOH, 98 : 2) to yield [2H3CO]-4′-
Compound 34. HPLC tR = 11.9 min, method B. UV (CH3CN–
methoxyindolyl-3′-acetonitrile (33a, 14 mg, 0.07 mmol, 62%). H2O) λmax (nm): 222, 280. FTIR (KBr, cm−1) νmax: 3389, 2889,
The corresponding non-deuterated compound 33 was prepared 1456, 1422, 1340, 1048, 745. 1H NMR δ (500 MHz, CD3OD) 7.64
similarly but using non-deuterated materials.
Compound 33. HPLC tR = 13.7 min, method A. UV (CH3CN– dd, J = 8, 7.5 Hz), 7.01 (1H, dd, J = 7.5, 7.5 Hz), 4.68 (1H, d, J =
H2O) λmax (nm): 220, 260. Other data as previously reported.39
10 Hz), 4.25 (1H, d, J = 16 Hz), 3.94 (1H, d, J = 16 Hz), 3.80 (1H,
(1H, d, J = 8 Hz), 7.33 (1H, d, J = 8 Hz), 7.13 (1H, s), 7.09 (1H,
Compound 33a. HPLC tR = 13.7 min, method A. UV d, J = 12 Hz), 3.59 (1H, m), 3.23 (1H, dd, J = 9.5, 9 Hz), 3.14
1
(CH3CN–H2O) λmax (nm): 220, 260. H NMR (500 MHz, CD3Cl) (1H, dd, J = 9, 9 Hz), 3.09–3.03 (2H, m). 13C NMR δ
δ 8.16 (1H, brs), 7.13 (1H, dd, J = 8, 8 Hz), 7.10 (1H, s), 6.98 (125.8 MHz, CD3OD) 154.3, 138.2, 128.3, 124.1, 122.7, 120.0,
(1H, d, J = 8 Hz), 6.51 (1H, d, J = 8 Hz), 4.06 (2H, s). HR-EI-MS 119.7, 112.4, 111.7, 83.0, 82.1, 79.5, 74.6, 71.3, 62.8, 30.4.
m/z 189.0977 [M]+, calcd for C11H7 H3N2O 189.0981 (100%), HR-EI-MS m/z 156.0685 [M − (S-glc) − OH]+, calcd for C10H8N2
2
1164 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
156.0687 (72%), 130.06 (100%). HPLC-ESI-MS m/z [M + 1]+
368.9 (42%), 207.0 (100%), 174.0 (27%), 130.0 (40%).
6 A. V. Morant, K. Jorgensen, C. Jorgensen, S. M. Paquette,
R. Sanchez-Perez, B. L. Moller and S. Bak, Phytochemistry,
2008, 69, 1795–1813.
7 S. I. Warwick, Brassicaceae in agriculture, in Genetics and
Genomics of the Brassicaceae, Plant Genetics and Genomics:
Crops and Models, ed. R. Schmidt, Is. Bancroft, Springer
Science, 2011, vol. 9, p. 34.
8 M. Takasugi, N. Katsui and A. Shirata, J. Chem. Soc., Chem.
Commun., 1986, 1077–1078.
9 M. S. C. Pedras and A. M. Adio, Phytochemistry, 2008, 69,
889–893.
Compound 34a. HPLC tR = 11.9 min, method A. UV
(CH3CN–H2O) λmax (nm): 222, 280. 1H NMR δ (500 MHz,
CD3OD) 7.11 (1H, s), 4.68 (1H, d, J = 10 Hz), 4.23 (1H, d, 16),
3.92 (1H, dd, 16, 1), 3.78 (1H, dd, 12, 2), 3.59 (1H, dd, 12, 6),
3.22 (1H, dd, 9.5, 9.5), 3.13 (1H, dd, 9.5, 9), 3.06–3.03 (2H, m).
HR-EI-MS m/z 160.0931 [M − (S-glc) − OH]+, calcd for
C10H42H4N2 160.0938 (74%), 134.09 (100%). HPLC-ESI-MS m/z
[M + 1]+ 373.1 (41%), 211.1 (100%).
10 J. Tsuji, M. Zook, S. Somerville, R. L. Last and
R. Hammerschmidt, Physiol. Mol. Plant Pathol., 1993, 43,
221–229.
Preparation of rutabaga roots, feeding experiments and
extractions
Rutabaga (B. napus ssp. rapifera) root tubers were cut horizon-
tally in 10–15 mm slices, cylindrical wells (ca. 16 mm in dia-
meter) were made on one side of the slices with a cork-borer
and slices were incubated for 24 h in the dark (Pedras et al.,
2004). Compounds (5 × 10−4 M) dissolved in H2O, H2O–
CH3OH–Tween 80 (95 : 5 : 0.05, v/v), or H2O–CH3OH (90 : 10, v/
v) were added to each well (500 μL per well) and the slices were
further incubated. All feeding experiments with labeled com-
pounds were conducted in triplicate. Following adsorption of
the solution, the wells were filled with distilled water. After
further incubation for 48 h, the aqueous solutions in wells of
each slice were collected, extracted (EtOAc) and the solvent was
removed under reduced pressure. The residue obtained was
dissolved in CH3CN and was analyzed by HPLC-DAD-ESI-MS.
Control experiments were similarly carried out by incubating
rutabaga root tubers with non-labeled precursors or with
carrier solution only. The tissue around each well was cut, was
ground and extracted with CH3OH; the MeOH extracts were
concentrated and analyzed by HPLC-DAD-ESI-MS.
11 G. R. Hicks and N. V. Raikhel, Nat. Chem. Biol., 2009, 5,
268–272.
12 H.-J. Wu, Z. Zhang, J.-Y. Wang, D.-H. Oh, M. Dassanayake,
B. Liu, Q. Huang, H.-X. Sun, R. Xia, Y. Wu, Y.-N. Wang,
Z. Yang, Y. Liu, W. Zhang, H. Zhang, J. Chu, C. Yan,
S. Fang, J. Zhang, Y. Wang, F. Zhang, G. Wanga, S. Y. Leed,
J. M. Cheesemanc, B. Yangb, B. Lib, J. Minb, L. Yangb,
J. Wangb, C. Chua, S.-Y. Chen, H. J. Bohnert, J.-K. Zhu,
X.-J. Wang and Q. Xie, Proc. Natl. Acad. Sci. U. S. A., 2012,
109, 12219–12224.
13 N. Agerbirk and C. E. Olsen, Phytochemistry, 2012, 77,
16–45.
14 B. A. Halkier and J. Gershenzon, Annu. Rev. Plant Biol.,
2006, 57, 303–333.
15 I. E. Sønderby, F. Geu-Flores and B. A. Halkier, Trends Plant
Sci., 2010, 15, 283–290.
16 K. Monde, M. Takasugi and T. Ohnishi, J. Am. Chem. Soc.,
1994, 116, 6650–6657.
17 M. S. C. Pedras and D. P. O. Okinyo, Org. Biomol. Chem.,
2008, 6, 51–54.
18 M. S. C. Pedras, D. P. O. Okinyo, K. Thoms and A. M. Adio,
Phytochemistry, 2009, 70, 1129–1138.
19 M. S. C. Pedras, E. E. Yaya and S. Hossain, Org. Biomol.
Chem., 2010, 8, 5150–5158.
20 M. S. C. Pedras and E. E. Yaya, Org. Biomol. Chem., 2012,
10, 3613–3616.
21 M. S. C. Pedras, Q. A. Zheng and R. S. Gadagi, Chem.
Commun., 2007, 368–370.
Acknowledgements
Support for the authors’ work was obtained from the Natural
Sciences and Engineering Research Council of Canada (Discov-
ery Research Grant to M.S.C.P.) and the University of Saskatch-
ewan (graduate assistantship to E.E.Y.). We acknowledge the
technical assistance of K. Thoms and K. Brown, Department of
Chemistry, in HR-MS and NMR determinations, respectively.
22 T. Mahmud, J. Labelled Compd. Radiopharm., 2007, 50,
1039–1051.
23 M. Somei, F. Yamada, M. Kunimoto and C. Kaneko, Hetero-
cycles, 1984, 22, 797–801.
24 K. C. Liu, B. R. Shelton and R. K. Howe, J. Org. Chem.,
1980, 45, 3916–3917.
References
1 T. Hartmann, Phytochemistry, 2007, 68, 2831–2846.
2 R. A. Dixon, Nature, 2001, 411, 843–847.
25 M. Klein and J. Papenbrock, Physiol. Plant., 2009, 135(2),
140–149.
3 J. W. Mansfield and J. A. Bailey, in Phytoalexins, 1982, 26 M. S. Kharasch, S. S. Kane and H. C. Brown, J. Am. Chem.
pp. 319–323. Soc., 1940, 62, 2242–2243.
4 H. D. VanEtten, J. W. Mansfield, J. A. Bailey and 27 N. K. Garg, R. Sarpong and B. M. Stoltz, J. Am. Chem. Soc.,
E. E. Farmer, Plant Cell, 1994, 6, 1191–1192. 2002, 124, 13179–13184.
5 M. S. C. Pedras, E. E. Yaya and E. Glawischnig, Nat. Prod. 28 P. Friis, P. O. Larsen and C. E. Olsen, J. Chem. Soc., Perkin
Rep., 2011, 28, 1381–1405.
Trans. 1, 1977, 661–665.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1149–1166 | 1165

View Article Online
Paper
Organic & Biomolecular Chemistry
29 S. F. Brady, L. Simmons, J. H. Kim and E. W. Schmidt, Nat. 35 N. K. Clay, A. M. Adio, C. Denoux, G. Jander and
Prod. Rep., 2009, 26, 1488–1503. F. M. Ausubel, Science, 2009, 323, 95–101.
30 M. Pfalz, M. D. Mikkelsen, P. Bednarek, C. E. Olsen, 36 P. Bednarek, M. Pislewska-Bednarek, A. Svatos,
B. A. Halkier and J. Kroymann, Plant Cell, 2011, 23, 716–729.
31 M. S. C. Pedras and V. K. Sarma-Mamillapalle, Bioorg. Med.
Chem., 2012, 20, 3991–3996.
B. Schneider, J. Doubsky, M. Mansurova, M. Humphry,
C. Consonni, R. Panstruga and A. Sanchez-Vallet, et al.,
Science, 2009, 323, 101–106.
32 R. Mithen, R. Bennett and J. Marquez, Phytochemistry, 37 L. Brochard, B. Joseph, M. C. Viaud and P. Rollin, Synth.
2010, 71, 2074–2086. Commun., 1994, 24, 1403–1414.
33 E. Pichersky and D. R. Gang, Trends Plant Sci., 2000, 5, 38 D. Coowar, J. Bouissac, M. Hanbali, M. Paschaki, E. Mohier
439–445. and B. Luu, J. Med. Chem., 2004, 47, 6270–6282.
34 J.-K. Weng, Y. Li, H. Mo and C. Chapple, Science, 2012, 337, 39 M. S. C. Pedras, P. B. Chumala and M. Suchy, Phytochemis-
960–964.
try, 2003, 64, 949–956.
1166 | Org. Biomol. Chem., 2013, 11, 1149–1166
This journal is © The Royal Society of Chemistry 2013