Non-symmetrically substituted phenoxazinones from laccase-mediated oxidative cross-coupling of aminophenols: An experimental and theoretical insight

Article abstract of DOI:10.1039/c1ob05795b

Oxidative cross-coupling reactions of substituted o-aminophenols were catalyzed by a commercial laccase to produce non-symmetrically substituted phenoxazinones for the first time. Identification by ¹H-, ¹³C- and ³¹P-NMR, a

Full text of DOI:10.1039/c1ob05795b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 1834
www.rsc.org/obc
PAPER
Non-symmetrically substituted phenoxazinones from laccase-mediated
oxidative cross-coupling of aminophenols: an experimental and theoretical
insight†
Frédéric Bruyneel,a Georges Diveb and Jacqueline Marchand-Brynaert*a
Received 20th May 2011, Accepted 3rd November 2011
DOI: 10.1039/c1ob05795b
Oxidative cross-coupling reactions of substituted o-aminophenols were catalyzed by a commercial laccase
to produce non-symmetrically substituted phenoxazinones for the first time. Identification by 1H-,
13C- and 31P-NMR, and by HPLC-PDA and HPLC-MS/MS of exclusively two kinds of substituted
phenoxazinones out of four potential heterocyclic frameworks was confirmed by a DFT study. The redox-
properties of the substrates, their relative rates of conversion and the rigid docking of selected substrates
led to a revisited mechanistic pathway for phenoxazinones biosynthesis. Our suggestions concern both the
first formal two-electron oxidation by laccase and the first intermolecular 1,4-conjugated addition which
secures the observed regioselectivity.
cycloheptenes were also recently synthesized.13 The main advan-
tage of those examples is the production by the laccase of an oxi-
Introduction
The phenoxazinone framework containing a tricyclic iminoqui-
none core structure is widely distributed in natural products dis-
playing remarkable biological activities and redox properties.1–3
Along with some of the most potent antineoplastic agents (acti-
nomycins), new antibiotics featuring a phenoxazinone structure
are still regularly isolated from numerous fungal, bacterial or
invertebrate sources (Fig. 1).4–7 Notwithstanding the wide pool
of natural organisms producing these heterocycles, a common
biosynthetic pathway has emerged which involves enzymes from
the class of multicopper oxidases (MCO).2a,5–10 Among them,
an increasing focus has been directed towards laccases [EC
1.10.3.2] because they are valuable biocatalysts in organic syn-
thesis and green chemical processes.11 Recent examples involve
the laccase-mediated synthesis of cinnabarinic acid analogues,
featuring phosphonylated or sulfonylated groups, as novel dyes
and fluorophores.9a,12 Some compounds endowed with tunable
water-solubility were successfully used in a live-cell imaging
application.12d Other heterocyclic frameworks displaying bio-
logical activities such as annulated benzofurans, phenazines and
dized electrophile on which any less oxidizable nucleophile can
aInstitute of Condensed Matter and Nanosciences (IMCN), Organic and
Medicinal Chemistry (CHOM), Université catholique de Louvain,
Bâtiment Lavoisier, Place Louis Pasteur L4.01.02, 1348, Louvain-la-
Neuve, Belgium. E-mail: jacqueline.marchand@uclouvain.be;
Fax: +32(10)474168
bCentre d’ingénierie des Protéines, Université de Liège, Bâtiment B6,
Allée du 6 Août, 4000, Sart-Tilman (Liège), Belgium
†Electronic supplementary information (ESI) available: voltammograms,
NMR spectra of isolated compounds, HPLC-PDA, MS, MS2 spectra and
identification description, docking results. See DOI: 10.1039/
c1ob05795b
Fig. 1 Structures of natural products featuring the 2-aminophenoxazi-
none framework.
1834 | Org. Biomol. Chem., 2012, 10, 1834–1846
This journal is © The Royal Society of Chemistry 2012

View Article Online
further react. Therefore the chemoselectivity of such catalytic
processes is secured in the first step and this ensures a good
yield of the final product. In the case of phenoxazinone deriva-
tives, the oxidized electrophile is generated by the laccase from
the nucleophilic partner itself which may reduce the molecular
diversity.
In nature, the structural variability within the phenoxazinone
series comes from the use of o-aminophenol precursors mainly
differing in their oxidation states. This is often followed by site-
selective acetylation and methylation reactions.5a,7 Further struc-
tural modifications may arise from rare N-acetylcysteine nucleo-
philic substitution, aminolytic oxirane ring opening or
transamination and dehydration steps.7b
remains to be addressed. The present study focuses on the che-
moselectivity of laccase-mediated coupling reactions of mixtures
of o-aminophenol substrates, considering their steric, redox and
electrophilic properties. The production of non-symmetrically
substituted phenoxazinone dyes, in a controlled manner, could
improve our understanding of the biosynthetic pathway mechan-
ism. The use of cyclic voltammetry allowed to determine the
ease of oxidation of the substrates. High-performance liquid
chromatography systems equipped with either photodiode array
detection (HPLC-PDA) or coupled to mass spectrometry
(HPLC-MS) were used to identify the novel dyes.21 The
HPLC-PDA method helped to delineate the selectivity and to
establish a relative scale for the conversion rates of selected sub-
strates by laccase. Using the Density Functional Theory (DFT)
method, we calculated how the regioselectivity may be secured
in the first 1,4-addition step on the iminoquinone intermediate.
The rigid docking of few substrates suggested a plausible expla-
nation for one precursor failing to give the symmetrically substi-
tuted phenoxazinone.
Regarding the actinomycin-like phenoxazinones or the related
xanthommatin framework (Fig. 1), the plausible synthetic
pathway involves two identical precursors i.e. o-substituted ami-
nophenols.2a,14 Their natural transformations are realized either
by a laccase, a phenoxazinone synthase or a tyrosinase.10 The
biosynthesis of grixazone, elloxazinones or bezerramycins A–C
is thought to be slightly different.5,7 In case of grixazone, part of
the o-aminophenol is submitted to oxidation and nucleophilic
substitution before reacting with the residual native o-amino-
phenol. Both elloxazinone B and bezerramycin C (Fig. 1) lack a
substituent at C-9 but are functionalized at the C-8 and C-1 pos-
itions. This necessarily implies a selective coupling of two
different precursors. Such transformation has not been realized
to date in a laboratory with a laccase.7
As members of the blue multicopper oxidases, laccases and
phenoxazinone synthases have been widely investigated to de-
lineate the structural features of their catalytic sites.15–17 It
appears now that the phenoxazinone synthase isolated from
Streptomyces antibiocus possesses a substrate binding pocket
and solvent channels more similar to those of laccases than any
other MCO.18 These structural characteristics concern more pre-
cisely the four copper atoms distributed into the three sites
named T1, T2 and T3.15 The blue Cu-site (T1) holds an intense
absorption around 600 nm due to the highly π covalent link
between S (Cys) and Cu2+. The copper is bound to two other
ligands (His) to form the trigonal plane, whereas a fourth ligand,
involved in the tuning of the enzyme activity, varies greatly
among proteins. The T1 site is the primary acceptor of electrons;
it is slightly buried in the enzyme core near the substrate binding
cavity where small molecules are oxidized.15,19 T2 and T3
copper sites form a trinuclear cluster where channelled electrons
from T1 are transferred to molecular oxygen which is reduced to
water. Recent X-ray crystallographic data of complexes contain-
ing both the enzyme and an aromatic substrate have shed some
light upon the binding mode of small molecules into the cavity
near the T1 site.20 Contrary to tyrosinase, the substrate in laccase
does not directly bind to the copper atom. The small molecule is
involved in hydrogen bonds with one of the His ligands
(His458) of the copper atom of the T1 site and probably to a
Asp residue (Asp206) in the vicinity.19 The rest of the cavity
mostly contains hydrophobic residues which may also be
involved in π–π stacking interactions.
Results and discussion
Oxidation potentials of the substrates
The set of substrates (i.e. substituted o-aminophenols 3–4)
chosen for this study was previously described regarding their
chemical synthesis and “dimerization” into the corresponding
symmetrically substituted phenoxazinones 5–6 (Table 1,
Scheme 1).12a,b,d A relative reactivity scale was previously given
based on isolated yields of oxidative dimerization products
obtained with a commercial semi-purified laccase (referred to as
Tv1) (Table 1, column 6). Moreover several kinetic data were
collected with a similar laccase isolated and totally purified
(Table 1, column 8). The intrinsic redox properties of com-
pounds 3–4 were assessed now (Table 1, column 4) along with
the conversion rates obtained using a surrogate commercial
laccase preparation of lower cost (referred to as Tv2) (Table 1,
column 7). Cyclic voltammetry was used to measure the relative
ease of oxidation of the substrates.21,22 The laccase natural sub-
strate i.e., 3-hydroxyanthranilic acid (3a, entry 1), well studied
by electrochemical methods,21,23 was included in the set as a
reference. The interpretation of the cyclic voltammograms (given
in ESI†) was based on previous studies dedicated to the redox-
properties of kynurenine species (oxidized metabolites of trypto-
phan degradation) in aqueous buffer (pH 7).23
Like kynurenines (e.g. (S)-2-amino-4-(2-amino-3-hydroxyphe-
nyl)- 4-oxo-butanoic acid), the aminophenols exhibited a distinct
redox-behavior. A positive shift towards higher potential was
observed here, due to the organic solvent used.24 All substrates
possess a marked oxidation peak and a very weak reduction one.
A single wave corresponding to a two-electron transfer is
observed along with the absence of a fully reversible redox be-
havior. The oxidation normally leads to a quinone-imine species
and the absence of significant reduction peak indicates that this
intermediate is highly reactive.23 The results of Table 1 can be
summarized as follows: all compounds exhibit similar oxidation
potentials which are close to the one of the reference (entry 1,
3a) and mostly fall within the experimental error. Sadly, neither
the oxidation potential measured or the conversion yields
The laccase-mediated synthesis of elaborated phenoxazinones
endowed with fluorescent properties has only been recently
investigated. The challenging problem of an oxidative cascade
leading to selected non-symmetrically substituted fluorophores
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1834–1846 | 1835

View Article Online
Table 1 The relative ease of oxidation of aminophenol precursors and their laccase-mediated conversions into phenoxazinones
Aminophenol precursors Laccase-mediated synthesized heterocycles
app e 12b
Entry
Fg
Cmpd
Epa/Va
Productsb cmpd
Conv% (Yield%)-Tv1c 12a,b,d
Conv%-Tv2d
Km
1
2
3
4
5
6
7
8
CO2H
PO3H2
SO3H
SO2NH2
C3
C3
C3
C3
C3
C3
C3
C3
C3
C3
C6
C6
C6
3a
3b
3c
3d
3e
3f
3g
3h
3i
1.03
0.98
1.00
1.03
1.03
1.04
1.05
0.98
1.10
1.03
0.96
1.02
0.98
5a
5b
5c
5d
5e
5f
5g
5h
5i
>70
>70
>90
>90
>80
Nd
>80
Nd
Nd
Nd
>80
<50
>60
∼0
Nd
Nd
>95-(90)
>95-(90)
>80-(70)
>70-(40)
>90-(86)
>70-(58)
>90-(90)
>90-(87)
>90-(90)
<20-(4)
>80-(73)
∼0-(0)
0.075 0.008
0.066 0.001
SO2NHPh
Nd
SO2NHC6H11
SO2NHCH2CO2Me
SO2NH(CH2)3–OH
SO2NH(CH2)2–NH2
SO2NH(CH2)3–NMe2
SO3H
Naf
Nd
Nd
Nd
9
10
11
12
13
3j
4c
4f
5j
6c
6f
0.069 0.014
1.899 0.313
0.042 0.003
Nd
SO2NHC6H11
SO2NH(CH2)3–NMe2
4j
6j
a Anodic peak of oxidation recorded by cyclic voltammetry in DMF with TBABF4 as supporting electrolyte ( 0.02 V); b See Scheme 1 for general
structure of the heterocyclic core; c Tv1 = commercial Trametes versicolor laccase (Bioscreen® laccase). Conversions were estimated by HPLC-PDA
and isolated yields of phenoxazinones were previously reported in literature.12a,b,d d Tv2 = commercial surrogate laccase from Trametes versicolor
;
(Aldrich). Conversions were estimated by HPLC-PDA; Nd = not determined; e Kinetic data collected with purified laccase from Pleurotus sajor caju
(28 U mL−1) at pH 6–6.5;12b f Na = not available as the precursor is not fully soluble above 2 mM.
sterically hindered substrates (entries 6, 10 and 12) whereas a
twofold increase is recorded for compound 4c (entry 11). Here
again, compound 4j (entry 13) failed to give a symmetrically
substituted phenoxazinone dye (general structure
6
in
Scheme 1). Based on these preliminary results, pairs of precur-
sors were selected to investigate the synthesis of non-symmetri-
cally substituted phenoxazinones. The surrogate laccase being
equally efficient for phenoxazinones synthesis, this laccase prep-
aration (Tv2) was used for the remaining part of the research.
Proof of concept for the synthesis of non-symmetrical
phenoxazinones
Scheme 1 Laccase-mediated synthesis of phenoxazinones. Fg = func-
tional group responsible for tunable solubility properties. (Fg = CO2H,
SO3H, SO2NH2, SO2NHR, PO3H2).12a,b,d
As non-symmetrical phenoxazinones have never been syn-
thesized in vitro with a laccase, a “proof of concept” reaction
was set up to validate our analytical method. Both the isolation
and structural elucidation of the representative novel dyes were
undertaken using appropriate substrates synthesized as pre-
viously described in the literature.12b,d Compound 3b, which
possesses a phosphorus substituent, was chosen to give reliable
structural data through coupling constants analysis (JH–P and
obtained with laccase could be experimentally linked to an
expected influence of the Hammet substituent effect.24b,c
Laccase-mediated oxidation of selected substrates
JC–P) in the NMR spectra. Compound 3f was selected because it
The results of the laccase-mediated oxidation of selected precur-
sors with the surrogate laccase Tv2 are given in Table 1 (column
7). The conversion yields were estimated using the HPLC-PDA
analytic method previously set up to follow the biotransform-
ations with the Bioscreen laccase Tv1.12a,b,d Recent studies show
that the laccase preparation Tv2 from a reliable commercial
source behaves similarly to few natural isozymes from Trametes
versicolor.25 This is a practical alternative as the laccases we
used previously were either uncompletely characterized (Bio-
screen, Tv1)26 or not yet commercially available (laccase from
Pleurotus sajor caju).27 In Table 1, the conversion percentages
into the corresponding phenoxazinone dyes (see Scheme 1) were
very similar to those collected with the Bioscreen laccase (Tv1)
(entries 1–4). A slight decrease is observed with the more
makes the LC-MS analysis easier and displays a typical fragmen-
tation pattern in MS/MS spectra (loss of cyclohexene, −82 m/z).
After 16 h of reaction (1 : 1 mixture of selected substrates, pH
6.5, 25 °C, laccase Tv2), HPLC analysis showed the conversion
of starting materials into four phenoxazinones in a ratio of about
55/6/4/35 deduced from the percentages of area detected at
410 nm. The first and last eluted dyes were identified as 5b (Fg
= PO3H2) and 5f (Fg = SO2NHC6H11, Scheme 1) by comparison
with references synthesized by known procedures from the litera-
ture.12b,d The crude reaction mixture was purified by preparative
HPLC and the four phenoxazinones (5b, 8bf, 7fb and 5f) were
isolated in relative amounts corresponding roughly to the molar
ratios observed by HPLC. The structural elucidation of the two
novel compounds, respectively 8bf (7.6% yield) and 7fb (3.8%
1836 | Org. Biomol. Chem., 2012, 10, 1834–1846
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 3 HPLC-PDA analysis of the reaction of the pair of substrates 3d
and 3f (1 mM) with the laccase Tv2. Top panel-left and right, 3d and 3f
uv-vis spectra. Bottom panel-left, HPLC-UV (310 nm) analysis at T0
(2 min), 3d (RT = 1.51 min) and 3f (RT = 3.88 min). Bottom panel-
right, HPLC-UV (310 nm) analysis at T3 (75 min of reaction).
1
Fig. 2 From the bottom: full H NMR spectrum of 7fb, 8bf followed
by expansion of aromatic region of 7fb, 8bf. On the top right, structures
of the phenoxazinones and atoms numbering for NMR attribution (see
Experimental).
yield) were obtained from NMR analyses and high resolution
mass spectra (see Experimental). In the top left panel of Fig. 2,
1
expansions of the aromatic regions of H NMR spectra clearly
show the significant difference on the aromatic proton signals
due to the coupling to the phosphorus atom. Compound 8bf is
characterized by a complex pattern (ddd) for H-8 due to the
1
3JH–P of 12.5 Hz whereas in compound 7fb, H NMR spectrum
only features JH–H couplings: H-8 and H-6 appear as doublet of
doublets (dd), H-7 as an apparent triplet. In 31P NMR, com-
pound 8bf is characterized by a signal at 9.5 ppm (phosphorus
on an aromatic) whereas compound 7fb appears at 12.4 ppm
(phosphorus anchored to a quinone ring). A second reaction was
realized with 3b (Fg = PO3H2) and 3g (Fg = SO2NHCH2CO2-
Me) which essentially gave similar HPLC elution profiles and
NMR signals for compounds 8bg (isolated in 6.8% yield) and
7gb (isolated in 4% yield) (see ESI† for structures and atom
numbering, and HPLC profiles). Here again the ratio of conver-
sion detected by HPLC roughly corresponds to the isolated
yields. It shows that even if the substrates are not significantly
distinct by their oxidation potentials, oxidation is faster for 3b
and homo-dimerization appears favoured versus cross-dimeriza-
tion (yield of 5b about 8 to 10-times higher than 8bf and 8bg).
The quick decrease of 3b concentration during the reaction
yields finally more of the symmetrical phenoxazinones 5f or 5g
than the non-symmetrical 7fb or 7fg.
Fig. 4 HPLC-PDA analysis of the reaction of the pair of substrates 3d
and 3f (1 mM) with the laccase Tv2. Panel-left, HPLC-UV (440 nm)
analysis at T3 (75 min of reaction). Panel-right, the uv-vis spectra of
peak 1 (A), peak 2 (B), peak 3 (C) and peak 4 (D).
phenoxazinones were further unambiguously identify by
HPLC-MS/MS as the analytic method (full identification data in
ESI†).‡ The relative rates of conversion were estimated from the
area decreases and increases of the respective precursors and
phenoxazinones in the first 75 min of reaction with laccase
(Table 2). This process is exemplified in Figs 3 and 4 with the
pair of precursors 3f and 3d; the detection of substrates (amino-
phenols) and products (phenoxazinones) is made respectively at
310 and 440 nm. Finally, compound 4j was included in the set
to determine by HPLC-MS/MS if it is oxidized or acts only as a
nucleophilic partner.
HPLC-MS/MS study of competitive laccase-mediated synthesis
of phenoxazinones
The competitive experiments were then conducted systematically
as above with 1 : 1 mixtures of two different aminophenol pre-
cursors, using HPLC to follow the reactions. Substrates bearing
primary or secondary sulfonamide moeties, synthesized as
described in precedent papers,12a,b were essentially chosen to
ensure optimal retention times and separations on the HPLC
column. The samples were analyzed by HPLC-PDA (retention
times given in the ESI†) and these non-symmetrically substituted
After 75 min, a faster consumption of 3d versus 3f is observed
along with the appearance of the corresponding phenoxazinones
5d, 5f, 7df and 8fd featuring the specific λmax at 220–240 nm
and 420–440 nm (see Fig. 4 and Scheme 2). From the four
peaks detected at 440 nm, peak 1 and 4 correspond, respectively,
to the known dyes 5d and 5f (see Scheme 1).12b Peaks 2 and 3
correspond to novel phenoxazinones (8df and 7fd, see ESI† for
identification) well distinct from the symmetrical dyes (5d, 5f)
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1834–1846 | 1837

View Article Online
Table 2 Relative area variations for the precursors and phenoxazinonesa
ΔArea (AT0 − AT3)/AT0
ΔA-(cmpd)
ΔArea (AT3 − AT1)/AT3
Entry
ΔA-(cmpd)
ΔA (dye 1)
ΔA (dye 2)
ΔA (dye 3)
ΔA (dye 4)
1
2
3
4
5
0.34-(3f)
0.30-(3f)
0.33-(3f)
0.05-(3f)
0.15-(3d)
0.43-(3d)
Na-(3c)
0.30-(3j)
0.40-(4f)
0.26-(4f)
0.55-(5d)
0.23-(5c)
0.19-(5j)
0.02-(10ff)
Nd-(5d)
0.54-(8df)
0.28-(8cf)
0.13-(8jf)
0.52-(9ff)
0.46-(9df)
0.48-(7fd)
0.28-(7fc)
0.07-(7fj)
0.38-(6f)
0.32-(5f)
0.50-(5f)
0.34-(5f)
0.12-(5f)
0.42-(6f)
0.03-(10fd)
a Precursors (1 mM) were mixed with laccase Tv2 (100 UL−1) in 1 mL of 0.2 M ammonium acetate buffer at 25 °C. Area for each precursor (310 nm)
and dyes (440 nm) were measured at T0 (2 min), T1 (25 min), T2 (50 min) and T3 (75 min). Na = not available as 3c was transformed in the first
25 min. Nd = not determined as the small amount of 5d detected at T1 did not increase further.
Scheme 2 Structures of non-symmetrically substituted phenoxazinones produced by oxidative cross-coupling reactions and identified from
HPLC-PDA and HPLC-MS/MS analyses (see ESI† for full description of identification data).
and close to each other in the elution profile (see Scheme 2, Figs
3 and 4). Results of the HPLC-MS/MS analysis in the identifi-
cation process for the pair 3f–3d (entry 1, Table 2) are exem-
plified in Fig. 5 (fully described in ESI†).
The relative rates of conversion for all precursors and the dis-
tinct heterocyclic dyes are presented in Table 2. Entry 1 of
Table 2 shows that the conversion of 3d into 5d is faster than 3f
into 5f in the earlier hours of reaction. Both independent laccase-
mediated syntheses of 5d and 5f (Table 1, entries 4 and 6) and
also the oxidation potential of 3d and 3f are very similar. Yet a
greater efficiency of the biocatalyst for 3d versus 3f is observed
here. Entry 2 shows again that despite being close regarding
their oxidation potential, a faster oxidation rate for 3c versus 3f
is still observed. Interestingly, the appearance of 5f is the most
important whereas all the three other dyes (5c, 7fc and 8cf) are
formed in lower proportions. The compound 3c may be a poor
nucleophile whereas 3f is a good one that reacts quickly with
both the oxidized 3c and 3f (Scheme 3). In entry 3, a similar rate
of conversion is observed for 3f and 3j (see Table 1, entries 6
and 10). Yet, the rate of formation for the dyes 1–4 differs sig-
nificantly, descending from 5f–5j–8jf to finally 7fj. It suggests a
preference for 3f reacting as a nucleophilic partner upon its own
oxidized intermediate (Scheme 3). This may be related to a
hydrophobic effect of the substrate since 3j is more polar around
1838 | Org. Biomol. Chem., 2012, 10, 1834–1846
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 5 HPLC-PDA analysis of the reaction of the pair of substrates 3d and 3f (1 mM) with the laccase Tv2 (100 UL−1) at 25 °C in 0.2 M ammonium
acetate buffer (pH 6). Top panel-left, HPLC-UV (440 nm) analysis after 24 h of reaction. Peak 1 (RT = 2.59 min) is 5d, peak 2 (RT = 10.62 min) and
peak 3 (RT = 11.06 min) are 8df and 7fd (Fg1 = SO2NH-cyclohexyl, Fg1′ = SO2NH2). Analytic condition with XTerra (2.1 × 50 mm, 2.5 μm), flow
rate = 0.2 mL min−1. Bottom panel-left, HPLC-MS-MS (ESI positive mode, m/z 453) analysis of the sample after 24 h of reactions. Divert valve was
set between 2.1 and 16.9 min, peak 2 eluted at 9.49 min and peak 3 at 10.04 min. Top panel-right, mass spectrum of peak 2 (9.49 min). Bottom panel-
right, mass spectrum of peak 3 (10.04 min). (see ESI† for full identification data).
pH 6 due to its protonated tertiary amino-group. Entry 4 is inter-
esting as the conversion of 4f is impressively quicker than 3f and
the formation of 6f is also faster than 5f. The most peculiar
result is the significant larger proportion of only one non-sym-
metrically substituted dye (9ff). This heterodimeric dye corre-
sponds to the residual nucleophilic 3f reacting with the larger
proportion of oxidized 4f (Scheme 3). Finally, entry 5 shows that
4f is converted more quickly than 3d. Again, two dyes are sig-
nificantly produced in larger proportions. A hydrophobic effect
may explain the larger quantity of 6f whereas the major decrease
of 3d is related to the nucleophilic addition of 3d upon the oxi-
dized 4f. From these results, an overall reactivity scale going
from 3c∼4f∼3b–3d–3j–3f may be proposed. It can be correlated
to the kinetic data obtained previously (4f–3d–3c–3j from
Table 1, column 8). In addition, by mixing 3f and the previously
inert 4j, a major new peak was detected in both HPLC-PDA and
HPLC-MS. The corresponding ion with m/z of 538 [M + H]+
(C23H31N5O6S2) gave in tandem mass spectrometry the typical
fragmentation along with a loss of neutral fragment of m/z 45
(HNMe2) indicative of the tertiary amino group. A second
compound with identical mass m/z 538 was detected in a tiny
amount. The same fragmentation pattern was observed. This
suggests that the major compound arises from laccase-mediated
oxidation of 3f and subsequent 1,4-addition of native 4j
(Scheme 2, phenoxazinone 10jf). The very minor compound
would arise from auto-oxidation of 4j and subsequent 1,4-
addition of native 3f (Scheme 2, phenoxazinone 9fj). The redox-
properties and apparent affinities for the enzyme seem to be
reflected in the formation of the phenoxazinones. Interestingly
Cambria et al. published in 2010 a report about a laccase from
Rigidiporus lignosus, highly similar to the inducible laccase
from Trametes versicolor and presenting similar kinetic proper-
ties.28 This suggests that a comparison between kinetic data col-
lected with laccases from Pleurotus sajor caju and Tv2 holds
pertaining information.12b The experimental results indicate that
non-symmetrically substituted phenoxazinones are indeed
formed in various proportions most probably depending on the
redox-properties of the precursors and their relative affinity for
the enzyme. We have proven that compound 4j acts as an
efficient nucleophilic partner and is weakly auto-oxidized in situ.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1834–1846 | 1839

View Article Online
Scheme 3 Proposed pathways for the intermolecular 1,4-addition of o-aminophenol substrates onto imino-quinone intermediates (first Michael
addition step).
Although it seems inert towards laccase-mediated oxidation,
recently 4j was successfully oxidized into phenoxazinone by
tyrosinase.29
adducts can adopt two conformations of the phenol (Scheme 3,
paths 1 and 2). This first reaction step has been investigated by
the location of the transition state (TS) structures at two compu-
tational levels using RHF/6-31+G and B3LYP/6-31+G(d) energy
functions (see ESI† for details). It can be noted that the partial
inclusion of the correlation energy with B3LYP and the
additional polarization functions to the double ζ basis set includ-
ing diffuse functions do not modify the relative energies of the
TS (Table 3 and details in ESI†). In each case, the imaginary fre-
quency is associated to the exchange of the H between the NH
Theoretical study of the regioselectivity of the 1,4-addition
The addition of the native o-aminophenol on the neutral imiqui-
none intermediate can be done following two paths depending
on the regioselectivity of the Michael addition. Each of these
Table 3 Ab initio and DFT relative energies (kcal mol−1) for transition statesa
Entry
1
Substrate
Transition state (TS)
RHF/6-31+G (ΔE)
B3LYP/6-31+G(d) (ΔE)
B3LYP/6-31+G(d) (ΔG)
o-aminophenol
R = H
TS1
TS2
TS3
TS4
TS1
TS2
TS3
TS4
TS1
TS2
TS3
TS4
77.293
72.889
61.251
64.754
58.515
49.953
69.060
60.735
50.394
52.352
80.126
85.572
55.344
57.787
45.300
47.574
35.659
40.851
51.373
43.401
39.285
41.274
60.936
Na
66.796
69.061
57.378
59.670
48.704
53.178
62.835
56.287
52.377
55.493
72.078
Na
2
3
3c
4c
a See Scheme 3 for the outline of the reaction paths involving the different transition states. Reactants are local minima energy obtained from RHF/6-
31+G or B3LYP/6-31+G(d) optimized geometries. ΔE and ΔG (kcal mol−1) are defined with respect to the sum of the lowest energies RHF/6-31+G or
B3LYP/6-31+G(d) for the reactants. Na = not located equilibrium structure
1840 | Org. Biomol. Chem., 2012, 10, 1834–1846
This journal is © The Royal Society of Chemistry 2012

View Article Online
group and the proximal carbon defining a pseudo four-membered
ring (see ESI† for representative description of equilibrium struc-
tures). The calculation results are summarized in Table 3. With R
= H, path 2 is energetically more favorable than path 1, the rela-
tive energies of the two conformers lying in a 2–3 kcal mole−1
range. The same type of calculations has been performed with R
= SO3H at the two positions in 3c and 4c. Remarkably, in
accordance with the experimental data, path 1 is energetically
less demanding than path 2. For 4c, the TS4 structure can be
trapped only at the RHF level. No corresponding equilibrium
structure can be located with the B3LYP function. Thus, the
favoured addition corresponds to the nucleophilic attack on the
β-carbon vs. the electron-withdrawing Fg substituent, in all
cases. This leads to zwitterionic intermediates with the negative
charge stabilized by Fg (see ESI† for representative equilibrium
structure).30
bonding, is the extraction of electrons from the substrate.19a The
transient intermediate iminoquinone flows out the binding cavity
and reacts with a nucleophilic partner in the bulk of the solution.
2-amino-3-oxo-3H-phenoxazine-8-sulfonic acid was reported to
be produced by a laccase in vitro without detection of any
radical intermediate by the EPR technique;12c this adds some
weight to our proposal. Indeed, the same technique successfully
identified radical intermediates in several related reactions.36 The
second step, i.e. the first 1,4-conjugated addition, would most
probably take place in the bulk of the solution and the regio-
chemistry is secured by the rules governing the Michael
addition. From the isolated yields in the “proof of concept” reac-
tions, the major path of conversion corresponds to the addition
of the best nucleophiles present in solution i.e. the aminophe-
nols. Yet side reactions could not be ruled out and by-products
or intermediates were indeed observed in preparative HPLC (see
ESI†). Nevertheless these results are similar to those obtained
with substituted catechols either in presence of laccase, or by
electrochemical oxidation.13,36 The subsequent steps of oxi-
dation, intramolecular 1,4-conjugated addition and final oxi-
dation are supposed to be identical to those previously described
for the phenoxazinone synthase.
Hypothesis for the laccase-mediated synthesis of phenoxazinones
Consistently with our experimental results, we propose a
mechanistic pathway for the production of substituted phenoxa-
zinones using laccase as catalyst. Our mechanistic proposal is
adapted from Barry et al. regarding the actinomycin synthesis
catalyzed by phenoxazinone synthase.8 Here, the heterocycle is
produced in the early hours of the reaction and the residual part
of the native precursor is still present in significant amounts
(HPLC analyses). This fact points to a non-electrophilic mechan-
ism in contrast to the formation of grixazone.5a The overall
process is depicted in Scheme 4. Briefly, the first two steps are
the ones revisited with our hypothesis. Depending on its affinity
for the enzyme, the o-aminophenol precursor is quickly oxidized
through two successive one-electron oxidations. Alternatively a
single laccase-mediated one-electron oxidation could produce
the anthranilyl radical which could undergo a further reaction.9
Indeed cinnabarinic acid (Fg = CO2H, formula 5 in Scheme 1) is
also produced in reactions involving radicals and molecular
oxygen.31–35 Yet, our suggestion that the laccase could directly
generate a quinone intermediate, in accordance with studies on
oxidase model,33 is compatible with our experimental results in
terms of selectivity and isolated yields. Oxidation at the laccase
Cu-site T1 is governed by the Marcus outer sphere mechan-
ism,19b and the rate limiting step, facilitated through hydrogen
Docking of selected substrates
One of the substrates (4j) failed to be transformed by the laccase
whereas it performed well as a nucleophilic partner. We envision
that a poor fitting of the substrate in the binding cavity could be
a reason for this behavior. At the pH of our experiments (pH of
6.5 for 0.2 M ammonium acetate), Asp206 is mostly deproto-
nated and aromatic substrates with –OH and –NH2 groups are
recognized and dragged inside the cavity by hydrogen bonding
with Asp206.19b,22,28 The known structure of Trametes versico-
lor enzyme complexed with p-xylidine (1kya) is regularly used
as a benchmark for docking studies of small molecules.19,28,37
Furthermore, the Bioscreen laccase has been experimentally
related to the inducible isozyme form (Lac2) from Trametes ver-
sicolor26 whereas the major constituent of the commercial
laccase preparation Tv2 seems related to the major isozyme
(Lac1) of Trametes versicolor.25a This prompts us to investigate
by simple rigid docking the steric factors which could possibly
prevent the oxidation of 4j by the laccase (Fig. 6). Before
Scheme 4 Mechanistic pathway for the synthesis of symmetrically and non-symmetrically substituted phenoxazinones from o-aminophenol
substrates.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1834–1846 | 1841

View Article Online
Table 4 Docking of selected substrates with Molegro virtual dockera
Mol dock score of docked poses (Productive)b
Entry
Ligand
Pose-1
Pose-2
Pose-3
Pose-4
Pose-5
1
2
3
4
5
6
7
p-xyd
3c
3f
3j
4c
4f
4j
−52.5 (Y)
−54.1 (Y)
−11.6 (N)
−50.8 (M)
−43.1 (Y)
−27.5 (N)
−72.3 (M)
−48.3 (N)
−48.3 (N)
−47.2 (N)
−54.8 (M)
−40.1 (N)
−79.3 (M)
−52.2 (Y)
−31.3 (N)
−53.1 (N)
−51.5 (M)
−8.2 (N)
−53.6 (Y)
−36.9 (Y)
−41.5 (M)
−52.8 (M)
−72.4 (N)
−57.3 (N)
—
−71.7 (N)
−47.5 (Y)
—
−22.8 (Y)
−67.0 (N)
−78.2 (N)
a Productive docking: Y = Yes (–OH group is directed towards Asp206 and His458 with a positioning at, or almost at, the expected hydrogen bonding
position); M = Medium (–NH2 group is directed towards Asp206 and His458 with a positioning at the expected hydrogen bonding position); N = No
(= sulfonamide moiety is either hydrogen bonded or pointing towards the inner of the cavity). See ESI† for complete listing of docked poses. b The
total MolDock Score energy (arbitrary units) is the sum of internal ligand energies, protein interaction energies and soft penalties. See ref. 39 for
further details.
docking, the conformations of selected substrates were energy
minimized. From the X-ray crystal structure (PDB ID-1kya) the
receptor pocket was prepared with the Molegro Virtual Docker
program.38,39 The software has a procedure allowing to create a
“template docking environment”. Briefly p-xylidine is taken as
the reference ligand; a cavity is detected around the ligand in
which steric factors and hydrogen bonds are identified along
with a preferential positioning of the aromatic ring (see ESI†).
Rotating/flexible bonds are identified for each ligand while the
receptor cavity is kept rigid. From there, a scoring function
(Moldock score)39 associated to the total energy of the system is
given to rank the different poses observed. As a reference ligand
is used, a RMSD constraint <1 Å (RMS deviation) is also taken
into account. For each substrate 10 runs (1500 iterative searches)
are undertaken from which only the 5 best docked conformations
are kept. In Table 4 we present the docked poses with an orien-
tation corresponding to either the aromatic –OH or –NH2 group
yielding productive hydrogen-bonds. The docked poses featuring
non productive hydrogen-bonds, with the sulfonate/sulfonamide
moeity, are also given. To validate the procedure, p-xylidine is
docked into the receptor cavity with a score taken as the refer-
ence (see Table 4, entry 1). It can be observed that compounds
3c, 3f and 3j gave at least one productive docking with the –OH
group forming hydrogen bonds with Asp206 and His458. As
expected, the smaller 3c gave the highest score with multiple
optimal docking whereas 3f and 3j only gave one productive
conformation with the side chains pointing towards the outside
of the cavity. For 4c, one conformation featured the –OH group
hydrogen-bonded to His458 whereas the sulfonic group pointed
partially into the cavity. An electrostatic repulsion could impair
the efficiency of this binding mode. The other docked confor-
mations showed the sulfonate pointing outside the cavity with
the –NH2 group hydrogen-bonded. One conformation for 4f also
featured the –OH group involved in hydrogen bonding whereas
for 4j, only one conformation with –NH2 and –OH groups point-
ing into the inner part of the cavity was found. In this docked
pose, the –NH2 group is at a distance of about 3 Å from both
His458 and Asp206. Further ligand energy minimization could
not improve the positioning of 4j. Docked pose 1 provided the
–OH group in the inner cavity albeit with a folding of the sulfo-
namide function towards the aspartate and this aromatic –OH
group. The absence of a docked conformation featuring the aro-
matic –OH group bonded to Asp206 and His458, is not per se a
proof that the laccase is not able to process 4j, but reinforces our
experimental observations.
Conclusion
Several non-symmetrically substituted phenoxazinones were iso-
lated and identified for the first time in laccase-mediated coup-
ling reactions of o-aminophenols in vitro. Taking advantage of
the previous isolation and structural characterization of symme-
trically substituted phenoxazinones, we have successfully
mimicked a natural transformation using mixtures of structurally-
close substrates. Redox properties were assessed and it was
shown that a careful selection of pairs of substrates could lead to
some control of the process by favoring one heterocycle versus
another. A qualitative correlation between both steric and redox
properties became apparent through the rates of conversion. The
chemoselectivity and regioselectivity observed could be
explained through DFT calculations, supporting our revisited
mechanism. We propose that the laccase-mediated oxidation of a
Fig. 6 A stereo view of docked pose 3 of 4j obtained with Virtual
Molegro Docker into the binding cavity. View elaborated with Pymol
software.40 The –NH2 group is at the expected position for the hydrogen
bonding with Asp206 and His458.
1842 | Org. Biomol. Chem., 2012, 10, 1834–1846
This journal is © The Royal Society of Chemistry 2012

View Article Online
given o-aminophenol is a formal two-electron oxidation that gen-
erates directly a short lived iminoquinone intermediate. This
highly electrophilic species then reacts with a nucleophilic
partner (i.e the same o-aminophenol or another similar substrate)
in the bulk of the solution following the chemical rules govern-
ing the classical 1,4-conjugated addition (Michael addition). One
compound (4j), inert towards laccase-mediated transformation,
could still yield a non-symmetrically substituted phenoxazinone
by acting as nucleophile versus another processable o-amino-
phenol. This observation strongly supports our original hypoth-
esis that the key-step of the oxidative dimerization, namely the
first Michael addition, proceeds outside the laccase cavity.
was added to a stirred solution (57 mL) of 0.2 M of ammonium
acetate at 25 °C. The reaction was started by adding a solution
(0.2 mL) of Laccase (1.5 mg mL−1, Trametes versicolor, Aldrich
53739, 20 Umg−1). HPLC analysis showed the total consump-
tion of the starting materials after 16 h. The solution was then
freeze-dried and the residual solid was dissolved in methanol
(1.2 mL), filtered on a PTFE syringe filter (SFPF02213-C1,
ROCC) and added to water (1.2 mL) before being filtered again.
The filtrate was directly used for purification by preparative
HPLC. The filters were rinsed several times with water and
methanol respectively. The respective washing solutions were
concentrated and dried under vacuo.
Experimental Details
2-Amino-3-oxo-3H-phenoxazine-9-phosphonic acid-1-sulfonic
acid cyclohexylamide (8bf)
Reagents and analyses of synthesized and isolated compounds
0.055 mmol of substrates (total mass of 27 mg) were engaged
and 30 mg of crude product were obtained. Concentrated
washing solutions (mainly symmetrical dyes 5b and 5f) gave
respectively 7 and 2 mg. 18 mg of the crude were purified by
HPLC-Prep yielding four fractions which, respectively, corre-
spond to 5b (5.8 mg), 8bf (2 mg), 7fb (1 mg) and 5f (8.5 mg).
The title product was isolated as an orange solid. Yield: 7.6%
(2 mg). RT: 3.37 (analytical) and 5.09 min (preparative). MW:
All aminophenols and aminophenoxazinones compounds were
synthesized as previously described12a,b,d except 3-hydroxy-
anthranilic acid, which is commercially available (Aldrich). All
the reagents and solvents used were of analytical grade and
obtained from either Acros or Aldrich company. Commercial
preparation of laccase Tv2 from Trametes versicolor was
obtained from Aldrich (38429 or 53739). The HPLC-PDA semi-
preparative system consisted of Waters Alliance 2699 separation
module, Waters 2998 photodiode array detector and Waters
Fraction collector III (Waters, Milford, Massachussets, USA).
Analytical separations were performed with Waters XBridge™
C18 column (4.6 × 50 mm, 2.5 μm) equipped with a precolumn
at 25 °C. Detection was performed at 410/440 nm and on-line
uv-visible scans were performed. Semi-preparative separations
were performed with Waters Xbridge™ Prep C18 column (10 ×
100 mm, 5 μm) equipped with a precolumn at 25 °C. The
mobile phase was water, ammonium acetate 50 mM/acetonitrile
(5%) and acetonitrile. A linear gradient (curve 6) was applied
from 0/90/10 to 0/10/90 for 7.5 min after 0.5 min at initial con-
ditions and before 2 min at final conditions (for 3b and 3f). A
linear gradient (curve 6) was applied from 10/80/10 to 10/10/80
for 7.5 min after 0.5 min at initial conditions and before 2 min at
final conditions (for 3b and 3g). The flow rate was set at 1.2 mL
min−1 for the analytical method (10 μL injection). For prepara-
tive separations, the flow rate was set at 4 mL min−1. A linear
gradient (curve 6) was applied from 10/80/10 to 10/10/80 for
9.5 min after 0.5 min at initial conditions and before 4 min at
final conditions (100 μL injection). An equilibration time of
1
C18H20N3O7PS, 453.40 g mol−1. H NMR (500 MHz, metha-
nol-d4) δ = 7.88 (1H, ddd, JH–H = 5.5, 3.0 and JH–P = 12.5 Hz,
H-8), 7.53–7.51 (2H, m, H-7 and H-6), 6.48 (1H, s, H-4), 3.09
(1H, tt, JH–H = 3.9, 11.3 Hz, H-15), 1.71 (2H, m, Ha-16), 1.62
(2H, m, He-16), 1.49–1.05 (6H, m, H-17,18) ppm. 31P NMR
(202.4 MHz) δ = 9.54 (Ar–PO3H2) ppm. 13C NMR (125 MHz,
BB-decoupling (48 h), missing C11,C13,C14) δ = 179.6 (C3),
151.2 (C12), 147.7, (C2), 134.4 (C9, JC–P = 162 Hz), 130.6 (C8,
JC–P = 14.5 Hz), 129.6 (C7, JC–P = 6.7 Hz), 118.6 (C6), 105.3
(C4), 55.0 (C15), 34.2 (C16), 30.7 (C18), 26.2 (C17) ppm (see
Fig. 2 for atoms numbering). UV/Vis (ammonium acetate
10 mM), λmax = 231, 430–439 nm. MS (ESI) m/z (%): 452.43
[M − H]− (100), 474.37 [M−2H + Na]− (70), 434.2 (25); MS2
of 452.3 m/z yields 434.3 [−18] and 370.38 [−82, –C6H10] m/z.
HRMS (ESI negative mode): clcd
C18H19N3O7PS; found 452.0698.
=
452.0681 for
2-Amino-3-oxo-3H-phenoxazine-9- sulfonic acid
cyclohexylamide-1-phosphonic acid (7fb)
1
4 min was used between successive injections. H, 31P and 13C
The title product was isolated as a red solid. Yield: 3.8% (1 mg).
RT: 3.65 (analytical) and 5.73 min (preparative). MW:
NMR spectra were recorded with a Bruker Avance 500 spec-
trometer. Spectra were obtained in methanol-d4 with a small
addition of D2O for 8bg. Chemical shifts are reported in ppm
relative to TMS or H3PO4 85% for 31P NMR. Low resolution
mass spectra were acquired using a Thermo Finnigan LCQ
spectrometer either in positive or negative mode ESI. High
Resolution Mass Spetrometry (HRMS) analyses using ESI were
performed at the University College of London (UK).
1
C18H20N3O7PS, 453.40 g mol−1. H NMR (500 MHz, metha-
nol-d4) δ = 7.85 (1H, dd, JH–H = 7.7, 1.1 Hz, H-8), 7.66 (1H, dd,
JH–H = 8.3, 1.1 Hz, H-6), 7.52 (1H, t, JH–H = 8.0 Hz, H-7), 6.41
(1H, s, H-4), 3.03 (1H, tt, JH–H = 3.9, 11.2 Hz, H-15), 1.62–1.48
(4H, m, H-16), 1.49–1.05 (6H, m, H-17,18) ppm. 31P NMR
(202.4 MHz) δ = 12.49 (PO3H2) ppm. 13C NMR (125 MHz,
BB-decoupling (24 h), missing C1, C11 and JC–P) δ = 182.0
(C3), 151.4 (C12), 149.8 (C2), 143.6 (C13), 138.7 (C9), 131.1
(C14), 128.4 (C7), 125.6 (C8), 121.2 (C6), 105.6 (C4), 55.0
(C15), 34.3 (C16), 30.7 (C18), 26.3 & 26.1 (C17) ppm (see
Fig. 2 for atoms numbering). UV/Vis (ammonium acetate
10 mM), λmax = 223, 423–440 nm. MS (ESI) m/z (%): 452.38
Laccase-mediated chemical synthesis: general procedure
A solution of 1 equivalent of each aminophenol in methanol
(3 mL, 0.055 mmol for 3b and 3f, 0.25 mmol for 3b and 3g)
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1834–1846 | 1843

View Article Online
[M − H]− (100), 474.43 [M−2H + Na]− (70); MS2 of 452.3 m/z
yields 434.3 [−18] and 370.51 [−82, –C6H10] m/z, MS2 of
454.40 [M+H] m/z yields 371.95 [−82, –C6H10] m/z. HRMS
(ESI negative mode): clcd = 452.0681 for C18H19N3O7PS;
found 452.0704.
electrode/Reference electrode: Ag/AgCl in EtOH sat LiCl. CVs
were recorded in triplicate at a scan rate of 150 mV s−1 with a
potential of −0.5 to 1.8 V. Oxidation potentials were obtained
from peak of maximum current intensity at the end of the oxi-
dation wave. Current polarity was defined according to the EU
convention.
{2-amino-1-[(2-methoxy-2-oxoethyl)sulfamoyl]-3-oxo-3H-
phenoxazin-9-yl}phosphonic acid (8bg)
High-pressure liquid chromatography-PDA instrumentation
0.25 mmol of substrates (total mass of 114 mg) were engaged
and 117 mg of crude product were obtained after freeze-drying.
90 mg of the crude were purified by HPLC-Prep yielding four
fractions which respectively correspond to 5b (48.7 mg), 8bf
(8 mg), 7fb (5 mg) and 5f (13.2 mg). The title product was iso-
lated as an orange solid. Yield: 6.8% (8 mg). RT: 2.91 (analyti-
cal) and 3.69 min (preparative). MW: C15H14N3O9PS, 443.30 g
System 1 used for Table 1 and Table 2: the HPLC-PDA system
consisted of Waters Alliance 2699 separation module, Waters
2998 photodiode array detector (Waters, Milford, Massachussets,
USA). Analytical separation were performed with Waters XTer-
raMsC18 column (4.6 × 100 mm, 5 μm) at 25 °C. Detection was
performed at 310/440 nm and on-line uv-visible scans were per-
formed. Flow rate was 1 mL min−1. Injection volume was of
10 μL. The mobile phase was water/formic acid (0.1%) and
acetonitrile/formic acid (0.1%). A hyperbolic gradient (curve 2)
was applied from 75/25 to 25/75 in 20 min before applying a
hyperbolic gradient (5 min) to the initial conditions.
System 2 used for Table 3: the HPLC-PDA system consisted
of Waters Alliance 2699 separation module, Waters 2998 photo-
diode array detector (Waters, Milford, Massachussets, USA).
Analytical separations were performed with Waters XTer-
raMsC18 column (2.1 × 50 mm, 2.5 μm) at 25 °C. Detection
was performed at 440 nm and on-line uv-visible scans were per-
formed. Flow rate was 0.2 mL min−1. Injection volume was of
10 μL. The mobile phase was water/formic acid (0.1%) and
acetonitrile/formic acid (0.1%). A linear gradient (curve 6) was
applied from 75/25 to 25/75 in 10 min before applying a hyper-
bolic gradient (curve 2) (15 min) to the initial conditions.
1
mol−1. H NMR (500 MHz, methanol-d4/D2O (9/1)) δ = 7.87
(1H, ddd, JH–H = 5.1, 3.5 and JH–P = 12.6 Hz, H-8), 7.55–7.53
(2H, m, H-7 & H-6), 6.49 (1H, s, H-4), 3.83 (2H, s, H-15), 3.55
(3H, s, H-17) ppm. 31P NMR (202.4 MHz) δ = 9.49 (Ar–
PO3H2) ppm. 13C NMR (125 MHz, BB-decoupling (24 h)) δ =
179.6 (C3), 171.6 (C16), 151.5 (C12), 147.8 (C11) 144.3 (C2),
143.6 (C13, JC–P = 12.6 Hz), 137.7 (JC–P = 171.6 Hz, C9), 134
(JC–P = 5.6 Hz, C14), 130.7 (JC–P = 15.1 Hz, C7), 130.0 (C1),
129.6 (JC–P = 6.6 Hz, C8), 118.8 (JC–P = 2.3 Hz, C6), 105.4
(C4), 52.6 (C17), 44.9 (C15) ppm (see ESI† for atoms number-
ing). UV/Vis (ammonium acetate 10 mM), λmax = 231, 438 nm.
MS (ESI) m/z (%): 442.23 [M − H]− (50), 464.15 [M−2H +
Na]− (48), 291.3 (100); MS2 of 442.3 m/z yields 355.09 and
291.34 m/z. HRMS (ESI negative mode): clcd = 442.0110 for
C15H13N3O9PS; found 442.0132.
{2-amino-9-[(2-methoxy-2-oxoethyl)sulfamoyl]-3-oxo-3H-
phenoxazin-1-yl}phosphonic acid (7gb)
High-pressure liquid chromatography-MS instrumentation
System 3 used for Table 3: the HPLC-MS analyses were per-
formed with a HPLC system consisting of a Spectrasystem pump
module P1000XR, autosampler AS3000 and uv-visible detector
UV6000P, coupled to a TSQ mass spectrometer (Finnigan Mat).
Analytical separations were performed with Waters XTer-
raMsC18 column (2.1 × 50 mm, 2.5 μm) at 25 °C. Detection
was performed at 440 nm. Flow rate was 0.2 mL min−1. Injec-
tion volume was of 100 μL. Mobile phase consisted of water
(98.9%)/formic acid (0.1%)/acetonitrile (1%) and acetonitrile. A
linear gradient was applied first from 75/25 to 25/75 in 10 min
then back to the initial conditions in 2 min before equilibrating
the column at the initial conditions. Mass spectra were acquired
by using electrospray ionization (ESI) method in positive ion
mode under the following conditions:
The title product was isolated as an orange solid. Yield: 4%
(5 mg). RT: 3.23 (analytical) and 4.09 min (preparative). MW:
1
C15H14N3O9PS, 443.30 g mol−1. H NMR (500 MHz, metha-
nol-d4) δ = 7.95 (1H, dd, JH–H = 7.4, 1.4 Hz, H-8), 7.57 (1H, dd,
JH–H = 8.3, 1.4 Hz, H-6), 7.53 (1H, apparent t, JH–H = 7.4 Hz,
H-7), 6.47 (1H, s, H-4), 3.80 (2H, s, H-15), 3.58 (3H, s, H-17)
ppm. 31P NMR (202.4 MHz) δ = 12.75 (PO3H2) ppm. 13C NMR
(125 MHz, BB-decoupling (24 h), missing C1 and JC–P) δ
=.179.5 (C3), 171.7 (C16), 150.8 (C12), 148.1 (C11), 145.1
(C2), 143.7 (C13), 142.7 (C14), 130.8 (C9), 129.9 (C7), 125.6
(C8), 119.5 (C6), 105.0 (C4), 52.4 (C17), 45.0 (C15) ppm (see
ESI† for atoms numbering). UV/Vis (ammonium acetate
10 mM), λmax = 231, 430–439 nm. MS (ESI) m/z (%): 442.27
[M − H]− (70), 464.15 [M−2H + Na]− (100), 291.4 (100); MS2
of 442.3 m/z yields 355.27 and 291.38 m/z. HRMS (ESI negative
mode): clcd = 442.0110 for C15H13N3O9PS; found 442.0128
Full mass spectra. Sheath gas pressure, 27 lb in−2; capillary
temperature 280 °C; spray voltage, 3.7 kV; and tube lens offset,
234 V. Parameters. Scan Mode Q1 MS; First mass 150 m/z and
last mass 1200 m/z; scan time 1 s; Q1 peak width 0.7.
Electrochemical instrumentation and measurements
Compounds were dissolved in a solution of N(nBu)4BF4
100 mM in DMF (10 mL) to reach a final concentration of
1 mM. Freshly made solutions were used directly. Cyclic voltam-
mograms (CVs) were recorded on a potentiostat EG&G model
283. Glassy carbon working electrode/Platinum foil counter
Tandem mass spectra. Sheath gas pressure, 27 lb in−2; capil-
lary temperature 280 °C; spray voltage, 3.7 kV; and tube lens
offset, 234 V. Parameters. Collision energy of 30 V; Q2 collision
gas pressure (mTorr) 1.00; scan time 1 s; Q1 and Q3 peak
widths 0.9.
1844 | Org. Biomol. Chem., 2012, 10, 1834–1846
This journal is © The Royal Society of Chemistry 2012

View Article Online
(d) H. Nakazawa, F. E. Chou, P. A. Andrews and N. R. Bachur, J. Org.
Chem., 1981, 46, 1493–1496.
Enzymatic oxidation for HPLC-MS studies
2 (a) G. W. Cavill, P. S. Clezy, J. R. Tetaz and R. L. Werner, Tetrahedron,
1959, 5, 275–280; (b) W. Schaefer, Prog. Org. Chem, 1964, 6, 135;
(c) M. Ionescu and H. Mantsch, Adv. Heterocycl. Chem., 1967, 8, 83;
(d) O. Crescenzi, G. Correale, A. Bolognese, V. Piscopo, M. Parrilli and
V. Barone, Org. Biomol. Chem., 2004, 2, 1577–1581 and references
therein.
3 (a) E. Delfourne, F. Darro, N. Bontemps-Subielos, C. Decaestacker,
J. Bastide, A. Frydman and R. Kiss, J. Med. Chem., 2001, 44, 3275;
(b) A. Bolognese, G. Correale, M. Manfra, A. Lavecchia, O. Mazzoni,
E. Novellino, V. Barone, P. La Colla and R. Loddo, J. Med. Chem., 2002,
45, 5217.
Briefly, the oxidations were run in 1 mL buffered water (0.2 M
ammonium acetate) containing 1 mM of substrates. Laccase
from Trametes versicolor (Aldrich, 38429) was purchased as a
brownish powder with an activity of 0.83 Umg−1. Unless other-
wise mentioned, laccase activity was of 100 UL−1. Samples
were periodically taken, diluted in acetonitrile (1/2, v/v) and
filtered on a syringe filter (0.22 μm) before injection in HPLC.
Typically reactions were ended after 24 h before being analyzed
by HPLC-MS and HPLC-PDA.
4 R. P. Maskey, F. C. Li, S. Quin and H. H. Fiebig, J. Antibiot., 2003, 56,
622–629.
5 (a) H. Suzuki, Y. Furusho, T. Higashi, Y. Ohnishi and S. Horinouchi,
J. Biol. Chem., 2006, 281, 824–833; (b) E. Graf, K. Schneider,
G. Nicholson, M. Strobele, A. L. Jones, M. Goodfellow, W. Beil, R.
D. Süssmuth and H. P. Fiedler, J. Antibiot., 2007, 60, 277–284.
6 G. Carr, W. Tay, H. Bottriell, S. K. Andersen, A. Grant Mauk and
R. J. Andersen, Org. Lett., 2009, 11, 2996–2999.
7 (a) P. B. Gomes, M. Nett, H. M. Dashe, I. Sattler, K. Martin and
C. Hertweck, Eur. J. Org. Chem., 2010, 231–235; (b) P. B. Gomes,
M. Nett, H. M. Dashe and C. Hertweck, J. Nat. Prod., 2010, 73, 1461–
1464.
RHF and DFT Computational methods
The internal coordinates of all the molecules have been fully
optimized at the RHF and B3LYP level using 6-31+G basis set
with added polarization functions (B3LYP/6-31+G(d).41,42 All
the calculations have been performed with the Gaussian 03
program.43
8 C. E. Barry, P. G. Nayar and T. P. Begley, Biochemistry, 1989, 28, 6323–
6333.
Rigid docking
9 (a) C. Eggert, U. Temp, J. F. D. Dean and K. E. L. Eriksson, FEBS Lett.,
1995, 376, 202–206; (b) J. Osiadacz, A. J. H. Al-Adhami,
D. Bajraszewska, P. Fischer and W. Peczynska-Czoch, J. Biotechnol.,
1999, 72, 141–149; (c) K. Li, P. S. Horanyi, R. Collins, R. S. Phillips and
K. E. L. Eriksson, Enzyme Microb. Technol., 2001, 28, 301–307.
10 M. Le Roes-Hill, C. Goodwin and S. Burton, Trends Biotechnol., 2009,
27, 248–258.
11 (a) S. Burton, Curr. Org. Chem., 2003, 7, 1317–1331; (b) S. Witayakran
and A. J. Ragauskas, Adv. Synth. Catal., 2009, 351, 1187–1209;
(c) F. Hollman, I. W. C. E. Arends, K. Buehler, A. Schallmey and
B. Bühler, Green Chem., 2011, 13, 226–265; (d) T. Kudanga,
G. S. Nyanhogo, G. M. Guebitz and S. Burton, Enzyme Microb. Technol.,
2011, 48, 195–208.
12 (a) F. Bruyneel, E. Enaud, L. Billottet, S. Vanhulle and J. Marchand-Bry-
naert, Eur. J. Org. Chem., 2008, 1, 71–79; (b) F. Bruyneel, O. Payen,
A. Rescigno, B. Tinant and J. Marchand-Brynaert, Chem.–Eur. J., 2009,
15, 8283–8295; (c) S. Forte, J. Polak, D. Valensin, M. Taddei, R. Basosi,
S. Vanhulle, A. Jarosz-Wilkolazka and R. Pogni, J. Mol. Catal. B:
Enzym., 2010, 63, 116–120; (d) F. Bruyneel, L. D’Auria, O. Payen, P.
J. Courtoy and J. Marchand-Brynaert, ChemBioChem, 2010, 11, 1451–
1457.
13 (a) S. Hajdok, H. Leutbecher, G. Greiner, J. Conrad and U. Beifuss, Tet-
rahedron Lett., 2007, 48, 5073–5076; (b) S. Witayakran and
A. J. Ragauskas, Green Chem., 2007, 9, 475–480; (c) S. Hajdok,
J. Conrad, H. Leutbecher, S. Strobel, T. Schleid and U. Beifuss, J. Org.
Chem., 2009, 74, 7230–7237; (d) V. Hahn, T. Davids, M. Lalk,
F. Schauer and A. Mikolasch, Green Chem., 2010, 12, 879–887.
14 (a) A. Rescigno and E. Sanjust, Atta-ur-Rahman (Ed.) Studies in Natural
Products Chemistry, Vol.26, Elsevier Science B. V., Amsterdam, 2002,
965–1028; (b) E. Sanjust, G. Cecchini, F. Sollai, N. Curreli and
A. Rescigno, Arch. Biochem. Biophys., 2003, 412, 272–278.
15 (a) E. I. Solomon, U. M. Sudaram and T. E. Machonkin, Chem. Rev.,
1996, 96, 2563–2605; (b) E. I. Solomon, R. K. Szilagi, S. D. George and
L. Basumallick, Chem. Rev., 2004, 104, 419–458; (c) L. Quintanar,
C. Stoj, A. B. Taylor, P. J. Hart, D. J. Kosman and E. I. Solomon, Acc.
Chem. Res., 2007, 40, 445–452 and reference herein (d) C. F. Thurston,
Microbiology, 1994, 140, 19–26.
The X-ray crystal structure of laccase from PDB ID 1kya (Tra-
metes versicolor) at a resolution of 2.4 Å was used for the active
site model. Molegro Virtual Docker program was used to prepare
the binding cavity. A docking template was set up to determine
the constraints (15 Å around the ligand of reference, the key
hydrogen bonds, the steric factors and the preferential position
of the aromatic ring). The docking was set up using the default
parameter (Moldock SE algorithm, Moldock score, similarity
score, RMSD of 1 Å). Ten runs were programmed for each
ligand from which the best 5 results were kept. The total
MolDock Score energy (arbitrary units) is the sum of internal
ligand energies, protein interaction energies and soft penalties.39
Acknowledgements
This research was supported in part by the European Commis-
sion, Sixth Framework Program (NMP2-CT2004-505899,
SOPHIED), and in part by the UCL (Belgium) with the Con-
certed Research Program ARC 08/13-009 and the Inter-Univer-
sity Attraction Pôle (IAP) program P6/19 PROFUSA. J.M.-B.
and G.D. are senior research associates of FRS-FNRS
(Belgium). The authors wish to thank R. Rozenberg (Laboratoire
de Spectrométrie de Masse, UCL) for technical assistance in the
acquisition of the MS and HPLC-MS spectra. G. D. thanks the
FRS-FNRS for access to HPC equipments installed in Liège and
LLN.
Notes and references
16 J. C. Freeman, P. G. Nayar, T. P. Begley and J. J. Villafranca, Biochemis-
try, 1993, 32, 4826–4830.
17 (a) D. J. Kosman, JBIC, J. Biol. Inorg. Chem., 2010, 15, 15–28;
(b) P. Giardina, V. Faraco, C. Pezzella, A. Piscitelli, S. Vanhulle and
G. Sannia, Cell. Mol. Life Sci., 2010, 67, 369–385.
18 (a) W. Smith, A. Camara-Artigas, M. Wang, J. P. Allen and W.
A. Francisco, Biochemistry, 2006, 45, 4378–4387.
19 (a) J. Reynisson and S. Steenken, Org. Biomol. Chem., 2004, 2, 578–
584; (b) M. A. Tadesse, A. D’Annibale, C. Galli, P. Gentili and F. Sergi,
Org. Biomol. Chem., 2008, 6, 868–878; (c) S. Bin Mohamad, A. Ling
Ong and A. Mat Ripen, Bioinformation, 2008, 2, 369–372.
‡As the HPLC-MS chromatographic system was equipped with a
smaller XTerraMsC18 column (2.1 × 50 mm, 2.5 μm), the same column
was used here with the HPLC-PDA systems. The reference dyes 5d, 5f,
6f, and 5j were eluted to check the consistency of the retention times
due to the decrease of both the flow rate and the column dimensions (see
Table 3, footnotes).
1 (a) S. A. Waksman and H. B. Woodruff, Proc. Soc. Exp. Biol. Med.,
1940, 45, 609–614; (b) H. Brockmann, Fortschr. Chem. Org. Naturst.,
1960, 18, 1–54; (c) H. Brockmann, Angew. Chem., 1960, 72, 939–947;
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1834–1846 | 1845

View Article Online
20 (a) K. Pionteck, M. Antorini and T. Choinowski, J. Biol. Chem., 2002,
277, 37663–37669; (b) T. Bertrand, C. Jolivalt, P. Briozzo, E. Caminade,
N. Joly, C. Madzak and C. Mougin, Biochemistry, 2002, 41, 7325–7333;
(c) F. J. Enguita, D. Marcal, L. O. Martins, R. Grenha, A. O. Henriques,
P. F. Lindley and M. A. Carrondo, J. Biol. Chem., 2004, 279, 23472–
23476; (d) N. Hakulinen, L.-L. Kiiskinen, K. Kruus, M. Sloheimo,
A. Paananen, A. Koivula and J. Rouvinen, Nat. Struct. Biol Letters,
2002, 9, 601–605; (e) I. Matera, A. Gullotto, S. Tilli, M. Ferraroni,
A. Scozzafava and F. Briganti, Inorg. Chim. Acta, 2008.
33 (a) T. M. Simandi, L. I. Simandi, M. Gyor, A. Rockenbauer and
A. Gomory, Dalton Trans., 2004, 1056–1060; (b) T. M. Simandi,
Z. May, I. Cs. Szigyarto and L. I. Simandi, Dalton Trans., 2005,
365–368.
34 O. Toussaint and K. Lerch, Biochemistry, 1987, 26, 5867–8571.
35 M. T. Wu and R. E. Lyle, J. Heterocycl. Chem., 1971, 8, 989–991.
36 (a) B. Brogni, D. Biglino, A. Sinicropi, E. J. Reijerse, P. Giardina,
G. Sannia, W. Lubitz, R. Basosi and R. Pogni, Phys. Chem. Chem. Phys.,
2008, 10, 7284–7292; (b) D. Nematollahi, D. Habibi, M. Rahmati and
M. Rafiee, J. Org. Chem., 2004, 69, 2637–2640.
21 H. Iwahashi, J. Chromatogr., Biomed. Appl., 1999, 736, 237–245.
22 M. Lahtinen, K. Kruus, H. Boer, M. Kemell, M. Andberg, L. Viikari and
J. Sipilä, J. Mol. Catal. B: Enzym., 2009, 57, 204–210.
37 F. D’Acunzo, C. Galli, P. Gentili and F. Sergi, New J. Chem., 2006, 30,
583–591.
23 G. I. Giles, C. A. Collins, T. W. Stone and C. Jacob, Biochem. Biophys.
Res. Commun., 2003, 300, 719–724.
24 (a) R. Sripriya, M. Chandrasekaran and M. Noel, Colloid Polym. Sci.,
2006, 285, 39–48; (b) C. Hansch, A. Leo and R. W. Taft, Chem. Rev.,
1991, 91, 165–195; (c) D. Job and H. B. Dunford, Eur. J. Biochem.,
1976, 66, 607–614.
38 (a) M. H. Hao, O. Haq and I. Muegge, J. Chem. Inf. Model., 2007, 47,
2242–2252; (b) K. A. Brameld, B. Kuhn, D. C. Reuter and M. Stahl,
J. Chem. Inf. Model., 2008, 48, 1–24.
39 Friesner et al., J. Med. Chem., 2004, 47, 1739–1749; see http://www.
molegro.com/mvd-technology.php for selected recent publications using
Molegro virtual docker program.
25 (a) M. Nakatani, M. Hibi, M. Minoda, J. Ogawa, K. Yokozeki and
S. Shimizu, New Biotechnol., 2010, 27, 317–323; (b) S. Kurniawati and
J. A. Nicell, Bioresour. Technol., 2008, 99, 7825–7834.
26 M. Trovaslet, E. Enaud, A. Baze, C. Jolivalt, F. Van Hove and
S. Vanhulle, Chemical Engineering Transaction, 2008, 14, 315.
27 (a) D. M. Soden and A. D. W. Dobson, Microbiology, 2001, 147, 1755–
1763; (b) P. Zucca, A. Rescigno, A. Olianas, S. Maccioni, F. A. Sollai
and E. Sanjust, J. Mol. Catal. B: Enzym., 2010, 68, 216–222.
28 M. T. Cambria, D. D. Marino, M. Falconi, S. Garavaglia and A. Cambria,
J. Biomol. Struct. Dyn., 2010, 27 (4), 501–509.
29 Rescigno, F. Bruyneel, A. Padiglia, F. Sollai, A. Salis, J. Marchand-Bry-
naert and E. Sanjust, Biochem. Biophys. Acta, 2011, 1810, 8, 799–807.
30 (a) J. J. Reddick, J. Cheng and W. Roush, Org. Lett., 2003, 5, 1967–
1970; (b) D. Enders and K. Hoffman, Eur. J. Org. Chem., 2009, 1665–
1668.
40 PyMOL(TM) Incentive Product - Copyright (C) 2006 DeLano Scientific
LLC.
41 M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, D. J. DeFrees, J.
A. Pople and M. S. Gordon, J. Chem. Phys., 1982, 77, 3654–3666.
42 T. Clark, J. Chandrasekhar, G. W. Spitznagel and P. v. R. Schleyer, J.
Comput. Chem., 1983, 4, 294–301.
43 Gaussian 03, Revision D.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma,
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,
S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaus-
sian Inc, Wallingford CT, 2004
31 L. R. Domingo, J. A. Saez, C. Palmucci, J. Sepulveda-Arques and
E. Gonzalez-Rosende, Tetrahedron, 2006, 62, 10408–10416.
32 (a) M. K. Manthey, S. G. Pyne and R. J. W. Truscott, Biochim. Biophys.
Acta, Gen. Subj., 1990, 1034, 207–212; (b) L. A. Hick, M. K. Manthey
and R. J. W. Truscott, J. Heterocycl. Chem., 1991, 28, 1157;
(c) S. Christen, P. T. Southwell-Keely and R. Stocker, Biochemistry, 1992,
31, 8090–8097; (d) K. Maruyama, T. Moriguchi, T. Mashino and
A. Nishinaga, Chem. Lett., 1996, 819–820.
1846 | Org. Biomol. Chem., 2012, 10, 1834–1846
This journal is © The Royal Society of Chemistry 2012