Synthesis and biological activity of analogues of vanchrobactin, a siderophore from Vibrio anguillarum serotype O2

Article abstract of DOI:10.1039/b719713f

Several analogues of vanchrobactin, a catechol siderophore isolated from the bacterial fish pathogen Vibrio anguillarum serotype O2 strain RV22, have been synthesized. The biological evaluation of these novel compounds showed that most of them are active as siderophores, as determined by growth promotion assays using the producer strain, as well as V. anguillarum serotype O1, Salmonella enterica, and Erwinia chrysanthemi. These compounds also gave a positive chrome azurol-S (CAS) test. On the basis of these results, we were able to deduce some structure-activity relationships. Furthermore, we found an analogue with siderophore activity that has appropriate functionality (an amino group) for use as an antibiotic vector to be employed in a "Trojan horse strategy". This journal is The Royal Society of Chemistry.

Full text of DOI:10.1039/b719713f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthesis and biological activity of analogues of vanchrobactin, a siderophore
from Vibrio anguillarum serotype O2
Raquel G. Soengas,a Marta Larrosa,a Miguel Balado,b Jaime Rodr´ıguez,*a Manuel L. Lemosb and
Carlos Jime´nez*a
Received 21st December 2007, Accepted 11th February 2008
First published as an Advance Article on the web 29th February 2008
DOI: 10.1039/b719713f
Several analogues of vanchrobactin, a catechol siderophore isolated from the bacterial fish pathogen
Vibrio anguillarum serotype O2 strain RV22, have been synthesized. The biological evaluation of these
novel compounds showed that most of them are active as siderophores, as determined by growth
promotion assays using the producer strain, as well as V. anguillarum serotype O1, Salmonella enterica,
and Erwinia chrysanthemi. These compounds also gave a positive chrome azurol-S (CAS) test. On the
basis of these results, we were able to deduce some structure–activity relationships. Furthermore, we
found an analogue with siderophore activity that has appropriate functionality (an amino group) for
use as an antibiotic vector to be employed in a “Trojan horse strategy”.
recognized serotypes, but those denoted O1 and O2 are the main
Introduction
ones implicated in vibriosis outbreaks.3 Although the virulence
All living creatures require iron in their cellular reactions. Since
iron is unavailable for uptake due to the low solubility of Fe3+
mechanisms of V. anguillarum are not fully understood, it is
known that the ability to scavenge iron through the utilization
at the neutral to alkaline pH conditions in soil or in aqueous
of siderophores contributes significantly to the virulence of this
bacterium.4 The best characterized siderophore produced by V.
media, organisms have developed different mechanisms to ac-
quire it from the environment. During infection, most bacteria
anguillarum is anguibactin, which was isolated from a strain
encounter within the host an iron-limited environment: higher
belonging to serotype O1. In the course of our studies into
eukaryotes contain substantial amounts of this metal but it is
tightly associated with transport and storage proteins and not
freely available for pathogens. Consequently, the level of free iron
in biological fluids is usually estimated to be only 10−18 M. In an
siderophores from marine microorganisms, a novel siderophore,
vanchrobactin (1c), was isolated (and characterized) from iron-
deficient cultures of V. anguillarum serotype O2 strain RV22.5
The stereochemistry of 1c was elucidated by chiral capillary
attempt to cope with this insolubility, many microorganisms have
electrophoresis analysis and total synthesis, showing the com-
pound to be N-[Nꢀ-(2,3-dihydroxybenzoyl)-D-arginyl]-L-serine.6
evolved specific iron uptake mechanisms. One of these mechanisms
involves the production of iron chelators named siderophores,
The increasing antibiotic-mediated selective pressure has led to the
which are low molecular weight (300 to 2000 Da) ferric-specific
emergence of multiresistant strains in many bacterial pathogens,
ligands. These molecules, which are generally excreted into the
and fish pathogens are no exception. Thus, more novel and
culture medium, are able to strongly chelate Fe3+ in a specific
efficient antimicrobial drugs are necessary to treat these emerging
manner to solubilize it and deliver it to the cells. In bacteria,
infections that are resistant to common antibiotics. One way as
the siderophore–iron(III) complex is recognized and transported
increase the efficiency of antimicrobials against bacterial infections
by specific outer membrane receptors. The uptake process is
could be the “Trojan horse strategy”, where antimicrobial drugs
thermodynamically driven from the cytoplasm through the TonB
are coupled to siderophores and transported across the bacterial
membranes through the iron uptake pathways.7 Such an approach
has already been developed and has given promising results with
the pyoverdine-mediated iron uptake system8 and conjugated
siderophore/b-lactamase inhibitors.9 Our working hypothesis was
that vanchrobactin from V. anguillarum could be a very interesting
candidate for such a prodrug strategy aimed at the development
of new antimicrobials against vibriosis and probably against other
related infectious diseases.
system that transduces the necessary energy from the cytosol to
the outer membrane receptor. Natural siderophores are usually not
very selective and many bacteria and fungi can utilize exogenous
siderophores made by different microorganisms.1
Vibrio anguillarum is the causative agent of vibriosis in fish, a
fatal hemorrhagic septicaemia. This disease affects marine and
freshwater fish species and results in considerable economic losses
in aquaculture farming worldwide.2 There are more than twenty
Previous examples of natural and synthetic catecholate-
type compounds based on amino acids or peptides, such as
chrysobactin,10 were shown to be active as siderophores. Although
those studies allowed some structure–activity correlations to be
deduced,11 very few systematic investigations into this type of
compound have been carried out.12 We report here the synthesis of
several analogues of vanchrobactin (1c) using the efficient, simple
and versatile synthesis that we reported recently.5 The aim was
aDepartamento de Qu´ımica Fundamental, Facultade de Ciencias, Universi-
dade da Corun˜a, Campus da Zapateira, 15071, A Corun˜a, Spain. E-mail:
carlosjg@udc.es, jaimer@ udc.es; Fax: +34 9811 67065; Tel: +34 9811
67065
bDepartamento de Microbiologia y Parasitologia, Instituto de Acuicultura,
Universidade de Santiago de Compostela, Santiago de Compostela, E-15782,
Spain. E-mail: mlemos@usc.es; Fax: +34 9815 47165; Tel: +34 9815 63100
1278 | Org. Biomol. Chem., 2008, 6, 1278–1287
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
to evaluate their siderophore activity and to study the structure-
siderophore activity relationships. Moreover, in view of its possible
use as an antibiotic vector, another goal was to obtain an analogue
with siderophore activity for V. anguillarum that incorporated
appropriate functionality (such as an amino group) for use as an
anchor to be attached directly (or via spacers) to other biologically
active agents, like antibiotics, for the design of iron transport-
mediated drug delivery.
ried out using Nd-Cbz-L-ornithine-OMe, which was prepared by
esterification of the commercially available Nd-Cbz-L-ornithine,16
to afford the enantiomeric acid 2a (79% yield). Both acids were
submitted to catalytic hydrogenation, giving rise to analogues
3a (DHBA-L-Orn) and 3b (DHBA-D-Orn) in quantitative yield.
Alternatively, protection of the acid 2b as the tert-butyl ester, cat-
alytic hydrogenation followed by the introduction of the guanidine
function by reaction with 1,3-di-Boc-2-(trifluoromethanesulfonyl)
guanidine,17 and acidic hydrolysis afforded analogue 4 (DHBA-D-
Arg) in 35% yield.
Both D- and L-serine tert-butyl esters were coupled with
compound 2a using TBTU, after removal of the benzyl and Cbz
protecting groups, to afford compounds 5a (28% yield) and 5b
(29% yield), respectively. Following the same procedure, L-serine
tert-butyl ester was coupled with 2b to give compound 5c in 82%
yield. Derivatives 5b and 5c were submitted to acidic hydrolysis
of the tert-butyl ester to afford analogues 6b (DHBA-L-Orn-L-
Ser) and 6c (DHBA-D-Orn-L-Ser), respectively, in quantitative
yields.
Results and discussion
In order to compare the influence of the configuration of the
asymmetric centres, the amino acid sequence and the presence of
the guanidine moiety on the siderophore activity of vanchrobactin
analogues, compounds 1–11 were prepared using a strategy similar
to that employed for the total synthesis of vanchrobactin.6
Synthesis of N-(2,3-dihydroxybenzoyl)-arginyl/ornithinyl
derivatives
Finally, guanidinylation of 5a, 5b and 5c as before followed
by acidic hydrolysis gave 1a (DHBA-L-Arg-D-Ser), 1b (DHBA-L-
Arg-L-Ser) and 1c (DHBA-D-Arg-L-Ser, vanchrobactin) in 73%,
62% and 61% yield, respectively.
Firstly, we studied the preparation of the analogues with the
DHBA directly linked to ornithine or arginine residues (Scheme 1),
i.e., analogues 1, 3, 4 and 6.
Nd-Cbz-D-ornithine-OMe, prepared from D-ornithine,13 was
coupled with 2,3-dibenzyloxybenzoic acid (DHBA)14 using TBTU
as the coupling agent to give the corresponding ester in very good
yield. The resulting ester was hydrolyzed with barium hydroxide,
giving the acid 2b in a 78% yield.15 The same sequence was car-
Synthesis of 2,3-dihydroxybenzoyl-serinyl derivatives
In an effort to evaluate the influence of the amino acid se-
quence on the siderophore activity, we designed the synthesis of
Scheme 1 Reagents and conditions: i: 1. SOCl2/MeOH (89%); 2. BnBr/K2CO3 (99%); 3. Ba(OH)2, THF, H2O, 50 ◦C, 12 h, quant.; 4.
Nd-Cbz-L-ornithine-OMe or Nd-Cbz-D-ornithine-OMe, TBTU, Et3N, DMF, rt, 16 h (78% and 79%, respectively); 5. Ba(OH)2, THF, H2O, 50 ◦C,
16 h, quant.; ii: H2, Pd/C, MeOH, rt, 3 h, quant.; iii: 1. t-BuOAc, HClO4; 2. H2, Pd/C, MeOH, rt, 3 h, (48%); 3. (NHBoc)2NTf, Et3N, CHCl3, rt, 1 h
(73%); 4. TFA/DCM 3:7, rt, 16 h, quant.; iv: 1. L-Ser-OtBu or D-Ser-OtBu, TBTU, Et3N, DMF, rt, 16 h; 2. H2, Pd/C, MeOH, rt, 3 h (28%, 29% and
82%); v: TFA/DCM 3:7, rt, 16 h, quant.; vi: 1. (NHBoc)2NTf, Et3N, CHCl3, rt, 1 h (73%, 62% and 61%); 2. TFA/DCM 3:7, rt, 16 h, quant.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1278–1287 | 1279
©

View Article Online
Scheme 2 Reagents and conditions: i: 1. SOCl2/MeOH (89%); 2. BnBr/K2CO3 (99%); 3. Ba(OH)2, THF, H2O, 50 ◦C, 12 h, quant.; 4. L-Ser-OMe or
D-Ser-OMe, TBTU, Et3N, DMF, rt, 16 h (84% and 82% respectively); 5. Ba(OH)2, THF, H2O, 50 ◦C, 16 h, quant.; ii: H2, Pd-C, MeOH, rt, 3 h, quant.; iii:
Nd-Cbz-L-ornithine-OtBu, TBTU, Et3N, DMF, rt, 16 h, (81% and 93% respectively); iv: 1. H2, Pd/C, MeOH, rt, 3 h, quant.; 2. TFA/DCM 3:7, rt, 16 h,
quant.; v: 1. (NHBoc)2NTf, Et3N, CHCl3, rt, 1 h (71% and 79% respectively); 2. TFA/DCM 3:7, rt, 16 h, quant.
2,3-dihydroxybenzoyl derivatives linked directly to a serine (in-
stead of arginine residues) using a similar strategy to that described
above (analogues 8, 10 and 11)
siderophore enterobactin but that can use vanchrobactin as an
exogenous siderophore).22 The results of these growth promotion
assays are given in Table 1.
Coupling of the L- and D-serine methyl esters with 2,3-
dibenzyloxybenzoic acid using TBTU, followed by hydrolysis
of the corresponding esters with barium hydroxide, afforded
compounds 7a (84% yield) and 7b (82% yield), respectively.
Catalytic hydrogenation of acids 7a and 7b gave analogues 8a
(DHBA-L-Ser)18 and 8b (DHBA-D-Ser) in quantitative yields.
The Nd-benzyloxycarbonyl-L- and D-ornithine tert-butyl esters19
were coupled with compound 7a using TBTU, and the resulting
products were submitted to catalytic hydrogenation to give com-
pounds 9a (81% yield) and 9b (83% yield), respectively. Acidic
hydrolysis of the tert-butyl ester afforded analogues 10a (DHBA-
L-Ser-L-Orn) and 10b (DHBA-L-Ser-D-Orn), both in quantitative
yield..
Growth promotion of Vibrio anguillarum strains
All analogues were tested for growth promotion under iron-
deficient conditions for two strains of V. anguillarum: (a) one
strain (RV22 from serotype O2) from which vanchrobactin was
isolated, and (b) one strain (775 from serotype O1) that produces
the siderophore anguibactin. All analogues stimulated growth of
both strains with the exception of 3a (DHBA-L-Orn), which was
inactive. These results are consistent with the catecholate and
amino acidic structural nature of the natural siderophores biosyn-
thesized by both strains. Furthermore, they indicate that these
microorganisms are not very specific in the use of siderophores
and they can use several different ones to supply iron to the cell,
even siderophores with very different structures.
Finally, guanidinylation of 9a and 9b as before followed by
acidic hydrolysis gave 11a (DHBA-L-Ser-L-Arg) and 11b (DHBA-
L-Ser-D-Arg) in 71% and 69% yield, respectively.
The natural siderophore 1c, along with its stereoisomers 1a and
1b, having the sequence DHBA-Arg-Ser, was found to be the most
active compound in the growth promotion of V. anguillarum RV22
(serotype O2). The similar values obtained for the three analogous
stereoisomers 1a–c indicate that the absolute configuration of the
amino acid moieties has little influence on the siderophore activity.
Interestingly, compounds 6b and 6c, which have an ornithine
instead of an arginine residue, also promoted the growth of both
strains under iron-deficient conditions. This result could be very
useful because these vanchrobactin derivatives possess an amino
group that can be attached to other bioactive agents in the Trojan
horse strategy. Compounds in which DHBA is linked directly
to serine and to D-ornithine were found to be the most active
analogues in the growth promotion of V. anguillarum 775 (serotype
O1). These differences between the two strains could be due to the
different outer membrane proteins involved in ferric-siderophore
transport in strains RV22 and 775.3
Biological evaluation
Compounds 1, 3, 4, 6, 8, 10 and 11 were evaluated for their
siderophore activities by growth promotion assays with several
bacteria that are well defined in their ability to transport and utilize
natural siderophores (siderophore indicator strains). The follow-
ing indicator strains were used: Vibrio anguillarum serotype O2
RV22 (strain producing vanchrobactin), V. anguillarum serotype
O1 775 (strain producing siderophore anguibactin, but that
can also use vanchrobactin as siderophore), Vibrio alginolyticus
TA1520 (strain producing a hydroxamate siderophore that cannot
use vanchrobactin as an exogenous siderophore), Erwinia chrysan-
themi PPV20 (strain that produces chrysobactin, a siderophore
structurally related to vanchrobactin)21 and Salmonella enter-
ica enb-1 (a mutant strain that does not produce the natural
1280 | Org. Biomol. Chem., 2008, 6, 1278–1287
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Growth promotion of Vibrio alginolyticus TA15
In order to test the specificity of each analogue, all compounds
were tested for growth promotion of a strain belonging to the
Vibrionaceae family, but producing a different type of siderophore.
The lack of biological activity in V. alginolyticus TA15 of all
analogues probably confirms the presence of a completely different
structure for the still unknown natural siderophore synthesized
by this pathogen. In fact, a recent study suggested a structure
that is closely related to vibrioferrin,23 which is a non-catecholate
siderophore belonging to a member of the carboxylate class of
siderophores and contains two a-hydroxy acid groups.24
Growth promotion of Erwinia chrysanthemi PPV20
This strain produces chrysobactin, a catechol siderophore bearing
lysine and serine residues linked to DHBA.21 Thus, the only
difference in the structure with respect to vanchrobactin is the
substitution of the arginine by a lysine residue. Due to these highly
related structures, we tested whether E. chrysanthemi could use
vanchrobactin and its analogues as siderophores. The synthetic
compounds in which the DHBA residue is directly linked to
ornithine or arginine were the most active analogues for the
growth promotion of E. chrysanthemi. However, those analogues
in which the DHBA is linked directly to serine showed weaker
growth promotion ability or were inactive. This result suggests
that the ferric-siderophore receptors present in E. chrysanthemi
are probably more structure-specific than those present in V.
anguillarum.
Growth promotion of Salmonella enterica serovar Typhimurium
enb-1
All salmonellas produce the well-known siderophore enterobactin,
which is a cyclic trimer of 2,3-dihydroxy-N-benzoyl-L-serine. Our
previous studies3 showed that the enterobactin-deficient mutant
enb-1 can use vanchrobactin to overcome iron limitation. We
wished to determine whether this utilization is exclusive for the nat-
ural siderophore. The results showed that besides vanchrobactin,
other derivatives containing ornithine instead of arginine were also
active as growth promoters under iron-limiting conditions.
Siderophore activity of vanchrobactin analogues in CAS assays
In parallel to the growth promotion assays, the relative iron com-
plexing capacities of the siderophore derivatives were evaluated
by reactivity with chrome azurol-S (CAS),25 and the results are
shown in Table 1. Most analogues gave a positive reaction in this
test with a range of values from −0.353 for 3a to −0.788 for 8b; the
exceptions are compounds 10a and 10b, which lack reactivity for
unknown reasons. It is worth noting that a high CAS reactivity
does not necessarily reflect high growth promotion ability. This
is explained by the fact that the CAS test is based on the iron
chelating activity of each compound, while the growth promotion
assays reflect the ability of each strain to actually transport the
different compounds into the cell, a process that is related to
the recognition of each ferric-compound by an appropriate outer
membrane receptor. Thus, there is generally a qualitative match
between the two methods but not a quantitative correlation, and
high biological activity can be seen along with relatively low CAS
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1278–1287 | 1281
©

View Article Online
reaction and vice versa. This same observation was previously
reported for analogues of exochelin from Mycobacterium.26
acid starting material (Rf 0.0) to a major product (Rf 0.32). The
solvent was removed in vacuo and the residue was preabsorbed
onto silica and purified by flash column chromatography (ethyl
acetate–hexane 1:2→1:1) to yield Na-2,3-dibenzyloxybenzoyl-Nd-
benzyloxycarbonyl-L-ornithine methyl ester (0.31 g) as a clear oil,
which was immediately dissolved in THF (10 mL) and water
(10 mL). To the resulting solution, barium hydroxide (0.27 g,
1.58 mmol) was added and the mixture was stirred for 16 h at
50 ◦C, when t.l.c. (ethyl acetate–hexane 1:1) indicated complete
conversion of starting material (Rf 0.32) to a major product (Rf
0.0). DOWEX 50 W (H+) was added to the solution to give a
pH = 6 and the resin was then removed by filtration. The filtrate
was evaporated in vacuo to give Na-2,3-dibenzyloxybenzoyl-Nd-
benzyloxycarbonyl-L-ornithine (2a) (0.29 g, 0.51 mmol, 79%),
which was used without further purification.
Conclusions
Very few systematic investigations into structure–activity corre-
lations of siderophores have been performed to date on amino
acids and dipeptides. We have prepared a new series of analogues
of vanchrobactin (1c) in order to determine the influence on the
iron transport activities of the configuration of the asymmetric
centres, the presence of the amino acid residues linked to the
dihydroxybenzoic acid moiety, and the presence of the guanidine
or amine moiety on the arginine or ornithine residues. The
results reveal the lack of stereoisomeric influence of the amino
acid scaffold on the siderophore activity in both serotypes of V.
anguillarum, although some differences can be observed in Erwinia
chrysanthemi PPV20 and Salmonella enterica serovar Typhimurium
enb-1.
Na-2,3-Dibenzyloxybenzoyl-L-ornithine (3a)
Na -2,3-Dibenzyloxybenzoyl-Nd -benzyloxycarbonyl-L-ornithine
(2a) (82 mg, 0.12 mmol) was dissolved in methanol (4 mL)
and the flask was flushed with argon. 5% Pd/C (0.01 g) was
added and the flask was again flushed with argon. The reaction
mixture was flushed with hydrogen and stirred under a hydrogen
atmosphere for 3 h, after which t.l.c. (dichloromethane–methanol
95:5) indicated complete consumption of the starting material.
The reaction mixture was filtered through Celite and the solvent
was removed in vacuo to afford Na-2,3-dibenzyloxybenzoyl-L-
ornithine (3a) (65 mg, 0.15 mmol, quantitative), which was used
without further purification; [a]D23: +2.3 (c, 0.8 in H2O); 1H NMR
(200 MHz, D2O) dH: 1.73–1.98 (m, 4H, OrnH-3, OrnH-3ꢀ, OrnH-
4, OrnH-4ꢀ), 2.91–2.99 (m, 2H, OrnH-2, OrnH-2ꢀ), 4.43–4.45 (m,
1H, OrnH-1), 6.86–7.42 (m, 3H, DHBH-4, DHBH-5, DHBH-6);
13C NMR (50 MHz, D2O): 25.33, 30.88, 40.44 (OrnC-3, OrnC-
4, OrnC-5), 49.88 (OrnC-2), 118.91, 128.79, 128.75 (DHBC-4,
DHBC-5, DHBC-6), 129.52, 135.12, 135.14 (DHBC-1, DHBC-
2, DHBC-3), 170.32 (CO2H); (+)-LRESIMS (m/z, %): 269.11
([M + H]+, 100%); (+)-HRESIMS: calculated for C12H17N2O7
[M + H]+ 269.1131, found 269.1129.
V. anguillarum and other Gram-negative bacteria tested were
able to use several different analogues as siderophores, with the
exception of V. alginolyticus TA15, which did not use any of
them. The siderophore activity found for most of the synthetic
analogues in the growth promotion of Erwinia chrysanthemi
PPV20 is consistent with the catecholate amino acid structure of
chrysobactin, its natural siderophore. This finding is in agreement
with previous observations concerning the importance of the
aromatic ring in catecholate siderophores binding to the outer
membrane receptors. In this regard, the Escherichia coli FepA
receptor transports certain catecholate ferric siderophores but not
others, nor any noncatecholate compounds.27.
Our observations confirm previous reports in the sense that
microorganisms are not always selective in the use of siderophores
and they can use several different ones to supply iron to the
cell, even siderophores with very different structures. This lack
of specificity could be an evolutionary adaptation of bacterial
pathogens to increase the number of possibilities to obtain iron
from the environment.
A vanchrobactin analogue has been prepared that has
siderophore activity and an amino functionality that could be
used as an anchor group to attach it to other bioactive agents.
This opens new and interesting possibilities for the use of the
Trojan horse strategy to develop new drugs that are effective
against vibriosis and other related infections. The relatively low
specificity of siderophore utilization could enable the design of
drugs that are effective against different Gram-negative bacteria.
Experiments aimed at this objective are currently under way.
Na-2,3-Dihydroxybenzoyl-L-ornithinyl-L-serine tert-butyl
ester (5b)
L-Serine tert-butyl ester (0.09 g, 0.44 mmol) was coupled to 2,3-
dibenzyloxybenzoyl-Nd-Cbz-L-ornithinic acid (2a) (0.06 g, 0.12
mmol) in anhydrous DMF (2 mL) and triethylamine (0.15 mL,
1.11 mmol) using TBTU as coupling agent (0.21 g, 0.67 mmol).
The solvent was removed in vacuo and the residue was preabsorbed
onto silica and purified by flash column chromatography (ethyl
acetate–hexane 1:2→2:3) to yield Na-2,3-dibenzyloxybenzoyl-Nd-
benzyloxycarbonyl-L-ornithinyl-L-serine tert-butyl ester, which
was hydrogenated in methanol (5 mL) using 5% Pd–C (0.01 g) as
catalyst. The reaction mixture was filtered through Celite and the
solvent was removed in vacuo to afford Na-2,3-dihydroxybenzoyl-
L-ornithinyl-L-serine tert-butyl ester (5b) (54 mg, 0.14 mmol, 28%),
which was used without further purification; [a]D25: +15.8 (c, 0.3 in
CHCl3); 1H NMR (200 MHz, CDCl3) dH: 1.42 (s, 9H, -C(CH3)3),
1.67–1.91 (m, 4H, OrnH-3, OrnH-3ꢀꢀ, OrnH-4, OrnH-4ꢀ), 2.94–
2.96 (m, 2H, OrnH-5, OrnH-5ꢀ), 3.52 (m, 1H, SerH-3), 3.76 (m,
1H, SerH-3ꢀ), 4.55–4.67 (m, 2H, OrnH-2, SerH-2), 6.61 (bs, 1H,
Experimental
Na-2,3-Dibenzyloxybenzoyl-Nd-benzyloxycarbonyl-L-ornithine
(2a)
Nd-Benzyloxycarbonyl-L-ornithine methyl ester hydrochloride
(0.19 g, 0.66 mmol) and TBTU (0.32 g, 0.99 mmol) were added to a
stirred solution of 2,3-dibenzyloxybenzoic acid (0.22 g, 0.66 mmol)
in anhydrous DMF (6 mL). After 10 min, triethylamine (0.23 mL,
1.65 mmol) was added and the reaction mixture was stirred at
room temperature for 20 h under an atmosphere of argon. T.l.c.
(ethyl acetate–hexane 1:1) indicated complete conversion of the
1282 | Org. Biomol. Chem., 2008, 6, 1278–1287
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
-NH-), 7.66–7.91 (m, 3H, DHBH-4, DHBH-5, DHBH-6), 8.22
(bs, 1H, -NH-); 13C NMR (50 MHz, CDCl3) dC: 23.21 (OrnC-
4), 27.30, 27.97 (-C(CH3)3), 27.97 (OrnC-5), 52.13 (OrnC-3),
53.15, 62.24 (OrnC-2, SerC-2), 73.31 (SerC-3), 82.11 (-C(CH3)3),
118.41, 119.00, 119.19 (DHBC-5, DHBC-6, DHBC-7), 145.25,
(+)-HRESIMS: calculated for C16H24N5O97 [M + H]+ 398.1670,
found 398.1674.
Na-2,3-Dihydroxybenzoyl-L-ornithinyl-D-serine tert-butyl ester
(5a)
148.37 (DHBC-2, DHBC-3, DHBC-4), 169.41, 169.80, 171.63
+
=
(3 × C O); (+)-LRFABMS (m/z, %): 434.2 [M + Na] , 412.2
D-Serine tert-butyl ester (0.10 g, 0.53 mmol) was cou-
pled with Na-2,3-dibenzyloxybenzoyl-Nd-benzyloxycarbonyl-L-
ornithine (2a) in a similar way as D-serine tert-butyl es-
ter to yield Na-2,3-dibenzyloxybenzoyl-Nd-benzyloxycarbonyl-L-
ornithinyl-D-serine tert-butyl ester (0.11 g), which was hydro-
genated in a similar way as the corresponding L-ornithinyl-L-
serine derivative to afford Na-2,3-dihydroxybenzoyl-L-ornithinyl-
D-serine tert-butyl ester (5a) (57 mg, 0.14 mmol, 29%), which was
used without further purification; [a]D24: −12.0 (c, 1.5 in H2O); 1H
NMR (200 MHz, CDCl3) dH: 1.40 (s, 9H, -C(CH3)3), 1.49–2.47
(m, 6H, OrnH-3, OrnH-3ꢀ, OrnH-4, OrnH-4ꢀ, OrnH-5, OrnH-5ꢀ),
3.40–3.71 (m, 2H, SerH-3, SerH-3ꢀ), 4.45–4.53 (m, 2H, SerH-2,
ArgH-2), 6.28–6.34 (m, 1H, DHBH-5), 6.71–7.27 (m, 2H, DHBH-
4, DHBH-6); (+)-LRFABMS (m/z, %): 434.2 [M + Na]+, 412.2
[M + H]+.
[M + H]+.
Na-2,3-Dihydroxybenzoyl-L-ornithinyl-L-serine (6b)
Na-2,3-Dihydroxybenzoyl-L-ornithinyl-L-serine tert-butyl ester
(5b) (31 mg, 0.075 mmol) was stirred at room temperature in a
solution of TFA and DCM (3:7, 1 mL). After 16 h, evaporation
of the solvent under reduced pressure and coevaporation with
toluene afforded Na-2,3-dihydroxybenzoyl-L-ornithinyl-L-serine
26
(6b) (25 mg, 0.07 mmol, quantitative) as a dark yellow oil; [a]D
:
1
−8.0 (c, 0.11 in H2O); H NMR (200 MHz, D2O) dH: 1.78–2.09
(m, 4H, OrnH-3, OrnH-3ꢀ, OrnH-4, OrnH-4ꢀ), 3.04–3.09 (m, 2H,
OrnH-5, OrnH-5ꢀ), 3.90 (dd, 1H, JH3,H2 = 4.0 Hz, JH3,H3 = 11.7 Hz,
ꢀ
ꢀ
ꢀ
ꢀꢀ
SerH-3), 3.99 (dd, 1H, JH3 ,H2 = 5.1 Hz, JH3,H3 = 11.7 Hz, SerH-3 ),
4.57–4.60, 4.68–4.70 (2 × m, 2H, SerH-2, OrnH-2), 6.89 (t, 1H,
JH4,H5 = JH5,H6 = 8.1 Hz, DHBH-5), 7.10, 7.30 (2 × d, 2H, JH4,H5
=
JH5,H6 = 8.1 Hz, DHBH-4, DHBH-6); 13C NMR (50 MHz,
D2O) dC: 23.93 (OrnC-4), 28.85 (OrnC-3), 39.56 (OrnC-5), 53.94,
55.49 (OrnC-2, SerC-2), 61.59 (SerC-3), 120.12, 120.33, 120.44
(DHBC-4, DHBC-5, DHBC-6), 117.44, 145.16, 147.20 (DHBC-1,
DHBC-2, DHBC-3), 170.61, 173.66 (2 × CONH), 174.18 (CO2H);
(+)-LRESIMS (m/z, %): 356.14 ([H]+, 100%); (+)-HRESIMS
calculated for C15H22N3O7 [M + H]+ 356.1452, found 356.1457.
2,3-Dihydroxybenzoyl-L-argininyl-D-serine (1a)
Na-2,3-Dihydroxybenzoyl-L-ornithinyl-D-serine tert-butyl ester
(5a) (57 mg, 0.15 mmol) was reacted with 1,3-di-tert-butoxycarbo-
nyl-2-trifluoromethanesulfonylguanidine (52 mg, 0.13 mmol) in a
similar way as 5b to yield Na-2,3-dihydroxybenzoyl-Nx-tert-bu-
toxycarbonyl-Nxꢀ-tert-butoxycarbonyl-L-argininyl-D-serine tert-
butyl ester (60 mg) which was hydrolysed in a similar way as the
L-argininyl-L-serine derivative to afford 2,3-dihydroxybenzoyl-L-
argininyl-D-serine (1a) (27 mg, 0.06 mmol, 62%) as a dark yellow
oil; [a]D24: −16.4 (c, 0.7 in MeOH); 1H NMR (200 MHz, D2O) dH:
1.56–1.94 (m, 4H, ArgH-3, ArgH-3ꢀ, ArgH-4, ArgH-4ꢀ), 3.10–3.14
ꢀ
2,3-Dihydroxybenzoyl-L-argininyl-L-serine (1b)
To a stirred solution of Na-2,3-dihydroxybenzoyl-L-ornithinyl-
L-serine tert-butyl ester (5b) (29 mg, 0.07 mmol) in
chloroform (1 mL) was added 1,3-di-tert-butoxycarbonyl-2-
(trifluoromethanesulfonyl)guanidine (26 mg, 0.065 mmol) and
triethylamine (0.01 mL, 0.07 mmol) and the resulting mix-
ture was stirred at room temperature. After 1 h, methanol
(0.5 mL) was added and the mixture was stirred for a further
30 min. The mixture was then preabsorbed onto silica and
purified by flash column chromatography (methanol–DCM 5%) to
yield Na-2,3-dihydroxybenzoyl-Nx-tert-butoxycarbonyl-Nxꢀ-tert-
butoxycarbonyl-L-argininyl-L-serine tert-butyl ester, which was
then stirred at room temperature in a solution of TFA and
DCM (3:7, 1 mL). After 16 h, evaporation of the solvent under
reduced pressure and coevaporation with toluene afforded 2,3-
dihydroxybenzoyl-L-argininyl-L-serine (1b) (25 mg, 0.06 mmol,
(m, 2H, ArgH-5, ArgH-5ꢀ), 3.80 (dd, 1H, JH3,H2 = 3.9 Hz, JH3,H3
=
ꢀ
ꢀ
10.1 Hz, SerH-3), 3.80 (dd, 1H, JH3 ,H2 = 5.1 Hz, JH3,H3 = 10.1 Hz,
SerH-3ꢀ), 4.47–4.57 (m, 2H, SerH-2, ArgH-2), 6.74 (t, 1H, JH4,H5
=
=
JH5,H6 = 8.1 Hz, DHBH-5), 6.96, 7.17 (2 × d, 2H, JH4,H5 = JH5,H6
8.1 Hz, DHBH-4, DHBH-6); 13C NMR (50 MHz, D2O) dC: 24.98
(ArgC-4), 29.02 (ArgC-3), 41.13 (ArgC-5), 54.13, 55.48 (ArgC-2,
SerC-2), 61.60 (SerC-3), 120.17, 120.32, 120.46 (DHBC-4, DHBC-
5, DHBC-6), 117.41, 145.19, 147.17 (DHBC-1, DHBC-2, DHBC-
3), 157.32 (C=N), 170.51, 173.72 (2 × CONH), 174.42 (CO2H);
(+)-LRESIMS (m/z, %): 398.10 ([M + H]+, 84%); (+)-HRESIMS:
calculated for C16H24N5O97 [M + H]+ 398.1670, found 398.1674.
Na-2,3-Dibenzyloxybenzoyl-Nd-benzyloxycarbonyl-D-ornithine
(2b)
73%) as a dark yellow oil; [a]D24: +12.0 (c, 0.7 in MeOH); H
1
NMR (200 MHz, D2O) dH: 1.61–1.95 (m, 4H, ArgH-3, ArgH-3ꢀ,
ArgH-4, ArgH-4ꢀ), 3.13–3.17 (m, 2H, ArgH-5, ArgH-5ꢀ), 3.78–
3.97 (m, 2H, SerH-3, SerH-3ꢀ), 4.43–4.59 (m, 2H, SerH-2, ArgH-
2), 6.77 (t, 1H, JH4,H5 = JH5,H6 = 8.1 Hz, DHBH-5), 6.99, 7.21 (2 ×
d, 2H, JH4,H5 = JH5,H6 = 8.1 Hz, DHBH-4, DHBH-6); 13C NMR
(50 MHz, D2O) dC: 24.92 (ArgC-4), 29.10 (ArgC-3), 41.16 (ArgC-
5), 54.04, 55.44 (ArgC-2, SerC-2), 61.63 (SerC-3), 117.35, 120.16,
120.41 (DHBC-4, DHBC-5, DHBC-6), 145.18, 147.25 (DHBC-1,
DHBC-2, DHBC-3), 157.30 (C=N), 170.39, 173.70 (2 × CONH),
174.39 (CO2H); (+)-LRESIMS (m/z, %): 398.10 ([M + H]+, 84%);
Nd-Benzyloxycarbonyl-D-ornithine methyl ester (0.40 g, 1.39
mmol) was coupled with 2,3-dibenzyloxybenzoic acid in
a
similar way as the corresponding L-ornithine deriva-
tive to yield Na-2,3-dibenzyloxybenzoyl-Nd-benzyloxycarbonyl-
D-ornithine methyl ester (0.65 g), which was hydrolysed in a
similar way as the corresponding L-ornithine derivative to give Na-
2,3-dibenzyloxybenzoyl-Nd-benzyloxycarbonyl-D-ornithine (2b)
(0.60 g, 1.06 mmol, 79%), which was used without further
purification.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1278–1287 | 1283
©

View Article Online
Na-2,3-Dihydroxybenzoyl-D-ornithine (3b)
in a similar way as 2a to yield Na-2,3-dibenzyloxybenzoyl-Nd-
benzyloxycarbonyl-D-ornithinyl-L-serine tert-butyl ester, which
was hydrogenated as previously reported for the L-ornithinyl-L-
serine derivative to afford Na-2,3-dihydroxybenzoyl-D-ornithinyl-
L-serine tert-butyl ester (5c) (0.33 g, 0.83 mmol, 82%), which was
used without further purification; [a]D23: +14.6 (c, 0.14 in Cl3CH);
1H NMR (200 MHz, CDCl3) dH: 1.40 (s, 9H, -C(CH3)3), 1.49–2.47
(m, 6H, OrnH-3, OrnH-3ꢀ, OrnH-4, OrnH-4ꢀ, OrnH-5, OrnH-5ꢀ),
3.40–3.71 (m, 2H, SerH-3, SerH-3ꢀ), 4.45–4.53 (m, 2H, SerH-2,
ArgH-2), 6.28–6.34 (m, 1H, DHBH-5), 6.71–7.27 (m, 2H, DHBH-
4, DHBH-6); (+)-LRFABMS (m/z, %): 434.2 [M + Na]+, 412.2
[M + H]+.
Na -2,3-Dibenzyloxybenzoyl-Nd-benzyloxycarbonyl-D-ornithine
(2b) (80 mg, 0.11 mmol) was hydrogenated in a similar way as
2a to afford Na-2,3-dihydroxybenzoyl-D-ornithine (3b) (65 mg,
0.11 mmol, quantitative), which was used without further pu-
rification; [a]D23: −4.6 (c, 0.55 in D2O); 1H NMR (200 MHz,
D2O) dH: 1.73–1.98 (m, 4H, OrnH-3, OrnH-3ꢀ, OrnH-4, OrnH-
4ꢀ), 2.91–2.99 (m, 2H, OrnH-2, OrnH-2ꢀ), 4.43–4.45 (m, 1H,
OrnH-1), 6.86–7.42 (m, 3H, DHBH-4, DHBH-5,DHBH-6); 13C
NMR (50 MHz, D2O) dC: 25.33, 30.88, 40.44 (OrnC-3, OrnC-
4, OrnC-5), 49.88 (OrnC-2), 118.91, 128.79, 128.75 (DHBC-4,
DHBC-5, DHBC-6), 129.52, 135.12, 135.14 (DHBC-1, DHBC-
2, DHBC-3), 170.32 (CO2H); (+)-LRESIMS (m/z, %): 269.11
([M + H]+, 100%); (+)-HRESIMS: calculated for C12H17N2O7 [M
+ H]+ 269.1131, found 269.1129.
Na-2,3-Dihydroxybenzoyl-D-ornithinyl-L-serine (6c)
Na-2,3-Dihydroxybenzoyl-D-ornithinyl-L-serine tert-butyl ester
(5c) (45 mg, 0.11 mmol) was hydrolysed in a similar way as
2,3-Dihydroxybenzoyl-D-arginine (4)
5a to afford Na-2,3-dihydroxybenzoyl-D-ornithinyl-L-serine (6c)
Na -2,3-Dibenzyloxybenzoyl-Nd -benzyloxycarbonyl-D-ornithine
(2b) (0.39 g, 0.69 mmol) was suspended in tert-butyl acetate
(2.30 mL, 16.70 mmol) and perchloric acid (70%, 0.07 mL, 0.77
mmol) was added. The mixture was stirred at room temperature
for 16 h and added to saturated aqueous sodium bicarbonate
(10 mL). After stirring for a further 30 min, the mixture was
extracted with diethyl ether (3 × 20 mL) and the combined
organic layers were dried, filtered and evaporated in vacuo
to obtain Na-2,3-dibenzyloxybenzoyl-Nd-benzyloxycarbonyl-D-
ornithine tert-butyl ester (0.44 g, 48%) as a clear oil, which was
hydrogenated using 5% Pd-C (0.07 g) in a similar way as 2a
to afford Na-2,3-dibenzyloxybenzoyl-D-ornithine tert-butyl ester,
which was used immediately in the next reaction. This D-ornithine
derivative ester was treated with 1,3-di-tert-butoxycarbonyl-2-
trifluoromethanesulfonyl guanidine (0.12 g, 0.30 mmol) in a
similar way as 5b. The product mixture was then preab-
sorbed onto silica and purified by flash column chromatogra-
phy (methanol/DCM 5%), to yield Na-2,3-dibenzyloxybenzoyl-
Nx-tert-butoxycarbonyl-Nxꢀ-tert-butoxycarbonyl-D-arginine tert-
butyl ester (0.133 g), which was then stirred at room temperature
in a solution in TFA and DCM (3:7, 1 mL). After 16 h the
solvent was evaporated under reduced pressure and coevapora-
tion with toluene afforded 2,3-dihydroxybenzoyl-D-arginine (4)
(74 mg, 0.25 mmol, 73%) as a dark yellow oil; [a]D24: +3.5 (c,
2.3 in H2O); 1H NMR (200 MHz, D2O) dH: 1.50–1.60, 1.68–
1.94 (2 × m, 4H, ArgH-3, ArgH-3ꢀ, ArgH-4, ArgH-4ꢀ), 3.03–
22
(36 mg, 0.11 mmol, quantitative) as a dark yellow oil; [a]D
:
+8.7 (c, 0.16 in H2O); 1H NMR (200 MHz, D2O) dH: 1.71–
2.04 (m, 4H, OrnH-3, OrnH-3ꢀ, OrnH-4, OrnH-4ꢀ), 3.00–3.05
(m, 2H, OrnH-5, OrH-5ꢀ), 3.86 (dd, 1H, JH3,H2 = 4.1 Hz, JH3,H3
ꢀ
ꢀ
ꢀ
= 12.0 Hz, SerH-3), 3.95 (dd, 1H, JH3 ,H2 = 5.0 Hz, JH3,H3
=
12.0 Hz, SerH-3ꢀ), 4.53–4.56, 4.62–4.65 (2 × m, 2H, SH-2, OrnH-
2), 6.81 (t, 1H, JH4,H5 = JH5,H6 = 8.0 Hz, DHBH-5), 7.03, 7.23
(2 × d, 2H, JH4,H5 = JH5,H6 = 8.0 Hz, DHBH-4, DHBH-6);
13C NMR (50 MHz, D2O) dC: 23.92 (OrnC-4), 28.85 (OrnC-3),
39.53 (OrnC-5), 53.94, 55.97 (OrC-2, SerC-2), 61.78 (SerC-3),
120.22, 120.35, 120.52 (DHBC-4, DHBC-5, DHBC-6), 117.45,
145.20, 147.07 (DHBC-1, DHBC-2, DHBC-3), 170.63, 174.06
(2 × CONH), 174.26 (CO2H); (+)-LRESIMS (m/z, %): 356.14
([M + H]+, 100%); (+)-HRESIMS calculated for C15H22N3O7
[M + H]+ 356.1452, found 356.1458.
2,3-Dihydroxybenzoyl-D-argininyl-L-serine (1c)
Na-2,3-Dihydroxybenzoyl-D-ornithinyl-L-serine tert-butyl ester
(5c) (95 mg, 0.25 mmol) reacted with 1,3-di-tert-butoxycarbonyl-
2-trifluoromethanesulfonylguanidine (85 mg, 0.21 mmol) in a
similar way as 5a to yield Na-2,3-dihydroxybenzoyl-Nx-tert-bu-
toxycarbonyl-Nxꢀ-tert-butoxycarbonyl-D-argininyl-L-serine tert-
butyl ester, which was hydrolysed in a similar way as the
corresponding L-argininyl-L-serine derivative to afford 2,3-
dihydroxybenzoyl-D-argininyl-L-serine (1c) (69 mg, 0.16 mmol,
61%) as a dark yellow oil; [a]D24: −20.8 (c, 0.1 in H2O) [natural
sample: [a]D24: −15.9 (c, 0.08 in H2O)]; 1H NMR (500 MHz, D2O)
dH: 1.56–1.94 (m, 4H, ArgH-3, ArgH-3ꢀ, ArgH-4, ArgH-4ꢀ), 3.10–
3.14 (m, 2H, ArgH-5, ArgH-5ꢀ), 3.80 (dd, 1H, JH3,H2 = 3.9 Hz,
3.07 (m, 2H, ArgH-5, ArgH-5ꢀ), 4.43–4.59 (dd, 1H, JH2,H3
=
=
ꢀ
5.4 Hz, JH2,H3 = 8.7 Hz, ArgH-2), 6.67 (t, 1H, JH4,H5 = JH5,H6
8.1 Hz, DHBH-5), 6.90, 7.12 (2 × d, 2H, JH4,H5 = JH5,H6 8.1 Hz,
DHBH-4, DHBH-6); 13C NMR (50 MHz, D2O) dc: 24.54 (ArgC-
4), 27.83 (ArgC-3), 40.58 (ArgC-5), 52.60 (ArgC-2), 116.15,
119.30, 119.71 (DHBC-4, DHBC-5, DHBC-6), 144.67, 147.15
(DHBC-1, DHBC-2, DHBC-3), 156.74 (C=N), 170.00, 1756.29
(2 × CONH), 174.39 (CO2H); (+)-LRESIMS (m/z, %): 311.13
([M + H]+, 81%); (+)-HRESIMS: calculated for C13H18N4O5 [M
+ H]+ 311.1354, found 311.1349.
ꢀ
ꢀ
ꢀ
JH3,H3 = 10.1 Hz, SerH-3), 3.90 (dd, 1H, JH3 ,H2 = 5.1 Hz, JH3,H3
=
10.1 Hz, SerH-3ꢀ),4.47–4.57 (m, 2H, SerH-2, ArgH-2), 6.74 (t, 1H,
JH4,H5 = JH5,H6 = 8.1 Hz, DHBH-5), 6.96, 7.17 (2 × dd, 2H, J =
1.5 Hz, J = 8.1 Hz, DHBH-4, DHBH-6); 13C NMR (120 MHz,
D2O) dC 24.98 (ArgC-4), 29.02 (ArgC-3), 41.13 (ArgC-5), 54.13,
55.48 (ArgC-2, SerC-2), 61.60 (SerC-3), 120.17, 120.32, 120.46
(DHBC-4, DHBC-5, DHBC-6), 117.41, 145.19, 147.17 (DHBC-1,
DHBC-2, DHBC-3), 157.32 (C=N), 170.51, 173.72 (2 × CONH),
174.42 (CO2H); (+)-LRESIMS (m/z, %): 398.10 ([M + H]+, 84%);
(+)-HRESIMS: calculated for C16H24N5O97 [M + H]+ 398.1670,
found 398.1674.
Na-2,3-Dihydroxybenzoyl-D-ornithinyl-L-serine tert-butyl
ester (5c)
Na -2,3-Dibenzyloxybenzoyl-Nd -benzyloxycarbonyl-D-ornithine
methyl ester (2b) (0.51 g, 0.90 mmol) was coupled with L-serine
1284 | Org. Biomol. Chem., 2008, 6, 1278–1287
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
2,3-Dibenzyloxybenzoyl-L-serine (7a)
2,3-Dihydroxybenzoyl-L-serinyl-L-ornithine (10a)
L-Serine hydrochloride methyl ester (0.34 g, 2.16 mmol) was
coupled to 2,3-dibenzyloxybenzoic acid (0.72 g, 2.16 mmol) in
anhydrous DMF (20 mL) and triethylamine (0.75 mL, 5.39 mmol)
using TBTU as the coupling agent (1.04 g, 3.23 mmol). The solvent
was removed in vacuo and the residue was preabsorbed onto
silica and purified by flash column chromatography (ethyl acetate–
hexane 1:1) to yield 2,3-dibenzyloxybenzoyl-L-serine methyl ester,
which was then dissolved in THF (40 mL) and water (80 mL).
To the resulting solution, barium hydroxide (0.79 g, 4.58 mmol)
was added and the mixture was stirred for 3 h at 50 ◦C, after
which t.l.c. (ethyl acetate–cyclohexane 2:3) indicated complete
conversion of starting material (Rf 0.28) to a major product (Rf
0.0). DOWEX 50 W (H+) was added to the solution to give a
pH = 6, the resin was removed by filtration and the filtrate was
evaporated in vacuo to give 2,3-dibenzyloxybenzoyl-L-serine (7a)
(0.84 g, 2.00 mmol, 84%), which was used for the next step without
further purification.
2,3-Dihydroxybenzoyl-L-serinyl-L-ornithine tert-butyl ester (9a)
(51 mg, 0.012 mmol) was stirred at room temperature in a solution
of TFA and DCM (3:7, 1 mL). After 16 h, evaporation of the
solvent under reduced presure and coevaporation with toluene
afforded 2,3-dihydroxybenzoyl-L-serinyl-L-ornithine (10a) (44 mg,
0.12 mmol, quantitative) as a clear oil; [a]D23: +3.9 (c, 1.1 in
MeOH); 1H NMR (200 MHz, D2O) dH: 1.32–1.87 (m, 4H, OrnH-3,
OrnH-3ꢀ, OrnH-4, OrnH-4ꢀ), 2.85–2.99 (m, 2H, OrnH-5, OrnH-
5ꢀ), 3.86–3.94 (m, 2H, SerH-3, SerH-3ꢀ), 4.51–4.61 (m, 2H, SerH-2,
OrnH-2), 6.64 (t, 1H, JH4,H5 = JH5,H6 = 7.8 Hz, DHBH-5), 7.04,
7.21 (2 × d, 2H, JH4,H5 = JH5,H6 = 7.8 Hz, DHBH-4, DHBH-6); 13
C
NMR (50 MHz, D2O) dC: 23.87 (OrnC-4), 29.17 (OrnC-3), 39.67
(OrnC-5), 49.51, 56.28 (OrnC-2, SerC-2), 62.04 (SerC-3), 117.73,
118.59, 120.52 (DHBC-4, DHBC-5, DHBC-6), 145.66, 147.01
(DHBC-1, DHBC-2, DHBC-3), 169.89, 171.44 (2 × CONH),
173.57 (CO2H).
2,3-Dihydroxybenzoyl-L-serinyl-L-arginine (11a)
2,3-Dihydroxybenzoyl-L-serinyl-L-ornithine tert-butyl ester (9a)
(64 mg, 0.15 mmol) was treated with 1,3-di-tert-butoxycarbonyl-
2-trifluoromethanesulfonylguanidine (52 mg, 0.13 mmol) in a
similar was as 5b. The product mixture was then preadsorbed
onto silica and purified by flash column chromatography
(methanol/DCM 5%), to yield Na-2,3-dihydroxybenzoyl-L-
serinyl-Nx-tert-butoxycarbonyl-Nxꢀ-tert-butoxycarbonyl-L-argin-
ine tert-butyl ester, which was hydrolysed with TFA as before to
afford 2,3-dihydroxybenzoyl-L-serinyl-L-arginine (11a) (40 mg,
0.10 mmol, 71%) as a dark yellow oil; [a]D20: +3.2 (c, 0.8 in
MeOH); 1H NMR (200 MHz, D2O) dH: 1.60–1.98 (m, 4H,
ArgH-3, OrnH-3ꢀ, OrnH-4, OrnH-4ꢀ), 3.12–3.18 (m, 2H, ArgH-5,
ArgH-5ꢀ), 3.78–3.84 (m, 2H, SerH-3, SerH-3ꢀ), 4.35 (dd, 1H,
2,3-Dihydroxybenzoyl-L-serine (8a)
2,3-Dibenzyloxybenzoyl-L-serine (7a) (0.04 g, 0.10 mmol) was
hydrogenated using 5% Pd/C (1 mg) in a similar was as 2b
to afford 2,3-dihydroxybenzoyl-L-serine (8a) (23 mg, 0.1 mmol,
1
quantitative); [a]D21: +12.2 (c, 0.6 in H2O); H NMR (200 MHz,
D2O) dH: 4.39 (m, 2H, SerH-3, SerH-3ꢀ), 5.04 (m, 1H, SerH-2), 7.14
(t, 1H, J = 7.0 Hz, DHBH-5), 7.35, 7.73 (2 × d, 2H, J = 7.0 Hz,
J = 7.2 Hz, DHBH-4, DHBH-6); 13C NMR (200 MHz, D2O) dC:
49.53 (SerC-2), 62.18 (SerC-3), 116.96, 120.02, 120.24 (DHBC-4,
DHBC-5, DHBC-6), 145.12, 147.54 (DHBC-1, DHBC-2, DHBC-
3), 170.21 (CONH), 177.64 (CO2OH); (+)-LRESIMS (m/z, %):
264.05 [M + Na]+, 242.06 [M + H]+; (+)-HRESIMS: calculated
for C10H12NO6 [M + H]+ 242.0659; found 242.0659.
ꢀ
JH2,H3 = 4.8 Hz, JH2,H3 = 9.0 Hz, SerH-2), 4.69–4.75 (m, 1H,
ArgH-2), 6.74 (t, 1H, JH4,H5 = JH5,H6 = 7.8 Hz, DHBH-5), 6.96,
7.18 (2 × d, 2H, JH4,H5 = JH5,H6 = 7.8 Hz, DHBH-4, DHBH-6);
13C NMR (50 MHz, CDCl3) dC: 24.75 (ArgC-4), 27.83 (ArgC-3),
40.25 (ArgC-5), 52.12, 55.43 (ArgC-2, SerC-2), 61.14 (SerC-3),
117.94, 119.06 (DHBC-4, DHBC-5, DHBC-6), 143.88, 146.75
(DHBC-1, DHBC-2, DHBC-3), 156.33 (C=N), 170.12, 171.35
(2 × CONH), 174.03 (CO2H); (+)-LRESIMS (m/z, %): 398.10
([M + H]+, 84%); (+)-HRESIMS: calculated for C16H24N5O97
[M + H]+ 398.1670, found 398.1678.
2,3-Dihydroxybenzoyl-L-serinyl-L-ornithine tert-butyl ester (9a)
Nd-Benzyloxycarbonyl-L-ornithine tert-butyl ester (0.07 g, 0.21
mmol) was coupled to 2,3-dibenzyloxybenzoyl-L-serine (7a)
(0.06 g, 0.12 mmol) in anhydrous DMF (2 mL) and triethylamine
(0.07 mL, 0.52 mmol) using TBTU as coupling agent (0.10 g, 0.30
mmol). The solvent was removed in vacuo and the residue was
preabsorbed onto silica and purified by flash column chromatogra-
phy (ethyl acetate–hexane 2:1) to yield 2,3-dibenzyloxybenzoyl-L-
serinyl-Nd-benzyloxycarbonyl-L-ornithine tert-butyl ester, which
was hydrogenated in methanol (6 mL) using 5% Pd/C (0.02 g) as
catalyst. The reaction mixture was filtered through Celite and the
solvent was removed in vacuo to afford 2,3-dihydroxybenzoyl-L-
serinyl-L-ornithine tert-butyl ester (9a) (64 mg, 0.16 mmol, 81%),
which was used without further purification; [a]D21: +15.5 (c, 0.55
in MeOH); 1H NMR (200 MHz, CD3OD) dH: 1.40 (s, 9H, -
C(CH3)3), 1.51–1.86 (m, 4H, OrnH-3, OrnH-3ꢀ, OrnH-4, OrnH-
4ꢀ), 2.78 (m, 2H, OrnH-5, OrnH-5ꢀ), 3.89–3.90 (m, 2H, SerH-3,
SerH-3ꢀ), 4.35 (m, 1H, SerH-2), 4.66 (bs, 1H, ArgH-2), 6.30–6.38
(m, 1H, DHBH-5), 6.76, 7.28 (2 × d, 2H, J = 7.0 Hz, J = 7.3 Hz,
DHBH-4, DHBH-6), (+)-LRFABMS (m/z, %): 434.2 [M + Na]+,
412.2 [M + H]+.
2,3-Dibenzyloxybenzoyl-D-serine (7b)
D-Serine methyl ester hydrochloride (0.34 g, 2.16 mmol) was cou-
pled with 2,3-dibenzyloxybenzoic acid (0.72 g, 2.16 mmol) in a sim-
ilar way as L-serine methyl ester to yield 2,3-dibenzyloxybenzoyl-
D-serine methyl ester, which was hydrolysed in a similar way as the
corresponding L-serine derivative to give 2,3-dibenzyloxybenzoyl-
D-serine (7b) (0.83 g, 1.97 mmol, 84%), which was used in the next
step without further purification.
2,3-Dihydroxybenzoyl-D-serine (8b)
2,3-Dibenzyloxybenzoyl-D-serine (7b) (31 mg, 0.07 mmol) was hy-
drogenated in a similar way as 7a to afford 2,3-dihydroxybenzoyl-
D-serine (8b) (17 mg, 0.07 mmol, quantitative); [a]D23: −10.8 (c,
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1278–1287 | 1285
©

View Article Online
0.8 in H2O); 1H NMR (200 MHz, D2O) dH: 4.39 (m, 2H, SerH-3,
SerH-3ꢀ), 5.04 (m, 1H, SerH-2), 7.14 (t, 1H, J 7.0 Hz, DHBH-
5), 7.35, 7.73 (2 × d, 2H, J = 7.0 Hz, J = 7.2 Hz, DHBH-4,
DHBH-6); 13C NMR (50 MHz, D2O) dC 49.53 (SerC-2), 62.18
(SerC-3), 116.96, 120.02, 120.24 (DHBC-4, DHBC-5, DHBC-6),
145.12, 147.54 (DHBC-1, DHBC-2, DHBC-3), 170.21 (CONH),
177.64 (CO2OH); (+)-LRESIMS (m/z, %): 264.05 [M + Na]+,
242.06 [M + H]+; (+)-HRESIMS: calculated for C10H12NO6 [M +
H]+ 242.0659; found 242.0659.
7.8 Hz, DHBH-4, DHBH-6); 13C NMR (50 MHz, D2O) dC: 24.26
(ArgC-4), 27.69 (ArgC-3), 40.36 (ArgC-5), 52.20, 55.62 (ArgC-2,
SerC-2), 61.02 (SerC-3), 118.02, 119.66 (DHBC-4, DHBC-5,
DHBC-6), 144.46, 146.75 (DHBC-1, DHBC-2, DHBC-3),
156.55 (C=N), 169.85, 171.84 (2 × CONH), 174.78 (CO2H);
(+)-LRESIMS (m/z, %): 398.10 ([M + H]+, 84%); (+)-HRESIMS:
calculated for C16H24N5O97 [M + H]+ 398.1670, found 398.1676.
Biological assays
The growth promotion assays were performed using the disc
diffusion method in Petri dishes. All tests were carried out
using CM9 minimal medium containing the non-assimilable iron
chelator 2,2ꢀ-dipyridyl at the appropriate concentration for each
test strain to make the medium iron depleted and to hinder
normal bacterial growth. Strains to be tested were suspended
in this medium before plate pouring. After agar solidification,
10 lL of siderophore solutions (all adjusted to a concentration of
1 mg mL−1) were applied to sterile paper discs of 6 mm diameter
onto the surface of the agar plates. The presence of growth zones
around discs after 24 h or 48 h incubation at 25 ◦C or 37 ◦C (for
Salmonella enterica) indicates utilization of the compound as an
iron source. Measurement of the growth halo diameter was used
as a quantitative estimation of utilization of each derivative. All
assays were repeated at least three times and the average values
are given in the results. A disc with 10 lL of a 10 lM solution of
Fe2(SO4)3 was used as the growth control.
In addition, the iron complexation capacity of each compound
was tested by the chrome azurol S (CAS) liquid assay developed by
Schwyn and Neilands. An aqueous solution of each compound at
a concentration of 1 mg mL−1 was diluted to 1:10 in purified water
and mixed with 1 vol of the CAS mixture. After 20 min the colour
change was measured in a UV-VIS spectrophotometer at 630 nm.
A positive siderophore activity produces a colour change from
blue to orange with the corresponding reduction in absorbance
values with respect to the blank.
2,3-Dihydroxybenzoyl-D-serinyl-L-ornithine tert-butyl ester (9b)
2,3-Dibenzyloxybenzoyl-D-serinic acid (7b) was coupled with
Nd-benzyloxycarbonyl-L-ornithine tert-butyl ester (0.07 g, 0.21
mmol) in a similar way as 7a to yield 2,3-dibenzyloxybenzoyl-D-
serinyl-Nd-benzyloxycarbonyl-L-ornithine tert-butyl ester, which
was hydrogenated in a similar way as the corresponding D-serinyl-
L-ornithine derivative to afford 2,3-dihydroxybenzoyl-D-serinyl-L-
ornithine tert-butyl ester (9b) (64 mg, 0.15 mmol, 81%), which was
used without further purification; [a]D21: +15.5 (c, 0.55 in MeOH);
1H NMR (200 MHz, CD3OD) dH: 1.40 (s, 9H, -C(CH3)3), 1.51–
1.86 (m, 4H, OrnH-3, OrnH-3ꢀ, OrnH-4, OrnH-4ꢀ), 2.78 (m, 2H,
OrnH-5, OrnH-5ꢀ), 3.89–3.90 (m, 2H, SerH-3, SerH-3ꢀ), 4.35 (m,
1H, SerH-2), 4.66 (bs, 1H, ArgH-2), 6.30–6.38 (m, 1H, DHBH-5),
6.76, 7.28 (2 × d, 2H, J = 7.0 Hz, J = 7.3 Hz, DHBH-4, DHBH-6);
(+)-LRESIMS: 434.2 [M + Na]+, 412.2 [M + H]+.
2,3-Dihydroxybenzoyl-D-serinyl-L-ornithine (10b)
2,3-Dihydroxybenzoyl-D-serinyl-L-ornithine tert-butyl ester (9b)
(45 mg, 0.01 mmol) was hydrolysed in a similar way as 9a
to give 2,3-dihydroxybenzoyl-D-serinyl-L-ornithine (10b) (39 mg,
0.01 mmol, quantitative) as a clear oil; [a]D23: −3.1 (c, 1.8 in
MeOH); 1H NMR (200 MHz, D2O) dH: 1.32–1.87 (m, 4H, ArgH-3,
OrnH-3ꢀ, OrnH-4, OrnH-4ꢀ), 2.85–2.99 (m, 2H, ArgH-5, ArgH-
5ꢀ), 3.86–3.94 (m, 2H, SerH-3, SerH-3ꢀ), 4.51–4.61 (m, 2H, SerH-2,
ArgH-2), 6.64 (t, 1H, JD4,D5 = JD5,D6 = 7.8 Hz, DHBH-5), 7.04,
7.21 (2 × d, 2H, JD4,D5 = JD5,D6 = 7.8 Hz, DHBH-4, DHBH-6); 13
C
NMR (50 MHz, D2O) dC: 23.87 (OrnC-4), 29.17 (OrnC-3), 39.67
(OrnC-5), 49.51, 56.28 (OrnC-2, SerC-2), 62.04 (SerC-3), 117.73,
118.59, 120.52 (DHBC-4, DHBC-5, DHBC-6), 145.66, 147.01
(DHBC-1, DHBC-2, DHBC-3), 169.89, 171.44 (2 × CONH),
173.57 (CO2H).
Acknowledgements
This work was financially supported by Xunta de Gali-
cia (PGIDIT05RMA10302PR, PGIDIT06PXIC103118PN and
PGIDIT04RMA261014PR) and Ministry of Education (MEC)
of Spain (cofunded by FEDER) (CQT2005-00793 and AGL2006-
00697). R.G.S. thanks the Parga Pondal Programme. M.B. is the
recipient of a FPI predoctoral fellowship from the MEC.
2,3-Dihydroxybenzoyl-D-serinyl-L-arginine (11b)
2,3-Dihydroxybenzoyl-D-serinyl-L-ornithine tert-butyl ester (9b)
(64 mg, 0.15 mmol) was reacted with 1,3-di-tert-butoxycarbonyl-2-
(trifluoromethanesulfonyl)guanidine in a similar way as 9a to yield
Na-2,3-dihydroxybenzoyl-D-serinyl-Nx-tert-butoxycarbonyl-Nxꢀ-
tert-butoxycarbonyl-L-arginine tert-butyl ester, which was
hydrolysed in a similar way as the corresponding L-serinyl-L-
arginine derivative to give 2,3-dihydroxybenzoyl-D-serinyl-L-
arginine (11b) (42 mg, 0.10 mmol, 71%) as a dark yellow oil;
References
1 (a) C. Ratledge and L. Dover G., Annu. Rev. Microbiol., 2000, 54, 881–
941; (b) C. Wandersman and P. Delepelaire, Annu. Rev. Microbiol.,
2004, 58, 611–647.
2 M. Balado, C. R. Osorio and M. L. Lemos, Microbiology, 2006, 152,
3517–3528.
3 R. F. Conchas, M. L. Lemos, J. L. Barja and A. E. Toranzo, Appl.
Environ. Microbiol., 1991, 57, 2956–2962.
4 M. K. Wolf and J. L. Crosa, J. Gen. Microbiol., 1986, 132, 2949–2952.
5 R. G. Soengas, C. Anta, A. Espada, V. Paz, I. R. Ares, M. Balado, J.
Rodr´ıguez, M. L. Lemos and C. Jime´nez, Tetrahedron Lett., 2006, 47,
7113–7116.
[a]D23: −6.2 (c, 0.4 in MeOH); H NMR (200 MHz, D2O) dH:
1
1.50–1.99 (m, 4H, ArgH-3, ArgH-3ꢀ, ArgH-4, ArgH-4ꢀ), 3.04–3.08
(m, 2H, ArgH-5, ArgH-5ꢀ), 3.89–3.91 (m, 2H, SerH-3, SerH-3ꢀ),
ꢀ
4.37 (dd, 1H, JH2,H3 = 4.8 Hz, JH2,H3 = 9.0 Hz, SerH-2), 4.69–4.71
ꢀ
(t, 1H, JH2,H3 = JH2,H3 = 5.2 Hz, ArgH-2), 6.74 (t, 1H, JH4,H5
JH5,H6 = 7.8 Hz, DHBH-5), 6.96, 7.19 (2 × d, 2H, JH4,H5 = JH5,H6
=
=
1286 | Org. Biomol. Chem., 2008, 6, 1278–1287
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
6 R. G. Soengas, C. Anta, A. Espada, R. M. Nieto, M. Larrosa, J.
Rodr´ıguez and C. Jime´nez, Tetrahedron Lett., 2007, 48, 3021–3024.
7 (a) M. J. Miller and F. Malouin, Accts. Chem. Res., 1993, 26, 241–249;
(b) A. Ghosh, M. Ghosh, C. Niu, F. Malouin, U. Moellmann and M. J.
Miller, Chem. Biol., 1996, 31, 1011–1019; (c) H. Budzikiewicz, Curr.
Top. Med. Chem., 2001, 1, 73–82; (d) F. Rivault, C. Liebert, A. Burger,
F. Hoegy, M. A. Abdallah, I. J. Schalk and G. L. Mislin, Bioorg. Med.
Chem. Lett., 2007, 17, 640–644.
8 C. Hennard, Q. C. Truong, J. F. Desnottes, J. M. Paris, N. J. Moreau
and M. A. Abdallah, J. Med. Chem., 2001, 44, 2139–2151.
9 J. D. Buynak, Curr. Med. Chem., 2004, 11, 1951–1964.
10 C. Lu, J. S. Buyer, J. F. Okonya and M. J. Miller, BioMetals, 1996, 9,
377–283.
15 W. H. Rastetter, T. J. Erickson and M. C. Venuti, J. Org. Chem., 1981,
46, 3579–3590.
16 T. T. Charvat, D. J. Lee, W. E. Robinson and A. R. Chamberlin, Bioorg.
Med. Chem., 2006, 14, 4552–4567.
17 K. Feichtinger, H. L. Sings, T. J. Baker, K. Matthews and M. Goodman,
J. Org. Chem., 1998, 63, 8432–8439.
18 I. G. OꢀBrien, G. B. Cox and F. Gibson, Biochim. Biophys. Acta, 1969,
177, 321–328.
19 M. J. Milewska and A. Chimiak, Synthesis, 1990, 3, 233–234.
20 J. Go´mez-Leo´n, L. Villamil, M. L. Lemos, B. Novoa and A. Figueras,
Appl. Environ. Microbiol., 2005, 71, 98–104.
21 M. Persmark, D. Expert and J. B. Neilands, J. Biol. Chem., 1989, 264,
3187–3193.
11 (a) M. Schnabelrauch, S. Wittmann, K. Rahn, U. Mo¨llmann, R.
Reissbrodt and L. Heinisch, BioMetals, 2000, 13, 333–348; (b) S.
Wittmann, L. Heinisch, I. Scherlitz-Hofmann, T. Stoiber, D. Ankel-
Fuchs and U. Mo¨llmann, BioMetals, 2004, 17, 53–64.
12 (a) A. J. Walz, U. Mo¨llmann and M. J. Miller, Org. Biomol. Chem., 2007,
5, 1621–1628; (b) R. Schobert Rainer, A. Stangl and K. Hannemann,
Tetrahedron, 2008, 64, 1711–1720.
22 J. R. Pollack, B. N. Ames and J. B. Neilands, J. Bacteriol., 1970, 104,
635–639.
23 Q. Wang, Q. Liu, Y. Ma, L. Zhou and Y. Zhang, Arch. Microbiol., 2007,
188, 433–439.
24 S. A. Amin, F. C. Ku¨pper, D. H. Green, W. R. Harris and C. J. Carrano,
J. Am. Chem. Soc., 2007, 129, 478–479.
25 B. Schwyn and J. B. Neilands, Anal. Biochem., 1987, 160, 47–56.
26 L. Dong, M. J. Miller and U. Mo¨llmann, BioMetals, 2004, 17,
99–104.
13 E. Masiukiewicz, B. Rzeszotarska and J. Szczerbaniewicz, Org. Prep.
Proc. Int., 1992, 24, 191–194.
14 S. K. Sharma, M. J. Miller and S. M. Payne, J. Med. Chem., 1989, 32,
357–367.
27 R. Annamalai, B. Jin, Z. Cao, S. M. Newton and P. E. Klebba,
J. Bacteriol., 2004, 186, 3578–3589.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1278–1287 | 1287
©