Total Synthesis of Petrobactin and Its Homologues as Potential Growth Stimuli for Marinobacter hydrocarbonoclasticus, an Oil-Degrading Bacteria

Article abstract of DOI:10.1021/jo049803l

A modular synthesis was developed to access petrobactin, a catechol-containing siderophore isolated from Marinobacter hydrocarbonoclasticus. A range of petrobactin homologues with differing dihydroxybenzamide motifs and in one case an increased number of

Full text of DOI:10.1021/jo049803l

Tota l Syn th esis of P etr oba ctin a n d Its Hom ologu es a s P oten tia l
Gr ow th Stim u li for Ma r in oba cter h yd r oca r bon ocla sticu s, a n
Oil-Degr a d in g Ba cter ia
Richard Andrew Gardner, Rebecca Kinkade, Chaojie Wang, and Otto Phanstiel IV*
Department of Chemistry, University of Central Florida, Orlando, Florida 32816-2366
ophansti@mail.ucf.edu
Received February 3, 2004
A modular synthesis was developed to access petrobactin, a catechol-containing siderophore isolated
from Marinobacter hydrocarbonoclasticus. A range of petrobactin homologues with differing
dihydroxybenzamide motifs and in one case an increased number of carbons in the polyamine
backbone were also synthesized. As such, these systems represent new isomeric probes to study
iron transport properties in M. hydrocarbonoclasticus. The synthesis of petrobactin and its
homologues and the first biological study of how these agents influence the growth of Mycobacter
hydrocarbonoclasticus are reported. New synthetic methods were developed to overcome issues
1
(imide formation) encountered in earlier syntheses. Both the H and 13C NMR of petrobactin were
consistent with the recently revised structure showing that petrobactin in fact contains a
3,4-dihydroxybenzene motif rather than a 2,3-dihydroxybenzene motif. The preliminary biological
studies suggested that using the native petrobactin 1b for M. hydrocarbonoclasticus-specific growth
stimulation may be a poor strategy for oil-spill cleanup.
In tr od u ction
Another strategy for bioremediation is the development
of new growth stimulants for bacteria based upon sid-
erophores. Unlike mammals, which use high molecular
weight proteins for iron transport, bacteria use low
molecular weight, iron-specific ligands (siderophores) to
acquire iron (a growth-limiting nutrient) from their
environment. While iron mainly exists as insoluble ferric
hydroxide polymers in the biosphere, siderophores solu-
bilize this essential micronutrient and allow for its use
by bacteria. Over the course of time, bacteria have
generated a variety of siderophore architectures contain-
ing iron-binding motifs, which typically include catechols,
N-hydroxyamides, and R-hydroxycarboxylic acids. Dif-
ferent bacteria often produce their own unique sidero-
phore to sequester iron(III) and presumably have an
uptake pathway to facilitate its import.6 Indeed, one may
be able to specifically feed certain bacteria in the presence
of others by using a native siderophore-iron complex.
Recently a new siderophore, petrobactin 1a , was
isolated and characterized from an oil-degrading marine
bacterium, Marinobacter hydrocarbonoclasticus,7 by But-
ler et al.8 Although there are many examples of sidero-
Oil spills by ocean-faring oil tankers not only result in
loss of valuable energy resources but also have disastrous
environmental consequences. After the March 1989
Exxon-Valdez oil spill, both Exxon and the EPA have
experimented with a shoreline cleanup technique involv-
ing bioremediation, which used bacteria to degrade oil
into harmless water, carbon dioxide, and fatty acids.
Exxon’s pilot project involved spraying fertilizers on 74
miles of the more than 900 miles of oiled shoreline in
Alaska in the hopes of spurring growth of naturally
occurring bacteria that eat oil.1 The fertilizers appear to
have accelerated natural biodegradation of the oil without
harmful side effects such as eutrophication, toxicity to
sensitive organisms, and release of untreated oil.1-5
Bioremediation can also be performed by introducing
genetically engineered microbes into the water, as was
done experimentally in the 1990 Mega Borg spill in the
Gulf of Mexico.2-5
* Address correspondence to this author. Phone: (407) 823-5410.
Fax: (407) 823-2252.
(1) Pritchard, P. H.; Costa, C. F. EPA’s Alaska Oil Spill Bioreme-
diation Project. Environ. Sci. Technol. 1991, 25, 372-379.
(2) Coping With an Oiled Sea: An Analysis of Oil Spill Response
Technologies. U.S. Congress, Office of Technology Assessment, Wash-
ington, DC, 1990.
(3) Baker, J . M.; Clark, R. B.; Kingston, P. F.; J enkins, R. H. Natural
Recovery of Cold Water Marine Environments After an Oil Spill;
presented at the Thirteenth Annual Arctic and Marine Oilspill
Program Technical Seminar, J une 1990.
(4) Head, I. M.; Swannell, R. P. J . Bioremediation of petroleum
hydrocarbon contaminants in marine habitats. Curr. Opin. Biotechnol.
1999, 10, 234-239.
(5) Prince, R. C.; Varadaraj, R.; Fiocco, R. J .; Lessard, R. R.
Bioremediation as an Oil Spill Response Tool. Environ. Technol. 1999,
20, 891.
(6) Neilands, J . B. Siderophores as Biological Deferration Agents
and the Regulation of their Synthesis by Iron. In The Development of
Iron Chelators for Clinical use; Bergeron, R. J ., Brittenhamm G. M.,
Eds.; CRC Press: Boca Raton, FL, 1994; p 165.
(7) Gauthier, M. J .; Lafay, B.; Christen, R.; Fernandez, L.; Acqua-
viva, M.; Bonin, P.; Bertrand, J . C. Marinobacter hydrocarbonoclasticus
gen. nov., sp. nov., a New, Extremely Halotolerant, Hydrocarbon-
Degrading Marine Bacterium. Int. J . Syst. Bacteriol. 1992, 42, 568-
576.
(8) Barbeau, K.; Zhang, G.; Live, D. H.; Butler, A. Petrobactin, a
Photoreactive Siderophore Produced by the Oil-Degrading Marine
Bacterium Marinobacter hydrocarbonoclasticus. J . Am. Chem. Soc.
2002, 124, 378-379.
10.1021/jo049803l CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/17/2004
3530
J . Org. Chem. 2004, 69, 3530-3537

Total Synthesis of Petrobactin and Its Homologues
F IGURE 1. Structure of citrate-containing chelators and the reported petrobactins, 1a ,b.
phores from soil bacteria, there are only a few examples
of marine-based siderophores.9-12 Upon inspection, 1a is
comprised of three subunits: a 2,3-dihydroxybenzoyl unit,
a spermidine unit, and a citric acid moiety. The citric acid
component is a common feature found in a number of
other siderophores such as nannochelin A 2 from Nan-
nocystis exedens,13 acinetoferrin 3 from Acinetobacter
haemolyticus,14 schizokinen 4 from Bacillus megate-
rium,15 and aerobactin 5 from Aerobacter strains.16
Utilizing earlier insights from the synthesis of nan-
nochelin A 2 and acinetoferrin 3, a modular synthesis
was developed to access this catechol-containing sidero-
phore. However, during our efforts to access a family of
petrobactin derivatives including petrobactin 1a , Berg-
eron et al.17 described its total synthesis and further
reported that the structure assigned to it by Butler et
al.8 was incorrect. These authors went on to synthesize
the corrected structure of petrobactin, which was re-
ported as 1b.17 In this report, we describe our synthesis
of petrobactin and its homologues as well as the first
biological study of how these agents influence the growth
of M. hydrocarbonoclasticus.
Resu lts a n d Discu ssion
(9) Reid, R. T.; Butler, A. Investigation of the Mechanism of Iron
Acquisition by the Marine Bacterium Aleteramonas luteoviolaceus:
Characterisation of Siderophore Production. Limnol. Oceanogr. 1991,
36, 1783-1792.
Syn th esis. The initial target 1a contained a 2,3-
dihydroxybenzoyl unit attached to spermidine via an
amide linkage. The other end of the polyamine was
attached via amide bonds to the terminal carboxyl groups
of citric acid. Spermidine is an unsymmetrical 3,4-
triamine and the strategy required regiospecific control
of amine reactivity. This issue was initially addressed
by sequential construction of the polyamine chain with
use of an amino-alcohol approach.18-20 The strategy began
with the generation of 2,3-bis(dibenzyloxy)benzoic acid
7a 21-23 (92%) from commercially available 6a . As shown
(10) Guofeng, X.; Martinez, J . S.; Groves, J . T.; Butler, A. Membrane
Affinity of the Amphiphilic Marinobactin Siderophores. J . Am. Chem.
Soc., 2002, 124, 13408-13415.
(11) Deng, J .; Hamada, Y.; Shioiri, T. Total Synthesis of Alterobactin
A, a Super Siderophore from an Open-Ocean Bacterium. J . Am. Chem.
Soc., 1995, 117, 7824-7825.
(12) Barbeau, K.; Rue, E. L.; Trick, C. G.; Bruland, K. W.; Butler,
A. Photochemical reactivity of siderophores produced by marine
heterotrophic bacteria and cyanobacteria, based on characteristic iron-
(III)-binding groups. Limnol. Oceanogr. 2003, 48, 1069-1078.
(13) Bergeron, R. J .; Phanstiel, O., IV The Total Synthesis of
Nannochelin: A Novel Cinnamoyl Hydroxamate-Containing Sidero-
phore. J . Org. Chem. 1992, 57, 7140-7143.
(14) Phanstiel, O., IV; Wang, Q. Total Synthesis of Acinetoferrin.
J . Org. Chem. 1998, 63, 1491-1495.
(15) Milewska, M. J .; Chimiak, A. I.; Glowacki, Z. Synthesis of
Schizokinen, Homoschizokinen, Its Imide and the Detection of Imide
with 13C-NMR Spectroscopy. J . Prakt. Chem. 1987, 329, No. 3, 447-
456.
(16) Maurer, P. J .; Miller, M. J . Microbial Iron Chelators: Total
Synthesis of Aerobactin and Its Constituent Amino Acid, N6-Acetyl-
N6-hydroxylysine. J . Am. Chem. Soc. 1982, 104, 3096-3101.
(17) Bergeron, R. J .; Huang, G.; Smith, R. E.; Bharti, N.; McManis,
J . S.; Butler, A. Total synthesis and structure revision of petrobactin.
Tetrahedron 2003, 59, 2007-2014.
(18) Wang, C.; Delcros, J .-G.; Cannon, L.; Konate, F.; Carias, H.;
Biggerstaff, J .; Gardner, R. A.; Phanstiel, O., IV Defining the Molecular
Requirements for the Selective Delivery of Polyamine Conjugates into
Cells Containing Active Polyamine Transporters. J . Med. Chem. 2003,
46, 5129-5138.
J . Org. Chem, Vol. 69, No. 10, 2004 3531

Gardner et al.
SCHEME 1 a
a
Reagents: (a) BnBr, K2CO3, acetone; (b) NaOH, MeOH; (c) 3-aminopropanol, HOBt, DCC; (d) TsCl, pyridine; (e) 1,4-diaminobutane.
SCHEME 2 a
a
Reagents: (a) BOC2O (0.33 equiv), TEA, MeOH; (b) BrCH2CH2CN or BrCH2CH2CH2CN, K2CO3, CH3CN; (c) BOC2O (1.5 equiv), TEA,
MeOH; (d) H2, Ra Ni, EtOH.
in Scheme 1, condensation of 3-aminopropanol and 7a
with use of HOBt/DCC gave the desired amide 8 (85%).
Previous experience19,20 suggested that polyamine 11
could be made via tosylate 9 and 1,4-diaminobutane.
However, tosylation of 8 led mainly to the 1,3-oxazine
10 rather than tosylate 9. Similar cyclizations have been
reported in other activated benzamide derivatives.23
Therefore, another route was selected.
diaminobutane gave the amine 12 in a good yield (71%).
The addition of bromopropionitrile gave 13a (70%) and
the protection of the newly formed secondary amine with
di-tert-butyl dicarbonate gave 14a (91%). Finally, the
reduction of the nitrile with Raney Ni gave bis-BOC-
protected polyamine 15a (91%).
Initial attempts to couple acid 7a with 15a by using
HOBt/DCC did not give the desired amide 17a . This was
interesting as amide 8 was previously made by this
method (Scheme 1). As shown in Scheme 3, successful
coupling to 15a was achieved via acid chloride 16a and
provided amide 17a in 86% yield.17,21
The synthesis of 15a (Scheme 2) has been described
by Nakanishi and Hu.24 The mono-BOC protection of
(19) Wang, C.; Delcros, J .-G.; Biggerstaff, J .; Phanstiel, O., IV
Synthesis and Biological Evaluation of N1-(Anthracen-9-ylmethyl)-
triamines as Molecular Recognition Elements for the Polyamine
Transporter. J . Med. Chem. 2003, 46, 2663-2671.
(20) Wang, C.; Delcros, J .-G.; Biggerstaff, J .; Phanstiel, O., IV
Molecular Requirements for Targeting the Polyamine Transport
System. Synthesis and Biological Evaluation of Polyamine-Anthracene
Conjugates. J . Med. Chem. 2003, 46, 2672-2682.
(21) Laursen, B.; Denieul, M.-P.; Skrydstrup, T. Formal total
synthesis of the PKC inhibitor, balanol: preparation of the fully
protected benzophenone fragment. Tetrahedron 2002, 58, 2231-2238.
(22) Rastetter, W. H.; Erickson, T. J .; Venuti, M. C. Synthesis of
iron chelators. Enterobactin, enantioenterobactin, and a chiral analog.
J . Org. Chem. 1981, 46, 3579-3590.
(23) (a) Dovlatyan, V. V.; Dovlatyan, A. D. Synthesis and thermal
decomposition of haloalkoxy-sym-triazines. 5. Thermal decomposition
of 2-methoxy(methylthio)-4-[1,3-dichloropropoxy(3-chloro-1-propoxy)]-
6-dimethylamino-sym-triazines. Chem. Heterocycl. Compd. (Engl.
Transl.) 1980, 16, No. 3, 316-320. (b) Dovlatyan, V. V.; Eliazyan, K.
A.; Azatyan, A. V. Synthesis and thermol decomposition of halogen-
alkoxy(thio)-sym-triazines. 12. Synthesis and certain transformations
of 2-(dialkylamino)-4-oxo-6-cyanodihydrothiazolo-sym-triazines. Chem.
Heterocycl. Compd. (Engl. Transl.) 1987, 23, No. 7, 803-805.
(24) (a) Goodnow, R., J r.; Konno, K.; Niwa, M.; Kallimopoulos, T.;
Bukownik, R.; Lenares, D.; Nakanishi, K. Synthesis of glutamate
receptor antagonist philanthotoxin-433 (PhTX-433) and its analogs.
Tetrahedron 1990, 46, 3267-3286. (b) Hu, W.; Hesse, M. Synthesis of
p-cumaroylspermidine. Helv. Chim. Acta 1996, 79, 548-559.
Several methods were tried for the Boc deprotection
of 17a to form the diaminoamide 18a (i.e., the TFA salt
of 11). Under the typical BOC-removal conditions (4 N
aq HCl, 100% TFA or 50% TFA/CH2Cl2), partial depro-
tection of the O-benzyl groups was observed by 1H NMR.
This phenomenon was monitored via the two benzylic
methylene singlets at δ 5.1 and δ 5.2, which were
converted to three singlets of similar intensity at δ 5.15,
δ 5.12, and δ 5.10 after Boc deprotection. A thermal Boc-
deprotection method showed that 17a was stable at 180
°C under argon, but formed complex products at 220 °C.
Finally, diaminoamide 18a was generated by using 10%
TFA/CH2Cl2 and no O-debenzylation was observed.
In earlier work with acinetoferrin 3,14 the synthesis of
the activated citrate derivative 19 (Figure 2) was achieved
with N-hydroxysuccinimide (NHS) and dicyclohexylcar-
bodiimide (DCC), albeit in 20% isolated yield of 19.
Subsequent coupling of 19 to a primary amine (80%)
gave a 16% yield for the tandem activation and coupling
3532 J . Org. Chem., Vol. 69, No. 10, 2004

Total Synthesis of Petrobactin and Its Homologues
SCHEME 3 a
a
Reagents: (a) 15a or 15b, NaOH, CH2Cl2; (b) 10% TFA, CH2Cl2, (c) 19, TEA, dioxane, CH2Cl2; (d) BOC2O, TEA, MeOH; (e) HCl/
AcOH (1:2); (f) H2, Pd/C, EtOH.
product.17 This compound was formed either during the
removal of the tert-butyl group with a solution of 20%
TFA in dichloromethane or during their removal of the
benzyl groups under acidic reductive conditions (PdCl2/
HOAc/water).
In our synthesis of petrobactin, we also encountered
F IGURE 2. Bis NHS ester 19.
the imide and have developed ways to minimize its
formation. Previously, the removal of the tert-butyl group
from a related citrate system, tert-butyl acinetoferrin 3a ,
had been achieved by using TFA (100%) at room tem-
perature without imide formation.14 Moreover, the imide
was only observed by treating acinetoferrin 3 with
refluxing TFA for 3 h. Therefore, TFA at mild tempera-
tures (0-25 °C) seemed to be the most reasonable
approach.14 However, even though TFA facilitated the
deprotection of the tert-butyl ester group in 20a , it also
generated the respective ammonium salts, which contain
the undesirable counterion (CF3COO-) for later biological
studies. A method was needed that would obviate the
need for a later anion exchange step and also produce a
salt form of petrobactin that could be used for biological
testing.
sequence. In more recent work, a number of different
reagent combinations were tried to improve this trans-
formation, including HOBt/DCC, HOBt/EDC, NHS/EDC,
and NHS/DCC. In each reaction the activated ester was
not isolated, but was used directly in the next step to
provide 20a . Ironically, the most efficient reagent was
the original NHS and DCC combination. Indeed, signifi-
cantly higher yields were obtained for the two-step
process, when 19 (Figure 2) was not isolated, but used
directly in the subsequent coupling step. Therefore, the
condensation of a solution of the TFA salt (18a ) and
triethylamine with a solution of crude 1914,15 gave the
corresponding tert-butylpetrobactin 20a (57%).
The final two steps involved the removal of the tert-
butyl group and the deprotection of the benzyl ethers to
give petrobactin 1a . In their synthesis of petrobactin,
Bergeron et al. reported the formation and isolation of
an imide of petrobactin (15%), 21a , alongside the natural
Several methods were tested involving different solu-
tions of HCl and glacial acetic acid. A solution of 6 M
HCl in glacial acetic acid removed both protecting
groups: the tert-butyl ester and the O-benzyl groups.
J . Org. Chem, Vol. 69, No. 10, 2004 3533

Gardner et al.
TABLE 1. Con ver sion of 20b to Acyclic Dia m id e X a n d
Im id e 21b w ith Differ en t HCl Con cen tr a tion s
F IGURE 3. The imides of petrobactin, 21a and 21b.
However, these conditions also perturbed other function-
alities (benzamide linkage) within the structure. A second
solution of 4 M HCl in glacial acetic acid proved to be
less severe and left the rest of the molecule intact,
acid conditions
X/Y ratio
1
although along with the removal of the tert-butyl, the H
4 M HCl in acetic acid
1M HCl in acetic acid
0.2M HCl in acetic acid
0.1 M HCl in acetic acid
4 M HCl in acetonitrile
7.4
8.8
14.6
8.9
NMR revealed 50% deprotection of the O-benzyl groups.
As these were to be removed in the next step, hydrogena-
tion of the resultant crude with H2 gas and 10% Pd/C in
ethanol provided petrobactin 1a in a yield of 53%.
9.8
1H NMR of the crude prior to isolation of 1a revealed
the formation of petrobactin ethyl ester (5%). While a
time-course study revealed that the ethyl ester was
suppressed to <1% after a 2-h hydrogenation period,
further studies found that performing the hydrogena-
tion (H2 gas, 10% Pd/C in ethanol) in the presence of
water (10 vol %) stopped the formation of the ethyl
ester altogether. Having developed a reasonable entry
to these systems, we synthesized the remaining deriva-
tives, 1b-e. Compound 1e required the use of a di-BOC-
protected 4,4-triamine: N5,N9-di(tert-butoxycarbonyl)-
homospermidine 15b. The synthesis of 15b is shown in
Scheme 2. The addition of bromobutyronitrile to the
mono-BOC-protected diaminobutane 12 gave 13b (61%)
and the protection of the newly formed secondary amine
gave 14b (87%). Reduction of the nitrile with Raney Ni
gave the bis-BOC-protected polyamine 15b (95%).
As shown in Table 1 decreasing the amount of HCl
decreased the formation of imide Y. Since the conversion
of 20b was typically 100%, the optimal conditions were
0.2 M HCl in acetic acid. Therefore, the imide could be
suppressed to e5% by this new method. Ironically, this
may be a moot point altogether as similar linear citrate
diamide systems have been shown to cyclize to their
imide form in MeOH at room temperature (79%).14
The synthesis of 1b-e has provided a range of petro-
bactin homologues with differing dihydroxybenzamide
motifs and in one case an increased number of carbons
in the polyamine backbone. As such, these systems
represent new isomeric probes to study iron transport
properties in M. hydrocarbonoclasticus.
A preliminary study of the growth stimulation ability
of these architectures upon M. hydrocarbonoclasticus was
conducted. Two systems, 1a (the 2,3 isomer) and 1b (3,4
isomer), were chosen as they represent petrobactin
derivatives with vicinal phenolic OH groups found to be
important in iron chelation.6
Biologica l Eva lu a tion . M. hydrocarbonoclasticus was
cultured in standard marine broth using the protocol of
Gauthier.7 The assay is described in detail in the Ex-
perimental Section. Siderophores 1a and 1b were dosed
and the absorbance at 490 nm measured as a function of
time. The time zero absorbance (an average of three
measurements) was subtracted from each successive data
point within each dosing series to allow for direct
comparisons within and between the two siderophore
data series.
Th e Im id e Issu e. The imide byproduct is typically
formed during the acid treatment of tert-butyl citrate
derivatives and has been observed by many investigators
during siderophore synthesis.14,25-28 In our studies with
1
petrobactin, formation of the imide was quantified by H
NMR integration of two doublets at δ 2.79 (1H) and δ
2.65 (1H), corresponding to diastereotopic protons of the
citrate methylene group present in the imide impurity.
Subsequent comparison of these integrations to that of
the doublet at δ 2.59 (2H, corresponding to the citrate
methylene groups of linear petrobactin) provided the
necessary quantification. For example, in the synthesis
of 1b, the imide was formed in 12% yield with 4 M HCl/
AcOH.
A further study was conducted to observe how chang-
ing the concentration of HCl during the removal of the
tert-butyl group from 20b effected the formation of the
imide 21b (Table 1).
As shown in Figure 4, the 2,3-isomer 1a provided a
dose-dependent effect on the growth of M. hydrocarbono-
clasticus. Since absorbance is directly related to bacterial
growth, the trend toward higher absorbance as a function
of increasing time and siderophore concentration is
consistent with growth stimulating activity for 1a . In this
regard, increasing doses of 1a (e.g., 100 and 500 µM 1a )
provided a 1.1- and 1.4-fold respective increase in bacte-
rial growth after 12 h (compared to the control with no
added siderophore). Therefore, while high doses of non-
native siderophore 1a (e.g. 100 or 500 µM) provided some
enhancement, it was not dramatic.
(25) Bergeron, R. J .; Xin, J .; Smith, R. E.; Wollenweber, M.;
McManis, J . S.; Ludin, C.; Abboud, K. A. Total Synthesis of Rhizoferrin,
An Iron Chelator. Tetrahedron 1997, 53, 427-434.
(26) Lee, B. H.; Miller, M. J . Natural Ferric Ionophores: Total
Synthesis of Schizochinen, Schizochinen A, and Arthrobactin. J . Org.
Chem. 1983, 48, 24-31.
(27) Nau, C. A.; Brown, E. B.; Bailey, J . R. Methylene-Citric
Anhydride Anhydride the Aniline Derivatives of Citric and Aconitic
Acids. J . Am. Chem. Soc. 1925, 47, 2596-2606.
(28) Kisfaludy, L.; Schon, I.; Renyei, M.; Gorog, S. Competitive
intramolecular displacement of the neutral amide group. Rearrange-
ment and dehydration reactions of asparagine and glutamine. J . Am.
Chem. Soc. 1975, 97, 5588-5589.
The native chelator 1b (i.e., the 3,4-isomer) provided
little to no growth enhancement even at very high con-
centrations (500 µM) of added siderophore 1b. Ironically,
3534 J . Org. Chem., Vol. 69, No. 10, 2004

Total Synthesis of Petrobactin and Its Homologues
needed, this is a “first glimpse” into the specificity of how
marine bacteria utilize isomeric iron binding ligands (1a ,
1b) for their growth processes. Indeed, to the best of our
knowledge, these are the first examples of how petro-
bactin analogues effect M. hydrocarbonoclasticus growth.
Con clu sion
In summary, the total synthesis of petrobactin 1b
showed that the structure previously assigned to it (1a )
by Butler et al. was incorrect and reconfirmed the revised
3,4-dihydroxybenzene structure of 1b.17 A modular syn-
thesis was developed that allowed facile access to a
number of petrobactin homologues, which differed in both
the polyamine and the dihydroxybenzene components.
New synthetic methods were developed to overcome the
issues (imide formation) encountered in earlier syntheses.
The preliminary biological studies suggested that using
the native petrobactin 1b for M. hydrocarbonoclasticus-
specific growth stimulation may be a poor strategy for
oil-spill cleanup. Future work will continue to evaluate
the role of siderophores in the growth processes of oil-
degrading bacteria and will be reported in due course.
Exp er im en ta l Section
Biologica l Eva lu a tion . Marinobacter hydrocarbonoclas-
ticus bacteria was grown for 24 h on a shaker at 32°C using
the methods of Gauthier.6 Siderophore stock solutions were
prepared in water to the desired concentrations and filtered
through a 0.2 µM filter to ensure sterility. The siderophore
stock solution was then added to sterile tubes containing only
marine broth. These tubes were then inoculated with 200 µL
of bacterial culture. All tubes were placed on the shaker for
24 h at 32 °C. Three tubes were taken out at the respective
time intervals and vortexed, and the absorbance of a sample
(200 µL) was detected at 490 nm on a microplate reader.
For the synthetic details and characterization of 3,4-
petrobactin (1b), 2,4-petrobactin (1c), 2,5-petrobactin (1d ), and
2,3-homopetrobactin (1e) and other intermediate compounds
such as dibenzyloxybenzoic acid derivatives (7b,31 7c,31 and
7d 32,33), 13b, 14b, 15b, 16b-d ,17,21,34 17b, 17c, 17d , 17e, 18b,
18c, 18d , 18e, 20b, 20c, 20d , and 20e, including detailed NMR
data, please see the Supporting Information.
2,3-P etr oba ctin (1a ). The petrobactin tert-butyl ester 20a
(0.3 g, 0.26 mmol) was dissolved in glacial acetic acid (5 mL)
and to this was added a mixture of concentrated HCl (12 M, 9
mL) and glacial acetic acid (13 mL). The solution was stirred
at room temperature for 45 min. The HCl and glacial acetic
acid were removed in vacuo and coevaporation with benzene
(3 × 10 mL) and chloroform (3 × 10 mL) removed any residual
acids to give a white foam. 1H NMR showed complete removal
of the tert-butyl group and removal of 50% of the benzyl groups.
This compound was used in the next step without further
purification. The free acid amine salt was dissolved in EtOH
(30 mL) and 10% Pd/C (75 mg) was added and the reaction
F IGURE 4. Growth curves of M. hydrocarbonoclasticus in the
presence of added siderophores: 2,3-petrobactin isomer 1a
(top) and 3,4-petrobactin isomer 1b (bottom). Data are normal-
ized to a starting “zero” absorbance value by subtracting the
time zero absorbance value from subsequent data points within
each series. The data shown are the average of triplicate
measurements. With a few exceptions, the typical error was
less than 4%.
the native architecture 1b provided virtually no advan-
tage in terms of growth stimulation. While the absor-
bance of both 1a and 1b is comparable at the respective
100 µM dosing after 12 h, this parallel trend in their
growth stimulation profile is inconsistent especially at
higher concentrations (e.g. 500 µM). Therefore, the
bacteria seem to be relatively insensitive to additional
native siderophore 1b, while it responded modestly to the
non-native 1a .
This result could be explained by the fact that M.
hydrocarbonoclasticus can synthesize its own 1b as
needed and that the uptake of the iron-1b complex may
be transport limited. In this regard, additional 1b may
not provide growth stimulation via an already saturated
uptake process. Indeed, since 1b is the native siderophore
for M. hydrocarbonclasticus, there are likely optimal
conditions for the uptake of its iron complex. Moreover,
this uptake may be tightly regulated as found in other
biosystems.29,30
(30) Kang, D. K.; J eong, J .; Drake, S. K.; Wehr, N. B.; Rouault, T.
A.; Levine, R. L. Iron regulatory protein 2 as iron sensor. Iron-
dependent oxidative modification of cysteine. J . Biol. Chem. 2003, 278,
14857-14864.
(31) Kanai, F.; Kaneko, T.; Morishima, H.; Isshiki, K.; Takita, T.;
Takeuchi, T.; Umezawa, H. J . Antibiot. 1985, 38 (No. 1), 39-50.
(32) Gil, S.; Sanz, V.; Tortjada, A. The synthesis of 2-hydroxy-5,6,7-
trimethoxyxanthone: a conformation of structure. J . Nat. Prod. 1987,
50, 301-304.
(33) Schild, H. G.; Kolb, E. S.; Mehta, P. G.; Ford M.; Gaudiana, R.
A. Polyesters with controlled thermal decomposition. Polymer 1994,
35, 1222-1228.
(34) Blagbrough, I. S.; Moya, E.; Walford, S. P. Practical, convergent
total synthesis of polyamine amide spider toxin NSTX-3. Tetrahedron
Lett. 1996, 37, 551-554.
On the other hand, non-native siderophores such as
1a may simply be supplying “solubilized” iron for utiliza-
tion by the bacteria by another uptake pathway, which
evokes a growth response. Although further study is
(29) Schalinske, K. L.; Blemings,K. P.; Steffen, D. W.; Chen, O. S.;
Eisenstein, R. S. Iron regulatory protein 1 is not required for the
modulation of ferritin and transferrin receptor expression by iron in a
murine pro-B lymphocyte cell line. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 10681-10686.
J . Org. Chem, Vol. 69, No. 10, 2004 3535

Gardner et al.
stirred under H2 at ambient pressure for 2 h. The mixture was
filtered through a bed of Sephadex to facilitate removal of the
palladium/charcoal and the solids further washed with EtOH.
The filtrate was concentrated in vacuo and subsequent co-
evaporation with CHCl3 gave a white foam. The crude solid
was purified by column chromatography, using LH-20 Sepha-
dex (20 g, 10% CH3CN/0.1% HCl/water), to give 1a as a white
foam (0.112 g, 53%). Rf 0.44 (35% CH3CN/0.1% TFA/water,
reverse phase silica). 1H NMR (500 MHz, DMSO) δ 8.98 (t,
2H), 8.70 (s, 4H), 8.04 (t, 2H), 7.32 (dd, 2H), 6.92 (dd, 2H),
6.68 (dd, 2H), 3.36 (q, 4H), 3.03 (q, 4H), 2.89 (m, 8H), 2.58 (d,
2H), 2.50 (d, 2H), 1.89 (quin, 4H), 1.59 (quin, 4H), 1.42 (quin,
4H); 13C NMR (300 MHz, DMSO) δ 174.8, 169.8, 169.4, 149.5,
146.1, 118.8, 117.9, 117.2, 114.8, 73.5, 46.5, 44.6, 43.3, 37.7,
36.1, 26.2, 25.7, 23.0; HRMS (FAB) m/z calcd for C34H51N6O11
(FAB) m/z calcd for C24H24NO3 (M + H)+ 374.1756, found
374.1750.
(4-Am in ob u t yl)ca r b a m ic Acid ter t-Bu t yl E st er (12).
1,4-Diaminobutane (4.4 g, 0.05 mol) was dissolved in a solution
of triethylamine and methanol (10% TEA in MeOH, 110 mL).
A solution of di-tert-butyl dicarbonate (3.63 g, 0.017 mol) in
methanol (10 mL) was added dropwise to this mixture with
vigorous stirring. The mixture was stirred at room tempera-
ture overnight. The methanol and TEA were removed in vacuo
to yield an oily residue that was dissolved in dichloromethane
(100 mL) and washed with a solution of 10% aq sodium
carbonate (2 × 100 mL). The organic layer was dried over
anhydrous Na2SO4 and filtered, the solvent removed in vacuo,
and the oily residue purified by flash column chromatography
(1:10:89 NH4OH:MeOH:CHCl3) to give the product 12 as a
clear oil (2.23 g, 71%). Rf 0.38 (1:10:89 NH4OH:MeOH:CHCl3);
1H NMR (300 MHz, CDCl3) δ 4.72 (br s, 1H, NHCO), 3.12 (q,
2H, CH2), 2.70 (t, 2H, CH2), 1.57-1.30 (m, 13H, 2 × CH2, 3 ×
CH3).
[4-(2-Cya n oeth yla m in o)bu tyl]ca r ba m ic Acid ter t-Bu -
tyl Ester (13a ). To a solution of the amine 12 (2.10 g, 11
mmol) in anhydrous acetonitrile (50 mL) was added potassium
carbonate (5.14 g) and the suspension was stirred at rt for 10
min. A solution of 3-bromopropionitrile (1.50 g, 11 mmol) in
acetonitrile (25 mL) was added and the resulting mixture
stirred at 50 °C for 24 h. The mixture was filtered to remove
most of the inorganic salts and the acetonitrile was removed
in vacuo to give a semisolid residue that was purified by flash
column chromatography (1:5:94 NH4OH:MeOH:CHCl3) to yield
the product 13a as a clear oil (1.88 g, 70%). Rf 0.45 (1:10:89
NH4OH:MeOH:CHCl3); 1H NMR (300 MHz, CDCl3) δ 4.73 (br
s, 1H, NHCO), 3.13 (q, 2H, CH2), 2.92 (t, 2H, CH2), 2.65 (t,
2H, CH2), 2.52 (t, 2H, CH2), 1.52 (m, 4H, 2 × CH2), 1.44 (s,
9H, 3 × CH3).
(M + H)+ 719.3610, found 719.3624. Anal. Calcd for C34H52
-
N6O11Cl2‚1.3H2O: C, 50.01; H, 6.75; N, 10.29. Found: C, 49.77;
H, 6.62; N, 10.06.
2,3-Bis(ben zyloxy)ben zoic Acid (7a ).21,23 A solution of
2,3-dihydroxybenzoic acid 6a (0.31 g, 2.0 mmol), benzyl
bromide (2.07 g, 12 mmol), and K2CO3‚1.5H2O (5 g) in acetone
(40 mL) was refluxed for 1 day under nitrogen. After filtration,
the solution was concentrated in vacuo to give the crude
product as a clear oil. The crude product was dissolved in
methanol (120 mL) and an aqueous solution of 5 M NaOH (30
mL) was added. The mixture was refluxed for 3 h and the
solution concentrated in vacuo to remove the methanol. The
residue was dissolved in water and extracted twice with
hexane. Then the water phase was acidified with 3 N HCl to
pH 2 and filtered to give the product 7a as a white solid (3.11
g, 93%). Rf 0.62 (40% acetone/hexane); 1H NMR (300 MHz,
CDCl3) δ 11.2 (br s, 1H), 7.70 (d, 1H), 7.4-7.2 (m, 12H), 5.23
(s, 2H), 5.19 (s, 2H); HRMS (FAB) m/z calcd for C21H18O4Na
(M + Na)+ 357.1097, found 357.1116.
(4-(ter t-Bu t oxyca r b on yla m in o)b u t yl)(2-cya n oet h yl)-
ca r ba m ic Acid ter t-Bu tyl Ester (14a ). The nitrile 13a (1.88
g, 7.8 mmol) was dissolved in a solution of triethylamine and
methanol (10% TEA in MeOH, 40 mL). A solution of di-tert-
butyl dicarbonate (2.55 g, 12 mmol) in methanol (20 mL) was
added dropwise to this mixture with vigorous stirring. The
methanol and TEA were removed in vacuo to yield an oily
residue that was dissolved in dichloromethane (100 mL) and
washed with a solution of sodium hydroxide (2.5 M, 2 × 50
mL) and water (50 mL). The organic layer was dried over
anhydrous sodium sulfate and filtered, the solvent removed
in vacuo, and the oily residue purified by flash column
chromatography (40% EtAc/hexane) to give the product 14a
as a clear oil (2.42 g, 91%). Rf 0.4 (40% EtAc/hexane); 1H NMR
(300 MHz, CDCl3) δ 4.60 (m, 1H, NHCO), 3.45 (t, 2H, CH2),
3.26 (t, 2H, CH2), 3.12 (q, 2H, CH2), 2.58 (m, 2H, CH2), 1.55
(quin, 2H, CH2), 1.50-1.38 (m, 20H, CH2, 6 × CH3); HRMS
(FAB) m/z calcd for C17H32N3O4 (M + H)+ 342.2393, found
342.2395.
(3-Am in op r op yl)(4-(ter t-bu toxyca r bon yla m in o)bu tyl)-
ca r ba m ic Acid ter t-Bu tyl Ester (15a ). The nitrile 14a (2.30
g, 6.7 mmol) was dissolved in ethanol (100 mL). NH4OH (10
mL) and Raney nickel (8 g) were added and ammonia gas was
bubbled through the solution for 20 min at 0 °C. The suspen-
sion was hydrogenated at 50 psi for 24 h. Air was bubbled
through the solution and the Raney nickel was removed by
filtering through a sintered glass funnel keeping the Raney
nickel residue moist at all times. The ethanol and NH4OH were
removed in vacuo and the oily residue dissolved in CH2Cl2 and
washed with sodium carbonate (10%, aq, 3 × 50 mL). The
organic layer was dried over anhydrous sodium sulfate and
filtered and the solvent removed in vacuo to give the product
15a as a clear oil without further purification (2.12 g, 91%).
Rf 0.4 (1:10:89 NH4OH:MeOH:CHCl3); 1H NMR (300 MHz,
CDCl3) δ 4.63 (m, 1H, NHCO), 3.29-3.02 (m, 6H, 3 × CH2),
2.64 (m, 2H, CH2), 1.64 (quin, 2H, CH2), 1.50 (m, 2H, CH2),
1.47-1.35 (m, 20H, CH2, 6 × CH3);13C NMR (300 MHz, CDCl3)
δ 156.2, 79.7, 46.8, 40.5, 40.0, 39.4, 28.83, 28.78, 27.8, 26.2,
2,3-Bis(ben zyloxy)-N-(3-h yd r oxyp r op yl)ben za m id e (8).
A solution of 2,3-bis(benzyloxy)benzoic acid 7a (0.424 g, 1.3
mmol), HOBt (0.027 g, 0.2 mmol), and DCC (0.308 g, 1.5 mmol)
in CH2Cl2 (20 mL) was stirred for 30 min at room temperature.
3-Amino-1-propanol (0.10 g, 1.3 mmol) was added dropwise
over 3 min and the mixture stirred overnight. The solution
was filtered to remove some of the dicyclohexyl urea (DCU).
The filtrate was concentrated in vacuo and the residue purified
by flash column chromatography (40% acetone/hexane) to give
the product 8 as a clear oil (85%). Rf 0.40 (40% acetone/hexane);
1H NMR (300 MHz, CDCl3) δ 8.10 (br s, 1H, NH), 7.72-7.20
(m, 13H), 5.16 (s, 2H), 5.08 (s, 2H), 3.48 (t, 2H), 3.40 (m, 2H),
2.80 (br s, 1H, OH), 1.53 (m, 2H); 13C NMR δ 166.6, 151.9,
147.1, 136.5, 129.03, 129.01, 128.95, 128.9, 128.5, 127.9, 126.9,
124.7, 123.5, 117.4, 76.9, 76.8, 71.6, 58.9, 36.0, 32.9; HRMS
(FAB) m/z calcd for C24H26NO4 (M + H)+ 392.1862, found
392.1862.
4-Meth oxyben zen esu lfon ic Acid 3-(2,3-Diben zyloxy-
b en zoyla m in o)p r op yl E st er (9)/2-(2,3-Bis(b en zyloxy)-
p h en yl)-5,6-d ih yd r o-4H-[1,3]oxa zin e (10). The amido al-
cohol 8 (1.07 g, 2.74 mmol) was dissolved in dry pyridine (15
mL) and stirred at 0 °C for 10 min. p-Toluenesulfonyl chloride
(TsCl, 1.10 g, 5.5 mmol) was added in small portions over 30
min. The mixture was stirred for an additional hour at 0 °C.
The reaction flask was then placed in a refrigerator (0-5 °C)
overnight. The mixture was poured into 100 mL of ice water,
and a viscous liquid typically precipitated. After decanting off
the upper aqueous layer, the remaining viscous liquid was
dissolved in methylene dichloride and washed with deionized
water several times. The organic layer was concentrated in
vacuo and the residue purified by flash column chromatogra-
phy (30% acetone/hexane) to give an unwanted side product
1
10 (60%). Rf 0.29 (30% acetone/hexane); H NMR (300 MHz,
CDCl3) δ 7.42-7.00 (m, 13H), 5.12 (s, 2H), 5.08 (s, 2H), 4.20
(t, 2H), 3.51 (t, 2H), 1.90 (m, 2H); 13C NMR δ 156.6, 152.4,
147.1, 138.2, 137.1, 131.3, 128.7, 128.5, 128.4, 128.1, 127.9,
127.7, 124.2, 122.3, 116.2, 75.9, 71.5, 65.5, 43.1, 22.2; HRMS
3536 J . Org. Chem., Vol. 69, No. 10, 2004

Total Synthesis of Petrobactin and Its Homologues
25.9; HRMS (FAB) m/z calcd for
346.2706, found 346.2718.
C
17H36N3O4 (M + H)+
mixture was then stirred at room temperature overnight. The
DCU was removed by filtration and the filtrate concentrated
in vacuo. The residue was then dissolved in chloroform and
washed twice with saturated aq Na2CO3. The organic layer
was separated, dried over anhydrous Na2SO4, filtered, and
concentrated in vacuo. The residue was purified by flash
chromatography (MeOH/CHCl3/NH4OH 10:88:2) on “iron-free”
silica gel to give the product 20a as a clear oil (0.269 g, 57%).
2,3-Bis(ben zyloxy)ben zoyl Ch lor id e (16a ).17,21 A solution
of 3,4-bis(benzyloxy)benzoic acid 7a (1.50 g, 4.5 mmol) in 2:1
dichloromethane/benzene (30 mL) was stirred at 0 °C for 10
min. Anhydrous DMF (3 drops) and oxalyl chloride (3.0 mL)
were added in sequence and the mixture was stirred for 30
min at 0 °C. The solution was concentrated in vacuo to give
the crude product 16a that was consumed immediately in the
next step.
1
Rf 0.26 (MeOH/CHCl3/NH4OH 10:88:2); H NMR (300 MHz,
CDCl3) δ 8.08 (br s, 2H), 7.63 (m, 2H), 7.38 (m, 20H), 7.11 (m,
6H, including NH2), 5.14 (s, 4H), 5.04 (s, 4H), 3.33 (m, 4H),
3.20 (m, 4H), 2.65-2.45 (m, 12H), 1.60-1.40 (m, 21H); 13C
NMR δ 172.9, 169.8, 165.6, 151.9, 146.8, 136.6, 136.5, 129.0,
128.9, 128.5, 127.9, 127.5, 124.7, 123.3, 117.1, 82.9, 76.7, 74.4,
71.5, 49.2, 47.2, 44.5, 39.3, 37.8, 29.5, 28.2, 27.3, 27.1; HRMS
(FAB) m/z calcd for C66H83N6O11 (M + H)+ 1135.6120, found
1135.6083. Anal. Calcd for C66H82N6O11‚1.1H2O: C, 68.62; H,
7.35; N, 7.28. Found: C, 68.36; H, 7.10; N, 7.65.
(4-{[3-(2,3-Bis(b en zyloxy)b en zoyla m in o)p r op yl]-ter t-
bu toxyca r bon yla m in o}bu tyl)ca r ba m ic Acid ter t-Bu tyl
Ester (17a ). A solution of amine 15a (1.66 g, 4.8 mmol) in
CH2Cl2 (20 mL) was mixed with 10% aqueous NaOH (10 mL).
The mixture was stirred at 0 °C for 20 min, then a solution of
2,3-dibenzyloxybenzoic acid chloride 16a (1.45 g, 4.1 mmol)
in CH2Cl2 (8 mL) was added dropwise with vigorous stirring.
The white cloudy solution was stirred at room temperature
for 2 h. The organic phase was separated and the water phase
extracted with CH2Cl2. The organic phases were combined,
dried over anhydrous Na2SO4, filtered, and concentrated in
vacuo. The residue was purified by flash column chromatog-
raphy (EtAc/CHCl3 8:92) to give the product 17a as a clear oil
P r ep a r a tion of 20f. A solution of 20e (0.776 g, 0.66 mmol)
in triethylamine-methanol (1:9 v/v, 20 mL) was stirred at 0
°C for 10 min. A solution of di-tert-butyl dicarbonate (0.58 g,
2.67 mmol) in methanol (5 mL) was added dropwise over 5
min. The mixture was stirred for one additional hour and the
temperature was allowed to warm to room temperature and
stirred overnight. The mixture was concentrated in vacuo and
the residue dissolved in CH2Cl2 and washed several times with
deionized water. The organic layer was separated, dried over
anhydrous Na2SO4, filtered, and concentrated in vacuo. The
residue was purified by flash column chromatography (30%
acetone/hexane) to give the product 20f as a clear oil (0.573 g,
63%). Rf 0.48 (50% acetone/hexane); 1H NMR (300 MHz,
CDCl3) δ 8.00 (t, 2H), 7.68 (m, 2H), 7.38 (m, 20H), 7.11 (m,
4H), 7.02 (br s, 2H), 5.14 (s, 4H), 5.04 (s, 4H), 3.22 (m, 8H),
3.05 (br s, 8H), 2.60 (dd, 4H), 1.42-1.30 (m, 43H); 13C NMR δ
172.7, 169.9, 165.3, 155.7, 151.9, 146.9, 136.5, 136.5, 128.9,
128.9, 128.6, 128.5, 127.9, 127.4, 124.7, 123.4, 117.1, 82.8,
79.6, 76.7, 74.4, 71.5, 46.84, 46.78, 44.1, 39.6, 39.4, 28.9, 28.2,
27.1, 26.8, 26.4, 26.0; HRMS (FAB) m/z calcd for C78H103N6O15
(M + H)+ 1364.6836, found 1364.6752. Anal. Calcd for
1
(2.33 g, 86%). Rf 0.33 (EtAc/CHCl3 8:92); H NMR (500 MHz,
CDCl3) δ 8.00 (br s, 1H), 7.66 (br s, 1H), 7.38 (m, 10H), 7.13
(m, 2H), 5.14 (s, 2H), 5.08 (s, 2H), 4.62 (br s, 1H), 3.23 (br s,
2H), 3.08 (m, 6H), 1.57 (br s, 2H), 1.40 (m, 22H); HRMS (FAB)
m/z calcd for C38H52N3O7 (M + H)+ 662.3805, found 662.3800.
Anal. Calcd for C38H51N3O7: C, 68.96; H, 7.77; N, 6.35.
Found: C, 68.87; H, 7.93; N, 6.32.
N-[3-(4-Am in obu tyla m in o)p r op yl]-2,3-bis(ben zyloxy)-
ben za m id e, Tr iflu or oa cetic Acid Sa lt (18a ). A solution of
17a (2.30 g, 3.48 mmol) in CH2Cl2 (50 mL) was stirred at 0 °C
for 30 min. A solution of trifluoroacetic acid (9 mL) in CH2Cl2
(40 mL) was added dropwise over 1 h with vigorous stirring.
The mixture was stirred at room temperature for 2 h. After
adding benzene (100 mL), the solution was concentrated to
half of the total volume. This process was repeated three more
times. The solution was then concentrated in vacuo and the
1
residue used in the next step without further purification. H
C78H102N6O15: C, 68.70; H, 7.54; N, 6.16. Found: C, 68.76; H,
NMR (300 MHz, DMSO-3 drops of D2O) δ 8.38 (br s, 1H),
7.44 (m, 2H), 7.38 (m, 9H), 7.13 (m, 2H), 5.19 (s, 2H), 5.00 (s,
2H), 3.23 (m, 2H), 2.80 (m, 6H), 1.78 (m, 2H), 1.57 (br s, 4H).
2,3-P etr oba ctin ter t-Bu tyl Ester (20a ). In an “iron-free”
round-bottom flask 2-tert-butyl citrate (0.10 g, 0.4 mmol) was
dissolved in anhydrous THF (5 mL). N-Hydroxysuccinimide
(0.127 g, 1.10 mmol) and dicyclohexylcarbodiimide (DCC, 0.25
g, 1.21 mmol) were added to this solution and the reaction was
stirred at room temperature for 3 h producing the activated
1,3-bis-N-hydroxysuccinimide ester 19 and a white precipitate.
The 1H NMR (acetone-d6) showed that the signal at ∼δ2.8 (dd)
corresponding to the two CH2 groups of 2-tert-butyl citrate had
disappeared and a new group of signals at ∼δ 3.3 (dd)
corresponding to the two CH2 groups in the activated NHS
ester 19 had appeared. The solution was concentrated in vacuo
giving a white solid residue that was used directly for the next
coupling step. The solution of crude 19 in dry dioxane (5 mL)
was stirred at 10 °C for 10 min. The solution of crude 18a (0.63
g, 0.90 mmol, 2.2 equiv) in CH2Cl2 (5 mL) was treated with
1.6 mL of TEA at 0 °C. This solution of the free amine of 18a
was added to the solution containing 19 over 2 min. The
7.56; N, 6.11.
Ack n ow led gm en t. The authors wish to thank Dr.
Saleh Naser (UCF Microbiology) for his oversight and
advice in performing the M. hydrocarbonoclasticus
studies and Dr. David Powell at the University of
Florida Department of Chemistry for obtaining the mass
spectra.
Su p p or tin g In for m a tion Ava ila ble: The experimental
details for 16b-d17,21,32 and the characterization (NMR, HRMS,
and C, H, and N analyses) of 3,4-petrobactin 1b, 2,4-petro-
bactin 1c, 2,5-petrobactin 1d , a n d 2,3-homopetrobactin 1e,
17b, 17c, 20b, 20c, and 20d ; the 1H NMR spectral data of
compounds 7b, 7c, 7d , 13b, 14b, 15b, 17d , 17e, 18b, 18c, 18d ,
18e, and 20e; HRMS of compounds 7b, 7c, 7d , 15b, 17d , 17e,
and 20e; C, H, and N analyses of compounds 17d and 17e;
the 13C NMR spectra of compounds 15b and 20e. This material
is available free of charge via the Internet at http://pubs.acs.org.
J O049803L
J . Org. Chem, Vol. 69, No. 10, 2004 3537