Solid-phase synthesis of methyl carboxymycobactin T 7 and analogues as potential antimycobacterial agents

Article abstract of DOI:10.1016/S0960-894X(03)00473-6

A solid-phase approach for the total synthesis of methyl carboxymycobactins 1a-d, with an on-resin cyclization leading to azopine 5 as the key step, was developed. The iron-affinity of these compounds was assessed by a competitive sulfoxine-binding assay, and antimycobacterial activity was tested against the growth of Mycobacterium avium.

Full text of DOI:10.1016/S0960-894X(03)00473-6

Bioorganic & Medicinal Chemistry Letters 13 (2003) 2553–2556
Solid-Phase Synthesis of Methyl Carboxymycobactin T 7
and Analogues as Potential Antimycobacterial Agents
Amruta R. Poreddy,a Otto F. Schall,a Garland R. Marshall,a,b
Colin Ratledgec and Urszula Slomczynskaa,*
aMetaPhore Pharmaceuticals, Inc., 1910 Innerbelt Business Center Drive, Saint Louis, MO 63114, USA
bDepartment of Biochemistry and Molecular Biophysics, Washington University School of Medicine,
660 South Euclid Avenue, Saint Louis, MO 63110, USA
cDepartment of Applied Biology, The University of Hull, Hull HU6 7RX, UK
Received 12 March 2003; accepted 30 April 2003
Abstract—A solid-phase approach for the total synthesis of methyl carboxymycobactins 1a–d, with an on-resin cyclization leading
to azopine 5 as the key step, was developed. The iron-affinity of these compounds was assessed by a competitive sulfoxine-binding
assay, and antimycobacterial activity was tested against the growth of Mycobacterium avium.
# 2003 Elsevier Ltd. All rights reserved.
Iron is a growth-limiting nutrient for microorganisms,
and sequestration of ferric ion is an essential component
for survival of many pathogens. These organisms have
evolved ligands known as siderophores with extremely
high affinity for ferric ion in order to sequester iron
from their surroundings. An elaborate mechanism has
co-evolved for pumping this siderophore–iron complex
back into cells and releasing the iron by intracellular
reduction. In mycobacteria, the three types of iron-
binding compounds produced are generally classified as
mycobactins, exochelins, and carboxymycobactins,
based on their relative lipophilicities, overall structure,
and location within or outside the cell.1 Mycobactins
are intracellular siderophores isolated from pathogenic
strains, and are water insoluble with a lengthy alkyl
component that confines them to the cell membrane.
While the exochelins are water-soluble extracelluar
siderophores synthesized by non-pathogenic strains, the
carboxymycobactins are mycobactin-like extracellular
siderophores from pathogenic strains. Prior to the
structural elucidation of carboxymycobactins,2 all
extracelluar mycobacterial siderophores were referred to
as exochelins. Even though the term exochelin is still
used,3 we prefer the term carboxymycobactins because
of their structural similarity to mycobactins.
Methyl carboxymycobactin T 7 (1a) (Fig. 1), containing
a saturated alkyl chain terminated with a methyl ester
group, was isolated from Mycobacterium tuberculosis,3a
and was later found effective in prevention of cardiac
reperfusion injury,4 ischemia/reperfusion injury,5 and
vascular injury by arresting smooth muscle cell pro-
liferation.6 Our interest was due, in part, to the hypo-
thesized therapeutic potential7 of modified mycobactins
that might competitively bind iron and inhibit its
assimilation by targeted forms of mycobacteria. Myco-
bactin analogues with moderate to high inhibition of the
growth of Mycobacterium tuberculosis H37Rv were
recently described by Miller’s group.8 In order to inves-
tigate these concepts, we developed a solid-phase
approach for the synthesis of 1a and its analogues, in
addition to known solution methods.8,9
Retrosynthetically, disconnection at the bonds depicted
(Fig. 1) gives the required synthons for stepwise con-
struction on the solid support. Key intermediates to these
synthons comprise commercially available acids 4 that
include 3-hydroxybutyric- and suitably protected
2,3-diaminopropionic acids (Bachem, CA), the known
oxazoline 2,8a and the monoester 6.10 The choice of Bpoc
[1-methyl-1-(4-biphenyl)ethoxycarbonyl] for temporary
protection11 of the resin-bound azopine 5 was compa-
tible with the acid-labile linker of the Wang resin. More-
over, Bpoc protection for the e-amino group in lysine
derivative 3 was compatible with 2,4-dimethoxybenzyl as
*Corresponding author. Tel.: +1-314-426-4803; fax: +1-314-426-
7491; e-mail: uslomczynska@metaphore.com
0960-894X/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00473-6

2554
A. R. Poreddy et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2553–2556
Scheme 1. Synthesis of intermediates 3 and 5: (a) (i) Triton B, MeOH,
rt, 30 min; (ii) 2-(4-biphenylyl)-prop-2-yl 40-methoxycarbonylphenyl
carbonate, DMF, 50 ꢂC, 4 h; (iii) N,N0-diisopropyl-O-methylisourea,
CH2Cl2, reflux, overnight; 60%; (b) (i) 2,4-(MeO)2BnO-NH-Ns, Ph3P,
DIAD, THF, 37 ꢂC, overnight; (ii) 1 N LiOH, THF, rt, 6 h; 92%; (c)
NsCl, 2,6-lutidine, 1,2-dichloroethane (DCE); (d) Ph3P, DIAD, THF,
37 ꢂC, overnight; (e) (i) HS(CH2)2OH, DBU, DMF, rt, 2ꢁ30 min; (ii)
0.5 N LiOH, THF, rt, 5 h; (f) PyAOP, DIPEA, NMP, rt, 2ꢁ3 h; (g) (i)
1% TFA/5% triethylsilane (TES) in DCE, rt, 2ꢁ15 min; (ii) 5%
DIPEA in DCE; (iii) PhCOCl, pyridine, DCE, rt, 4 h; (iv) TFA–
CH2Cl2 (4:1), rt, 30 min.
Figure 1. Retrosynthetic analysis of methyl carboxymycobactin T 7
(1a) and analogues.
the O-protecting group,12 since the former can be selec-
tively cleaved with the nosyl (2-nitrobenzenesulfonyl,
Ns)13 as the N-protecting group on the hydroxylamine
moiety. The nosyl group was favored here based on our
own experience with its utility in the synthesis of poly-
hydroxamate analogues.14
acid (4a, XR=OH, R1=b-Me, R2=H) sodium salt in
the presence of HOBT/HBTU to give alcohol 15. The
acylation of the secondary alcohol group of 15 with
lysine derivative 3 under Mitsunobu conditions gave the
desired ester 16 (>50%) with inversion of configuration
at the methyl group, along with the by-product formed
by dehydration (presumably olefinic derivative, LC-MS
data). In order to improve the yield, several other
methods for activation of the carboxyl group were tes-
ted. EDC/HOAt in DMF (with or without DMAP) was
the only reagent that gave small amounts of the desired
ester (ꢀ30%, opposite stereochemistry at the methyl
group). Since the Mitsunobu reaction gave better yield
of the product and the by-product would not interfere
in subsequent steps, the Mitsunobu method was used
for generating 16. Intermediate 16, after Bpoc depro-
tection and neutralization, was coupled with oxazoline 2
in the presence of EDC/HOAt to give 17 (>50%). After
deprotection of the nosyl group, the resulting
hydroxylamine was then coupled to monoester 6 in the
presence of EDC/HOAt. Subsequent on-resin deprotec-
tion of the 2,4-dimethoxybenzyl group, followed by
cleavage from the resin provided crude methyl
carboxymycobactin T 7 (1a) with ꢀ50% purity. This
material was purified by RP-HPLC [YMC CombiPrep
ODS AQ C18 20ꢁ50 mm, 218 nm, 25 mL/min, 0–100%
B/10 min (A: water; B: acetonitrile)], and 1a was
obtained in 14% overall yield (87% pure).
The N,C-bisprotected 6-hydroxynorleucine 8 was pre-
pared by reacting a preformed carboxylate salt of l-6-
hydroxynorleucine (7) with Bpoc reagent, followed by
esterification with N,N0-diisopropyl-O-methylisourea.15
Reaction of 8 with O-[(2,4-dimethoxyphenyl)methyl]-N-
(2-nitrophenylsulfonyl)-hydroxylamine14 under Mitsu-
nobu conditions, followed by saponification afforded 3
(Scheme 1). The on-resin cyclization strategy leading to
azopine 5, constitutes the most important aspect of our
synthesis. Thus, Wang resin (1.1 mmol/g)-bound
hydroxylamine 916 was activated with the nosyl group
for alkylation with alcohol 8 under Mitsunobu condi-
tions to give intermediate 11. The nosyl group was
deprotected using thiolate anion17 and the methyl ester
hydrolyzed with 0.5 N LiOH/THF to insure proper
swelling of the resin. The resulting carboxylate salt 12
was activated with 7 - azabenzotriazol - 1 - yloxytris
(pyrrolidino)phosphonium
hexafluorophosphate
(PyAOP) and cyclized to the azopine 5, which was
characterized as its N-benzoyl derivative 13. The MS
(ESI) analysis of 13 [m/z 271 (M+Na)+] confirmed
formation of the azopine 5 with no linear byproduct,
indicating a high yield of cyclization (78%) based on
HPLC.
Methyl carboxymycobactin T 7 (1a) and its analogues
were assembled on the solid support as depicted in
Scheme 2. Each intermediate was evaluated by
RP-HPLC with UV detection and MS (ESI), after deri-
vatization of an aliquot of the resin with benzoyl chlo-
ride in the presence of pyridine, followed by TFA
cleavage. The resin-bound azopine derivative 14,
obtained from 5 by deprotection of the Bpoc-group and
neutralization, was coupled with (S)-3-hydroxybutyric
Methyl carboxymycobactin 1b, having an opposite ste-
reochemistry at the methyl group (S), was synthesized in
19% overall yield (86% pure) by using (R)-3-hydroxy-
butyric acid (4b, XR=OH, R1=a-Me, R2=H) sodium
salt early in the coupling with azopine 14 and repeating
all subsequent reactions.

A. R. Poreddy et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2553–2556
2555
Both amide analogues 1c and 1d (amide linkage instead
of central ester linkage) with pendant S- and R-amino
groups were synthesized utilizing suitably protected
1,3-diaminopropionic acid derivatives, Boc-Dap(Fmoc)-
OH (4c) and Boc-d-Dap(Fmoc)-OH (4d) respectively
(only 1c is shown in Scheme 2). Thus, resin-bound azo-
pine 14 was coupled with 4c in the presence of HATU to
give 18. After deprotection of the Fmoc group, the free
amine was coupled with lysine derivative 3 in the pre-
sence of EDC/HOAt to give 19, which was then trans-
formed to 1c (58%;ꢀ2:1 diastereomeric mixture;
LC-MS evidence) as described above in the preparation
of 1a. Similarly, 1d (63%, ꢀ2:1 diastereomeric mixture)
was synthesized by initial coupling of 4d with 14, and
subsequently repeating the later steps. The crude pro-
ducts were purified by preparative RP-HPLC as descri-
bed earlier, to obtain 1c and 1d in overall yields of 19%
and 24%, with purities of 93% and 91%, respectively.18
The formation of diastereomers was not expected. We
assume epimerization occurred during the coupling
reaction leading to the intermediate 19 rather than the
lysine derivative 3 being not optically pure, as this
problem was not observed during the synthesis of 1a
and 1b. The stated HPLC purities of the methyl car-
boxymycobactins 1a–d, did not include 3–6% of com-
pounds complexed with iron during isolation.
The iron-binding affinity of methyl carboxymycobactin
T 7 (1a) and its analogues was assessed using a compe-
titive spectrophotometric assay with 8-hydroxyquino-
line-5-sulfonic acid (sulfoxine).18b The measurement
reflects the relative affinities of the ligands, compared
with the sulfoxine-iron complex. The screening results
are summarized in Table 1 with desferrioxamine B
(DFO)18b as a control. Methyl carboxymycobactin T 7
(1a) shows better affinity than the corresponding S-iso-
mer 1b though somewhat inferior to DFO itself. The
selectivity of one isomer over the other is worth noting,
considering similar profiles obtained for isomers from
Table 1. Relative Fe(III) affinity of methyl carboxymycobactin T 7
(1a) and its analogues in sulfoxine binding assay
Compd
% Fe removed from pfSa
% Fe removed from pfLb
DFO
1a
1b
1c
1d
75
54
21
49
52
75
54
23
49
52
aPreformation of the sulfoxine–iron complex with subsequent addi-
tion of the test ligand.
bPreformation of the ligand–iron complex with subsequent addition
of sulfoxine.
Scheme 2. Synthesis of methyl carboxymycobactin T 7 (1a) and analogues: (a) (i) 1% TFA/5% TES in DCE, rt, 2ꢁ15 min; (ii) 5% DIPEA in DCE;
(b) (i) (S)-3-hydroxybutyric acid (4a) sodium salt, HOBT, HBTU, DMA, rt, 2 h; (c) 3, Ph3P, DEAD, THF, rt, 20 h; (d) (i) 1% TFA/5% TES in
DCE, rt, 2ꢁ15 min; (ii) 5% DIPEA in DCE; (iii) 2, EDC, HOAt, DMF, rt, 7 h; (e) (i) HS(CH2)2OH, DBU, DMF, rt, 2ꢁ30 min; (ii) 6, EDC, HOAt,
DMF, rt, 8 h; (iii) 1% TFA/5% TES in DCE, rt, 2ꢁ30 min; (iv) TFA–CH2Cl2 (4:1), rt, 30 min; (f) Boc-Dap(Fmoc)-OH (4c), HATU, DIPEA,
DMA, rt, 6 h; (g) (i) 25% piperidine in DMF, rt, 15 min; (ii) 3, EDC, HOAt, DMF, rt, 8 h.

2556
A. R. Poreddy et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2553–2556
competition experiments within the carboxymycobactin
family.9 Both isomers of amide analogues 1c and 1d dis-
played similar iron affinities, comparable to that of 1a.
Ewing, M. Tetrahedron Lett. 1995, 36, 4129. (b) Lane, S. J.;
Marshall, P. S.; Upton, R. J.; Ratledge, C. Biometals 1998, 11,
13. (c) Ratledge, C.; Dover, L. G. Annu. Rev. Microbiol. 2000,
54, 881.
3. (a) Gobin, J.; Moore, C. H.; Reeve, J. R., Jr.; Wong, D. K.;
Gibson, B. W.; Horwitz, M. A. Proc. Natl. Acad. Sci. U.S.A.
1995, 92, 5189. (b) Wong, D. K.; Gobin, J.; Horwitz, M. A.;
Gibson, B. W. J. Bacteriol. 1996, 178, 6394.
4. Horwitz, L. D.; Sherman, N. A.; Kong, Y.; Pike, A. W.;
Gobin, J.; Fennessy, P. V.; Horwitz, M. A. Proc. Natl. Acad.
Sci. U.S.A. 1998, 95, 5263.
5. Amersi, F.; Dulkanchainun, T.; Nelson, S. K.; Farmer,
D. G.; Kato, H.; Zaky, J.; Melinek, J.; Shaw, G. D.; Kupiec-
Weglinski, J. W.; Horwitz, L. D.; Horwitz, M. A.; Busuttil,
R. W. Transplantation 2001, 71, 112.
6. (a) Pahl, P. M. B.; Yan, X. D.; Hodges, Y. K.; Rosenthal,
E. A.; Horwitz, M. A.; Horwitz, L. D. J. Biol. Chem. 2000,
275, 17821. (b) Rosenthal, E. A.; Bohlmeyer, T. J.; Monnet,
E.; MacPhail, C.; Robertson, A. D.; Horwitz, M. A.; Burche-
nal, J. E. B.; Horwitz, L. D. Circulation 2001, 104, 2222.
7. Snow, G. A. Bacteriol. Rev. 1970, 34, 99.
8. (a) Hu, J.; Miller, M. J. J. Am. Chem. Soc. 1997, 119, 3462.
(b) Xu, Y.; Miller, M. J. J. Org. Chem. 1998, 63, 4314.
9. Geraci, L. S.; Levy, S. G.; Hudspeth, J. P.; Buswell, R. L.;
Stearns, J. F. U.S. Patent 5,952,492, 1999; Chem. Abstr. 1999,
131, 214556.
10. Newton, R. F.; Reynolds, D. P.; Greenwood, J.; Schein-
mann, F. J. Chem. Soc., Perkin Trans. 1 1980, 2346.
11. Kuisle, O.; Quinoa, E.; Riguera, R. A. J. Org. Chem.
1999, 64, 8063.
The methyl carboxymycobactins 1a–d were screened for
anti-mycobacterial activity against Mycobacterium
avium, one of the most common opportunistic patho-
gens associated with AIDS. Each compound was tested
at four different concentrations ranging from 2–20 mg/
mL and added at the time of inoculation to Myco-
bacterium avium. Growth was followed visually, and
after one week, all compounds showed strong inhibition
of growth (10–15% of the controls). However, growth
was about 75–80% of the controls after 2 weeks, and no
differences were seen after 3 and 4 weeks. At lower
concentrations (0.1–1 mg/mL), no inhibition was seen
with any of the compounds after one week. All com-
pounds were then tried in multiple doses with 20 mg/mL
each being added at time zero, and 7 and 14 days later.
While all compounds retarded growth after one week,
there was no discernible difference after 2 weeks, and
there was no inhibition after 3 weeks. This could be
explained by an initial slow degradation of the drug by
bacteria, which is sufficient to decrease the effective
concentration of the drug, eventually leading to the
bacterial growth returning to the level of controls.
In conclusion, a solid-phase method for the synthesis of
methyl carboxymycobactin T 7 (1a) and its analogues
was developed. These compounds were found to be very
effective iron-chelators, but the inhibitory activity
against Mycobacterium avium growth was found to
diminish after the first week of culture.
12. Barlaam, B.; Hamon, A.; Maudet, M. Tetrahedron Lett.
1998, 39, 7865.
13. Fukuyama, T.; Jow, C.-K.; Cheung, M. Tetrahedron Lett.
1995, 36, 6373.
14. Reddy, P. A.; Schall, O. F.; Wheatley, J. R.; Rosik, L. O.;
McClurg, J. P. M.; Marshall, G. R.; Slomczynska, U. Synth-
esis 2001, 1086.
15. Mathias, L. J. Synthesis 1979, 561.
16. Floyd, C. D.; Lewis, C. N.; Patel, S. R.; Whittaker, M. A.
Tetrahedron Lett. 1996, 37, 8045.
Acknowledgements
17. Miller, S. C.; Scanlan, T. S. J. Am. Chem. Soc. 1997, 119,
2301.
We gratefully acknowledge National Institutes of
Health for partial financial support (SBIR 1 R43
AI44584-01).
18. Published as a part in: (a) Slomczynska, U.; Reddy,
P. A.; Schall, O. F.; Osiek, T.; Naik, A.; Edwards, W. B.;
Wheatley, J.; Marshall, G. R. Peptides: The Wave of the
Future. In 2nd International/17th American Peptide Sympo-
sium, Lebl, M., Houghten, R. A., Eds.; American Peptide
Society: San Diego, 2001; pp 177–179 (b) Marshall, G. R.;
Reddy, P. A.; Schall, O. F.; Naik, A.; Beusen, D. D.; Ye, Y.;
Slomczynska, U. In Advances in Supramolecular Chemistry,
Vol. 8, Gokel, G. W., Ed.; Cerberus Press, Inc, 2002; pp 174–
243.
References and Notes
1. Vergne, A. F.; Walz, A. J.; Miller, M. J. Nat. Prod. Rep.
2000, 17, 99.
2. (a) Lane, S. J.; Marshall, P. S.; Upton, R. J.; Ratledge, C.;