Structure-activity relationships of the peptide deformylase inhibitor BB-3497: Modification of the metal binding group

Article abstract of DOI:10.1016/S0960-894X(02)00790-4

A series of analogues of the potent peptide deformylase (PDF) inhibitor BB-3497 containing alternative metal binding groups was synthesised. Enzyme inhibition and antibacterial activity data for these compounds revealed that the bidentate hydroxamic acid and N-formyl hydroxylamine structural motifs represent the optimum chelating groups on the pseudopeptidic BB-3497 backbone.

Full text of DOI:10.1016/S0960-894X(02)00790-4

Bioorganic & Medicinal Chemistry Letters 12 (2002) 3595–3599
Structure–Activity Relationships of the Peptide Deformylase
Inhibitor BB-3497: Modification of the Metal Binding Group
Helen K. Smith, R. Paul Beckett, John M. Clements, Sheila Doel, Stephen P. East,*
Steven B. Launchbury, Lisa M. Pratt, Zoe M. Spavold, Wayne Thomas,
Richard S. Todd and Mark Whittaker
British Biotech Pharmaceuticals Limited, Watlington Road, Oxford OX4 6LY, UK
Received 2 May 2002; revised 28 June 2002; accepted 11 September 2002
Abstract—A series of analogues of the potent peptide deformylase (PDF) inhibitor BB-3497 containing alternative metal binding
groups was synthesised. Enzyme inhibition and antibacterial activity data for these compounds revealed that the bidentate hydrox-
amic acid and N-formyl hydroxylamine structural motifs represent the optimum chelating groups on the pseudopeptidic BB-3497
backbone.
# 2002 Elsevier Science Ltd. All rights reserved.
Bacterial peptide deformylase (PDF) is a metalloenzyme
responsible for the removal of an N-formyl group from
the terminal methionine residue following protein
synthesis.1 PDF is essential forbacteiral gorwth and,
consequently, it has been widely recognised as an
attractive novel target for antibacterial chemotherapy.2
structural motifs. These included an N-acetyl hydro-
xylamine, an N-formylamine, a hydrazide, an amidox-
ime and two N-hydroxyureas. Additionally, we elected
to prepare other metal binding groups used in the MMP
field such as a carboxylic acid, a thiol, an amino-
carboxylate and some phosphorus-based chelating
groups. During the course of this work, significant con-
tributions in the area of PDF inhibitors containing
hydroxamic acids5 oraltneartive metal binding
groups5c,6 were reported from other laboratories,
underlining the current interest in PDF as an anti-
bacterial target.
We reported previously3 that the N-formyl hydro-
xylamine BB-3497 (Fig. 1) is an effective inhibitor
(IC50=7 nM) of the Escherichia coli PDF.Ni enzyme,
exhibiting potent antibacterial activity both in vitro and
in vivo. As part of an ongoing research programme, we
undertook a comprehensive SAR study of BB-3497
observing the effects on enzyme inhibition and anti-
bacterial activity. Herein, we describe the initial findings
of this study in which analogues of BB-3497 containing
alternative metal binding groups were synthesised and
tested in ourassays.
Forthe synthesis of compounds erlated to N-formyl
hydroxylamine, we were able to exploit a precursor used
7
in ourasymmetric route to BB-3497.
Thus, acetylation
of 1, followed by O-benzyl deprotection afforded the
N-acetyl hydroxylamine 2 (Scheme 1). Alternatively, 1
was treated with benzyloxycarbonyl isocyanate (pre-
pared from benzyl carbamate and oxalyl chloride8),
followed by hydrogenolysis to give the N-hydroxyurea
On the basis of ourextensive knowledge of erlevant
metal binding groups for the inhibition of matrix
metalloproteinases (MMPs),4 we synthesised a variety
of analogues of BB-3497 capable of eithermonodentate
or bidentate chelation to a metal centre. In particular,
we concentrated on functional groups that resembled
the N-formyl hydroxylamine and hydroxamic acid
*Corresponding author. Tel.: +44-1865-748747; fax: +44-1865-
781001; e-mail: east@britbio.co.uk
Figure 1. BB-3497 and metal binding group (MBG) analogues.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00790-4

3596
H. K. Smith et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3595–3599
Scheme 3. (i) MeCN, LDA, THF; (ii) NH2OH.HCl, K2CO3,
MeOH:H2O 54% (2 steps), then separate diastereoisomers (pre-
parative HPLC); (iii) CH3P(O)(OMe)2, n-BuLi, THF, 26%; (iv)
TMSBr, BSTFA, CH2Cl2, 41%, then separate diastereoisomers (pre-
parative HPLC).
Scheme 1. (i) Ac2O; (ii) H2, 10% Pd/C, MeOH, 36% (2 steps); (iii)
BnOCONCO, CH2Cl2, 76%; (iv) H2, Pd(OH)2/C, MeOH, 75%; (v)
H2, 10% Pd/C, EtOH, 32%; (vi) BtCHO, THF, 40%.
product 3. Additionally, the N-formamide compound
4 was synthesised from 1 by reduction of the N–O
bond and subsequent formylation with N-formyl-
benzotriazole9 (BtCHO).
12 with benzyl bromoacetate, and subsequent hydro-
genolysis provided the aminocarboxylate 14.
To draw analogy with compounds 6 and 13, we synthe-
sised 18, which contained an oxygen atom adjacent to
the metal binding group. Compound 15, derived from
racemic 2-hydroxycaproic acid, was reacted with
4-nitrophenyl chloroformate to give the activated car-
bonate 16. This was treated with O-benzyl hydroxyl-
amine to furnish 17. Hydrogenolysis under standard
conditions afforded the target compound 18 as a mix-
ture of diastereoisomers (Scheme 5).
The hydroxamic acid analogue 6 of BB-3497 and the
related hydrazide 7 were prepared from the succinate-
derived carboxylic acid 510 understandard peptide cou-
pling conditions (Scheme 2).
The amidoxime 10 and phosphonic acid 11 were pre-
pared as mixtures of separable diastereoisomers from
the a-bromo derivative 811 via nucleophilic displace-
ment (Scheme 3). Target 10 was synthesised through
reaction of 8 with the anion of acetonitrile to provide
the cyano derivative 9. Subsequent treatment of 9 with
hydroxylamine under basic conditions gave the required
amidoxime 10. Alternatively, reaction of the anion of
dimethyl methylphosphonate with
8 followed by
removal of the methyl groups using TMSBr-bis(-
trimethylsilyl)trifluoroacetamide (BSTFA)12 gave phos-
phonic acid 11.
l-Norleucine was used as the chiral building block to
access analogues of BB-3497 where the methylene unit
adjacent to the metal binding group was replaced with a
nitrogen atom. This strategy generated peptide 12,
which was treated with 4-nitrophenyl chloroformate
followed by reaction with hydroxylamine to give N-
hydroxyurea 13 (Scheme 4). Additionally, alkylation of
Scheme 4. (i) 4-Nitrophenyl chloroformate, pyridine, CH2Cl2; (ii)
NH2OH.HCl, Et3N, DMF, 30% (2 steps); (iii) BrCH2CO2Bn, K2CO3,
acetone, 25%; (iv) H2, 10% Pd/C, MeOH, 85%.
Scheme 5. (i) 4-Nitrophenyl chloroformate, pyridine, CH2Cl2; (ii)
NH2OBn, Et3N, CH2Cl2, 28% (2 steps); (iii) H2, 10% Pd/C, MeOH,
98%.
Scheme 2. (i) NH2OH, EDC, HOBt, DMF, 42%; (ii) N2H4.H2O,
EDC, HOBt, DMF, 40%.

H. K. Smith et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3595–3599
3597
The thiol analogue of BB-3497 was prepared from the
b-hydroxy carboamide 1913 (Scheme 6). Displacement
of the mesylate derived from alcohol 19 with potassium
thioacetate was followed by saponification with aqueous
LiOH to give thiol 20.
Deletion of a chelating N-hydroxyl group as in 4 com-
pletely abolished activity against PDF.
The hydroxamic acid 6 is an excellent inhibitorof the
enzyme and also exerted a strong antibacterial effect.
However, replacement of the hydroxyl group in 6 with
an amino functionality gave a hydrazide 7 which was
200 times less active against PDF and did not show
antibacterial activity. The amidoxime 10 was inactive
against the enzyme. The parent carboxylic acid 5
showed moderate activity against the enzyme but no
antibacterial activity. Substituting heteroatoms (oxygen
and nitrogen) in place of the methylene spacer in 6 pro-
vided compounds 13 and 18, with weak activity against
PDF. This result is consistent with observations by
Thorarensen et al.5c where an N-hydroxyurea was
identified as a weak PDF inhibitor(IC 50=2.2 mM),
except that ourinhibitor 13 did not exhibit antibacterial
activity.
Finally, the phosphinic acid 22 was prepared as a mix-
ture of separable diastereoisomers from the a,b-unsatu-
rated amide 217 by a modification of a procedure
described by Regan et al.14 (Scheme 7). Thus, conjugate
addition of bis(trimethylsilyl)phosphonite (prepared in
situ from ammonium phosphinate and BSTFA) to 21
provided phosphinic acid 22.
All compounds were screened for activity against the E.
coli PDF.Ni enzyme and Gram-positive and Gram-
negative in-house bacterial strains (Table 1).15
Replacement of the formyl proton with a methyl group
as in 2 oran amino substituent as in 3 greatly reduced
PDF activity. Examination of the E. coli PDF X-ray
structure16 and associated molecularmodelling of the
PDF enzyme active site suggested that these substrates
were too bulky to bind effectively to the metal centre.
The thiol compound 20, which would be expected to
chelate in a monodentate fashion to the Ni centre,
showed only weak activity against the PDF enzyme and
no antibacterial activity. Meinnel et al.6b described a
related thiol PDF inhibitor 23 (Fig. 2) that also
demonstrated weak activity against PDF (Ki (E. coli
PDF.Ni)=2.5 mM), howeverno micorbiological data
were provided. Additionally, Pei and co-workers6c,d
prepared compounds containing a thiol metal bind-
ing group (e.g., 24), which inhibited PDF in the low
nanomolararnge ( Ki (E. coli PDF.Fe)=19 nM) and
exerted a moderate antibacterial effect (E. coli MIC
Scheme 6. (i) MsCl, Et3N, CH2Cl2, 86%; (ii) HSAc, K2CO3, DMF,
44%; (iii) LiOH, H2O, 31%.
Scheme 7. (i) H2PO2NH4, BSTFA, 17%, then separate diastereoi-
somers (preparative HPLC).
Figure 2. Thiol PDF inhibitors.
Table 1. PDF enzyme inhibition and antibacterial activity of BB-3497 and its metal binding group analogues
E. coli
PDF.Ni IC50 (nM)
E. coli
MIC (mM)
S. capitis
MIC (mM)
Compd
MBG
BB-3497
2
3
4
6
CH2N(OH)CHO
CH2N(OH)COCH3
CH2N(OH)CONH2
CH2NHCHO
CH2CONHOH
CH2CONHNH2
CH2C(NH2)=NOH
CH2COOH
NHCONHOH
OCONHOH
NHCH2CO2H
CH2SH
7
60%@1 mMa
>1000
12.5
>200
>200
>200
25
>200
>200
>200
>200
>200
>200
>200
>200
>200
25
>200
>200
>200
100
>200
>200
>200
>200
>200
>200
>200
>200
>200
>1000
1
200
>1000
7
10b
5
200
13
18c
14
20
22b
11b
40%@1 mMa
20%@1 mMa
>1000
10%@1 mMa
>1000
>1000
CH2PO(OH)H
CH2PO(OH)2
Staphylococcus capitis (S. capitis).
aIC50 expressed as percentage inhibition at a concentration of 1 mM.
bBoth diastereoisomers were inactive.
cMixture of diastereoisomers.

3598
H. K. Smith et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3595–3599
75–100 mM). These data suggest that the nature of the
peptidic backbone contributes to inhibitory activity.
antibacterial activity but we have shown elsewhere3 that
improvements in antibacterial activity cannot necessa-
rily be attributed to improvements in enzyme inhibition.
Other factors that affect drug distribution such as bac-
terial membrane permeability, efflux and metabolism are
clearly important in determining antibacterial activity.
The phosphinic acid 11 and phosphinic acid 22, which
could potentially bind in a bidentate fashion to the
metal centre thus forming a 4-membered ring, or alter-
natively behave as a transition-state mimic, as proposed
by Pei et al.6a with theirphosphonates, were inactive in
ourassays.
Acknowledgements
Overall, the results provided in Table 1 demonstrate
that the N-formyl hydroxylamine and hydroxamic acid
represent the optimum metal chelating groups for the
pseudopeptidic backbone of BB-3497 in terms of PDF
inhibition and antibacterial activity. Other metal bind-
ing group analogues such as the hydrazide 7 and the
carboxylic acid 5, which were weak PDF inhibitors,
were not potent enough to exert an antibacterial effect.
We thank Drs. Andy Ayscough, Mario Lobell and Jac
Wijkmans of British Biotech for useful discussions.
References and Notes
1. (a) Adams, J. M.; Capecchi, M. Proc. Natl. Acad. Sci.
U.S.A. 1966, 55, 147. (b) Adams, J. M. J. Mol. Biol. 1968, 33,
571.
A diagrammatic representation of the X-ray structure
forBB-3497 bound to the active site of the E. coli PDF
enzyme is shown in Figure 3.3 Based on this and our
structure of the PDF-actinonin complex,3 we predict
that 6 binds in a similarbidentate mode.
2. (a) Pei, D. Emerg. Ther. Targets 2001, 5, 23. (b) Giglione,
C.; Meinnel, T. Emerg. Ther. Targets 2001, 5, 41. (c) Yuan, Z.;
Trias, J.; White, R. J. Drug Discov. Today 2001, 6, 955. (d)
Clements, J. M.; Ayscough, A. P.; Keavey, K.; East, S. P.
Curr. Med. Chem. – Anti-Infective Agents 2002, 1, 239.
3. Clements, J. M.; Beckett, R. P.; Brown, A.; Catlin, G.;
Lobell, M.; Palan, S.; Thomas, W.; Whittaker, M.; Wood, S.;
Salama, S.; Baker, P. J.; Rodgers, H. F.; Barynin, V.; Rice,
D. W.; Hunter, M. G. Antimicrob. Agents Chemother. 2001,
45, 563.
In both BB-3497 and 6, the two oxygen atoms of the
metal binding group establish a bidentate coordination
to the metal centre. Additionally, they can hydrogen
bond with amino acid residues in the active site, enhan-
cing theirbinding potential to the enzyme. These
observations correlate well with the structure reported
forthe PDF.Ni enzyme bound to the substarte, Met-
Ala-Ser,17 in which two watermolecules at the active
site occupy similarpositions to the oxygen atoms in our
inhibitors. Metal binding groups other than the N-for-
myl hydroxylamine or the hydroxamic acid are weak
PDF inhibitors. In some cases this may be explained by
a relatively weak monodentate interaction whilst in
otherexamples the steirc erquierments forbackbone
and sidechain interactions may not be compatible with
the orientation necessary for efficient bidentate chela-
tion to the metal.
4. Whittaker, M.; Floyd, C. D.; Brown, P.; Gearing, A. J. H.
Chem. Rev. 1999, 99, 2735.
5. (a) Chen, D. Z.; Patel, D. V.; Hackbarth, C. J.; Wang, W.;
Dreyer, G.; Young, D. C.; Margolis, P. S.; Wu, C.; Ni, Z. J.;
Trias, J.; White, R. J.; Yuan, Z. Y. Biochemistry 2000, 39,
1256. (b) Apfel, C.; Banner, D. W.; Bur, D.; Dietz, M.; Hirata,
T.; Hubschwerlen, C.; Locher, H.; Page, M. G. P.; Pirson, W.;
Rosse, G.; Specklin, J.-L. J. Med. Chem. 2000, 43, 2324. (c)
Thorarensen, A.; Deibel, M. R.; Rohrer, D. C.; Vosters, A. F.;
Yem, A. W.; Marshall, V. D.; Lynn, J. C.; Bohanon, M. J.;
Tomich, P. K.; Zurenko, G. E.; Sweeney, M. T.; Jensen, R. M.;
Nielsen, J. W.; Seest, E. P.; Dolak, L. A. Bioorg. Med. Chem.
Lett. 2001, 11, 1355.
6. (a) Hu, Y.-J.; Rajagopalan, P. T. R.; Pei, D. Bioorg. Med.
Chem. Lett. 1998, 8, 2479. (b) Meinnel, T.; Patiny, L.; Ragusa,
S.; Blanquet, S. Biochemistry 1999, 38, 4287. (c) Huntingdon,
K. M.; Yi, T.; Wei, Y.; Pei, D. Biochemistry 2000, 39, 4543. (d)
Wei, Y.; Yi, T.; Huntingdon, K. M.; Chaudhury, C.; Pei, D. J.
Comb. Chem. 2000, 2, 650. (e) Jayasekera, M. M. K.; Kendall,
A.; Shammas, R.; Dermeyer, M.; Tomala, M.; Shapiro, M. A.;
Holler, T. P. Arch. Biochem. Biophys. 2000, 381, 313.
7. Pratt, L. M.; Beckett, R. P.; Davies, S. J.; Launchbury,
S. B.; Miller, A.; Spavold, Z. M.; Todd, R. S.; Whittaker, M.
Bioorg. Med. Chem. Lett. 2001, 11, 2585.
In conclusion, a series of BB-3497 analogues with alter-
native metal binding groups have been synthesised. The
nature of the metal chelating group was found to be
critical for PDF inhibitory activity and the optimum
groups on the pseudopeptidic backbone of BB-3497
were the N-formyl hydroxylamine and the hydroxamic
acid. These two compounds also demonstrated the best
8. Grehn, L.; Almeida, M. L. S.; Ragnarsson, U. Synthesis
1988, 992.
9. Katritzky, A. R.; Chang, H.-K.; Yang, B. Synthesis 1995,
503.
10. (a) The succinate portion of compound 5 was prepared by
employing an asymmetric alkylation on an Evans N-hexanoyl
oxazolidinone with t-butyl bromoacetate. See: (a) Beckett,
R. P.; Crimmin, M. J.; Davis, M. H.; Spavold, Z. Synlett 1993,
137. (b) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am.
Chem. Soc. 1982, 104, 1737.
11. Compound 8 was prepared from 2-bromohexanoyl bro-
mide and H-t-Leu-NMe2.
12. Szardenings, A. K.; Gordeev, M. F.; Patel, D. V. Tetra-
hedron Lett. 1996, 37, 3635.
Figure 3. Binding mode of BB-3497 to the active site of E. coli PDF
from X-ray data and predicted binding of hydroxamic acid 6.

H. K. Smith et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3595–3599
3599
13. The carboxylic acid precursor to compound 19 was pre-
pared from the Evans N-hexanoyl oxazolidinone via a TiCl4-
promoted asymmetric alkylation using 1,3,5-trioxane. See:
Takacs, J. M.; Jaber, M. R.; Vellekoop, A. S. J. Org. Chem.
1998, 63, 2742.
15. PDF inhibition assays and MICs were performed as
described previously.3
16. Chan, M. K.; Gong, W.; Rajagopalan, P. T. R.; Hao, B.;
Tsai, C. M.; Pei, D. Biochemistry 1997, 36, 13904.
17. Becker, A.; Schlichting, I.; Kabsch, W.; Groche, D.;
Schultz, S.; Wagner, A. F. V. Nat. Struct. Biol. 1998, 5,
1053.
14. Boyd, E. A.; Corless, M.; James, K.; Regan, A. C. Tetra-
hedron Lett. 1990, 31, 2933.