Complexation and proton transfer by hydroxamic acids in model inhibited metallohydrolases: Formation of metal hydroxamate trimers

Article abstract of DOI:10.1016/j.inoche.2003.12.038

Reaction between the dimeric metal complexes, [M₂(μ-OAc _F)₂(OAc_F)₂(μ-H ₂O)(tmen)₂], M=Co(II), Ni(II), and aceto- and benzohydroxamic acids (AHA, BHA) gives the novel trimeric com

Full text of DOI:10.1016/j.inoche.2003.12.038

Inorganic Chemistry Communications 7 (2004) 495–498
www.elsevier.com/locate/inoche
Complexation and proton transfer by hydroxamic acids
in model inhibited metallohydrolases: formation
of metal hydroxamate trimers q
a,
b
a,
a
D.A. Brown *, G.J. Clarkson , N.J. Fitzpatrick *, W.K. Glass ,
a
A.J. Hussein , T.J. Kemp , H. Muller-Bunz
b
a
€
a
Department of Chemistry, Centre for Synthesis and Chemical Biology, Conway Institute of Biomolecular and Biomedical Research,
University College Dublin, Belfield, Dublin 4, Ireland
b
Department of Chemistry, University of Warwick, Coventry CV4 7AL, UK
Received 29 November 2003; accepted 17 December 2003
Published online: 4 March 2004
Abstract
Reaction between the dimeric metal complexes, [M2(l-OAcF)2(OAcF)2(l-H2O)(tmen)2], M ¼ Co(II), Ni(II), and aceto- and
benzohydroxamic acids (AHA, BHA) gives the novel trimeric complexes, [M3(l-OAcF)4(l-RA)2(tmen)2], M ¼ Co(II), Ni(II);
RA ¼ AA, BA, in which each hydroxamate bridges two metal centres via its deprotonated hydroxyl and the carbonyl oxygen bonds
one metal centre only together with the doubly protonated salt [(tmen) ꢀ 2H][OAcF]2. In contrast, self-assembly from
M(OAcF)2 ꢀ 4H2O, tmen and RHA gives the dibridged hydroxamate dimers [M2(l-OAcF)(l-RA)2(tmen)2] [OAcF].
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Crystal structures; Trinuclear cobalt; Nickel hydroxamates; Model inhibited complexes
Dinuclear metal hydrolases which catalyse the hy-
drolysis of a range of peptide and phosphate ester bonds
contain a dinuclear metal site featuring Zn(II), Ni(II),
Co(II) and Mn(II) with carboxylate bridges [1]. Hy-
droxamic acids are ubiquitous bioligands and, through
chelation of the zinc centre in the matrix metallopro-
teinases (MMP) by the hydroxamate anion, form the
basis of a wide range of therapeutic inhibitors [2]. In
contrast to these chelates, the acetohydroxamate-inhib-
ited C319A variant of Klebsiella aerogenes urease
(KAU) contains a novel l2–g1:g2 hydroxamate with the
deprotonated hydroxyl oxygen of the hydroxamic acid
bridging the two nickel atoms in the active site and the
carbonyl oxygen bonding one nickel atom only [3].
Inhibition of enzymes in biological systems by hy-
droxamic acids is generally assumed to involve initial
attack by a neutral hydroxamic acid molecule [4] but it is
not clear where the proton bonds on formation of the
chelates [2,3]. One suggestion based on the calculation of
proton dissociation energies is that amino acid side
chains in proteins might serve as proton acceptors [4].
In our previous model studies [5] reaction of the di-
meric nickel complex [Ni2(l-OAc)2(OAc)2(l-H2O)
(tmen)2] with a neutral hydroxamic acid (RHA ¼
MeHA, PhHA) gave attack by RHA at the central
water-dicarboxylate bridge with proton transfer to the
g1 acetates and formation of a dibridged hydroxamate
Abbreviations: OAc, CH3COOꢁ; OAcF, CF3COOꢁ; tmen,
N,N,N0,N0-tetramethylethylenediamine; AHA, acetohydroxamic acid;
AA, acetohydroxamate anion; BHA, benzohydroxamic acid; BA,
benzohydroxamate anion.
q Supplementary data associated with this article can be found, in the
online version, at doi:10.1016/j.inoche.2003.12.038.
E-mail address: Noel.Fitzpatrick@ucd.ie (N.J. Fitzpatrick).
* Corresponding authors. Tel.: +35317162297; fax: +35317162127
(D.A. Brown). Tel.: +353-1-7062283; fax: +353-1-7062127 (N.J.
Fitzpatrick).
complex [Ni2(l-OAc)ðl-RA)2(tmen)2][OAc] ꢀ AcOH,
with a very similar structure to that of the C319A
E-mail address: Noel.Fitzpatrick@ucd.ie (N.J. Fitzpatrick).
1387-7003/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.inoche.2003.12.038

496
D.A. Brown et al. / Inorganic Chemistry Communications 7 (2004) 495–498
variant of KAU(3). The formation of these bridged/
chelated hydroxamate model inhibited complexes is a
general reaction and occurs also for the imidazole series
in which each tmen capping ligand is replaced by two
imidazoles and the pivalate series in which the bridging
acetates are replaced by bridging pivalates [6]. However,
in contrast to the acetate and pivalate series, the present
note reports different reactions of the analogous triflu-
oroacetate bridged dimers [M2(l-OAcF)2(OAcF)2(l-
H2O)(tmen)2], M ¼ Co(II), Ni(II), with aceto- and
benzohydroxamic acids (AHA, BHA) which give proton
transfer to the tmen nitrogen atoms and formation of
the doubly protonated salt, [(tmen) ꢀ 2H][OAcF]2 (I) with
H-bonding between the protons and the trifluoroace-
tates as shown in Fig. 1 together with the novel tri-
meric complexes [M3(l-OAcF)4(l-RA)2(tmen)2] (II),
(M ¼ Co(II), Ni(II); RA ¼ AA, BA) in which the central
metal atom is linked to the outer metal atoms by hy-
droxamates in the bridging/chelating bonding mode and
bridging trifluoroacetates with the outer metal atoms
retaining their capping tmen ligands (Scheme 1, path c).
The structure of the cobalt-acetohydroxamate trimer is
shown in Fig. 2 and similar structures occur in the other
three trimers (M ¼ Co(II), RA ¼ BA; M ¼ Ni(II),
RA ¼ AA, BA).
F
O
F
F
R
O
H
N
O
N
N
N
N
Path a
MeOH, RHA,tmen
O
O
M(OAcF)2.4H2
M
M
F3
C6
O
O
N
F
O
H
F1
O
O2
C7
MeOH,
tmen
Path b
R
F
N2
C5
(III)
C8
F
F
F
F2
F
F
O1
F
F
O
O
O
O
H
O
N
N
N
N
C2
C1
O
M
O
F
F
M
H
O
N
N
H
O
O
F
F4
F
O3
C9
O
F
(I)
MeOH, RHA
Path c
O
F
F
F
F
C4
H
F
N1
F
F
F5
C10
F6
F
F
F
F
O
O4
R
O
F
F
O
H
O
N
O
O
N
N
O
C3
N
N
O
M
M
M
O
O
O
M=Ni(II), Co(II)
RHA=MeHA, PhHA
O
Fig. 1. The molecular structure of [(tmen) ꢀ 2H][OAcF]2, (I). Selected
F
F
O
F
N
F
F
H
R
ꢀ
bond lengths (A) and angles (°): N(1)–C(4), 1.474(4); N(1)–C(3),
F
1.484(4); N(1)–C(1), 1.497(3); N(1)–H(1)ꢀ ꢀ ꢀO(4), 2.708(3); C(1)–C(2),
1.501(4), C(9)–O(4), 1.248(3); C(9)–O(3), 1.212(3), C(4)–N(1)–C(3),
110.2(2); C(4)–N(1)–C(1), 115.7(2); O(3)–C(9)–O(4), 129.1(2).
(II)
Scheme 1.
N2
C3
O3
O4
H1
O1
Co2
Co1
N3
O5
N1
O4
O2
C
O6
C1
H
C5
N
O
F
Co
ꢀ
Fig. 2. The molecular structure of [Co3(l-OAcF)4(l-AA)2 (tmen)2], (II). Selected bond lengths (A) and angles (°): Co(1)–Co(2), 3.550(5); Co(1)–O(1),
2.0481(13); Co(2)–O(1), 2.0655(12); Co(2)–O(5), 2.1121(15); Co(1)–O(6), 2.0956(13); Co(1)–O(4), 2.1605(14); Co(2)–O(3), 2.0795(15); Co(2)–N(3),
2.1796(17); Co(2)–N(2), 2.205(2); N(1)–H(1)ꢀ ꢀ ꢀO(4)#1, 2.690(2); Co(1)–O(1)–Co(2), 119.31(6); O(1)#1–Co(1)–O(1), 180.00; O(1)–Co(2)–O(5),
89.83(5); O(1)–Co(1)–O(6), 89.03(5).

D.A. Brown et al. / Inorganic Chemistry Communications 7 (2004) 495–498
497
In contrast, self-assembly from M(OAcF)2 ꢀ 4H2O
(M ¼ Co(II), Ni(II)), tmen and RHA (RHA ¼ AHA,
BHA) gives the dibridged hydroxamate dimers [M2(l-
OAcF)(l-RA)2(tmen)2][OAcF] with very similar struc-
tures to those reported previously for the acetate and
pivalate series [5,6] (Scheme 1, path a). The structure of
the cobalt-acetohydroxamate dimer (III) is shown in
Fig. 3 with similar structures for the other three dimers.
Experimental details for the preparation of I, II and
III are given in the supporting information.
Crystallographic data for I, II and III are given in
Table 1 and selected bond angles and distances in Figs.
1–3 respectively. Crystal data for I and II were collected
using a Bruker SMART APEX CCD area detector
diffractometer and for III using a Siemens SMART
CCD area-detector diffractometer. A full sphere of the
reciprocal space was scanned by phi–omega scans for I
and II and a full hemisphere for III. Numerical ab-
sorption correction was performed by the program
SADABS [7]. The structures were solved by direct
methods using SHELXTL-PC [8] and refined by full
matrix least-squares on F 2 for all data using SHELXL-
97 [9]. For I and II, hydrogens attached to disordered
carbons were added at calculated positions and refined
using a riding model. Their isotropic temperature fac-
tors were fixed to 1.2 (1.5 for methyl hydrogens) times
the equivalent isotropic displacement parameter of the
carbon atom the H-atom is attached to. All other hy-
Fig. 3. The molecular structure of [Co2 (l-OAcF) (l-AA)2
(tmen)2][OAcF], (III). Selected bond lengths (A) and angles (°): Co1–
Co2, 3.0280(0.0015); Co1–O21, 2.080(5); Co1–O31, 2.102(5); Co1–
O23, 2.081(5); Co1–O41, 2.110(5); Co2–O40, 2.121(5); Co2–N15,
2.175(6); Co2–N11, 2.236(7); N(32)–H(32A)ꢀ ꢀ ꢀO(51), 2.843(9); O21–
Co1–O31, 87.2(2); O23–Co1–O31, 94.33(19); Co1–O21–Co2,
91.74(19); N05–Co1–N02, 82.1(2).
ꢀ
Table 1
Crystal data and structure refinement for complexes I, II, III
Parameters
I
II
III
Formula
C10H18F6N2O4
344.26
Monoclinic
C24H40Co3F12N6O12
1009.41
Triclinic
C20H40Co2F6N6O8
724.44
Orthorhombic
Molecular weight
Crystal system
Space group
ꢁ
P21=n
P1
P212121
ꢀ
a/A
9.3892(11)
15.8868(18)
10.5003(12)
90
9.0953(13)
9.8345(14)
13.0368(18)
71.571(2)
75.979(2)
69.309(2)
1023.7(3)
293(2)
13.269(2)
15.244(2)
15.260(2)
90
ꢀ
b/A
ꢀ
c/A
a/°
b/°
c/°
102.080(2)
90
90
90
3
ꢀ
V /A
T/K
Z
1531.6(3)
293(2)
3086.7(8)
180(2)
4
0.71073
1
0.71073
4
0.71073
ꢀ
h/A
q
ðcalcÞ/mg mꢁ3
1.493
0.70 ꢂ 0.40 ꢂ 0.16
0.158
1.637
0.63 ꢂ 0.60 ꢂ 0.15
1.315
1.559
0.38 ꢂ 0.08 ꢂ 0.08
1.161
Crystal size (mm3)
l/mmꢁ1
hk l Ranges
)11,11; )18,18; )12,12
21373
)11,12; )12,12; )17,16
8800
)15,15; )18,11; )18,18
15961
Number of reflections collected
Independent reflections
Max. and min. transmission
RðF Þ½I > 2rðIÞꢃ/%
2703 (Rint ¼ 0:0267)
0.9752 and 0.8975
6.63
4604 (Rint ¼ 0:0160)
0.8272 and 0.4914
3.18
5421 (Rint ¼ 0:1148)
0.93 and 0.64
6.34
RwðF 2Þ (all data)/%
Goodness-of-fit on F 2
Largest peak and hole/e A
7.65
1.062
3.77
1.037
12.88
0.973
ꢁ3
ꢀ
0.655, )0.294
0.281 and )0.331
3.550(5)
0.613 and )0.812
3.0280(0.0015)
ꢀ
M–M distance (A)
[tmen ꢀ 2H][OAcF]2, I; [Co3(AA)2(OAcF)4(tmen)2], II; [Co2(AA)2(OAcF)(tmen)2][ OAcF], III.

498
D.A. Brown et al. / Inorganic Chemistry Communications 7 (2004) 495–498
drogens were located in the difference fourier map and
allowed to refine freely including isotropic temperature
factors. In III, all hydrogens were treated as the hy-
drogens attached to disordered carbons in I and II.
Anisotropic temperature factors were used for all non-
hydrogen atoms.
This result suggests that in related enzymes, e.g. ureases,
protonation may occur at the nitrogen atoms of ami-
noacid side chains in proteins as suggested previously [4].
Crystallographic data for the structures of I, II and
III reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre, CCDC.
A comparison of the structures of the trimeric and
dimeric cobalt hydroxamates II and III (Figs. 2 and 3,
respectively) shows the presence of bridging/chelating
hydroxamates in both cases with two bridges in the case
of the dimer III and single hydroxamate bridges between
each pair of cobalt atoms in II with the central cobalt
occupying an inversion centre with O(1)# 1–Co(1)–O(1)
at 180.000°. The Co(1)–Co(2) distance of 3.028(2) in III
is shorter than the corresponding distance Co(1)–Co(2)
of 3.550(5) in II, due to the presence of a double hy-
droxamate bridge in III compared with one hydroxa-
mate bridge for each pair of cobalt atoms in II. Small
lengthenings are present in some comparable distances
e.g. Co(1)–O(1) of 2.048(1) in II and Co(1)–O(21) of
2.080(5) in III reflecting the linear Co–O–Co moiety in
II compared with the bent bridging Co–O bonds in III,
where Co(1)–O(21)–Co(2) is 91.74 (19)°. In both com-
plexes the cobalt atoms are in near octahedral environ-
ments e.g. O(21)–Co(1)–O(31) is 87.2(2)° in III
compared with O(1)–Co(2)–O(5) is 89.83(5)° in II.
The different behaviour between the trifluoroacetate
and acetate bridged dimers is a consequence of their
different electronic properties. Thus OAcF is a weaker
donor then OAc and so the bridging water–carboxylate
bonds are more easily broken by the neutral hydroxamic
acid ligand for the former series, but the resulting tri-
fluoroacetic acid thereby formed is more acidic than
acetic acid and so protonates the tmen nitrogen atoms.
Acknowledgements
We thank the EU COST D21 programme, Project
D21/0001/00 and Dublin City Council for support.
References
[1] D.E. Wilcox, Chem. Rev. 96 (1996) 2435–2458.
[2] E.M.F. Muri, M.J. Nieto, R.D. Sindelar, J.S. Williamson, Curr.
Med. Chem. 9 (2002) 1631–1653.
[3] (a) M.A. Pearson, L.O. Nickel, R.P. Hausinger, P.A. Karplus,
Biochemistry 36 (1997) 8164–8172;
(b) S. Ciurli, S. Benini, W.R. Rypniewski, K.S. Wilson, S. Miletti,
S. Mangani, Coord. Chem. Rev. 190-192 (1999) 331–355.
[4] J. El Yazal, Y.P. Pang, J. Phys. Chem. 104 (2000) 6499–6504.
[5] M. Arnold, D.A. Brown, O. Deeg, W. Errington, W. Haase, K.
Herlihy, T.J. Kemp, H. Nimir, R. Werner, Inorg. Chem. 37 (1998)
2920–2925.
[6] D.A. Brown, W.K. Glass, N.J. Fitzpatrick, T.J. Kemp, W.
Errington, G. Clarkson, W. Haase, F. Karsten, A. Mahdi, Inorg.
Chim. Acta, in press; doi:10.1016/j.ica.2003.08.021.
[7] G.M. Sheldrick, SADABS, Empirical Absorption Corrections
Program, University of Gottingen, Germany, 1996.
[8] Siemens SHELXTL-PC Version 5.0 Reference Manual, Siemens
Industrial Automation, Inc., Analytical Instrumentation, Madison,
WI, 1994.
[9] G.M. Sheldrick, SHELXL97 Program for Crystal Structure
Refinement, University of Gottingen, Germany, 1997.