Synthesis, structure, and DNA cleavage properties of copper(II) complexes of 1,4,7-triazacyclononane ligands featuring pairs of guanidine pendants

Article abstract of DOI:10.1021/ic1018136

Two new ligands, L¹ and L², have been prepared via N-functionalization of 1,4,7-triazacyclononane (tacn) with pairs of ethyl- or propyl-guanidine pendants, respectively. The X-ray crystal structure of [CuL¹](ClO₄)₂ (C1) isolated from basic solution (pH 9) indicates that a secondary amine nitrogen from each guanidine pendants coordinates to the copper(II) center in addition to the nitrogen atoms in the tacn macrocycle, resulting In a five-coordinate complex with intermediate square-pyramidal/trigonal bipyramidal geometry. The guanidines adopt an unusual coordination mode in that their amine nitrogen nearest to the tacn macrocycle binds to the copper(II) center, forming very stable five-membered chelate rings. A spectrophotometry pH titration established the pK_app for the deprotonation and coordination of each guanidine group to be 3.98 and 5.72, and revealed that [CuL¹]²⁺ Is the only detectable species present in solution above pH ～8. The solution speciation of the CuL ² complex (C2) is more complex, with at least 5 deprotonation steps over the pH range 4-12.5, and mononuclear and binuclear complexes coexisting. Analysis of the spectrophotometric data provided apparent deprotonation constants, and suggests that solutions at pH ～7.5 contain the maximum proportion of polynuclear complexes. Complex C1 exhibits virtually no cleavage activity toward the model phosphate diesters, bis(p-nitrophenyl)phosphate (BNPP) and 2-hydroxypropyl-p-nitrophenyl phosphate (HPNPP), while C2 exhibits moderate activity. For C2, the respective k_obs values measured at pH 7.0 (7.24 (±0.08) × 10^-5S^-1 (BNPP at 50°C) and 3.2 (±0.3) × 10^-5 s^-1 (HPNPP at 25°C)) are 40- and 10-times faster than [Cu(tacn)(OH₂) ₂]²⁺ complex. Both complexes cleave supercoiled pBR 322 plasmid DNA, indicating that the guanidine pendants of [CuL¹] ²⁺ may have been displaced from the copper coordination sphere to allow for DNA binding and subsequent cleavage. The rate of DNA cleavage by C2 Is twice that measured for [Cu(tacn)(OH₂)₂]²⁺, suggesting some degree of cooperativity between the copper center and guanidinium pendants In the hydrolysis of the phosphate ester linkages of DNA. A predominantly hydrolytic cleavage mechanism was confirmed through experiments performed either in the presence of various radical scavengers or under anaerobic conditions.

Full text of DOI:10.1021/ic1018136

Inorg. Chem. 2011, 50, 621–635 621
DOI: 10.1021/ic1018136
Synthesis, Structure, and DNA Cleavage Properties of Copper(II) Complexes
of 1,4,7-Triazacyclononane Ligands Featuring Pairs of Guanidine Pendants
†
†
‡,§
,^
,†
Linda Tjioe, Tanmaya Joshi, Jo €e l Brugger, Bim Graham,* and Leone Spiccia*
†
‡
School of Chemistry, Monash University, Vic 3800, Australia, School of Earth & Environmental Science,
§
University of Adelaide, Adelaide, SA 5005 Australia, South Australian Museum, Adelaide, SA 5000, Australia,
and Medicinal Chemistry and Drug Action, Monash Institute of Pharmaceutical Sciences, Monash University,
^
Parkville, Vic 3052, Australia
Received September 3, 2010
1
Two new ligands, L and L , have been prepared via N-functionalization of 1,4,7-triazacyclononane (tacn) with pairs of
2
1
ethyl- or propyl-guanidine pendants, respectively. The X-ray crystal structure of [CuL ](ClO ) (C1) isolated from basic
4
2
solution (pH 9) indicates that a secondary amine nitrogen from each guanidine pendants coordinates to the copper(II)
center in addition to the nitrogen atoms in the tacn macrocycle, resulting in a five-coordinate complex with intermediate
square-pyramidal/trigonal bipyramidal geometry. The guanidines adopt an unusual coordination mode in that their
amine nitrogen nearest to the tacn macrocycle binds to the copper(II) center, forming very stable five-membered
chelate rings. A spectrophotometric pH titration established the pK for the deprotonation and coordination of each
app
1
guanidine group to be 3.98 and 5.72, and revealed that [CuL ] is the only detectable species present in solution
2þ
2
above pH ∼8. The solution speciation of the CuL complex (C2) is more complex, with at least 5 deprotonation steps
over the pH range 4-12.5, and mononuclear and binuclear complexes coexisting. Analysis of the spectrophotometric
data provided apparent deprotonation constants, and suggests that solutions at pH ∼7.5 contain the maximum
proportion of polynuclear complexes. Complex C1 exhibits virtually no cleavage activity toward the model phosphate
diesters, bis(p-nitrophenyl)phosphate (BNPP) and 2-hydroxypropyl-p-nitrophenyl phosphate (HPNPP), while C2
-
5
-1
2þ
exhibits moderate activity. For C2, the respective k values measured at pH 7.0 (7.24 ((0.08) ꢀ 10
obs
s
(BNPP at
-
5 -1
5
0 ꢀC) and 3.2 ((0.3) ꢀ 10
s
(HPNPP at 25 ꢀC)) are 40- and 10-times faster than [Cu(tacn)(OH ) ] complex.
2
2
1
2þ
Both complexes cleave supercoiled pBR 322 plasmid DNA, indicating that the guanidine pendants of [CuL ] may
have been displaced from the copper coordination sphere to allow for DNA binding and subsequent cleavage. The rate
2
of DNA cleavage by C2 is twice that measured for [Cu(tacn)(OH ) ] , suggesting some degree of cooperativity
þ
2
2
between the copper center and guanidinium pendants in the hydrolysis of the phosphate ester linkages of DNA.
A predominantly hydrolytic cleavage mechanism was confirmed through experiments performed either in the presence
of various radical scavengers or under anaerobic conditions.
5
2
Introduction
DNA (10 years at pH 7, 25 ꢀC). However, processes that
either attach or detach phosphate groups from biological
molecules underpin many fundamental biological processes.
Since the uncatalyzed rates of cleavage of such groups from
biological molecules are too slow to sustain life, enzymes
The Human Genome Project has found that the human
genome consists of 20 000-25 000 protein-coding genes,
which are transcribed into the messenger ribonucleic acids
3-6
1
(mRNA) responsible for protein synthesis. Since even a
(
protein-based, as well as ribozymes) are used to promote
single mutation in any of these genes, or loss of an mRNA or
protein, can have serious implications, nature utilizes kineti-
cally inert linkages to connect the key components, viz.,
phosphodiester and peptide bonds in RNA/DNA and pro-
teins, respectively. This inertness is highlighted by the half-life
for the uncatalyzed cleavage of phosphate ester bonds in
7,8
phosphoryl transfer reactions in the biological arena.
(
3) Trawick, B. N.; Daniher, A. T.; Bashkin, J. K. Chem. Rev. 1998, 98,
939–960.
4) Gani, D.; Wilkie, J. Structure and Bonding; Springer: Berlin, Germany,
997.
5) Strater, N.; Lipscomb, W. N.; Klabunde, T.; Krebs, B. Angew. Chem.,
(
1
(
*
To whom correspondence should be addressed. E-mail: Bim.Graham@
Int. Ed. Engl. 1996, 35, 2024–2055.
pharm.monash.edu.au (B.G.); Leone.Spiccia@monash.edu (L.S.). Fax: þ61
(6) Hegg, E. L.; Burstyn, J. N. Coord. Chem. Rev. 1998, 173, 133–165.
(7) Cotton, F. A.; Hazen, E. E. J.; Legg, M. Proc. Natl. Acad. Sci. U.S.A.
1979, 76, 2551–2555.
(8) Weber, D. J.; Meeker, A. K.; Mildvan, A. S. Biochemistry 1991, 30,
6103–6114.
3
9903 9582 (B.G.); þ61 3 9905 4597 (L.S.).
1) International Human Genome Sequencing Consortium Nature 2001,
09, 860–921.
2) Radzicka, A.; Wolfenden, R. Science 1995, 267, 90–93.
(
4
(
r 2010 American Chemical Society
Published on Web 12/09/2010
pubs.acs.org/IC

6
22 Inorganic Chemistry, Vol. 50, No. 2, 2011
Tjioe et al.
A key feature of many (ribo)nucleases and phosphatases is
find utility as artificial restriction enzymes in molecular bio-
logy research, and as nucleic acid-targeting therapeutics.
16-36
the presence of one or more metal ions within the active sites
of these enzymes that is crucial for activity. These ions faci-
litate phosphate cleavage/phosphoryl group transfer in a
variety of ways: (i) Lewis acid activation of the substrate
through coordination to the metal ion(s); (ii) provision of
metal-bound hydroxide or alkoxide groups to serve as
nucleophiles or bases; (iii) stabilization of transition states;
and (iv) assisting the departure of leaving groups. Functional
groups of amino acid side chains present within the enzyme
active sites serve to synergistically reinforce the catalytic
action of these metal ions. A good example is provided by
the well-studied enzyme alkaline phosphatase (AP), which
uses two zinc centers, in conjunction with key serine and
arginine residues, to help promote the rapid cleavage of
Many of the first generation enzyme mimics were simple
mononuclear chelate complexes featuring bidentate or tri-
dentate ligands, such as bipyridine (bipy), 1,4,7-triazacyclo-
nonane (tacn), and bis(2-pyridylmethyl)amine (BPA or
36-47
DPA).
More recently, increasing attention has beenpaid
to the design and synthesis of more sophisticated supporting
ligand structures featuring auxiliary amino, ammonium and
34,39,40,48-59
guanidinium groups.
These yield complexes that
mimic the cooperativity between metal ions and key amino
acid residues found within the active sites of the metallo-
enzymes, and which generally cleave model phosphate esters
or nucleic acids significantly more rapidly than their non-
39,40
functionalized counterparts. Kr €a mer and co-workers,
for
9
phosphate monoesters. The positively charged guanidinium
example, showed that the copper(II) complex of a bipy-based
ligand with inbuilt ammonium groups (Figure 1A) cleaves a
model phosphate diester 2900-times faster than the complex
of the corresponding ligand without hydrogen bond donors.
group present in arginine is postulated to assist with substrate
activation and transition state stabilization, while the serine
residue provides the attacking nucleophile, leading to the
formation of a phosphoserinyl intermediate during AP’s
catalytic cycle.
4
1
More recently, Chin and co-workers have reported a
copper(II) complex with two amino groups (Figure 1B) in
close proximity to the metal center which accelerates the rate
Inspired by the occurrence of metal-containing (ribo)-
nucleases and phosphatases, many research groups have
sought to develop low-molecular weight metal complexes
that are able to cleave biologically important phosphate
0
0
of hydrolysis of 2 -3 -cyclic adenosine monophosphate
7
(cAMP) by a factor of 2 ꢀ 10 . Our group has studied the
DNA cleavage activity of a series of copper(II) complexes of
6,10-15
esters.
This research has been stimulated by the reali-
zation that small, hydrolytically active metal complexes (and
their conjugates with various targeting agents) may potentially
(
36) Qian, J.; Gu, W.; Liu, H.; Gao, F.; Feng, L.; Yan, S.; Liao, D.;
Cheng, P. Dalton Trans. 2007, 1060–1066.
37) Li, J.-H.; Wang, J.-T.; Zhang, L.-Y.; Chen, Z.-N.; Mao, Z.-W.; Ji,
L.-N. Inorg. Chim. Acta 2009, 362, 1918–1924.
38) Li, J.-H.; Wang, J.-T.; Hu, P.; Zhang, L.-Y.; Chen, Z.-N.; Mao,
Z.-W.; Ji, L.-N. Polyhedron 2008, 27, 1898–1904.
39) Kovari, E.; Heitker, J.; Kramer, R. J. Chem. Soc. Chem. Commun.
(
(
(
9) Wilcox, D. E. Chem. Rev. 1996, 96, 2435–2458.
10) Yun, J. W.; Tanase, T.; Lippard, S. J. Inorg. Chem. 1996, 35, 7590–
(
7
600.
(
(
11) Gomez-Tagle, P.; Yatsimirsky, A. K. J. Chem. Soc., Dalton Trans.
1
995, 1205–1206.
2
001, 2663–2670.
(40) Kovari, E.; Kramer, R. J. Am. Chem. Soc. 1996, 118, 12704–12709.
(41) Wall, M.; Linkletter, B.; Williams, D.; Lebuis, A.-M.; Hynes, R. C.;
(
(
(
(
12) Hay, R. W.; Govan, N. Polyhedron 1998, 17, 463–468.
13) Hegg, E. L.; Burstyn, J. N. J. Am. Chem. Soc. 1995, 117, 7015–7016.
14) Hegg, E. L.; Burstyn, J. N. Inorg. Chem. 1996, 35, 7474–7481.
15) Hegg, E. L.; Deal, K. A.; Kiessling, L. L.; Burstyn, J. N. Inorg. Chem.
Chin, J. J. Am. Chem. Soc. 1999, 121, 4710–4711.
42) Sheng, X.; Guo, X.; Lu, M.-X.; Lu, G., -Y.; Shao, Y.; Liu, F.; Xu, Q.
Bioconjugate Chem. 2008, 19, 490–498.
43) Kim, J. H.; Youn, M. R.; Lee, Y.-A.; Kim, J. M.; Kim, S. K. Bull.
Korean Chem. Soc. 2007, 28, 263–270.
44) Iranzo, O.; Elmer, T.; Richard, J. P.; Morrow, J. R. Inorg. Chem.
003, 42, 7737–7746.
45) Zhao, Y. M.; Zhu, J. H.; He, W. J.; Yang, Z.; Zhu, Y. G.; Li, Y. Z.;
(
1
997, 36, 1715–1718.
16) Erkkila, K. E.; Odom, D. T.; Barton, J. K. Chem. Rev. 1999, 99,
777–2796.
(
(
2
(
(
(
(
(
(
(
17) Rajski, S. R.; Williams, R. M. Chem. Rev. 1998, 98, 2723–2796.
2
18) Burger, R. M. Chem. Rev. 1998, 98, 1153–1170.
(
19) Sherman, S. E.; Lippard, S. J. Chem. Rev. 1987, 87, 1153–1181.
20) Cowan, J. A. Curr. Opin. Chem. Biol. 2001, 5, 634–642.
21) Morrow, J. R.; Iranzo, O. Curr. Opin. Chem. Biol. 2004, 8, 192–200.
22) Baker, B. F.; Lot, S. S.; Kringle, J.; Cheng-Flournoy, S.; Villiet, P.;
Zhang, J. F.; Guo, Z. J. Chem.;Eur. J. 2006, 12, 6621–6629.
(46) Wahnon, D.; Hynes, R. C.; Chin, J. J. Chem. Soc. Chem. Commun.
1994, 1441–1442.
Sasmor, H. M.; Siskowski, A. M.; Chappell, L. L.; Morrow, J. R. Nucl.
Acids. Res. 1999, 27, 1547–1551.
(47) Belousoff, M. J.; Tjioe, L.; Graham, B.; Spiccia, L. Inorg. Chem.
2008, 47, 8641–8651.
(48) Jubian, V.; Dixon, R. P.; Hamilton, A. D. J. Am. Chem. Soc. 1992,
114, 1120–1121.
(49) Dixon, R. P.; Geib, S. J.; Hamilton, A. D. J. Am. Chem. Soc. 1992,
114, 365–366.
(50) Jubian, V.; Veronese, A.; Dixon, R. P.; Hamilton, A. D. Angew.
Chem., Int. Ed. 1995, 34, 1237–1239.
(
23) Perreault, D. M.; Ansyln, E. V. Angew. Chem., Int. Ed. 1997, 36,
4
32–450.
(
24) Cheng, C. C.; Rokita, S. E.; Burrows, C. J. Angew. Chem., Int. Ed.
1
993, 32, 277–278.
(
25) Friedel, M. G.; Pieck, J. C.; Klages, J.; Dauth, C.; Kessler, H.; Carell,
T. Chem.;Eur. J. 2006, 12, 6081–6094.
(
(
26) Mancin, F.; Tecilla, P. New. J. Chem. 2007, 31, 800–817.
(51) Muche, M.-S.; Goebel, M. W. Angew. Chem., Int. Ed. 1996, 35, 2126–
2129.
27) Noll, D. M.; Mason, T. M.; Miller, P. S. Chem. Rev. 2006, 106,
2
5
5
77–301.
(52) Ariga, K.; Ansyln, E. V. J. Org. Chem. 1992, 57, 417–419.
(53) Ait-Haddou, H.; Sumaoka, J.; Wiskur, S. J.; Folmer-Andersen, J. F.;
Ansyln, E. V. Angew. Chem., Int. Ed. 2002, 41, 4013–4016.
(54) Feng, G.; Mareque-Rivas, J. C.; de-Rosales, R. T. M.; Williams,
N. H. J. Am. Chem. Soc. 2005, 127, 13470–13471.
(55) Feng, G.; Mareque-Rivas, J. C.; Williams, N. H. Chem. Commun.
2006, 1845–1847.
(56) Chen, X. Q.; Wang, J. Y.; Sun, S. G.; Fan, J. L.; Wu, S.; Liu, J. F.;
Ma, S. J.; Zhang, L. Z.; Peng, X. J. Bioorg. Med. Chem. Lett. 2008, 18,
109–113.
(57) He, J.; Hu, P.; Wang, Y. J.; Tong, M.-L.; Sun, H. Z.; Mao, Z.-W.; Ji,
L.-N. Dalton Trans. 2008, 3207–3214.
(58) Sheng, X.; Lu, X. M.; Chen, Y. T.; Lu, G., -Y.; Zhang, J. J.; Shao, Y.;
Liu, F.; Xu, Q. Chem.;Eur. J. 2007, 13, 9703–9712.
(59) Shao, Y.; Sheng, X.; Li, Y.; Jia, Z. L.; Zhang, J. J.; Liu, F.; Lu, G. -Y.
Bioconjugate Chem. 2008, 19, 1840–1848.
(
28) Hettich, R.; Schneider, H.-J. J. Am. Chem. Soc. 1997, 119, 5638–
647.
(
29) Ren, R.; Yang, P.; Zheng, W.; Hua, Z. Inorg. Chem. 2000, 39, 5454–
463.
(
30) Komiyama, M.; Kina, S.; Matsumura, K.; Sumaoka, J.; Tobey, S.;
Lynch, V. M.; Ansyln, E. V. J. Am. Chem. Soc. 2002, 124, 13731–13736.
31) Yang, M.-Y.; Richard, J. P.; Morrow, J. R. Chem. Commun. 2003,
832–2833.
32) Worm, K.; Chu, F.; Matsumoto, K.; Best, M. D.; Lynch, V. M.;
Ansyln, E. V. Chem.;Eur. J. 2003, 9, 741–747.
(
2
(
(
(
33) Jin, Y.; Cowan, J. A. J. Am. Chem. Soc. 2005, 127, 8408–8415.
34) An, Y.; Tong, M.-L.; Ji, L.-N.; Mao, Z.-W. Dalton Trans. 2006,
2
066–2071.
35) Corneillie, T. M.; Whetstone, P. A.; Lee, K. C.; Wong, J. P.; Meares,
C. F. Bioconjugate Chem. 2004, 15, 1389–1391.
(

Article
Inorganic Chemistry, Vol. 50, No. 2, 2011 623
Instrumentation and Methods. Infrared spectra were recorded
as KBr disks using a Bruker Equinox FTIR spectrometer fitted
-
1
with an ATR platform at 4.0 cm resolution. Microanalyses
were performed by the Campbell Microanalytical Service, Otago,
1
13
New Zealand. H and C NMR spectra were recorded at 25 ꢀC
in D O or CDCl (as listed) on a Bruker AC200, AM300 or
2
3
DX400 spectrometer. The chemical shifts, δ, were recorded on
the δ scale in parts per million (ppm) and were calibrated using
either tetramethylsilane (TMS) or signals because of the residual
protons of deuterated solvents. Abbreviations used to de-
Figure 1. Examples of metal-based synthetic nucleases.
1
scribe H NMR resonances are s (singlet), d (doublet), t (triplet),
q (quartet), m (multiplet), and br s (broad singlet). Low-resolution
electrospray ionization mass spectra (ESI-MS) were obtained
with a Micromass Platform II Quadrupole Mass Spectrometer
fitted with an electrospray source. The capillary voltage was set
at 3.5 eV and the cone voltage at 35 V. Thin Layer Chromatog-
raphy (TLC) was performed using silica gel 60 F-254 (Merck)
plates with detection of species present by UV irradiation or
1
2
Figure 2. Ligands, L and L , prepared in this study as tetrahydrochlor-
ide salts.
KMnO oxidation. UV-vis spectra were recorded in 1 cm
4
DPA ligands bearing guanidinium pendants of varying
quartz cuvettes using Varian Cary Bio 300 or 5G spectrophoto-
meters. Agarose gel electrophoresis of plasmid DNA cleavage
products was performed using a Biorad Mini-Protean 3 Elec-
trophoresis Module. Bands were visualized by UV light irradia-
tion, fluorescence imaged using an AlphaImager, and photo-
graphed with a CCD camera. The gel photographs were
analyzed with the aid of the program ImageQuaNT version
47
lengths (Figure 1C). Again, the combination of a copper(II)
center and guanidinium group in close proximity was found
to result in an enhanced ability to cleave plasmid DNA and,
in some cases, an RNA mimic (UpNP).
In this paper, we report the synthesis and cleavage proper-
ties of two further copper(II)-based synthetic nucleases
featuring in-built cationic hydrogen-bond donors. The li-
gands used in this study are based on the well-known tacn
macrocycle. Pioneering work by Burstyn and co-workers
demonstrated that the copper(II) complex of tacn itself is
capable of cleaving simple activated phosphate esters, DNA,
4
.1. Intensity values for supercoiled pBR 322 DNA were cor-
rected by a factor of 1.42 to account for the decreased ability of
ethidium bromide to intercalate into supercoiled DNA (Form I)
3
4
compared to nicked DNA (Form II). The relative proportions
of each different form of DNA were determined by dividing the
fluorescence intensity of each band by the sum of fluorescence
intensities of all bands in the same lane. Background fluores-
cence was determined by reference to a lane containing no DNA.
Caution: Although no problems were encountered in this work,
perchlorate salts are potentially explosive. They should be pre-
pared in small quantities and handled with care.
6
RNA and peptides. Building on this seminal study, a num-
ber of research groups, including our own, have shown that
N-functionalization of tacn with various groups can enhance
3
6,42-44
cleavage activity.
Here, we have prepared two new
tacn-based ligands featuring pairs of guanidinium pendants,
L and L (Figure 2), determined the crystal structure of the
copper(II) complex of L , and studied the solution speciation
1
2
Syntheses. Ethyl-4,7-bis(tert-butoxycarbonyl)-1,4,7-triaza-
cyclononane (1). A solution of iodoethane (1.04 g, 6.68 mmol)
in acetonitrile (50 mL) was added dropwise to a mixture of
1
of the two complexes. The kinetics of cleavage of three dif-
ferent phosphate diesters, namely pBR 322 plasmid DNA,
bis(p-nitrophenyl)phosphate (BNPP) and 2-hydroxypropyl-
p-nitrophenyl phosphate (HPNPP), by the two complexes
have also been investigatedto gain some insight into theeffect
of the appended positively charged guanidinium groups on
cleavage activity.
1
6
,4-bis(tert-butoxycarbonyl)-1,4,7-triazacyclononane (2.20 g,
.68 mmol), K CO (0.92 g, 13.4 mmol), and KI (0.12 g,
2
3
0.60 mmol) in acetonitrile (50 mL) and the resulting mixture
stirred for 1 h at room temperature, then refluxed overnight.
After cooling to room temperature, the inorganic salts were
filtered off and the solvent removed from the filtrate under
reduced pressure to yield 1 as a yellow oil. Yield: 2.01 g (84%).
1
H NMR (300 MHz, CDCl
3
): δ 1.00 (t, 3H, J=7.2 Hz, ethyl
), 2.58 (m, 6H, ethyl CH and tacn
ring CH ), 3.25 (m, 4H, tacn ring CH ), 3.47 (m, 4H, tacn ring
t
CH
3
), 1.47 (s, 18H, Bu CH
3
2
Experimental Section
2
2
Materials and Chemicals. Chemicals and solvents were of
reagent or analytical grade, and were used as received unless
13
t
CH ). C NMR (75 MHz, CDCl ): δ 12.6 (ethyl CH ), 28.3 ( Bu
CH
CH
2
3
2
3
3
), 49.6, 49.8 (tacn CH
), 53.0, 53.2 (tacn CH
2
), 50.1, 50.6 (tacn CHt
), 52.7 (ethyl
), 79.0 (quaternary Bu C), 155.4
2
2
otherwise indicated. Distilled H O and HPLC grade chloroform
2
˚
-1
were used throughout. THF was dried over 4 A molecular sieves
and then freshly distilled from Na/benzophenone prior to
(
(
CdO). IR (neat), υ(cm ) 2973s (υC-H), 2932s (υC-H), 1694s
υ
CdO), 1462s, 1413s, 1365s (υC-O), 1250s, 1175s, 988 m, 860 m,
60
use. 1,4-bis(tert-butoxycarbonyl)-1,4,7-triazacyclononane and
the sodium salt of 2-hydroxypropyl-p-nitrophenylphosphate
þ
7
73 m, 733 m. ESI-MS (m/z): 358.2 (100%) [M þ H] .
1-Ethyl-1,4,7-triazacyclononane (2). Compound 1 (1.02 g,
.86 mmol) was dissolved in a 1:1 v/v mixture of trifluoroacetic
61,62
(
NaHPNPP)
were synthesized according to literature pro-
2
acid (TFA)/CH Cl and the solution was stirred at room
temperature overnight. The pH was then slowly adjusted to 14
by the careful addition of 5 M NaOH and the product extracted
with CH
were dried over MgSO , and following filtration, the solvent was
removed under reduced pressure to yield 2 as a yellow oil. Yield:
0
cedures. pBR 322 plasmid DNA was purchased from Promega
Corporation. Milli-Q water used for DNA cleavage was ster-
ilized by autoclaving and all reaction solutions were prepared
according to standard sterile techniques. Deoxygenated water
was prepared by boiling distilled water under nitrogen for 4 h
and cooling while bubbling with nitrogen gas.
2
2
Cl
2 2
(3 ꢀ 100 mL). The combined organic fractions
4
1
.39 g (85%). H NMR (300 MHz, CDCl
3
): δ 1.05 (t, 3H, J =
(
60) Kimura, S.; Bill, E.; Bothe, E.; Weyhermuller, T.; Wieghardt, K.
7
.2 Hz, ethyl CH ), 2.59-2.70 (m, 6H, ethyl CH and tacn ring
3
2
J. Am. Chem. Soc. 2001, 123, 6025–6039.
CH ), 2.78-2.85 (m, 8H, tacn ring CH ), 3.15 (s br, 2H, NH).
(
(
61) Brown, D. M.; Usher, D. A. J. Chem. Soc. 1965, 6558–6564.
2
2
1
3
62) Tsang, J. S.; Neverov, A. A.; Brown, R. S. J. Am. Chem. Soc. 2003,
C NMR (75 MHz, CDCl
3
): δ 12.6 (ethyl CH
CH ), 46.2 (tacn CH ), 50.7 (tacn CH ), 51.4 (tacn CH ).
2 2 2 2
3
), 45.6 (ethyl
1
25, 1559–1566.

6
24 Inorganic Chemistry, Vol. 50, No. 2, 2011
Tjioe et al.
IR (neat), υ(cm-1) 3312 m br (υN-H), 1199 m (υC-N), 1078 m,
removed under reduced pressure to yield crude 7 as a yellow oil.
The crude product was purified by column chromatography
þ
7
29 m. ESI-MS (m/z): 158.0 (100%) [M þ H] .
1-Ethyl-4,7-bis(tert-butoxycarbonyl ethyl carbamate)-1,4,7-
triazacyclononane (3). A solution of tert-butyl-(2-bromoethyl)-
(Merck neutral alumina gel 90, eluent: 2% MeOH/CHCl
the desired fraction having an R = 0.35. Yield: 0.95 g (55%). H
NMR (300 MHz, CDCl ): δ 1.02 (t, 3H, J = 7.2 Hz, ethyl CH ),
3
), with
1
f
carbamate (1.90 g, 8.49 mmol) in acetonitrile (30 mL) was added
dropwise to a mixture of 2 (0.67 g, 4.24 mmol), K CO (1.46 g,
3
3
t
1.49 (s, 36H, Bu CH ), 2.55 (m, 4H, ethyl CH
(m, 14H, tacn CH and ethyl CH
3
2
), 2.65-2.79
2
3
1
t
3
1
0.6 mmol) and KI (0.12 g, 0.60 mmol) in acetonitrile (30 mL).
2
2
), 3.47 (m, 4H, ethyl CH
2
).
C
NMR (75 MHz, CDCl ): δ 12.7 (ethyl CH ), 27.4, 28.0 ( Bu
The mixture was stirred for 1 h at room temperature and then
refluxed for 3 days. After the mixture was cooled to room
temperature, the inorganic salts were removed by filtration.
3
3
CH
5
3
), 39.1 (ethyl CH
6.1, 57.1 (tacn CH
2
), 51.2 (ethyl CH
2
), 54.3 (ethyl CH
2
), 55.6,
), 78.6, 82.3 (quaternary Bu C), 152.5
t
2
-
1
(
CdO), 155.7 (CdO), 163.3 (CdN). IR (neat), υ(cm ) 3331s
Evaporation of the solvent from the filtrate in vacuo yielded 3 as
1
(
υN-H), 2976s (υC-H), 1732s (υCdO), 1651s (υCdN), 1557s,
a yellow oil. Yield: 1.42 g (76%). H NMR (300 MHz, CDCl ): δ
3
t
), 1.46 (s, 18H, Bu CH
1
7
416s, 1167s (υC-O), 1057s, 1026s, 913 m. ESI-MS (m/z):
1
.02 (t, 3H, J = 7.2 Hz, ethyl CH
3
3
),
þ
28.5 (100%) [M þ H] .
2
.52-2.68 (m, 18H, tacn CH and ethyl CH
2
2
), 3.17 (m, 4H, ethyl
1
3
1-Ethyl-4,7-bis[3-bis(tert-butoxycarbonyl)guanidinopropyl]-
,4,7-triazacyclononane (8). Compound 8 was prepared using
CH ), 5.29 (s br, 2H, NH). C NMR (75 MHz, CDCl ): δ 12.4
2
3
t
1
the same method as for 7 by dissolving compound 6 (0.54 g, 1.99
mmol) in THF (20 mL), followed by the addition of N,N -Boc
1H-pyrazole-1-carboxamidine (1.23 g, 3.98 mmol) in THF
(
ethyl CH
3
), 28.2 ( Bu CH
), 56.9 (ethyl CH
quaternary Bu C), 155.9 (CdO). IR (neat), υ(cm ) 3354s
br (υN-H), 2977s (υC-H), 2933s (υC-H), 1684s (υCdO), 1455s,
3
), 38.7 (ethyl CH
2
), 52.3, 54.6 (tacn
CH
(
2
), 56.7 (ethyl CH
2
2 2
), 57.2 (tacn CH ), 78.3
t
-1
0
2
-
(
20 mL). The desired product has an R =0.32. Yield: 0.85 g
f
1
251s (υC-O), 1170s, 922s, 732s. ESI-MS (m/z): 444.5 (100%)
1
þ
3
(57%). H NMR (300 MHz, CDCl ): δ 0.98 (m br, 2H, NH),
[
M þ H] .
1
ethyl CH ), 1.47 (s, 36H, Bu CH ), 1.72 (m, 4H, propyl CH ),
.00 (t, 3H, J = 7.2 Hz, ethyl CH
3
), 1.25 (m, 6H, propyl CH
2
and
1-Ethyl-4,7-bis(tert-butoxycarbonyl propyl carbamate)-1,4,7-
t
2
3
2
triazacyclononane (4). Compound 4 was synthesized following
the same procedure as for 3, using 2 (0.76 g, 4.83 mmol), tert-
butyl-(3-bromopropyl)carbamate (2.30 g, 9.65 mmol), K CO
2 3
2
(
CH
.50-2.60 (m, 12H, tacn CH ), 3.47 (m, 4H, propyl CH ), 8.34
2
2
1
3
t br, 2H, NHBoc). C NMR (75 MHz, CDCl
3
): δ 12.8 (ethyl
t
3 3 2
), 27.4, 28.0 ( Bu CH ), 28.3, 29.6 (propyl CH ), 39.2 (ethyl
(
(
1.67 g, 12.1 mmol) and KI (0.11 g, 0.60 mmol). Yield: 1.98 g
t
1
CH ), 52.2, 55.0, 56.1 (tacn CH ), 79.0, 82.9 (quaternary Bu C),
2
153.2 (CdO), 156.1 (CdO), 163.6 (CdN). IR (neat), υ(cm
2
88%). H NMR (300 MHz, CDCl
3
): δ 1.05 (t, 3H, J = 7.2 Hz,
), 1.50 (m, 4H, propyl CH ),
and ethyl CH ), 2.58 (m, 12H, tacn
-
1
t
)
ethyl CH
2
3
), 1.46 (s, 18H, Bu CH
3
2
3
334s (υN-H), 2934s (υC-H), 1715s (υCdO), 1644s (υCdN), 1555s,
.45 (m, 6H, propyl CH
2
2
1
3
1455s, 1252s (υ
), 1133s, 1048s, 913 m. ESI-MS (m/z): 756.3
C-O
CH ), 3.05 (m, 4H, propyl CH ), 3.25 (m, 2H, NH). C NMR
2
2
þ
(
100%) [M þ H] .
(
75 MHz, CDCl
3
): δ 12.4 (ethyl CH
3
), 27.8 (propyl CH
2
), 28.2
t
1-Ethyl-4,7-bis(ethylguanidinium)-1,4,7-triazacyclononane tetra-
hydrochloride (L 4HCl). A solution of the Boc-protected amine
7 (0.68 g, 0.93 mmol) was dissolved in a 1:1 v/v mixture of TFA/
(
Bu CH
3
), 39.5 (propyl CH
2
), 41.2 (ethyl CH
), 78.4 (quaternary Bu C), 160
CdO). IR (neat), υ(cm ) 3335s (υN-H), 2932s (υC-H), 2808s,
2
), 52.1 (propyl
1
t
CH
(
2
), 55.2, 55.9, 57.1 (tacn CH
2
3
-
1
2 2
CH Cl and the solution stirred at room temperature overnight.
The solvent was then removed under reduced pressure and the
residual brown oil taken up in a mixture of EtOH (5 mL) and
1
694s (υCdO), 1520s, 1391s, 1252s(υC-O), 1171s, 922w, 732s.
þ
ESI-MS (m/z): 472.4 (100%) [M þ H] .
1-Ethyl-4,7-bis(2-aminoethyl)-1,4,7-triazacyclononane (5). A
solution of the Boc-protected amine 3 (0.57 g, 1.29 mmol) was
2
concentrated HCl (2 mL). Addition of Et O (5 mL) produced a
pale yellow precipitate, which was collected by filtration, dis-
solved in a small volume of water and then freeze-dried to yield
the product as a yellow solid. Yield: 0.23 g (75%). Microana-
dissolved in a 1:1 v/v mixture of TFA/CH Cl and the solution
2
2
left stirring at room temperature overnight. Following adjust-
ment of the pH to 14 by the cautious addition of 5 M NaOH, the
product was extracted with CH Cl (3 ꢀ 100 mL). Workup
lysis: Calcd for C14
Found C 33.3, H 8.1, N 24.3, Cl 27.6%. H NMR (300 MHz,
33 9 4
H N Cl C 33.0, H 8.1, N 24.8, Cl 27.9%;
2
2
1
1
as for 2 gave 6 as a yellow oil. Yield: 0.28 g (88%). H NMR
D O): δ 1.50 (t, 3H, J = 7.2 Hz, ethyl CH ), 2.96 (m, 4H, tacn
(
1
300 MHz, CDCl
3
): δ 1.04 (t, 3H, J = 7.2 Hz, ethyl CH
3
), 2.61 (m,
2
3
1
3
CH
2
), 3.13 (m, 8H, tacn CH
2
2
), 3.43 (m, 6H, ethyl CH ), 3.51 (m,
6H, tacn CH and ethyl CH ), 2.80 (m, 6H, ethyl CH ). CNMR
2 2
2
1
3
(
75 MHz, CDCl ): δ 9.74 (ethyl CH ), 37.7 (ethyl CH ), 52.4 (ethyl
4H, ethyl CH
37.1 (ethyl CH
55.4 (ethyl CH
(υN-H), 2977s (υC-H), 2960s (υC-H), 1664s (υCdN), 1450 m,
2
). C NMR (75 MHz, D
2
O): δ 9.30 (ethyl CH
3
),
3
3
2
CH
(
2
), 52.9 (tacn CH ), 53.2 (tacn CH
2
neat), υ(cm ) 3003w (υN-H), 2975s (υC-H), 2932s (υC-H), 1432
2
), 54.8 (ethyl CH ). IR
2
2
), 49.2, 49.6, 50.1 (tacn CH
2
), 54.3 (ethyl CH
2
),
-1
-1
) 157.7 (CdN). IR (KBr disk), υ(cm ): 3404s
2
þ
m, 1182s, 799s. ESI-MS (m/z): 244.2 (100%) [M þ H] .
2þ
1
230w, 1178w, 940w. ESI-MS (m/z): 164.7 (90%) [M þ 2H]
,
1
-Ethyl-4,7-bis(3-aminopropyl)-1,4,7-triazacyclononane (6).
þ
Compound 6 was synthesized from 4 (0.48 g, 1.01 mmol)
following the same procedure as for 5. This yielded the desired
product as a brown oil. Yield: 0.21 g (75%). H NMR
328.4 (10%) [M þ H] .
1-Ethyl-4,7-bis(propylguanidinium)-1,4,7-triazacyclononane
tetrahydrochloride (L 4HCl). Deprotection of compound 8
1
2
3
(
300 MHz, CDCl
3
): δ 1.02 (t, 3H, J = 7.2 Hz, ethyl CH
3
), 1.62
(0.57 g, 0.90 mmol) was carried out in the same fashion as for
compound 7 to yield the product as a yellow solid. Yield: 0.16 g
(
m, 6H, propyl CH and ethyl CH ), 2.51-2.60 (m, 8H, propyl
2
2
1
3
CH
2
), 2.70-2.78 (m, 12H, tacn CH
): δ 11.7 (ethyl CH ), 29.6 (propyl CH
CH ), 51.5 (ethyl CH ), 53.2 (propyl CH ), 53.9, 56.4, 66.7 (tacn
2
). C NMR (75 MHz,
(78%). Microanalysis: Calcd for C16
23.9, Cl 26.9%; Found C 36.4, H 8.2, N 24.0, Cl 27.1%. H
NMR (300 MHz, D O): δ 1.47 (t, 3H, J = 7.2 Hz, ethyl CH ),
1.97 (m, 4H, propyl CH ), 3.23 (m, 8H,
tacn CH ), 3.37 (m, 6H, propyl CH
4H, propyl CH
37 9 4
H N Cl C 36.4, H 8.3, N
1
CDCl
3
3
2
), 40.0 (propyl
2
2
2
2
3
-1
CH ). IR (neat), υ(cm ) 3334s (υN-H), 2930s (υC-H), 1463 m,
1
2
), 3.08 (m, 4H, tacn CH
2
2
þ
173s, 1128s, 923w, 731s. ESI-MS (m/z): 271 (100%) [M þ H] .
2
2
and ethyl CH
2 2 3
). C NMR (75 MHz, D O): δ 9.43 (ethyl CH ),
2
), 3.53 (m,
1
3
1-Ethyl-4,7-bis[2-bis(tert-butoxycarbonyl)guanidinoethyl]-
,4,7-triazacyclononane (7). To a stirred solution of 5 (0.57 g,
2
(propyl CH
3.6 (propyl CH ), 39.0 (propyl CH ), 49.1 (ethyl CH ), 49.3
1
2
1
2
2
2
0
.35 mmol) in THF (20 mL) was added a solution of N,N -Boc -
2
2
), 50.0, 53.9, 54.1 (tacn CH
2
), 157.2 (CdN). IR
-
1
H-pyrazole-1-carboxamidine (1.47 g, 4.71 mmol) in THF
(KBr disk), υ(cm ): 3412s (υN-H), 2954s (υC-H), 1668s (υCdN),
þ
(
20 mL). The resulting solution was stirred under nitrogen at
1436 m, 1172w, 1018w. ESI-MS (m/z): 356.4 (100%) [M þ H] .
1
1
room temperature for 2 days. The solvent was then evaporated
under reduced pressure, and the residue taken up in DCM
4 2
[CuL ](ClO ) (C1). To a stirred solution of L .4HCl (0.093 g,
0.18 mmol) in water (4 mL) was added a solution of Cu-
(ClO .6H O (0.081 g, 0.22 mmol) in water (4 mL). The pH
of the solution was adjusted to 9 with 1 M NaOH, resulting in a
(
the organic fraction was dried over Na SO and the solvent
50 mL). After it was washed with 0.1 M NaOH (3 ꢀ 30 mL),
4
)
2
2
2
4

Article
Inorganic Chemistry, Vol. 50, No. 2, 2011 625
color change to deep blue. An off-white precipitate of Cu(OH)
2
2-Hydroxypropyl-p-nitrophenylphosphate (HPNPP). These
experiments were carried out in a similar manner to the BNPP
experiments. The rate of cleavage of HPNPP by the Cu(II)
complexes was measured at pH 7.0 (HEPES buffer) and T =
25 ꢀC, again by following the release of the highly colored
p-nitrophenoxide ion (vide supra), in solutions containing
0.1 mM HPNPP, 2 mM Cu(II) complex, and 0.15 M NaClO4.
was removed by filtration, and the deep blue solution left to
slowly evaporate in a crystallization dish, resulting in the for-
mation of blue crystals of the product. Yield: 0.043 g (41%).
Microanalysis: Calc for Cu
1
C
14
H
33
N
9
O
8
Cl
2
: C 28.5, H 5.6,
O): λmax
nm) [ε ] (M cm ): 600 [250], 750 [85], 985 [125]. Selected IR
N 21.4%, Found: C 28.2, H 5.7, N 21.4%. UV-vis (H
2
-
1
-1
(
max
-
1
bands (ATR) υ(cm ): 3445 m, 3348 m, 3252 m (υ
), 2967w,
The observed rate constants were determined by fitting the data
t.
N-H
-
-kobs
2862w (υC-H), 1584s (υCdN), 1031s br, 618s (υClO4 ).
to the expression, Abs = A þ Be
Solution Speciation Studies. pH titrations of C1 and C2. pH
DNA Cleavage Experiments. Reaction mixtures (total volume
15 μL) containing pBR 322 supercoiled plasmid DNA (38 μM
base pair concentration) and copper(II) complex (75, 100, 150,
300, and 600 μM) dissolved in 40 mM buffers (MES, HEPES,
TAPS, and CHES) at pH 6.0, 6.5, 7.0, 7.5, 8.0 and 9.0, were
incubated in a water bath at 37 ꢀC for periods of up to 48 h.
Loading buffer (0.1 M EDTA, 40% (w/v) sucrose, 0.05% (w/v)
bromophenol blue, and 0.5% (w/v) sodium lauryl sulfate, SDS)
(5 μL) was added to stop the reactions at defined time periods,
and the resulting solutions stored at -20 ꢀC until just prior
to analysis. The solutions were then loaded onto 1% agarose
titrations were carried out by adding 1 μL aliquots of 10 M HCl
to solutions of the Cu(II) complexes ([C1] = 9.28 mM, [C2] = 7.2
mM, total volume 10 mL). Stock copper(II) complexes solutions
were prepared by mixing equimolar amounts of the desired ligands
2 2
and CuCl .2H O in water. The solutions were stirred for 1 min
after each addition to ensure that the pH was stable, and then the
background-corrected UV-vis-NIR spectrum measured.
The series of spectra were analyzed using the BEEROZ soft-
63
ware. Briefly, this method employs constrains from mass
balance and mass action equations in order to test speciation
models for spectral data sets containing many species, and
to derive formation constants for the spectroscopically active
-
3
gels containing 1.0 μg dm ethidium bromide, and the DNA
fragments separated by gel electrophoresis (70 V for 2 h in 1 x
tris-acetate EDTA (TAE) buffer). Ethidium-stained agarose
gels were imaged, and the extent of supercoiled DNA cleavage
determined via densitometric analysis of the visualized bands
using the volume quantification method. Experiments were
performed at least in duplicate. The variation in the amount
of DNA cleavage with time was modeled as a first-order process,
64,65
species.
The analysis provides the calculated spectra based on
the model, the spectra for the individual complexes, and formation
constants for all species included in the model. For the present
study, activity coefficients were fixed to unity (fixed ionic strength).
Titration of C1 with (4-Nitrophenyl)phosphate (NPP). Ali-
quots (1 μL) of a 1.39 M NPP solution were added to solutions of
complex C1 (9.28 mM, final volume 3 mL), buffered at pH 6.0, 7.0,
and 9.0 with 1 M MES, HEPES and CHES, respectively. A back-
ground-corrected UV-vis-NIR spectrum was measured after each
-k
obs
Abs = A þ Be
t and observed rate constants determined by
fitting the decrease in the relative intensity of Form I band as
well as the appearance of the Form II band to this expression.
DNA Cleavage Experiments in the Presence of Radical Sca-
addition of NPP. The NPP binding constant, K , was be determined
b
from the variation in absorbance with [NPP] by using the equa-
vengers. Aliquots (5 μL) of aqueous solutions of scavenging
t
agents (30 mM KI, DMSO, BuOH or NaN
tion (A- A
where A = absorbance at a particular [NPP], A
bance, [CuL] = initial copper(II) complex concentration, and
0
)/[NPP] =-K
b
(A- A
0
) þ K
b
[CuL]
0
(εCuLNPP - εCuL),
0
3
in 40 mM HEPES
= initial absor-
buffer at pH 7.0) were added to the solutions of supercoiled DNA
(5 μL, 113.5 μM base pair concentration) prior to the addition of
complexes C1 and C2. The final reaction conditions were 150 μM
for copper(II) complexes, 10 mM concentration for the scavenging
agents and 38 μM base pair concentration for supercoiled plasmid
DNA. Each solution was incubated at 37 ꢀC for 6 h, quenched and
analyzed according the procedures described above.
0
εCuL and εCuLNPP = molar extinction coefficients for copper
complex and complex with NPP bound, respectively. The plot of
(
the graph corresponding to the binding constant.
A - A )/[NPP] vs (A - A ) was found to be linear with the slope of
0
0
Cleavage of Model Phosphate Esters. Bis(p-nitrophenyl)phos-
phate (BNPP). These experiments were conducted using estab-
DNA Cleavage under Anaerobic Condition. Experiments un-
der anaerobic condition were performed following the protocol
6
6,67
lished procedures.
Cu(II) complexes was measured at pH 7.0 (HEPES buffer) and T =
0 ꢀC, by following the formation of p-nitrophenoxide ion (λmax=
Briefly, the rate of cleavage of BNPP by the
1
4
reported by Burstyn and co-workers. Deoxygenated water was
prepared by five freeze-pump-thaw cycles. Before each cycle
the water was equilibrated by bubbling nitrogen through the
solution. It was used immediately in the preparation of anaero-
bic stock solutions in a nitrogen-filled environment. Reaction
mixtures were prepared immediately by addition of the appro-
priate amounts of stock solutions to reaction tubes, which were
subsequently sealed and incubated at 37 ꢀC. All other conditions
were as for the cleavage experiments performed under aerobic
condition in which final concentrations for the reaction mixtures
were 150 μM for complexes, 40 mM HEPES buffer, and 38 μM
base pair concentration for plasmid DNA.
5
4
-
1
-1
00 nm, λmax=18,700 M cm ) in solutions containing 0.1 mM
BNPP, 2 mM Cu(II) complex and 0.15 M NaClO . Stock copper(II)
complexes solutions were prepared by mixing equimolar amounts of
the desired ligands and CuCl 2H O in water and adjusting the pH
to 7 with addition of NaOH. Absorbance measurements were
commenced 2 min after mixing and were continued for 8000 min,
with a reading taken every 15 min. As the complex was in large
excess compared to BNPP, the time dependence of the appearance
of NP (and cleavage of BNPP) was modeled as a first-order process.
Observed rate constants were determined by fitting the data to the
4
2
3
2
-k
obs
equation, Abs=Aþ Be t, whereA and Bare constants, or by the
initial rate method (i.e., directly from the plot of the increase of
X-ray Crystallography. Intensity data for a blue crystal of C1
(0.20 ꢀ 0.20 ꢀ 0.05 mm) were collected at 123 K on a Bruker
Apex II CCD fitted with graphite monochromated Mo KR
p-nitrophenolate concentration versus time). The latter gave linear
plots with typical R values >0.995.
2
68
˚
radiation (0.71073 A). The data was collected to a maximum 2θ
value of 55ꢀ and processed using Bruker Apex II software
package. Crystal parameters and details of the data collection
are summarized in Table S01 in Supporting Information. The
structure of C1 was solved using SHELX-97 and expanded
using standard Fourier transform routines. All hydrogen atoms
(
(
63) Brugger, J. Comput. Geosci. 2007, 33, 248–261.
64) Brugger, J.; McPhail, D. C.; Black, J.; Spiccia, L. Geochim. Cosmo-
69,70
chim. Acta 2001, 65, 2691–2708.
65) Liu, W.; Etschmann, B.; Brugger, J.; Spiccia, L.; Foran, G.; McInnes,
B. Chem. Geol. 2006, 231, 326–349.
66) Fry, F.; Fischmann, A. J.; Belousoff, M. J.; Spiccia, L.; Brugger, J.
Inorg. Chem. 2005, 44, 941–950.
67) Belousoff, M. J.; Duriska, M. B.; Graham, B.; Batten, S. R.;
Moubaraki, B.; Murray, K. S.; Spiccia, L. Inorg. Chem. 2006, 47, 3746–3755.
68) Deal, K. A.; Burstyn, J. N. Inorg. Chem. 1996, 35, 2792–2798.
(
(
(69) Sheldrick, G. M. SHELXL-97; University of G €o ttingen: G €o ttingen,
(
Germany, 1997.
(70) Sheldrick, G. M. SHELXS-97; University of G €o ttingen: G €o ttingen,
(
Germany, 1997.

6
26 Inorganic Chemistry, Vol. 50, No. 2, 2011
Tjioe et al.
1
Scheme 1. Synthesis of Hydrochloride Salts of Ligands L and L
2
were placed in idealized positions, except for the hydrogens on
the nitrogen atoms, which were located on the Fourier difference
map and refined with restrained N-H distances. The position of
the N-H hydrogen atom, H8N, located as a Fourier difference
peak, could not be modeled unambiguously, the hydrogen
Microanalytical data for C1 were consistent with the
proposed formula and the IR spectrum of the com-
plex shows bands due to NH stretches of the guanidine
-1
groups in the region of 3252-3444 cm , a sharp band at
-1
1
584 cm attributed to the υ(CdN) of the guanidine
bonding interaction with the neighboring perchlorate anions
˚
-1
groups, and bands at ∼1031 and 618 cm due to the
perchlorate counterions. The electronic spectrum of C1
exhibits bands at 600 and 985 nm, which are typical for
a copper(II) complex with square pyramidal geometry. In
addition, there is a band at 750 nm, which could arise from
significant distortion of the Cu(II) coordination sphere, as
identified in the X-ray structure determination (vide infra).
Crystallography. The complex C1 crystallizes in the
present in the lattice ((N(8)-H(8N) O(3)#3 = 3.160(7) A,
3
3 3
1
-
48(5)ꢀ, #3 denoting the symmetry operator (-x þ 1, -y þ 1,
z þ 1), Table 3), as well as the differences in the corresponding
guanidine C-N bond distances forming the basis for its assign-
ment. The isotropic thermal parameters for N-H hydrogen
atoms were fixed at 1.2 times that of the respective nitrogen
atom. All non-hydrogen atoms were refined anisotropically.
Results and Discussion
space group P2 /c. The molecular structure consists of
1
1
2þ
1
discrete five-coordinate [CuL ] cations and noncoordi-
nating perchlorate anions. A thermal ellipsoid represen-
tation of the complex cation is shown in Figure 3, while
selected bond lengths and angles are listed in Table 1. The
copper(II) coordination sphere is occupied by the three
nitrogen atoms of the tacn ring and by a nitrogen donor
from each of the guanidine groups, which are present in
their neutral form. The guanidine groups bind via the non-
terminal amine nitrogen, forming five-membered chelate
rings. A search of the Cambridge Crystallography Data-
base revealed that C1 is the first example of a pendant
guanidine group adopting this type of chelation mode.
Binding of a guanidine group appended to a macro-
cycle has been previously reported by Kimura and co-
Synthesis. The synthesis of the new tacn derivatives, L
and L , featuring pairs of guanidinium pendants followed
2
the route shown in Scheme 1. This involved, first, reac-
tion of 1,4-bis(tert-butoxycarbonyl)-1,4,7-triazacyclono-
nane with iodoethane to alkylate one of the three
nitrogens of the tacn ring. The Boc protecting groups
were then removed through treatment with TFA, and
guanidinium pendants introduced at the two exposed
secondary nitrogens via a four-step procedure involving
the addition of aminoalkyl pendants, followed by con-
version of the terminal amino groups into guanidinium
0
groups via treatment with N,N -Boc -1H-pyrazole-1-car-
2
boxamidine. The new compounds were characterized by
H and C NMR spectroscopy, ESI mass spectrometry,
1
13
7
1
workers, for a zinc(II)-cyclen compound, however, in
this case, binding was through one of the terminal amines
and led to formation of a seven-membered chelate ring.
Deprotonation and coordination of guanidine at near
neutral pH was somewhat unexpected, given the strongly
and IR spectroscopy. Satisfactory microanalyses were
obtained for the final ligands, isolated as their tetra-
hydrochloride salts.
1
1
The copper(II) complex of L , [CuL ](ClO ) (C1), was
4
2
isolatedfollowingslowevaporation ofanaqueoussolution
of the ligand and Cu(ClO ) 6H O adjusted to pH 9.
Efforts to isolate C2 were unsuccessful, probably because
4
2
3
2
(71) Aoki, S.; Iwaida, K.; Hanamoto, N.; Shiro, M.; Kimura, E. J. Am.
of the complexity of solution speciation (vide infra).
Chem. Soc. 2002, 124, 5256–5257.

Article
Inorganic Chemistry, Vol. 50, No. 2, 2011 627
Figure 3. Thermal ellipsoid representation of the complex cation unit in
C1 (ellipsoids drawn at 50% probability; selected hydrogen atoms and
perchlorate counterions are omitted for clarity).
˚
Table 1. Selected Bond Lengths [A] and Angles [deg] for C1
Figure 4. Hydrogen-bonding interactions (shown as dashed bonds)
between the complex cation and the perchlorate anions present in the
crystal lattice of C1, as viewed along the b axis.
Cu(1)-N(1)
Cu(1)-N(2)
Cu(1)-N(3)
Cu(1)-N(4)
Cu(1)-N(7)
2.215(4)
2.067(4)
2.092(4)
2.001(4)
1.997(4)
N(7)-Cu(1)-N(4)
N(7)-Cu(1)-N(2)
N(4)-Cu(1)-N(2)
N(7)-Cu(1)-N(3)
N(4)-Cu(1)-N(3)
N(2)-Cu(1)-N(3)
N(7)-Cu(1)-N(1)
N(4)-Cu(1)-N(1)
N(2)-Cu(1)-N(1)
N(3)-Cu(1)-N(1)
103.01(2)
85.2 (2)
168.0 (2)
136.7(2)
83.2 (2)
84.8(2)
136.4(2)
97.8(2)
81.5(2)
a
˚
Table 2. Hydrogen Bonding Interactions in C1 [A and deg]
D-H
3 3 A
d(D-H) d(H
3 3 3 A)
d(D
3 3 3 A) —(DHA)
3
N(6)-H(2N) O(6)
0.885(19)
0.885(19)
2.47(4)
2.30(3)
3.221(6)
3.139(6)
3.116(6)
3.334(6)
3.185(6)
2.973(6)
3.099(7)
3.311(6)
3.096(7)
3.378(6)
3.160(7)
143(5)
159(5)
171(5)
155(5)
132(5)
116(4)
117(4)
152(5)
149(5)
153(5)
148(5)
3
3 3
N(6)-H(2N) O(7)
3
3 3
83.2(2)
N(6)-H(1N) O(8)#1 0.88(2)
2.25(2)
2.51(3)
2.52(4)
2.48(5)
2.60(5)
2.51(3)
2.31(3)
2.57(3)
2.38(3)
3
3 3
N(5)-H(3N) O(7)
0.883(19)
0.883(19)
0.87(2)
3
3 3
3 3
basic character of guanidine (viz., guanidinium groups
typically have pK values >12 ). The strength of binding
N(5)-H(3N) O(1)
3
7
2
N(5)-H(4N) O(4)
a
3 3 3
N(9)-H(5N) O(5)#2 0.88(2)
to the copper(II) center clearly overrides the proton affi-
nity of guanidine (see speciation studies). This can be
attributed in part to the formation of five-membered chelate
rings. The high affinity of the deprotonated guanidines
for copper(II) is highlighted by the fact that the two
Cu-N(pendant) distances are significantly shorter than
those for the macrocyclic nitrogens, av. 1.999(4) A vs
.067(4)-2.215(4) A.
The presence of five-membered chelate rings in the C1
complex cation (in particular the three fused edge-sharing
chelate rings of the tacn moiety) leads to considerable
distortion from an idealized coordination geometry, with
all intra-ring N-Cu-N angles well below 90ꢀ (Table 1).
An effective means of quantifying the degree of distortion
in five-coordinate copper(II) complexes was developed by
Addison et al., which applies the parameter, τ, defined
as {(θ - φ)/60} ꢀ 100, where θ and φ are the largest and
smallest basal angle, respectively (τ is zero for a square
pyramidal geometry and one for a trigonal bipyramidal
geometry). For C1, the relevant angles, θ = 168.0ꢀ and
φ=136.7ꢀ, yield a τ value of 52%, which implies a Cu(II)
geometry intermediate between the two regular extremes.
3 3 3
N(9)-H(6N) O(1)#2 0.875(19)
3
3 3
N(8)-H(7N) O(2)#2 0.87(2)
3
3 3
N(8)-H(7N) O(1)#2 0.87(2)
3
3 3
N(8)-H(8N) O(3)#3 0.88(2)
3
3 3
a
Symmetry transformations used to generate equivalent atoms: #1
x, y þ 1/2, -z þ 1/2; #2 x, -y þ 1/2, z þ 1/2; #3 -x þ 1, -y þ 1, -z þ 1.
-
˚
˚
In C1, the Cu-N bond lengths to the coordinating
2
guanidine nitrogens are similar to those reported in the
literature for other Cu(II)-tacn complexes bearing pen-
dant amino arms with coordinating nitrogen atoms (av.
˚
ca. 2.0 A, see Table 3). The Cu(1)-N(4) and Cu(1)-N(5)
bonds in C1 are virtually identical in length to the Cu-
2
N(pendant) bond in Cu[tacn(CH CH NH )] , where
þ
2
2
2
only one aminoethyl pendant coordinates to the copper-
7
3
(
0
II) center. However, this bond distance is increased by ca.
2
þ
˚
.02 A in Cu[tacn(CH CH NH ) ] , which features an
2 2 2 2
7
4
additional coordinating aminoethyl pendant.
The
Cu-N(pyridine) bond distances seen in Cu(II)-tacn com-
plexes containing pyridyl pendants are also similar in size
7
5
˚
(
1.99-2.02 A). Overall, this analysis is indicative of
strong binding between the Cu(II) center and the guani-
dine groups in C1. This could be contributing to the high
distortion of the copper(II) coordination sphere. The τ of
1
The guanidine groups of L are involved in weak hydro-
gen bonding interactions with the two non-coordinated
perchlorate anions, leading to an extended intermolecular
hydrogen bonding network within the crystal lattice of C1
52% indicates that the degree of geometrical distortion
is much greater than generally found for Cu(II)-tacn
(
see Figure 4 and Table 2).
(74) Tei, L.; Bencini, A.; Blake, A. J.; Lippolis, V.; Perra, A.; Valtancoli,
(
72) Riordan, J. F.; McElvany, K. D.; Borders, C. L. Science 1977, 195,
84–886.
73) Addison, A. W.; Rao, T. N.; Reedjik, J.; Rijin, J. V.; Verschoor, G. C.
B.; Wilson, C.; Schroder, M. Dalton Trans. 2004, 1934–1944.
(75) Gasser, G.; Tjioe, L.; Graham, B.; Belousoff, M. J.; Juran, S.;
Walther, M.; Kunstler, J.-U.; Bergmann, R.; Stephan, H.; Spiccia, L.
Bioconjugate Chem. 2008, 19, 719–730.
8
(
J. Chem. Soc., Dalton Trans. 1984, 1349–1356.

6
28 Inorganic Chemistry, Vol. 50, No. 2, 2011
Tjioe et al.
a
Table 3. Comparison of Selected Structural Parameters for C1 with Those for Other Cu(II)-tacn Complexes Featuring Pendants with N Donors
˚
˚
˚
complex
Cu-N(tacn) (axial) (A)
Cu-N(tacn) (equat.) (A)
Cu-N(pendant) (A)
τ (%)
chelate ring
refs
C1
Cu(dmptacn)
2
Cu (tmpdtne)
2.215(4)
2.162(8)
2.205(4)
2.119(10)
2.206(2)
2.356(3)
2.215(2)
2.067(4); 2.092(4)
2.071(8); 2.000(8)
1.978(4); 2.009(4)
2.100(9); 2.119(10)
2.054(2); 2.033(2)
2.062(4); 2.082(4)
2.026(2); 2.075(2)
1.997(4); 2.001(4)
2.015(8); 2.016(8)
1.993(4); 2.001(4)
1.995(9)
2.018(3); 2.019(3)
2.032(3); 2.039(3)
2.008(2)
52
32
11
3
19
20
18
5
5
5
5
5
6
6
this work
2
þ
þ
76
76
74
74
77
77
2
2
þ
Cu[tacn(CH
2
CH
CH
2
NH
NH
2
)]
)
2þ
Cu[tacn(CH
Cu[ Prtacn(CH
Cu[Bn Prtacn(CH
2
2
2
2
]
,
i
2þ
2
CH
2
CH
2þ
2 2 2
CH CH NH )]
2
NH
2
)
2
]
i
2
a
0
Abbreviations: dmptacn =2-[4,7-bis(2-pyridylmethyl)-1,4,7-triazacyclononane; tmpdtne =1,2-bis[N,N -bis(2-pyridylmethyl)-1,4,7-triazacyclono-
nyl]ethane; tacn(CH
Prtacn(CH
2
CH
NH
2
NH
)
2
) = 1-(2-aminoethyl)-1,4,7-triazacyclononane,; tacn(CH
= 1,4-bis(3-aminopropyl)-7-isopropyl-1,4,7-triazacyclononane; Bn Prtacn(CH
2
CH
2
NHi
2
)
2
=1,4-bis(2-aminoethyl)-1,4,7-triazacyclononane;
CH CH NH ) = 1-(3-aminopropyl)-4-
i
2
CH
2
CH
2
2
2
2
2
2
2
benzyl-7-isopropyl-1,4,7-triazacyclononane.
complexes with pendant coordinating groups (range of
3-32%, Table 3), whose geometry is closer to square
pyramidal.
Solution Speciation Studies. Spectrophotometric pH
Titrations. To investigate the solution speciation of com-
plexes C1 and C2, a series of spectrophotometric pH
titrations were performed, which monitored changes in
the electronic spectrum of each complex as a function of
pH (Figures 5 and 6). The titrations were commenced at
pH ∼12 and spectra were measured after addition of
aliquots of acid. In the case of C1, the complex with both
guanidine pendants coordinated (C1) was anticipated to
be the only species at pH 12, given that this complex
crystallizes from solution at pH 9. This was confirmed by
the fact that the UV-vis spectra remained constant over
the pH range 8-12 (Figure 5b). As the pH decreased,
major changes in spectrum were observed below pH 8, the
most significant being the loss of the peak at ∼750 nm, but
the peaks at ∼600 and 985 nm also shifted to ∼670 and
1
050 nm. These changes in spectrum can be attributed to a
shift to a more ideal square pyramidal geometry and a
weakening of the ligand field around the copper(II)
center, most likely arising from protonation of the co-
ordinated guanidines and displacement from the coordi-
nation sphere by water, namely, CuN is converted to
5
CuN O . Principal component analysis (PCA) revealed
2
3
that three species are required to describe the spectral
changes with pH (Figure S01 in Supporting Information),
and a model based on two successive protonation pro-
cesses was successfully used to fit the data, from which
apparent binding constants (Kapp1 and Kapp2) could be
determined, corresponding to deprotonation of each
guanidinium group and coordination of the resultant
guanidine (Scheme 2).
The pKapp1 and pKapp2 values were found to be 3.98
and 5.72, respectively (Figure 5c). This speciation model
provides an extremely good fit to the data, with a maxi-
mum difference of (0.012 absorbance units between the
experimental and calculated data. In a study of the zinc-
Figure 5. Changes in the UV-vis-NIR spectrum of complex C1 as a
function of pH, measured on a 9.28 mM aqueous solution of [C1]. (a)
Series of UV-vis spectra; titration was commenced at pH 12, top
spectrum, and finished at pH 4.0, bottom spectrum. (b) Evolution of
molar absorbance at three wavelengths as a function of pH. Full lines
show the absorbance calculated with the proposed speciation model
(Scheme 2). (c) Species distribution diagram showing the variation in
species present with pH for C1. Crosses at the top show the pH of
measured solutions.
(
Kimura and co-workers were unable to determine the
II) complex of a cyclen monoguanidinium derivative,
7
1
pK of the pendant guanidium group from pH titrations
a
2
þ
on the ligand in the absence of Zn (but estimated it to
be >12), yet the apparent binding constant for deproto-
nation/coordination of the guanidine groups was deter-
mined to be 5.9. This implies a remarkable binding affi-
nity of zinc(II) for the guanidine group (log K > 6), even
though a seven-membered chelate involving the terminal
nitrogen was formed. In our case, an even higher binding
affinity is apparent (log K > 6.5), which probably arises
from the formation of more stable five-membered chelate
rings involving the nonterminal amine nitrogen of the
guanidine groups.
The species distribution profile calculated for the ap-
1
parent stability constants indicates that for the Cu(II)-L

Article
Inorganic Chemistry, Vol. 50, No. 2, 2011 629
Scheme 2. Plausible Model for Rationalizing the Changes in the Electronic Spectrum of C1 with pH
1
2þ
system, [CuL ] is the only species present in significant
concentration above pH 8 (Figure 5c). One explanation
for the observed change in species distribution is that as
the pH is decreased, detachment and protonation of one
guanidine group occurs, possibly leading to a monoaqua
1
3þ
complex, [Cu(L H)(OH )] , with one protonated gua-
2
nidine (dominant at ∼pH 5), followed by formation of
1
the complex with two protonated guanidines, [Cu(L H )-
2
4
OH ) ] , at the lower pH. We note, however, that the
þ
(
2
2
short Cu-N distance to the pendant guanidine nitrogens
implies that these groups bind strongly to the copper(II)
center.
The speciation for C2 is more complex than for C1, as
illustrated by Figure 6 and the principal component
analysis (Figure S01, Supporting Information). In these
plots, each factor corresponding to an actual absorbing
species in solution causes a dramatic decrease in residuals,
63,78
whereas additional factors only cause a small decrease.
In the case of the C2 complex, the analysis of data sets
obtained at constant ligand concentrations clearly shows
the need for at least five factors to explain the series of
spectra (Figure S01b, Supporting Information). The five
species requiredtoexplain data set(b) may be related tothe
deprotonation of H O (three species) and NH2 (two
2
species) groups, as per Scheme 3. Note that this analysis
cannot distinguish complexes of identical composition,
e.g., L Cu(NH ) (OH)(OH ) from a complex in which
2
3þ
2
2
2
pendant guanidinium groups hydrogen bond to the hydroxo
group.
To test for the presence of polynuclear complexes, a
second series of spectra were measured at a higher com-
plex concentration (15 mM, see Figure 6a). The addition
2
of a second data set at higher [CuL ] results in an increase
in the number of factors from 5 to 6, suggesting the pre-
sence of polynuclear complexes (Figure S01c, Supporting
Information). In contrast, additional factors were not
required to describe the concentration dependence of C1,
indicating the absence of polynuclear species. Analysis of
the 5 mM data set using the five possible mononuclear
species delivers an excellent fit (not shown). Different
combinations of mono- and polynuclear species were
Figure 6. (a) Changes in the UV-vis-NIR spectrum of complex C2
as a function of pH, measured on 5 and 15 mM aqueous solutions
of C2. (b,c) Evolution of molar absorbance at two different wave-
lengths as a function of pH (the full lines show the absorbance
calculated with the proposed speciation model, vide supra). (d)
Speciation diagram showing the variation in species present with
pH, crosses at the top show the pH of measured solutions, while the
solid and dashed lines are for the 5 mM and 15 mM data sets. (e) Plot
of the percent of polynuclear species as a function of pH and
concentration of C2.
(
76) Brudenell, S. J.; Spiccia, L.; Tieknik, E. D. T. Inorg. Chem. 1996, 35,
974–1979.
77) Belousoff, M. J.; Battle, A. R.; Graham, B.; Spiccia, L. Polyhedron
007, 26, 344–355.
78) Meloun, M.; Capek, J.; Miksik, P.; Brereton, R. G. Anal. Chim. Acta
000, 423, 51–68.
1
2
2
(
(

6
30 Inorganic Chemistry, Vol. 50, No. 2, 2011
Tjioe et al.
Scheme 3. Plausible Model Rationalizing the Changes in the Electronic Spectrum of C2 with pH and [C2]
tested to explain the combined data set (5 and 15 mM),
and the best results were obtained with a model including
the five mononuclear species and three polynuclear spe-
2
2þ
2
2
L CuðNH2Þ ðOHÞ
2
2
4þ
¼
L Cu ðNH Þ ðNHÞ ðOHÞ
þ 2H O log K ¼ 1:65
2
5þ
2
2
2
2 2
2
2
2
d
cies, L Cu (NH ) (NH)(OH) , L Cu (NH ) (NH) -
2
2
2
2 3
2
2
2 2
2
4
OH)2 , and L Cu (NH )(NH) (OH) . The analysis
þ
2
3þ
(
provides the following constants:
2
2þ
þ L CuðNH2ÞðNHÞðOHÞ2þ
2
2
2
2
3
2
L CuðNH2Þ ðOHÞ
2
2
2
3þ
2
4þ
2
¼
L Cu ðNH ÞðNHÞ ðOHÞ
þ 2H O log K ¼ 2:58
L CuðNH2Þ ðOH2Þ
2
2
2
3
2
2
d
2
Note that this model includes eight species, while PCA
shows that the data set can be explained using only six.
This is only an apparent contradiction, since the distribu-
tion of the polynuclear species is highly correlated to that
of the mononuclear species (Figure 6d). The fit quality is
good, as shown in Figures 6b,c. On average, the difference
between calculated and measured absorbances is 1%, and
the highest discrepancies are 0.06 absorbance units. Care-
ful observation of the data shows that these relatively high
values are not related to the quality of the fit, but to higher
experimental noise during C2 titration (see, for example,
Figure 6c).
2
3þ
¼
L CuðNH2Þ ðOHÞðOH2Þ þ Hþ pKa1 ¼ 7:00
2
2
3þ
L CuðNH2Þ ðOHÞðOH2Þ
2
2
2þ
þ Hþ
¼
L CuðNH2Þ ðOHÞ
pKa2 ¼ 7:68
pKa3 ¼ 11:5
2
2
2
2þ
L CuðNH Þ ðOHÞ
2
2
2
2
þ
¼
L CuðNH2ÞðNHÞðOHÞ2 þ Hþ
þ
The resulting distribution of species (Figure 6d) shows
significant amounts of polynuclear complexes, with the
highest proportion found at pH ∼7.5, and increasing
proportions also at high pH (>11) (Figure 6e).
2
L CuðNH ÞðNHÞðOHÞ
2
2
2
þ
¼
L CuðNHÞ ðOHÞ ðaqÞ þ H
pK ∼13:4
a4
2
2
Binding of Model Phosphate Esters. To establish the
effect of pH on the binding of model phosphate esters, the
electronic spectrum of C1 was recorded as a function of
concentration of added p-nitrophenylphosphate (NPP) at
2
2þ
2
3þ
L CuðNH2Þ ðOHÞ
þ L CuðNH2Þ ðOHÞðOH2Þ
2
2
2
2
5þ
¼
L Cu ðNH Þ ðNHÞðOHÞ
þ 2H O log K ¼ 2:97
2
2
2 3
2
2
d

Article
Inorganic Chemistry, Vol. 50, No. 2, 2011 631
Table 4. First-Order Rate Constants for Hydrolysis of BNPP by Copper(II)
a
Complexes of Tacn and Tacn Derivatives at pH 7.0 (Standard Error in Parentheses)
-1
compound
kobs (s
)
refs
-
10
control
3.00 ꢀ 10
this work
this work
this work
this work
67
II
-6
-8
Cu -tacn
C1
C2
1.71((0.01) ꢀ 10
1.00((0.03) ꢀ 10
7.24((0.08) ꢀ 10
1.24 ꢀ 10
b
-5
II
c
c
-5
-5
-5
-5
-5
-5
Cu -Me
2
tacn
tacn
II
Cu -Me
3
3.7 ꢀ 10
66
II
i
c
Cu - Pr
3
tacn
c
4.3 ꢀ 10
66, 79
77
77
II
i
Cu -Et Prtacn
1.43 ꢀ 10
II
i
c
Cu -Bn Prtacn
1.53 ꢀ 10
II
c
Cu -BnMe
2
tacn
7.01 ꢀ 10
67
a
Abbreviations: Me
2
tacn = 1,4-dimethyl-1,4,7-triazacyclononane,
i
Me
triisopropyl-1,4,7-triazacyclononane, Et Prtacn = 1-ethyl-4-isopropyl-
3
tacn = 1,4,7-trimethyl-1,4,7-triazacyclononane, Pr tacn = 1,4,7-
3
i
i
1
,4,7-triazacyclononane, Bn Prtacn = 1-benzyl-4-isopropyl-1,4,7-tria-
zacyclononane, Bn(Me )tacn=1-benzyl-4,7-dimethyl-1,4,7-triazacyclo-
nonane. Conditions used: [complex] = 2 mM, [BNPP] = 0.1 mM,
2
Figure 7. pH dependence of the rate of hydrolysis of BNPP by C2 at
50 ꢀC ([C2] = 4 mM, [BNPP] = 0.1 mM).
b
[
using the initial rate method, yielding kobs indirectly, which was further
HEPES] = 50 mM, I = 0.15 M, T = 50 ꢀC). Data were analyzed
c
converted to a first order rate constant. Aqua complexes formed on
dissolution in aqueous solution.
three pH values (6.0, 7.0 and 9.0). The spectrophoto-
metric titrations for NPP are summarized in Supporting
Information (Figure S02). At pH 9.0, no changes in the
d-d electronic transitions of the complex were observed,
even when a large excess of NPP was added. This in-
dicates that NPP cannot displace the coordinated nitro-
gens from the copper(II) coordination sphere. At the
lower pH values, a significant change in spectrum was
observed, indicating that NPP is able to bind when aqua
ligands occupy one or more metal coordination sites.
From the variation in absorbance with [NPP] at the
two wavelength maxima, 625 and 980 nm, the binding
constants were calculated (see Experimental Section).
Figure 8. Complex concentration dependence profile for the hydrolysis
of BNPP by C2 at pH 7.0 and T = 50 ꢀC ([C2] = 1-10 mM, [BNPP] =
0.1 mM).
-
1
At pH 6.0, the K value was found to be 34 ((1) M
b
while the corresponding K value at pH 7.0 was lower
(
b
detailed than for C2. The greater activity of C2 could
be the result of activation of the phosphate ester or
stabilization of transition states through electrostatic/
hydrogen-bonding interactions with the guanidine pen-
dants, or a restriction in the formation of inactive dihy-
-
1
14 ((2) M ) due to the lower concentration of the aqua
complex available for complexation by NPP at this pH.
In the case of C2, the change in spectrum on addition
of NPP was small and did not allow the determination
of Kb.
droxo-bridged copper(II) dimers by N-functionalization
of the tacn macrocycle.
Hydrolysis of Model Phosphate Esters. Bis(p-nitro-
66,67,77,79
In comparison with
phenyl)phosphate (BNPP). To investigate the phosphate
ester cleavage properties of C1 and C2, their reactivity
toward the widely employed activated phosphodiester,
BNPP, was explored by monitoring the release of the
highly colored p-nitrophenoxide anion, (λmax = 400 nm).
The rates of BNPP cleavage by 2 mM solutions of C1, C2,
N-alkylation of the tacn macrocycle, the introduction of
positively charged guanidinium pendants (C2) does not
further enhance the rate of BNPP hydrolysis, suggesting
that the pendant guanidinium groups have only marginal
influence (Table 4).
Further investigation of the kinetics of BNPP hydro-
lysis by C2 was undertaken to examine the effect of
complex concentration and pH on the rate of hydrolysis.
The pH dependence of the rate of BNPP hydrolysis
2
þ
and, for comparison, [Cu(tacn)(OH ) ] , measured
2
2
at pH 7.0 and T = 50 ꢀC, are summarized in Table 4.
Interestingly, C1 was found to be significantly less effec-
tive in cleaving BNPP than the parent complex, [Cu-
follows a bell-shaped profile (Figure 7), similar to the
As in previous
2
þ
(
tacn)(OH ) ] , under the conditions of our study. At
2
80-82
2
profiles reported by other groups.
studies, this pH dependence can be proposed to arise from
the relative activity of the species formed on successive
lower and higher pH values, no release of NP could be
detected by spectrophotometry even after seven days, which
meant that the reaction was too slow for rate constants to
be determined. This was anticipated from the speciation
profile, which showed that the coordination sphere of the
dominant complex is occupied by strong donor atoms. In
contrast, the rate of BNPP cleavage by C2 is 300-times
faster than for C1 and 30-times faster than for [Cu(tacn)-
(79) Deck, K. M.; Tseng, T. A.; Burstyn, J. N. Inorg. Chem. 2002, 41,
6
69–677.
(80) Chapman, J., W. H.; Breslow, R. J. Am. Chem. Soc. 1995, 117, 5462–
5
469.
(
81) Young, M. J.; Chin, J. J. Am. Chem. Soc. 1995, 117, 10577–10578.
82) Mancin, F.; Rampazzo, E.; Tecilla, P.; Tonellato, U. Eur. J. Org.
(
2
þ
(
OH ) ] . Consequently, investigation of C1 was less
2
Chem. 2004, 2004, 281–288.
2

6
32 Inorganic Chemistry, Vol. 50, No. 2, 2011
Tjioe et al.
Table 5. First-Order Rate Constants for Hydrolysis of HPNPP by Copper(II)
Complexes at pH 7.0 (Standard Error in Parentheses)
faster than the control. By comparison, C2 accelerates the
a
rate of cleavage of HPNPP 10-fold when compared to
2
þ
[Cu(tacn)(OH ) ] and 3000 times relative to the control
2 2
-
1
compound
control
Cu -tacn
k
obs (s
)
refs
at the same pH. As for the hydrolysis of BNPP, C1 was
found to be significantly less effective in cleaving HPNPP
than [Cu(tacn)(OH2)2] because of its inability to co-
-
8
1.2 ((0.7) ꢀ 10
3.58 ((0.2) ꢀ 10
1.3 ((0.6) ꢀ 10
3.2 ((0.3) ꢀ 10
1.53 ꢀ 10
this work
this work
this work
this work
84
II
-6
2þ
-
7
5
C1
C2
-
ordinate the substrate, a critical first step in the activation
of the phosphate diester toward hydrolysis. The increased
rate of cleavage of HPNPP by C2 compared to [Cu(tacn)-
II
b
-5
-5
-5
Cu -D
II
b
Cu -E
3.89 ꢀ 10
84
54
II
c
Zn -F
4.6 ꢀ 10
2þ
(
OH ) ] reflects similar findings reported in the litera-
2
2
a
84
Conditions used: [complex] = 2 mM, [HPNPP] = 0.1 mM,
ture. For instance, Hamilton and co-workers found that
introduction of peripheral tertiary amino groups into a
Cu(II)-terpy complex (Figure 9, complexes D and E)
increased HPNPP cleavage activity by 3-7-fold. More
[
HEPES] = 50 mM, I = 0.15 M, T = 25 ꢀC. Data were analyzed using
first-order analysis, yielding kobs directly. [complex]=2 mM, [HPNPP]
b
c
=
0.2 mM at 25 ꢀC, pH 7.0. [complex] = 1 mM.
5
4
deprotonations of the two coordinated aqua ligands. The
first deprotonation (K ) forms a monohydroxo mono-
impressively, Williams et al. have reported a zinc(II)
complex with appended amino groups (Figure 9, complex
a1
3
aqua complex, which is proposed to be active and re-
sponsible for the increase in reaction rate to a maximum
at pH 7.4, but exists in equilibrium with an inactive
dihydroxo-bridged dimer, whose concentration depends
F) that cleaves HPNPP 1.5 ꢀ 10 -times faster than the
parent complex lacking any amino groups. These findings
clearly indicate that cooperativity between metal ions and
proximal functional groups can effectively enhance the
rate of phosphodiester hydrolysis.
on the magnitude of the dimerization constant, KDIM
The second deprotonation (K ) generates a mononuclear
.
In our case, the speciation data for C2 allows the
estimation of a second-order rate constant of 0.9 (
a2
dihydroxo complex which is poorly active as coordina-
tion and activation of the phosphate ester involves
displacement of a hydroxo ligand. Consequently, a sig-
nificant decrease in rate is observed above pH 7.4.
-
1
-1
2
0.1 M
(OH)(OH2)]
Cleavage of Plasmid DNA. The ability of the copper(II)
s
for the cleavage of HPNPP by [L Cu(NH ) -
2 2
3
þ
.
1
2
The rate of BNPP hydrolysis increases with C2 con-
centration (Figure 8), reaching saturation at the higher
concentrations (10 mM). This behavior is similar to that
exhibited by a variety of other cis-diaqua copper(II)
complexes of L and L to cleave nucleic acids was probed
using plasmid pBR 322 DNA as a model system. DNA
cleavage was measured by the conversion of the super-
coiled form of the plasmid DNA (Form I) to the nicked
circular form (Form II). The doubly nicked linear form
(Form III) was not observed in this study. Experiments
were conducted in which pBR 322 plasmid DNA (38 μM
base pair concentration) was incubated with an excess of
C1 and C2 (75-600 μM) (pseudo-first-order conditions)
for varying time intervals, under near physiological con-
ditions (pH 7.0 and T = 37 ꢀC) and at variable pH (see
Figure 10A and Supporting Information Figures S04-S08).
Cleavage experiments were also performed with [Cu-
4
6,83
complexes,
where the reactivity of phosphate ester
hydrolysis levels off with increasing concentrations of
complex, due to the dimerization of metal complexes.
Spectrophotometric pH titration of C2 indicates the
2
presence of the dihydroxo-bridged dimer, L Cu -
2
5
þ
NH ) (NH)(OH)
2
(
at pH 7.0, which was found to
2
3
increase with increasing complex concentration (see
Figure 6d,c). The speciation data for C2 was used to
estimate the concentration of the monohydroxo complex,
2
3þ
2þ
[
L Cu(NH ) (OH)(OH )] , as a function of [C2]. The
2
(tacn)(OH2)2] , to assess whether the guanidinium
2
2
linear dependence of kobs on the concentration of this
species supports the view that this is the active species in
BNPP hydrolysis (see Figure S03 and Table S02, Sup-
porting Information), from which the second order rate
groups in C1 and C2 lead to enhanced DNA cleavage
activity (see Figure 10A and Supporting Information
Figure S08).
Typical images of the electrophoresis gels obtained from
the incubated reaction mixtures are shown in Figure 10A.
For all complexes, a decrease in the intensity of the
band due to Form I with incubation time was accompa-
nied by the appearance and intensification of a band
corresponding to Form II of the plasmid DNA. This
demonstrates the ability of the complexes to create single
nicks in plasmid pBR 322 DNA. The extent of cleavage
was found to be somewhat greater for the complexes
bearing guanidinium pendant groups. After 24 h incuba-
tion at pH 7.0, 60-75% of the initial pBR 322 plasmid
DNA added to solutions containing 150 μM of C1 and C2
was converted to Form II, cf., 40% for the non-function-
2
constant for cleavage of BNPP by [L Cu(NH ) (OH)-
2
2
3
OH2)] was estimated. The value of 1.5 M
þ
-1 -1
(
higher than the values of 0.135, 0.060, and 0.0002 M
s
is
-
1
-1
s
þ
determined at 50 ꢀC for [(Me tacn)Cu(OH)(OH )] ,
3
2
i
þ
( Pr tacn)Cu(OH)(OH )] and [(tacn)Cu(OH)(OH )] ,
3 2
þ
[
2
6
6
respectively.
-Hydroxypropyl-p-nitrophenylphosphate (HPNPP).
2
The rate of cleavage of an RNA model, HPNPP, by C1
and C2 was measured at pH 7.0 at 25 ꢀC (Table 5). As for
RNA, this compound undergoes hydrolysis via an intra-
molecular transesterification reaction, involving nucleo-
0
philic attack of the 2 -OH, to form a cyclic diester and the
2
þ
alized [Cu(tacn)(OH ) ] complex. This enhanced DNA
2 2
highly colored p-nitrophenolate chromophore, which can
be monitored by spectrophotometry (see Experimental
Section). The nonfunctionalized parent complex, [Cu(tacn)-
cleavage activity is reflected in the higher observed first-
order rate constants, kobs, for C1 and C2 {for typical
examples of the data and kinetic profiles see in Support-
ing Information Figures S09-S11 and Tables S03-S05,
2
þ
OH ) ] was itself found to cleave HPNPP 300 times
2 2
(
(
83) Sissi, C.; Mancin, F.; Gatos, M.; Palumbo, M.; Tecilla, P.; Tonellato,
U. Inorg. Chem. 2005, 44, 2310–2317.
(84) Liu, S.; Hamilton, A. D. Tetrahedron Lett. 1997, 38, 1107–1110.

Article
Inorganic Chemistry, Vol. 50, No. 2, 2011 633
Figure 9. Examples of complexes with amino group pendants that accelerate HPNPP cleavage.
Table 6. Observed Rate Constant, kobs, for Single-Strand Cleavage of pBR
22 Plasmid DNA by Copper(II) Complexes at pH 7.0 (Standard Error in
3
a
Parentheses)
-1
Compound
k
obs (s
)
refs
II
-5
-5
Cu -tacn
1.23 ((0.37) ꢀ 10
1.5 ꢀ 10
this work
14
this work
this work
59
85
57
58
II
b
Cu -tacn
-
-
5
5
C1
C2
1.54 ((0.32) ꢀ 10
2.60 ((0.71) ꢀ 10
3.61 ꢀ 10
II
c
-6
-5
-4
-4
Cu -G
II
d
Zn -H
1.13 ꢀ 10
II
Cu -I
8.50 ꢀ 10
II
Cu -J
2.39 ꢀ 10
a
Conditions used: [complex] = 150 μM, [pBR 322 plasmid] = 38 μM
(
per bp), [HEPES] = 40 mM (pH 7.0 at 37 ꢀC. Data were analyzed using
b
first-order analysis, yielding kobs directly. [complex] = 25 μM at 50 ꢀC,
pH 7.8. [complex] = 100 μM. [complex] = 144 μM.
c
d
that the species active in DNA cleavage are (or are
derived) from the metal complexes, C1 and C2.
In contrast to the BNPP and HPNPP cleavage results,
complex C1 was found to cleave DNA at a similar rate
to [Cu(tacn)(OH ) ] . Synthetic metallonucleases are
2
þ
2
2
suggested to promote phosphate ester hydrolysis via
intramolecular attack on the substrate by a metal-
coordinated hydroxide. Therefore, to be an effective
nuclease, the metal complex should have sites available
for substrate binding as well as for binding water or hy-
6
droxide. The speciation distribution curve (Figure 5c)
shows that at pH 7.0, C1 (or [CuL ] ) constitutes 95% of
1
2þ
the total complex concentration at pH 7, indicating that
,
1
only 5% of the monoaqua complex, [Cu(L H)(OH )]
3þ
2
is present. Assuming that C1 itself is inactive, and given
that there is a large excess of complex present relative to
the DNA concentration, it can then be postulated that
1
3þ
Cu(L H)(OH )] is the active DNA cleavage agent. The
2
[
1
4þ
2 2
involvement of [Cu(L H )(OH ) ] is unlikely due to the
2
very low concentration of this species expected at pH 7.
On introduction of the DNA, the negatively charged
phosphate groups on the DNA backbone will interact
with the pendant guanidinium groups and a series of
complex equilibria will be established in solution. De-
pending on the strength of the DNA phosphate-guanidi-
nium interactions and the DNA phosphate-Cu(II) coor-
dinative interaction, detachment and protonation of the
coordinated guanidine could occur, which would have
the consequence of “activating” the cleavage process. We
note that the initial DNA cleavage process involves a
single nick that converts the supercoiled DNA into the
relaxed circular form (Form II), which could potentially
be achieved by the action of a single metal center.
Figure 10. (A) Agarose gels showing cleavage of pBR 322 plasmid
DNA (38 μM bp) by Cu -tacn, C1, and C2 (150 μM) in HEPES buffer
II
(
40 mM, pH 7.0) at 37 ꢀC for various time intervals. Lane 1 and 2: DNA
control, lane3 and 4:1h, lane5and 6:2 h, lane7 and8:4 h, lane9 and10: 6
h, lane 11 and 12: 8 h, lane 13and 14: 16h, lane 15and 16: 24h, lane17and
18: 48 h. (B) The pH dependence of the cleavage of pBR 322 DNA by C1
(a) and C2 (b) after 4 h of incubation ([DNA] = 38 μM bp, [complex] =
1
50 μM, I = 0.15 M and T = 37 ꢀC). (C) Concentration dependence of
II
cleavage of pBR 322 DNA promoted by Cu -tacn, C1 and C2 in 40 mM
HEPES (pH 7.0) at 37 ꢀC over a defined time intervals.
and see Table 6 and Supporting Information Tables S06-
S07 for summaries of k values}. Control experiments
confirmed that no measurable DNA cleavage occurs
when pBR 322 plasmid DNA is incubated with either
The rate of DNA cleavage by C2 was faster than that
obs
2
þ
for both C1 and [Cu(tacn)(OH ) ] . As indicated above,
2 2
the solution speciation of C2 is much more complex, with
mononuclear and binuclear complexes coexisting in the
same pH range, as has been established previously for
150 μM of the non-metalated ligands or and 150 μM CuCl2
(
see Supporting Information Figure S12). This indicates

6
34 Inorganic Chemistry, Vol. 50, No. 2, 2011
Tjioe et al.
sphere and is therefore incapable of coordinating the
phosphate groups on DNA. For C2, mononuclear com-
plexes will be converted into the corresponding dihy-
droxo complexes as the pH increases (see Scheme 3),
which for similar reasons as for C1 are expected to be
inactive.
Effect of Complex Concentration on DNA Cleavage.
The concentration dependence of DNA cleavage by C1,
C2 and [Cu(tacn)(OH ) ] at the pH of optimum activity
2
þ
2
2
is shown in Figure 10c (for typical DNA electrophoresis
gels, see Figures S06-S08 in Supporting Information). At
lower concentrations (<100 μM), all three complexes
showed similar cleavage rates, indicating minimal effect
from the guanidine pendants. At 150 μM, however, C2
showed significantly greater cleavage activity (k =2.60ꢀ
obs
Figure 11. Selected ligands with guanidine pendants whose complexes
accelerate DNA cleavage.
-
5
-1
-5 -1
10
s
) than both C1 (1.54 ꢀ 10
þ
2
s ) and [Cu(tacn)-
2
-5 -1
(
OH ) ] (1.23 ꢀ 10
s ). Increasing the concentration
2
6
6,67,77,79
these types of Cu(II)-tacn complexes.
The higher
beyond 150 μM produced little change in the activity of
C1 or C2, while a significant decrease in rate was observed
for [Cu(tacn)(OH ) ] . The saturation profile seen for
rates of cleavage of the model phosphate esters and DNA
by this complex, therefore, suggests that either more
active monomeric species are present in solution or that
the cleavage reactions are promoted by cooperative inter-
actions with the charged guanidinium groups.
2
þ
2
2
C1 and C2 is observed quite often for biomolecular reac-
tions, and has been previously reported for the cleavage of
8
6-88
DNA by copper(II) complexes.
the profile measured for [Cu(tacn)(OH ) ] has pre-
On the other hand,
2
þ
A modest DNA cleavage activity was achieved by
complexes C1 and C2 compared to cleavage agents re-
ported by other groups (Table 6 and Figure 11).
2
2
viously been tentatively attributed to increased formation
of dihydroxo-bridged dimers at higher concentration,
which bind to the DNA and block access of the cleavage-
5
7-59,81
Both cleave DNA an order of magnitude faster than the
Cu(II) complex of ligand G, which bears pairs of guani-
6
6,67,77,79
active monomeric species.
Effect of Reactive Oxygen Species on DNA Cleavage.
5
9
dinoethyl pendants on an acyclic backbone. A tacn
8
5
macrocycle with guanidinoethyl and alcohol pendants (H)
exhibited slightly higher cleavage rate (3.64 ꢀ 10
DNA cleavage by copper(II) complexes can be achieved
92-94
-
5 -1
8
3,89-91
s ),
in the absence of any metal, compared to both C1 and C2.
However, the cleavage activity was lower for the corre-
through hydrolytic
as well as oxidative
path-
ways. Reactive oxygen species (ROS) generated from the
interaction of the copper(II) complexes with dioxygen are
8
5
2
1,95,96
sponding Zn(II)-H complex. This highlights the un-
certainty that exists about the factors responsible for
promoting metal-guanidine assisted DNA cleavage. Fur-
ther, incorporation of the guanidine pendant into a rigid
believed to be a key factor in DNA cleavage.
To
investigate whether ROS, such as singlet oxygen and
hydroxyl radicals, were involved in DNA cleavage by
C1 and C2, reactions were carried out in the presence of
II
57
aromatic scaffold (Cu -I complex) rather than linking
them by a flexible alkyl chain, as in C1 and C2 or on a lar-
scavengers for hydroxyl radical (10 mM KI, DMSO or
BuOH) or singlet oxygen (10 mM NaN ), as has been
t
3
2,58,59
II
58
4
ger macrocycle (Cu -J) results in faster DNA cleavage.
This suggests that both the rigidity of the guanidine pendant
arm and the stability of the Cu(II)-guanidine complex are
crucial factors in determining the cleavage activity.
reported previously in the literature.
As shown in
Supporting Information Figures S13-S15 and Tables
S08-S09, no obvious inhibition of cleavage activity was
observed in the presence of the singlet oxygen scavenger,
sodium azide, or the hydroxyl radical scavengers, DMSO
pH-Dependence of DNA cleavage. To investigate the
effect of solution pH on the rate of DNA cleavage,
cleavage experiments were performed in various buffer
solutions (40 mM) over the pH range of 6.0-9.0. Typical
DNA electrophoresis gels are shown in Figures S04 and
S05 in the Supporting Information. For both C1 and C2,
the variation in cleavage with pH was measured at a com-
plex concentration of 150 μM and after a reaction time
of 4 h. The plot of percentage cleavage versus pH dis-
plays a bell-like profile (Figure 10b), with an optimal pH
for DNA cleavage of 7.0 for both C1 and C2. This pH-
dependent cleavage behavior is very similar to that re-
t
and BuOH, which indicates that these species are not
likely to be involved in DNA cleavage. However, the
(
86) Schwarzenbach, G.; Boesch, J.; Egli, H. J. Inorg. Nucl. Chem. 1971,
3, 2141–2156.
87) Burstyn, J. N.; Deal, K. A. Inorg. Chem. 1993, 32, 3585–3586.
3
(
(88) Basile, L. A.; Raphael, A. L.; Barton, J. K. J. Am. Chem. Soc. 1987,
09, 7550–7551.
(89) Gupta, T.; Dhar, S.; Nethaji, M.; Chakravarty, A. R. Dalton Trans.
1
2
004, 1896–1900.
(90) Kirin, S. I.; Happel, C. M.; Hrubanova, S.; Weyhermuller, T.; Klein,
C.; Metzler-Nolte, N. Dalton Trans. 2004, 1201–1207.
(91) Zhu, L.; dos Santos, O.; Koo, C. W.; Rybstein, M.; Pape, L.; Canary,
J. W. Inorg. Chem. 2003, 42, 7912–7920.
(92) Singh, U. S.; Scannell, R. T.; An, H.; Carter, B. J.; Hecht, S. M.
J. Am. Chem. Soc. 1995, 117, 12691–12699.
(93) Abraham, A. T.; Zhou, X.; Hecht, S. M. J. Am. Chem. Soc. 2001, 123,
3
7,58
ported in the literature for related complexes.
In the
case of C1, the decrease in the cleavage activity at high pH
can be attributed to conversion of the complex to the in-
active [CuL ] species (see speciation profile in Figure 5
and Scheme 2), which has a fully saturated coordination
1
2þ
5
167–5175.
94) Wagner, C.; Wagenknecht, H., -A. Chem.;Eur. J. 2005, 11, 1871–
876.
95) Thyagarajan, S.; Murthy, N. N.; Narducci, S., A. A.; Karlin, K. D.;
(
1
(
(
85) Sheng, X.; Lu, X. M.; Zhang, J. J.; Chen, Y. T.; Lu, G., -Y.; Shao, Y.;
Rokita, S. E. J. Am. Chem. Soc. 2006, 128, 7003–7008.
(96) Pyle, A. M.; Barton, J. K. Prog. Inorg. Chem. 1990, 38, 413.
Liu, F.; Xu, Q. J. Org. Chem. 2007, 72, 1799–1802.

Article
Inorganic Chemistry, Vol. 50, No. 2, 2011 635
Figure 12. Extent ofaerobic and anaerobic cleavage of pBR 322 plasmid DNA byC1 (a) and C2 (b) [C1] and [C2] = 150 μM in 40 mM HEPES (pH7.0) at
7 ꢀC over defined time intervals.
3
addition of the hydroxyl radical scavenger, KI, resulted in
substantial inhibition of DNA cleavage. On the basis of
this observation, the mechanism of DNA cleavage
mediated by C1 and C2 may be similar to that proposed
Conclusion
Two new derivatives of 1,4,7-triazacyclononane featuring
pairs of ethylguanidine (L ) and propylguanidine (L ) pen-
1
2
dant arms are reported. An X-ray structure determination of
9
7-100
for other multinuclear copper complexes,
whereby
1
[
CuL ](ClO ) , revealed the unexpected coordination of the
4
2
DNA cleavage is redox-mediated - Cu(II) centers are
initially reduced to Cu(I) species and, subsequently,
react with dioxygen to form peroxo-dicopper(II) deriva-
tives, which in turn generate active oxygen species neces-
saryforcleavage. However, although widely used it should
be noted that an excess of iodide is capable of reducing the
copper(II) complexes to the corresponding copper(I)
complexes, and then displacing the ligands from the
pendant guanidine groups. The quite short Cu-N distances
˚
of 2.0 A, compare and estimated binding constants of
6
>
10 for each guanidine indicate remarkably stable Cu-
N(guanidine) bonds. Of the two complexes, C2 was a more
effective cleavage agent for the model phosphate esters,
BNPP and HPNPP, than the non-functionalized parent
2þ
complex, [Cu(tacn)(OH ) ] , while C1 was essentially
2
2
inactive. In contrast, DNA cleavage experiments indi-
cated that both complexes enhance the cleavage of DNA
1
01
complex, leading to the precipitation of CuI.
Anaerobic Cleavage of DNA. To further probe the
effects of molecular oxygen on the degradation of DNA
2þ
relative to [Cu(tacn)(OH ) ] , but it is unclear whether
2
2
this is a direct result of the introduction of the positively
charged guanidinium groups. The complexes showed
cleavage activity toward plasmid DNA under both aero-
bic and anaerobic conditions, with a ∼30% decrease in
reactivity observed under anaerobic condition, indicating
that cleavage occurs through both hydrolytic and redox
pathways.
1
2
by the copper(II) complexes of L and L , reactions were also
performed under rigorously anaerobic condition. Both C1
and C2 were found to still effectively cleave supercoiled
plasmid pBR 322 plasmid DNA, although the rate of the
hydrolysis was significantly reduced (∼30%) when com-
pared to that observed under aerobic condition (Figures 12
and Supporting Information Figures S16-S17 and
Tables S10-S11). These results are similar to those obtained
Acknowledgment. This work was supported by the
Australian Research Council through the Discovery
Program. L.T. is the recipient of a Monash Graduate
Scholarship.
2þ
14
for [Cu(tacn)(OH ) ] by Burstyn and co-workers, indi-
2
2
cating the simultaneous occurrence of a prevailing hydrolytic
cleavage pathway and an oxidative cleavage pathway.
(
97) Humphreys, K. J.; Karlin, K. D.; Rokita, S. E. J. Am. Chem. Soc.
002, 124, 6009–6019.
98) Humphreys, K. J.; Karlin, K. D.; Rokita, S. E. J. Am. Chem. Soc.
001, 123, 5588–5589.
99) Humphreys, K. J.; Karlin, K. D.; Rokita, S. E. J. Am. Chem. Soc.
002, 124, 8055–8066.
100) Humphreys, K. J.; Johnson, A. E.; Karlin, K. D.; Rokita, S. E.
J. Biol. Inorg. Chem. 2002, 7, 835–842.
101) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; 6th ed.;
John Wiley & Sons: New York, 1999.
Supporting Information Available: Crystallographic file in
CIF format, table of crystallographic data (Table S01), principal
component analysis (PCA) (Figure S01), UV-vis-NIR spectra
of complex C1 titrated against NPP (Figure S02), analysis of
kinetics of BNPP cleavage (Figure S03 and Table S02), and
agarose gel images of DNA cleavage and kinetic profiles
(Figures S04-S17 and Tables S03-S11). This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
2
2
2
(
(
(
(