Polymethacrylates with a covalently linked CuII-cyclen complex for the in situ generation of nitric oxide from nitrosothiols in blood

Article abstract of DOI:10.1002/anie.200503588

(Figure Presented) In the matrix: A new cross-linked hydrogel polymer, poly(2-hydroxyethyl methacrylate) (pHEMA), containing appended Cu ^II-cyclen sites (cyclen = 1,4,7,10-tetraazacyclododecane) was prepared and shown to generate nitric oxide from endogenous S-nitrosothiol species (RSNO) typically found in circulating whole blood.

Full text of DOI:10.1002/anie.200503588

Angewandte
Chemie
Biosensors
DOI: 10.1002/anie.200503588
Polymethacrylates with a Covalently Linked
Cuii–Cyclen Complex for the In Situ Generation
of Nitric Oxide from Nitrosothiols in Blood**
Sangyeul Hwang, Wansik Cha, and
Mark. E. Meyerhoff*
Nitric oxide produced by endothelial cells (EC) which line the
walls of all blood vessels plays an important role in preventing
platelet activation, adhesion, and, ultimately, thrombosis at
the EC/blood interface.[1,2] We have reported previously that
[*] S. Hwang, W. Cha, Prof. M. E. Meyerhoff
Department of Chemistry
The University of Michigan
930 North University Avenue
Ann Arbor, MI, 48109-1055 (USA)
Fax: (+1)734-647-4865
E-mail: mmeyerho@umich.edu
[**] We thank the NIH (EB-000783 and EB-004527) for the financial
support of this research.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 2745 –2748
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2745

Communications
plasticized polymeric films
doped with a lipophilic cop-
per(ii) complex decompose
S-nitrosothiols
(RSNO),
such as S-nitrosogluta-
thione (GSNO), to NO at
their interface in the pres-
ence of naturally occurring
reducing equivalents (e.g.,
glutathione
(GSH)).[3]
Indeed, it is well known
that Cuii/Cui ions can cata-
lyze the generation of NO
from RSNO species (see
the Supporting Informa-
tion).[4,5] Furthermore, it
was reported that peptide-
and protein-bound Cuii can
be reduced to Cui by a thiol
to generate NO from
RSNO.[6] It now appears
that appropriate Cuii/Cui
organic ligand complexes
Scheme 1. The synthesis of the cross-linked hydrogel CuII–cyclen–pHEMA polymer 7. a) Boc2O (3 equiv), Et3N
(3 equiv), CHCl3, RT, 12 h (76%); b) 3-bromopropanol (3.4 equiv), Na2CO3 (1.2 equiv), NaI (1.3 equiv),
CH3CN, 808C, 12 h (88%); c) methacryloyl chloride (5.4 equiv), Et3N (9.5 equiv), THF, ꢀ208C, RT, 1 h
(59%); d) trifluoroacetic acid (TFA), CH2Cl2, 08C, RT, 12 h (quantitative yield); e) 1. HEMA (66.0 wt%),
ethyleneglycol dimethacrylate (EGDM; 2.1 wt%), azobisisobutyronitrile (AIBN; 0.4 wt%), 65 8C, 12 h;
2. soaked and washed with a solution of aqueous NaOH/EtOH (1:10, v/v, pH 11–12); f) CuCl2·2H2O
(1 equiv vs 5) in EtOH, 658C, 4 h.
can carry out similar redox chemistry.[3] Hence, it should be
possible to utilize endogenous reducing equivalents (e.g.,
potential NO generation by the resulting polymeric material
could take place within the bulk of a polymer coating, not
merely at the polymer/solution interface.
AHCTREUNG
ascorbate, GSH, and cysteine (CySH)) and the endogenous
NO precursors (e.g., GSNO, S-nitrosocysteine (CysNO), and
S-nitrosoalbumin (AlbSNO)) in blood to generate locally
enhanced NO levels at a polymer/blood interface by immo-
bilizing catalytic CuII/CuI sites within the polymeric material.
Although the exact concentrations of RSNOs in normal
human blood are in debate,[7,8] the levels of RSNOs and
required reducing equivalents are thought to be in the
nanomolar to micromolar range.[9,10] As very low levels of
NO (< 1 nm) can be effective in inhibiting platelet function,[11]
even a small fractional conversion of endogenous RSNO
species into NO at a polymer/blood interface could be
beneficial in lowering the risk of thrombus formation at
such interfaces.
The film of CuII–cyclen–pHEMA 7 displayed a lmax value
in the range 621–644 nm in the visible region of the UV/Vis
spectrum after being fully hydrated with deionized water or
phosphate-buffered saline (PBS) (10 mm, pH 7.4). This
absorption is within the range known for square-pyramidal
Cuii complexes (550–670 nm).[14] In addition, electron para-
magnetic resonance (EPR) spectra of 7 in both frozen
aqueous and PBS-buffered solutions exhibited the typical
four-line patterns expected for a CuII–cyclen complex (see the
Supporting Information).[15] Moreover, the extent of the blue
color of 7 and the corresponding copper content, as deter-
mined by atomic absorption (AA) spectrophotometric anal-
ysis after dissolution of a given mass of film in a solution of
sulfuric acid, correlated very closely with the molar ratio of
the modified cyclen monomer 5/HEMA used in the polymer-
ization reaction and the theoretical amount of copper
expected (see the Supporting Information). Taken together,
the data suggest that most cyclen sites covalently linked to the
polymer are complexed with Cuii ions in the final cross-linked
pHEMA films.
Figure 1illustrates the NO generation that is achieved
when a small piece of the CuII–cyclen–pHEMA film is placed
in and then removed from a solution of GSNO and GSH in
deoxygenated PBS buffer. The NO generated is monitored
continuously with a chemiluminescence NO analyzer (NOA).
As shown in Figure 1a, a blank experiment in which a
hydrogel film of 6 (not treated with copper) was placed in the
test solution did not generate any detectable NO. This
experiment demonstrates that, as expected, copper ions are
key for NO generation. In a second blank experiment, a cross-
linked pHEMA film of the same size was prepared by the
reaction of HEMA, EGDM, and AIBN, but without copoly-
merization of 5, and further treated with copper ions by using
Herein, we report the synthesis of a modified CuII–cyclen
(cyclen = 1,4,7,10-tetraazacyclododecane) complex cova-
lently attached to a cross-linked poly(2-hydroxyethyl meth-
acrylate) (pHEMA) and demonstrate that this new material
can catalytically generate NO from naturally occurring
RSNOs at the physiological pH value. Further, it will be
shown that spontaneous NO generation from RSNOs which
exist in fresh sheepꢀs blood can be achieved with this novel
material.
Cyclen is known to have very high binding constant with
[12]
Cuii (1023 mꢀ1
)
and square-pyramidal structure,[12,13] thus
allowing this complex to possess an open site(s) at which
interactions with reducing equivalents and RSNO species are
possible. Herein, a propylmethacrylate derivative of cyclen
was prepared for copolymerization with 2-hydroxyethyl
methacrylate (HEMA; see Scheme 1). pHEMA is a high-
water-uptake polymer network into which endogenous reduc-
ing agents (CySH, GSH, ascorbate, etc.) and low-molecular-
weight RSNOs (CySNO and GSNO) can readily penetrate.
Thus, in contrast to the system described previously,[3]
2746 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2745 –2748

Angewandte
Chemie
leaching of copper ions, as measured by AA spectrophoto-
metric analysis (the films soaked in the solutions still retained
more than 64% of their original copper content after 15 days
at 37.58C; see the Supporting Information), and this leaching
resulted in a decrease in the apparent level of NO generation
(25–50% decrease) relative to a control of the same film (not
soaked in solutions of GSH and/or GSNO). However, the
lmax value of the film, as shown by UV/Vis spectroscopic
analysis, did not change in location before and after such
soaking, thus suggesting that GSH (a reducing agent) and
glutathione disulfide (GSSG, a by-product of the NO-
generation reaction), both strong ligands for copper
ions,[16,17] do not bind irreversibly as ligands for the covalently
linked complex. A more critical test was the examination of
NO generation by films of the new CuII–cyclen–pHEMA 7
material (radius: 2.8 mm; thickness: 0.2 mm; type: disk) from
GSNO before and after the exposure to animal blood for a
finite period. It was found that NO generation from GSNO is
maintained after soaking such films in 10 mL of the sheep
plasma for 3 days at 48C or whole blood for 1day at 4 8C (see
the Supporting Information). These results indicate that the
deactivation of 7 caused by endogenous free thiols, proteins,
and other blood components may not be a significant issue.
Clearly, a large number of in vivo studies over varying time
periods will be required to fully assess the true NO-generation
ability and resultant thromboresistance of the new polymer
under physiological conditions.
Figure 1. The profiles of NO generation for the following disk polymers
(thickness: 0.3 mm; radius: 2.0 mm) in deoxgenated PBS buffer
(2 mL, 10 mm; pH 7.4), monitored with a chemiluminescence NOA at
room temperature after the following steps: a) A: Addition of EDTA
(10 mm); B: addition of GSH (30 mm); C: addition of GSNO (10 mm);
D: insertion of the cross-linked film 6; E: removal of the cross-linked
film 6; F: insertion of pHEMA; G: removal of pHEMA. b) H: Addition
of EDTA (10 mm); I: addition of GSH (5 mm); J: addition of GSNO
(5 mm); K: insertion of CuII–cyclen–pHEMA 7; L: removal of 7.
To further demonstrate that the CuII–cyclen–pHEMA
polymer can generate NO spontaneously in fresh whole
blood, an amperometric RSNO sensor was constructed using
an interpenetrating network (IPN) film of 7 and hydrophilic
polyurethane (HPU, Tecophilic, SP-93A-100) as an outer film
on a NO-selective electrochemical sensor (see the Supporting
Information). The response of a control NO sensor without
the polymer containing the CuII–cyclen complex (but with an
HPU outer membrane) was also monitored in the same blood
samples. Each sensor was precalibrated for an intrinsic direct
amperometric response to NO (see the Supporting Informa-
tion), and then the NO and RSNO levels were measured,
respectively, in the fresh sheepꢀs blood diluted with PBS
buffer at 378C. Figure 2 shows that the NO sensor without the
catalytic CuII–cyclen–pHEMA polymer site in the outer
membrane has relatively little amperometric response when
placed in the blood sample, whereas the difference in NO
responses between the RSNO and NO sensors (equivalent to
ꢁ 25 nm NO, calculated from calibration curves) appears to be
due to the decomposition of RSNO species in the sheepꢀs
blood by the IPN film of 7 and HPU. Therefore, this
experiment confirms that the CuII–cyclen complex covalently
attached to pHEMA generates NO from endogenous sub-
strates in the sheep blood in vitro.
the same method employed for the preparation of 7; this film
yielded only a small NO flux upon the very first immersion in
the solution of GSNO. This behavior is due to nonspecifically
adsorbed CuII ions adhered to the pHEMA matrix. Indeed,
little or no NO generation was observed for subsequent
immersions of the same blank films (see Figure 1a). In
contrast, the first immersion of a similarly sized film of 7 into
the solution of GSNO provided a slow increase in the NO flux
which reached a steady state (see Figure 1b). The NO flux
decreased to the baseline quickly after this film was removed
from the reaction cell. Subsequent immersion/removal cycles
of the same film showed that 7 could reversibly achieve nearly
the same steady-state NO flux on sequential immersions into
fresh solutions of GSNO/GSH. When CysNO was examined
as the RSNO substrate, higher fluxes of NO generation were
achieved over a relatively short time periods (see the
Supporting Information), but the NO generation decreased
quickly, as the bulk CysNO substrate was consumed by a
rapid catalytic reaction. These observations might relate to
the relative reaction rates of the two species with immobilized
Cuii sites. As the concentration of CySNO in blood is higher
than GSNO,[10] CySNO is more likely to contribute toward a
larger fraction of the potential NO-generating capability by
the new CuII–cyclen–pHEMA polymers when they are in
contact with fresh blood.
In conclusion, a CuII–cyclen complex was covalently to
attached pHMEA, and the resulting material is capable of
generating NO from naturally occurring RSNO species, both
in buffer solution and directly when in contact with fresh
whole blood. The presence of endogenous reducing equiv-
alents is required for such catalytic chemistry to take place.
Assessment of the stability of the new CuII–cyclen–pHEMA 7
material suggests that the slow leaching of copper ions from
Several soaking experiments with the new polymeric films
were conducted in the presence of excess amounts of
solutions of GSH and/or GSNO in PBS buffer to investigate
the potential leaching of copper ions and the deactivation of
7. The extraction solutions were shaken continuously, and
replaced with fresh solution daily. There was substantial
Angew. Chem. Int. Ed. 2006, 45, 2745 –2748
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2747

Communications
[14] M. C. Styka, R. C. Smierciak, E. L. Blinn, R. E. DeSimone, J. V.
Passariello, Inorg. Chem. 1978, 17, 82 – 86.
[15] M. Soibinet, I. Dechamps-Olivier, E. Guillon, J.-P. Barbier, M.
Aplincourt, F. Chuburu, M. L. Baccon, H. Handel, Eur. J. Inorg.
Chem. 2003, 10, 1984 – 1994.
[16] H. K. Baek, R. L. Cooper, R. A. Holwerda, Inorg. Chem. 1985,
24, 1077 – 1081.
[17] B. C. Gilbert, S. Silvester, P. H. Walton, J. Chem. Soc. Perkin
Trans. 2 1999, 1115 – 1121.
Figure 2. The direct amperometric detection of endogenous RSNOs in
fresh sheep’s blood by using a NO sensor A) as a control and a RSNO
sensor and B) upon injection of 6 mL of fresh sheep’s blood (!) into
PBS buffer (30 mL, 10 mm; pH 7.4) at 378C under a stream of
nitrogen.
the polymer film may cause a measurable decrease in NO
generation with time, but copper deactivation is not a
significant problem. Although fundamental questions with
respect to the exact mechanism by which 7 generates NO
remain, the data presented herein suggest that this new
material may prove beneficial as a potentially new thrombo-
A
tions with contact times of hours to several days. Experiments
in vivo to demonstrate the effectiveness of this material in
reducing thrombus formation will begin shortly.
Received: October 11, 2005
Revised: January 12, 2006
Published online: March 20, 2006
Keywords: biosensors · copper · N ligands · nitrogen oxides ·
.
polymers
[1] R. T. Mathie, T. M. Griffith, The Haemodynamic Effects of
Nitric Oxide, Imperial College Press, Singapore, 1999.
[2] A. O. Mateo, A. A. D. Artinano, Pharmacol. Res. 2000, 42, 421–
427.
[3] B. K. Oh, M. E. Meyerhoff, J. Am. Chem. Soc. 2003, 125, 9552 –
9553.
[4] S. C. Askew, D. J. Barnett, J. McAninly, D. L. H. Williams, J.
Chem. Soc. Perkin Trans. 2 1995, 741– 745.
[5] A. P. Dicks, P. H. Beloso, D. L. H. Williams, J. Chem. Soc. Perkin
Trans. 2 1997, 1429 – 1434.
[6] A. P. Dicks, D. L. H. Williams, Chem. Biol. 1996, 3, 655 – 659.
[7] R Rossi, D. Giustarini, A. Milzani, R. Colombo, I. Dalle-Donne,
P. Di Simplicio, Circ. Res. 2001, 89, 47e.
[8] J. S. Stamler, Circ. Res. 2004, 94, 414 – 417.
[9] D. Giustarini, A. Milzani, R. Colombo, I. Dalle-Donne, R. Rossi,
Clin. Chim. Acta 2003, 330, 85 – 98.
[10] M. Kelm, Biochim. Biophys. Acta 1999, 1411, 273 – 289.
[11] M. W. Vaughn, L. Kuo, J. C. Liao, Am. J. Physiol. 1998, 274,
H2163 – H2176.
[12] V. J. Thom, G. D. Hosken, R. D. Hancock, Inorg. Chem. 1985, 24,
3378 – 3381.
[13] V. J. Thom, C. C. Fox, J. C. A. Boeyens, R. D. Hancock, J. Am.
Chem. Soc. 1984, 106, 5947 – 5955.
2748 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2745 –2748