Preparation and characterization of an improved Cu2+-cyclen polyurethane material that catalyzes generation of nitric oxide from S-nitrosothiols

Article abstract of DOI:10.1039/c2jm32726k

A new, stable and highly efficient Cu²⁺-cyclen-polyurethane material is described and shown to exhibit an improved performance compared to that of prior materials for the catalytic decomposition of S-nitrosothiols to physiologically active nitric oxide. The Royal Society of Chemistry.

Full text of DOI:10.1039/c2jm32726k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 18784
www.rsc.org/materials
COMMUNICATION
Preparation and characterization of an improved Cu2+–cyclen polyurethane
material that catalyzes generation of nitric oxide from S-nitrosothiols†
*
Kun Liu and Mark E. Meyerhoff
Received 30th April 2012, Accepted 1st August 2012
DOI: 10.1039/c2jm32726k
polymer that possesses a covalently bound Cu2+–ligand complex (1,
Fig. 1) is developed as an alternative to the nanoparticle-based system
to avoid potential cytotoxicity due to copper leaching from the
polymer. Although two analogous polymers (2 and 3, Fig. 1)8,9 were
previously reported to exhibit catalytic decomposition activity toward
RSNOs, these materials had limitations with respect to their future
biomedical application. For example, the preparation of polymer 2 is
quite complicated and the catalyst content is difficult to control.
Moreover, the amide bond between the catalyst and the modified
polymer is not stable and therefore is susceptible to acid, base, or
enzymatic degradation. The structure of polymer 3 is too hydro-
phobic due to the absence of a hydrophilic linker when tethering the
Cu(II)–cyclen catalyst to the polymer backbone. Indeed, the hydro-
phobicity of polymer 3 might compromise its true biocompatibility,
owing to a higher degree of protein adsorption.
A new, stable and highly efficient Cu2+–cyclen–polyurethane
material is described and shown to exhibit an improved performance
compared to that of prior materials for the catalytic decomposition
of S-nitrosothiols to physiologically active nitric oxide.
Nitric oxide (NO) generating polymeric materials have been investi-
gated recently and demonstrated to produce NO locally from S-
nitrosothiols (RSNOs), such as S-nitrosoglutathione (GSNO), S-
nitrosocysteine (CySNO) and S-nitrosoalbumin (AlbSNO).1 These
materials may continuously produce NO under physiological
conditions from the pool of endogenous S-nitrosothiols present in
blood, and thereby prevent the activation of platelets and concomi-
tant formation of thrombus on blood contacting medical devices
coated with such materials. Indeed, it is known that certain metal ions
(Cu2+, Fe2+, Hg2+, and Ag+)2 as well as organometallic species
(organodiselenides (RSeSeR)3,4 and organoditellurides (RTeTeR)5)
can catalyze the decomposition of RSNOs to NO. Hence, new
biomaterials being developed, incorporating such catalysts on their
surfaces, have the potential to sustain NO generation utilizing
endogenous RSNOs as substrates or by infusing additional amounts
of these species6 into blood.
Among materials reported to date, Cu2+-mediated NO generating
polymers are attractive because Cu2+ is a well-known and very
effective catalyst for RSNO decomposition.7 The mechanism involves
Cu2+ reduction by a thiolate anion (RSꢀ) to yield Cu+ and a disulfide
(RSSR), and Cu+ then binds and transfers an electron to the RSNO
species, yielding NO and a thiolate anion along with Cu2+. The thi-
olate anion can then reduce Cu2+ to the active Cu+ species, creating a
catalytic cycle.
In this communication, the optimized preparation, stability and
catalytic activity of the new Cu2+–cyclen mediated NO generating
material are examined. Compared to the published polymers, poly-
mer 1 is stable and potentially more thromboresistant because of an
established covalent bonding (C–C) and the hydrophilic PEG
(tetraethylene glycol) linker that attaches the catalytic complex to the
polyurethane backbone. To synthesize polymer 1, a short amino-
terminated PEG linker (4) was first prepared in good yield following
a modified Gabriel synthesis10 (ESI, Scheme S1†). This LMW PEG is
soluble in non-polar organic solvents (e.g., chloroform, methylene
chloride, ether) and therefore it is easy to wash away
unreacted materials during the workup. In contrast, the HMW
In recent work, a NO generating polymer containing copper
nanoparticles coupled with RSNO infusion6 was successfully applied
to a rabbit model for extracorporeal circulation (ECC) (with the NO
generating coating on the inner walls of the ECC tubing). The results
suggested that significantly reduced thrombosis and prevention of
platelet aggregation after 4 h of blood exposure can be achieved via
the NO generation chemistry. In this work, a new NO generating
Department of Chemistry, University of Michigan, 930 N. University Ave,
Ann Arbor, MI, 48109, USA. E-mail: mmeyerho@umich.edu; Fax: +1-
734-6474865; Tel: +1-734-7642169
† Electronic supplementary information (ESI) available: Experimental
details for the synthesis of polymer 1, characterization spectra and
additional chemiluminescence data. See DOI: 10.1039/c2jm32726k
Fig. 1 Comparison of the newly synthesized (1) with the published (2
and 3) Cu2+-mediated NO generating polymers.
18784 | J. Mater. Chem., 2012, 22, 18784–18787
This journal is ª The Royal Society of Chemistry 2012

View Article Online
diamino-terminated polyethylene glycol (dipropylamine-PEO, DPA-
400E)8 used in the preparation of polymer 2 is difficult to remove
from the reaction mixture using a Soxhlet extraction owing to its high
polarity. Secondly, a stable cyclen derivative (5) was synthesized by
covalently binding cyclen and the new PEG linker (4) through a
three-step synthesis (ESI, Scheme S2†). Finally, compound 5 was
applied to prepare the target polymer, TPU-PEG–cyclen–CuII (1)
(Scheme 1), in accordance with previous methods.8,11–13 IR and NMR
spectroscopy were employed to monitor and characterize each step of
the syntheses (ESI, Fig. S1 and S2†). Finally, it was found that the
final polymer 1 contained 0.2–0.5 wt% copper, as quantitated by ICP-
AES analysis. The presence of the Cu(II)–cyclen complex within the
isolated polymer was also characterized by UV-Vis spectroscopy via
the absorbance band with a l at 674 nm (ESI, Fig. S3†).
Cu(II)-catalyzed decomposition of RSNO has previously been
investigated, including the mechanism, proposed structures of active
intermediates, rate constants and rate limiting steps.7,14,15 All of
these factors have an influence on the NO generation observed with
polymers possessing tethered Cu2+–ligand complexes. When films
of the new polymer were cast (using THF as solvent) and the
solvent removed, the dried films were tested for NO generation
from RSNOs in a solution of PBS (10 mM, pH 7.4) containing
3 mM EDTA (used to chelate trace metal ion contaminates in the
buffer) using real-time detection of NO by chemiluminescence. The
new polymer (1) exhibited excellent NO generating activity, even
when using a low concentration of GSNO (5 mM), with apparent
fluxes of 6–10 ꢁ 10ꢀ10 mol cmꢀ2 minꢀ1 (see Fig. 2(a)). This
is higher than the biologically relevant NO fluxes (ca. 0.5–4.0 ꢁ
10ꢀ10 mol cmꢀ2 minꢀ1) known to occur from the surface of
endothelial cells that line the inner walls of healthy blood
vessels.16,17 Repeated reinsertion/removal of the film from the test
solution indicated that a relatively constant NO flux of 6 ꢁ
10ꢀ10 mol cmꢀ2 minꢀ1 can be achieved when the polymer is placed
into the GSNO solution. When the film is removed from the
buffer, the NO signal gradually decreases to the baseline, suggesting
that the copper ion coordinated within cyclen does not significantly
leach out of the polymer after participating in the redox cycle that
generates NO from the RSNO species in solution (also, see the
results below from a copper leaching study in blood). This reaction
likely takes place at the polymer film/solution interface, although it
is possible that the test RSNO species (GSNO) may have some
capability to diffuse into the polyurethane polymer, given the
hydrophilic nature of the base polymer employed in this work (i.e.,
water uptake value of 100 wt%).
Fig. 2 The measurements of catalytic NO generation by the small films
of TPU-PEG–cyclen–CuII (1) (thickness: 30 mm, 0.47 wt% copper) in a
solution of (a) 5 mM GSNO/GSH (area ¼ 0.13 cm2) or (b) 0.5 mM
CySNO/CySH (area ¼ 0.31 cm2) in PBS buffer (10 mM, pH 7.4) con-
taining 3 mM EDTA via chemiluminescence NOA at RT. Arrows indicate
addition or removal of the given species into the mixture. All spikes in
figures are artifacts due to opening the reaction vessel.
To confirm the catalytic mechanism of NO generation from
GSNO by the new polymer appended Cu(II)–cyclen complex, the
decomposition of RSNOs was studied using a much higher concen-
tration of GSNO (1 mM). After incubation with a small piece of the
polymer film, a 24-fold greater amount of NO (1.5 ꢁ 10ꢀ6 moles)
was generated by the new polymer containing 6.3 ꢁ 10ꢀ8 moles of
Cu(II) sites in the GSNO solution prepared in 10 mM PBS (pH 7.4)
with 1 mM reducing agent present. In this experiment, the catalytic
amount of Cu(II) present was only 3% of the total GSNO used. In
contrast, when no Cu(II) was incorporated into the polyurethane
material, very little NO (7.89 ꢁ 10ꢀ9 moles) was observed using
the same concentration of GSNO with a 0.4% conversion rate
Scheme 1 Synthesis of the new Cu2+-mediated NO generating polymer: TPU-PEG–cyclen–CuII (1).
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 18784–18787 | 18785

View Article Online
(see ESI, Fig. S4†). These data clearly demonstrate that the Cu(II)–
cyclen–PU material can catalytically decompose RSNO to generate
NO, and the polymer alone cannot produce a significant amount of
NO without the presence of the catalytic Cu(II) sites.
at a constant NO level with repeated insertion/removal of the film
from the test solution (see ESI, Fig. S6(b)†).
Another blood-contact study was performed with the previously
published polymer (2). Unfortunately, it loses nearly all the NO
generating activity under the same experimental conditions (see
Fig. 3(b) and S7†), clearly suggesting that polymer (1) is a more stable
catalyst that is less poisoned by the presence of species in blood.
Regardless of what type of RSNO species serves as the NO source,
1–5 ꢁ 10ꢀ10 mol cmꢀ2 minꢀ1 NO flux can be generated from the new
polymer (1) even after contacting with blood/plasma for 24 h.
A prolonged copper leaching study of the new material as well as
polymer 2 was conducted by soaking polymer films in fresh whole
sheep blood for 2 days. The copper concentrations in the plasma
phase before and after contacting with the materials were analyzed
via ICP-AES (see ESI, Fig. S8†). Results indicated that 25% of the
copper leached from the new polymer 1 and that 49% was lost from
the previously reported polymer 2 material. This result is consistent
with the above in vitro evaluation of the NO generating activities of
the two different polymers after contact with whole blood, and
provides a reasonable explanation for the above results.
A much higher surface flux of NO, 16 ꢁ 10ꢀ10 mol cmꢀ2 minꢀ1
(Fig. 2(b)), was observed when a much lower concentration of
CySNO (0.5 mM) (rather than GSNO) was used in the test solution.
Askew et al.14 studied the rate constants of Cu(II)-catalyzed decom-
position of a variety of RSNOs by forming a five or six-membered
ring intermediate via the amino or carboxyl group of the RSNOs.
The rate constant for free Cu(II) for decomposing CySNO is 24 500 ꢂ
500 molꢀ1 dm3 sꢀ1, which may explain why the polymer (1) could
produce more NO even when a 10-fold lower concentration of
CySNO than that of GSNO was used.
A NO generation study was also conducted under physiological
conditions18 at a temperature of 37 ꢃC with 1 mM GSNO and 30 mM
GSH present in the test solution. A flux of 5 ꢁ 10ꢀ10 mol cmꢀ2 minꢀ1
of NO was generated from a small film of the new polymer (see ESI,
Fig. S5†). The total generated NO over 12 min was found to be 9 ꢁ
10ꢀ10 mol, equivalent to a 45% conversion rate of the GSNO species.
This low catalytic efficiency is assumed to be due to a low second-
order rate constant of Cu(II)-catalyzed decomposition of GSNO,14
and also to the inefficient reducing ability of GSH (CySH is regarded
as a more favorable reducing agent at physiological pH than GSH8).
Depending on the structure (and hence reactivity) of the given
RSNO, either Cu(I) formation or its reaction with RSNO can be rate-
limiting.7
The in vivo stability of the new NO generating material was also
investigated using a rabbit model or extracorporeal circulation (ECC)
in which the NO generating material is coated on the inner walls of
PVC tubing, as described previously.6 The polymer 1 material still
preserved its NO generating ability after 4 h of blood exposure with a
surface flux of 13 ꢁ 10ꢀ10 mol cmꢀ2 minꢀ1 using SNAP as the test
RSNO species (see ESI, Fig. S9†). In addition, the plasma copper
assay prior to and post-ECC exposure did not show a significant
amount of copper leaching (ESI, Fig. S10†). These data clearly
demonstrate that the new NO generating material is relatively stable
in a physiological environment.
Beyond testing in PBS buffer, it is critical to also assess the NO
generating ability of the new polymer after contacting with blood.
Therefore, small films of the polymer were soaked in whole sheep
blood or platelet-rich sheep plasma for 24 h. The levels of NO
generated per cm2 of film were detected via chemiluminescence both
prior to and after soaking in the blood/plasma. Both GSNO and
CySNO were employed as substrates to evaluate any changes in
activity owing to the exposure to blood. As shown in Fig. 3(a) and
S6,† a significant portion of the NO generating activity could be
preserved after contacting whole blood (30–50%) or platelet-rich
plasma (50–70%). A surface flux of 2–3 ꢁ 10ꢀ10 mol cmꢀ2 minꢀ1 of
NO is generated in the presence of 5 mM GSNO (ESI, Fig. S6(a)†).
More NO is produced (5–10 ꢁ 10ꢀ10 mol cmꢀ2 minꢀ1 NO flux) from
0.5 mM CySNO (Fig. 3(a)). Moreover, the ability of the polymer film
to generate NO from CySNO after exposure to blood is reproducible
Protein adsorption is the first of a complex series of events trig-
gered by material–blood interactions. Adsorption of proteins leads to
activation of platelets and leukocyte adhesion, and finally coagula-
tion, which normally occurs within seconds of material–blood
contact.19 Thus, it is also important to assess whether new materials
can cause increased protein adsorption when in contact with blood. A
control blood-contact study with the unmodified TPU (SP-93A-100)
was carried out using the same blood samples as described above.
The NOA data for the control films did not show any obvious change
before and after blood exposure, with all the measurements only
detecting baseline NO signals (see ESI, Fig. S11†). This result
suggests all the generated NO after contact with blood/plasma
(shown in Fig. 3(a) and S6†) is derived from decomposition of RSNO
catalyzed by the Cu(II)–cyclen structure bound to the new polymer
(1), not from the adsorbed metallo-proteins on the surface of the
material.
Another method to determine whether the new NO generating
material exhibits appreciable protein adsorption involves assessing, in
vitro, fibrinogen adsorption.20 Fibrinogen is a crucial clotting factor
within the blood-coagulation cascade and it plays a major role in
hemostatic plug formation by acting as a cell adhesion molecule for
platelets21 and leukocytes.22 It participates in platelet aggregation via
specific inducible receptors (glycoprotein IIb–IIIa, Gp IIb–IIIa)23,24
on the surface of platelets.25 Platelets have also shown an increase in
adhesion on fibrinogen-immobilized surfaces.26,27 A fluorescent assay
takes advantage of a fluorescein-labeled goat anti-fibrinogen IgG
(polyclonal) antibody to quantitatively detect surface-adsorbed
fibrinogen by fluorescence measurements. Using such an assay, the
Fig. 3 The evaluation of NO generation catalyzed by the same sized
films (area ¼ 0.23 cm2) of (a) TPU-PEG–cyclen–CuII (1) (thickness ¼
30 mm, Cu content ¼ 0.47 wt%) and (b) polymer 2 (0.11 wt% Cu) in a
solution of 0.5 mM CySNO/CySH in 10 mM PBS buffer (pH 7.4, 3 mM
EDTA) before and after contacting with whole sheep blood or platelet-
rich sheep plasma at 4 ꢃC for 24 h via a chemiluminescence NOA.
18786 | J. Mater. Chem., 2012, 22, 18784–18787
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Cu(II)–cyclen modified polyurethane (1) does not exhibit any signif-
icant increase in fibrinogen adsorption compared to that of the
unmodified TPU (P > 0.1, see ESI, Fig. S12†). This result suggests
that the new derivatization chemistry to tether the Cu2+–cyclen
catalyst to the TPU material does not alter the fundamental protein
adsorption properties of the material compared to those of the base
polymer. Hence, with the added NO generation activity, the new
polymer could further enhance thromboresistance of this biomedical
grade polyurethane.
for carrying out the preliminary rabbit ECC experiments with the
new polymer described in this work.
Notes and references
1 D. Giustarini, A. Milzani, R. Colombo, I. Dalle-Donne and R. Rossi,
Trends Pharmacol. Sci., 2004, 25, 311.
2 P. G. Wang, M. Xian, X. Tang, X. Wu, Z. Wen, T. Cai and
A. J. Janczuk, Chem. Rev., 2002, 102, 1091.
3 W. Cha and M. E. Meyerhoff, Biomaterials, 2007, 28, 19.
4 J. Yang, J. L. Welby and M. E. Meyerhoff, Langmuir, 2008, 24,
10265.
5 S. Hwang and M. E. Meyerhoff, J. Mater. Chem., 2007, 17,
1462.
Conclusions
6 T. C. Major, D. O. Brant, C. P. Burney, K. A. Amoako,
G. M. Annich, M. E. Meyerhoff, H. Handa and R. H. Bartlett,
Biomaterials, 2011, 32, 5957.
In conclusion, a new NO generating polymer, Cu(II)–cyclen–PEG-
TPU (1), was successfully prepared. This material exhibits greatly
improved catalytic efficiency for generating NO from RSNO species
compared to those of analogous materials reported previously. This
improvement is achieved by optimizing the synthesis and tethering a
more active catalyst to a commercial PU structure. Based on the
pendant isocyanate sites created on the isocyanated polymer (TPU-
NCO) and the final copper content, a mean yield of 50% for each step
was obtained during the entire synthesis of the polymer. The new
polymer is able to generate 6–10 ꢁ 10ꢀ10 mol cmꢀ2 minꢀ1 surface
flux of NO in the presence of 5 mM GSNO and 16 ꢁ 10ꢀ10 mol cmꢀ2
minꢀ1 NO flux from 0.5 mM CySNO in PBS buffer (10 mM, pH 7.4).
No evidence of significant leaching of copper ions is observed.
Exposure to whole blood or plasma reduces the rate of NO
7 A. P. Dicks, H. R. Swift, D. L. H. Williams, A. R. Butler, H. H. Al-
Sa’doni and B. G. Cox, J. Chem. Soc., Perkin Trans. 2, 1996, 481.
8 S. Hwang and M. E. Meyerhoff, Biomaterials, 2008, 29, 2443.
9 S. C. Puiu, Z. Zhou, C. C. White, L. J. Neubauer, Z. Zhang,
L. E. Lange, J. A. Mansfield, M. E. Meyerhoff and
M. M. Reynolds, J. Biomed. Mater. Res., Part B, 2009, 91B, 203.
10 L. Wang, S. Wang and J. z. Bei, Polym. Adv. Technol., 2004, 15, 617.
11 Z. Zhou and M. E. Meyerhoff, Biomaterials, 2005, 26, 6506.
12 K. D. Park, A. Z. Piao, H. Jacobs, T. Okano and S. W. Kim,
J. Polym. Sci., Part A: Polym. Chem., 1991, 29, 1725.
13 M. C. Styka, R. C. Smierciak, E. L. Blinn, R. E. DeSimone and
J. V. Passariello, Inorg. Chem., 1978, 17, 82.
14 S. C. Askew, D. J. Barnett, J. McAninly and D. L. H. Williams,
J. Chem. Soc., Perkin Trans. 2, 1995, 741.
15 D. L. H. Williams, Acc. Chem. Res., 1999, 32, 869.
16 M. W. Vaughn, L. Kuo and J. C. Liao, Am. J. Physiol., 1998, 274,
H1705.
17 M. W. Vaughn, L. Kuo and J. C. Liao, Am. J. Physiol., 1998, 274,
H2163.
18 D. Giustarini, A. Milzani, R. Colombo, I. Dalle-Donne and R. Rossi,
Clin. Chim. Acta, 2003, 330, 85.
19 M. B. Gorbet and M. V. Sefton, Biomaterials, 2004, 25, 5681.
20 T. C. Major, D. O. Brant, M. M. Reynolds, R. H. Bartlett,
M. E. Meyerhoff, H. Handa and G. M. Annich, Biomaterials, 2010,
31, 2736.
generation, but physiological fluxes still remain (1–5
ꢁ
10ꢀ10 mol cmꢀ2 minꢀ1). This value is much higher than that which
can be obtained from an earlier polymer (2) that utilizes different
linking chemistry. The new polymer reported here is a very attractive
material for future studies to assess the improved thromboresistance
achieved with and without supplementation of exogenous RSNOs to
the bloodstream of the animals, and potentially, in humans.
21 D. A. Cheresh, S. A. Berliner, V. Vicente and Z. M. Ruggeri, Cell,
1989, 58, 945.
Acknowledgements
22 L. Tang and J. W. Eaton, J. Exp. Med., 1993, 178, 2147.
23 J. S. Bennett and G. Vilaire, J. Clin. Invest., 1979, 64, 1393.
24 G. A. Marguerie, T. S. Edgington and E. F. Plow, J. Biol. Chem.,
1980, 255, 154.
25 G. A. Marguerie, N. Thomas-Maison, M. H. Ginsberg and
E. F. Plow, Eur. J. Biochem., 1984, 139, 5.
26 B. Savage and Z. M. Ruggeri, J. Biol. Chem., 1991, 266, 11227.
27 T. Zaidi, L. McIntire, D. Farrell and P. Thiagarajan, Blood, 1996, 88,
2967.
This work is financially supported by the National Institutes of
Health fund (Program no. EB-004527). The authors acknowledge
Joshua Pohlmann in the ECMO research lab at the University of
Michigan Medical School, for providing the fresh sheep blood, and
Dr Ted Huston in the Department of Geology at the University of
Michigan, for conducting the copper analysis using ICP-HRMS. In
addition, the authors thank Dr Terry Major in the ECMO laboratory
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 18784–18787 | 18787