Organoditelluride-tethered polymers that spontaneously generate nitric oxide when in contact with fresh blood

Article abstract of DOI:10.1039/b715523a

An organoditelluride (RTeTeR) species (5,5′-ditelluro-2,2′- dithiophenecarboxylic acid, DTDTCA) that catalytically decomposes endogenous S-nitrosothiols (RSNOs) to nitric oxide (NO) in the presence of free thiols (RSHs) is used to prepare new types of NO generating polymers (NOGPs). In this work, DTDTCA is covalently linked to poly(allylamine hydrochloride) (PAH) to form a water soluble NOGP (polymer 2). This polymer is further crosslinked within a cellulose membrane to yield an interpenetrating polymer network (IPN) (polymer 3). The catalytic NO generation ability of these NOPGs is clearly observed via chemiluminescence detection of liberated NO. The observed NO fluxes of the IPNs in the presence of given concentrations of RSNOs and reducing thiols can be tuned by adjusting the DTDTCA loading of polymer 2 crosslinked within the cellulose support. Polymer 3 is shown to spontaneously generate NO when in contact with fresh blood via amperometric sensor measurements. Hence, these new types of NOGPs are potentially useful as catalytic layers to create new amperometric RSNO sensors and possibly blood contacting medical devices that exhibit improved hemocompatiblity. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b715523a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Organoditelluride-tethered polymers that spontaneously generate
nitric oxide when in contact with fresh blood†
Sangyeul Hwang and Mark E. Meyerhoff*
Received 9th October 2007, Accepted 14th February 2008
First published as an Advance Article on the web 26th February 2008
DOI: 10.1039/b715523a
0
0
An organoditelluride (RTeTeR) species (5,5 -ditelluro-2,2 -dithiophenecarboxylic acid, DTDTCA)
that catalytically decomposes endogenous S-nitrosothiols (RSNOs) to nitric oxide (NO) in the presence
of free thiols (RSHs) is used to prepare new types of NO generating polymers (NOGPs). In this work,
DTDTCA is covalently linked to poly(allylamine hydrochloride) (PAH) to form a water soluble
NOGP (polymer 2). This polymer is further crosslinked within a cellulose membrane to yield an
interpenetrating polymer network (IPN) (polymer 3). The catalytic NO generation ability of these
NOPGs is clearly observed via chemiluminescence detection of liberated NO. The observed NO fluxes
of the IPNs in the presence of given concentrations of RSNOs and reducing thiols can be tuned by
adjusting the DTDTCA loading of polymer 2 crosslinked within the cellulose support. Polymer 3 is
shown to spontaneously generate NO when in contact with fresh blood via amperometric sensor
measurements. Hence, these new types of NOGPs are potentially useful as catalytic layers to create new
amperometric RSNO sensors and possibly blood contacting medical devices that exhibit improved
hemocompatiblity.
Introduction
release for long periods of time. Although it has been suggested
that very low surface levels of NO (<1 nM) can be effective in
9
inhibiting platelet function, the therapeutic window of NO
The discovery of nitric oxide (NO) production in the human
1
body has greatly expanded related research in many fields,
surface concentrations (ranges between effective and toxic
dosage) for these materials is unknown and possibly variable
depending on the given application.
2,3
including biomaterials.
Numerous investigations on the
physiological roles of NO have elucidated its unique dual func-
tionality as both a cytoprotective and a cytotoxic agent. Among
the many physiological functions of NO, the relaxation of smooth
Nitric oxide generating polymers (NOGPs) have been
suggested to overcome the limitations of NO-releasing
10
4
muscle cells and the inhibition of platelet aggregation are well
materials. NOGPs possess catalytic sites that can decompose
known cytoprotection processes that occur at the endothelium
surface within all blood vessels. At the same time, excess NO
can be toxic, forming high levels of methemoglobin, dramatically
lowering blood pressure (e.g., septic shock), and enhancing its
reaction with superoxide to create peroxynitrite, a powerful
endogenous S-nitrosothiols (RSNOs) in blood, such as S-nitro-
soglutathione (GSNO) and S-nitrosocysteine (CySNO), to NO
in the presence of naturally occurring reducing equivalents
such as free thiols (RSHs; glutathione (GSH), cysteine (CySH),
etc.) or ascorbate. Hence, NOGPs can locally produce enhanced
levels of NO at the polymer/blood interface for prolonged time
periods. To date, to develop NOGPs, the known chemistries of
5
oxidizing and nitrosating reagent that can lead to neurotoxicity.
NO is also involved in killing some types of bacteria and cancer
6
cells, perhaps through generation of peroxynitrite.
11a
11b
copper ions and organodiselenides (RSeSeR) as catalysts
for the decomposition of RSNOs to NO have been exploited
to prepare novel polymeric materials. For example, polymers
In order to exploit the benefits of NO in the biomedical area,
NO-releasing polymeric materials have been developed and
2
studied to prevent fungal infections as well as to enhance the
10b
10c
with covalently linked Cu(II)-cyclen or selenocysteine, have
been reported to generate NO catalytically from various RSNOs
that exist in fresh animal blood. Recently, a new chemistry for
RSNO decomposition using an organoditelluride (RTeTeR) as
3
hemocompatibility of blood contacting biomedical devices such
as catheters, vascular grafts, extracorporeal circuits, and
implanted sensors. However, the ultimate biomedical applica-
tions of NO-releasing polymers may be limited to short-term
applications owing to the difficulty in creating large reservoirs
12
the catalytic species has been reported. It is too early to deter-
mine which NOGP is more useful because they have not been
examined clinically. However, the organochalcogenide catalysts
are likely to have advantages in terms of reduced catalyst
leaching because of the inherent kinetic lability of metal ion–
organic ligand complexes. In addition, aromatic dichalcogenides
7
8
of NO donors (e.g., diazeniumdiolates, S-nitrosothiols, etc.)
within very thin polymeric coatings that must maintain NO
Department of Chemistry, The University of Michigan, 930 North
University Ave., Ann Arbor, MI, 48109-1055, USA. E-mail:
mmeyerho@umich.edu; Fax: +1-734-647-4865; Tel: +1-734-763-5916
1
3
0
are known to be more stable than aliphatic ones. Hence, 5,5 -
0
ditelluro-2,2 -dithiophenecarboxylic acid (DTDTCA, 1; see
†
Electronic supplementary information (ESI) available: Scheme of NOA
Scheme 1) reported as a useful catalyst for NO generation
measurement, additional NOA data, configuration of amperometric
NO/GSNO sensor, and NO/GSNO calibration data for the sensors. See
DOI: 10.1039/b715523a
12
from RSNOs in a previous communication may be advanta-
geous over an aliphatic organodiselenide such as selenocysteine.
1
784 | J. Mater. Chem., 2008, 18, 1784–1791
This journal is ª The Royal Society of Chemistry 2008

View Article Online
unless otherwise noted. Distillation was employed for the purifi-
cation of Et N and THF prior to use. Deionized (DI) water was
3
ꢀ1
prepared by a Milli-Q filter system (18 MU cm : Millipore
Corp., Billerica, MA). Cellulose membranes were purchased
from Spectrum Laboratory Inc. (Rancho Dominguez, CA).
Measurements
FTIR spectra were collected with a Perkin-Elmer spectrum BX
FT-IR system. UV-Vis spectra were recorded by a Perkin-Elmer
Lambda 35 UV/VIS spectrometer. The tellurium (Te) content of
the polymers was determined by inductively coupled plasma-
high resolution mass spectrometry (ICP-HRMS) using a Thermo
Finnigan Element instrument. NO measurements were made
using a nitric oxide analyzer (NOA) (Sievers, model 280). The
current signals of amperometric NO/RSNO sensors were
monitored by a microchemical sensor analyzer (Diamond
General, Ann Arbor, MI).
Synthesis
Polymer 2. An organoditelluride compound (DTDTCA, 1, see
Scheme 1, 82 mg, 0.16 mmol), which was synthesized as
1
2
described previously, was dissolved in a solution of Et
(130 mg, 1.28 mmol), EDC (76 mg, 0.40 mmol), and NHS
50 mg, 0.43 mmol) in THF–water (2 mL–0.1 mL). To a stirred
solution of these reagents was added a solution of PAH (0.3 g,
equivalent to 3.2 mmol of amine groups in the polymer, M
5 kDa) and Et N (130 mg, 1.28 mmol) in DI water (8 mL). After
3
N
Scheme 1 The preparation of interpenetrating polymer network (IPN),
polymer 3, by crosslinking polymer 2 consisting of poly(allylamine
(
0
0
hydrochloride) (PAH) with covalently linked 5,5 -ditelluro-2,2 -
dithiophenecarboxylic acid (DTDTCA, 1) within a cellulose membrane
n
¼
(see Experimental for details).
1
3
the reaction mixture was stirred overnight at RT, the mixture
was filtered with a membrane filter (molecular weight cut-off
(MWCO) ¼ 5 kDa) and washed with DI water several times
using a centrifuge to remove any low molecular weight
compounds. The filter-cake was collected and freeze-dried to
give polymer 2 (light reddish brown color). ICP-HRMS analysis
yielded a Te content of 7.5 wt% in polymer 2, which corresponds
to ca. 0.3 mmol organoditelluride 1 per g of polymer 2 (y ¼ 56%,
see Fig. 1 for IR and UV spectra).
Herein, we employ this tellurium chemistry to synthesize and
characterize novel polymeric materials in which an organodi-
telluride is linked covalently to polymers and these polymers are
subsequently crosslinked within another support polymer to
create NO generating surfaces. Specifically, polymer 2 (poly(allyl-
amine hydrochloride), PAH, modified with organoditelluride 1)
and polymer 3 (crosslinked polymer 2 within a cellulose
membrane) are prepared and characterized (see Scheme 1 and
Experimental for details).
ꢀ1
ꢀ1
IR (KBr) ¼ 3424 cm (COO–H, N–H), 2924, 2864 cm (–C–
ꢀ
1
ꢀ1
ꢀ1
The catalytic NO generation of these new materials (polymers
H), 1635 cm (C–O), 1559 cm (HNCONH), 1457 cm (C–N,
ꢀ
1
2
and 3) from endogenous RSNOs and RSHs at physiological
N–H), 1406 cm (–C–H).
pH is demonstrated. Further, it is shown that the NO generation
rate of polymer 3 can be tuned by varying the amount of
catalytic sites (polymer 2) loaded within the cellulose membrane
matrix. Catalyst leaching from the polymer is examined and
shown to be quite low. Finally, we demonstrate that polymer 3
spontaneously generates NO when in contact with fresh
sheep blood via use of the polymer as an outer coating on an
amperometric NO sensor.
Polymer 3. A cellulose membrane (Spectra/Porꢀ, MWCO ¼
50 kDa) was cut into smaller square films (1 cm ꢁ 1 cm). After
these films were extensively washed with 0.1 M EDTA in DI
water to remove any metal ion impurities, they were separately
put into three different concentrations of polymer 2 solutions
(0.9, 2.7, and 9 wt% in 1 mL of DI water) and shaken overnight
at RT. Each film was then placed on a glass slide. Three different
solutions of polymer 2 (0.2 mL of 0.9, 2.7, and 9 wt% solution,
respectively) containing triethylamine (ca. 10 wt% were placed
onto each film, followed by adding glutaraldehyde solution
Experimental
(
0.1 mL of 5 wt% in DI water). The resultant films were allowed
Materials
to stand for 2 d at RT. Each film was then washed extensively
with DI water, and shaken in a NaBH4 solution (0.1 M in
5 mL DI water) for 1 h at RT. The films were again stirred in
a buffered solution (pH 4.3) overnight at RT to remove any
Glutathione (GSH), cysteine (CySH), N-hydroxysuccinimide
0
(
NHS), and N-(3-dimethylaminopropyl)-N -ethylcarbodiimide
hydrochloride (EDC$HCl) were products of Sigma (St. Louis,
MO). Other reagents from Aldrich and solvents from Fisher
Scientific (Fair Lawn, NJ) were used without further purification
residual NaBH
4
, then washed and stored in PBS buffer (pH
7.4) prior to NOA experiments.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1784–1791 | 1785

View Article Online
General NOA experimental procedures
To a reaction cell (home-made 3 neck amber vial to avoid
ambient light-induced RSNO decomposition) was added 2 mL
of PBS buffer, pH 7.4, containing 0.5 mM EDTA (EDTA is
necessary to prevent any RSNO decomposition to NO by
a tiny amount of metal ion impurities present in PBS buffer
and elsewhere). Typically, one neck was connected to a nitrogen
gas supply (positive pressure) that flows into the reaction
solution through a glass pipette in order to efficiently purge
any generated NO from the reaction mixture. The second neck
was linked to a branched nitrogen flow of the same gas supply,
which passes above the reaction solution to carry all gas phase
species to the NOA. The third neck was connected to the
16
pre-calibrated NOA system (via negative pressure) (see Fig. 1s
in ESI† for the scheme of NOA measurement). Hence, any small
amount of NO generated in the solution flows into the chamber of
the NOA, where NO reacts with ozone to yield a chemilumines-
cence signal, which is converted to NO concentration equivalents
ꢀ
1
(
ppb s ) continuously. A conversion constant for the mole
ꢀ1
ꢀ1
number of NO (mol ppb s ) can be pre-calibrated by injecting
a known amount of NaNO into a solution of excess KI in dilute
2
sulfuric acid solution. Prior to the experiment, each measurement
was also calibrated by an internal two-point calibration (the
nitrogen flow as zero gas and 45 ppm NO gas). Therefore, the
number of moles of NO evolved during any experiment can
Fig. 1 IR (A) and UV (B) spectra of polymer 2, PAH, and DTDTCA
1); (A) The IR spectrum of polymer 2 (a) matches with the absorption
(
ꢀ1
be calculated by multiplying the peak area (ppb s ) by the pre-
bands in DTDTCA (1) (b) (dashed lines 1, 2, and 5 for the aromatic
C–H stretching and bending; 3 for the C–O stretching), PAH (c), as
well as a new band (dashed line 4) indicating the amide bond formation
between PAH and DTDTCA; (B) UV-Vis spectra of polymer 2 (a; 1 mg
per 10 mL in DI water containing ꢂ30 mM organoditelluride 1) shows the
distinctive absorption band (shoulder, l ¼ 306 nm) of DTDTCA (1) (b;
ꢀ1
ꢀ1
calibrated conversion constant (mol ppb s ). The measured
number of moles of NO by the NOA was compared with the
expected amount of NO that should evolve from a given concen-
tration of RSNO solution used in the experiment. Thus, the initial
steady-state NO amounts already present in the absence of the
catalytic species, which are the basal NO values, are subtracted
to obtain the NO levels produced by the presence of the given
catalyst. Once the baseline chemiluminescence signal was stabi-
lized, a given amount of RSH solution was added into the test
solution. Then, a given amount of fresh RSNO solution was
added. After attaining a new baseline of NO (usually slightly
higher than the original baseline due to the NO originating
from RSNO), a given test polymer solution was added into the
mixture until all the RSNO was consumed. In order to demons-
trate catalytic behavior of a given polymer film, a small piece of
the polymer was pierced by a stainless steel needle (which was
proved not to affect any NO generation from the mixture), and
the polymer portion was fully soaked into the mixture until the
observed NO flux reached a steady-state. Then, the film with
the needle was removed from the reaction cell. After the NO
signal returned to the baseline, repeated insertion/removal cycles
of the same film were performed to demonstrate the catalytic
RSNO decomposition.
22 mM in PBS buffer, pH 7.4) in the UV-Vis spectrum after tethering
DTDTCA into PAH (c; 1 mg per 10 mL in DI water).
Sample preparation for analyzing Te content in polymers 2 and 3
A given amount of polymer was dried using a vacuum pump and
completely dissolved in 0.5 mL of concentrated HNO
3
in
a Teflon-sealed glass bottle, and then diluted with DI water to
yield 5 mL of solution. Each solution was quickly filtered with
a syringe filter (0.45 mm PTFE, a National Scientific Company
product), and the filtrate was analyzed by ICP-HRMS for Te
content.
Preparation of RSNOs
CySNO and GSNO solutions were prepared by a modified
14
method previously reported. Ten mM of L-cysteine or L-gluta-
thione solution in DI water containing 110 mM H SO were well
mixed with 10 mM of NaNO in DI water for 5–10 min to yield
2
4
2
the corresponding 5 mM acidic RSNO solution. The reaction
vial was completely wrapped with aluminium foil to minimize
Te leaching experimemts
15
light-induced decomposition. Because a tiny amount of NO
below pmol) that naturally decomposes from the original
An H film of polymer 3 was fully immersed in a solution of
10 mM GSH/GSNO solution in 5 mL of PBS buffer, pH 7.4,
for 1 d at RT. A small amount of sample (30 mL) was then with-
drawn from the mixture, and a fresh solution of GSH and GSNO
was added into the mixture (the total volume was maintained).
The processes of sampling and injecting the fresh GSH/GSNO
(
RSNO can be detected by the NOA, every RSNO solution was
freshly prepared from corresponding stock solutions (10 mM
of NaNO
2
and acidified GSH) for each experiment. The stock
solutions were prepared daily as needed.
1
786 | J. Mater. Chem., 2008, 18, 1784–1791
This journal is ª The Royal Society of Chemistry 2008

View Article Online
solutions were repeated daily for 15 d. Te content in all sample
solutions was determined by ICP-HRMS.
low molecular weight RSNOs and RSHs can diffuse into the
porous cellulose structure, any NO generation is likely to occur
throughout the film. The polymer film was then treated with
Amperometric NO detection in blood
4
NaBH solution for a short time (1 h) to reduce the imines to
amines and form a more stable covalent linkage. This process
may also be helpful in removing any non-covalently linked
Fresh sheep blood (obtained from the Extracorporeal
Membrane Oxygenation (ECMO) laboratory at the University
of Michigan Medical School) was heparinized (to achieve
halves of the DTDTCA structure by reducing any ditelluride
ꢀ
(
RTeTeR) linkages to tellurolate (RTe ) species, in which the
ꢀ
>
250 s of activated clotting time post-heparin) and protected
small molecular RTe species not linked to the polymer are
ꢀ
from ambient light by entirely wrapping with aluminium foil.
ꢃ
washed out, but the covalently bonded RTe species stay in
The blood sample was kept at 35 C before and during the
the polymer and are reoxidized to RTeTeR species by oxygen
12,18
experiments. Two amperometric sensors (a control ampero-
metric NO sensor mounted with a cellulose film and a RSNO
sensor mounted with the polymer 3, H film) were assembled
under ambient conditions.
A longer treatment time with
NaBH4 (at least overnight) is required to fully reduce all
RTeTeR species within such a hydrogel type of the polymer
and wash out all of the small molecule tellurolate species from
17
according to reported methods (see Fig. 4s in ESI†), and pre-
calibrated with respect to their intrinsic NO responses in the
4
this film. However, such conditions with excess NaBH and
ꢃ
stirred PBS buffer (pH 7.4) at 35 C (see Fig. 5s (A) in ESI†)
high pH also cause the degradation of the cellulose membrane
17
using a standard stock solution of NO. Both the NO and
RSNO sensors were concurrently immersed into a new PBS
buffer (70 mL) in the presence of 50 uM GSH (the final concen-
tration after adding a blood sample later) prior to use, and then
the amperometric signals from both sensors were monitored
after adding the fresh sheep blood (30 mL) into the solution.
Finally, a known amount of fresh GSNO was added into the
same blood sample to obtain the sensitivity of the sensor toward
GSNO in the presence of the blood (see Fig. 5s (B) in ESI†).
Current signals of two sensors were converted into the NO
and/or GSNO equivalent concentrations based on their pre
and/or post-calibrations.
via reduction/hydrolysis reactions. Therefore, the short-term
4
treatment with NaBH carried out in this work is not likely
adequate to efficiently eliminate all partially linked DTDTCA
species, Hence, any remaining non-covelently bound halves of
DTDTCA may slowly leach out from polymer 3 during the
experiments (see below).
By varying the concentration of polymer 2 (0.9 wt% (low; L):
2
.7 wt% (medium; M): 9.0 wt% (high; H) ¼ 1 : 3 : 10) when
loaded in the same size cellulose membrane, the resulting
polymer 3 can possess different densities of polymer 2 as deter-
mined by the amount of tellurium (Te) found within the resultant
IPN films (see Table 1). The analyzed ratios of Te within samples
of polymer 3 suggest that the yield of the loading process is
slightly decreased at higher loading concentrations of polymer
Results and discussion
2
, which is reasonable given the limited pore size and total
Synthesis and characterization
internal volume of the cellulose membrane matrix.
The organoditelluride species, DTDTCA (1), was prepared by
The images of the intersected parts of the IPN films (blank, L,
M, and H) as captured by scanning electron microscopy (SEM)
indicate that polymer 2 density is scattered (as the observed
brighter islands in the films; see Fig. 2), while the blank cellulose
12
a previously reported method and was then covalently coupled
to poly(allylamine hydrochloride) (PAH, M
¼ 15 000) via
carbodiimide/N-hydroxysuccinimide mediated coupling
n
a
chemistry (see synthesis part in Experimental for the details),
resulting in polymer 2. This polymer was found to possess
7
.5 wt% of tellurium (Te) as determined by inductively coupled
Table 1 Comparison of three different polymer 3 preparations obtained
by altering the amounts of polymer 2 (0.9 wt% (L): 2.7 wt% (M): 9.0 wt%
(H) ¼ 1 : 3 : 10) loaded in the cellulose membrane (size ¼ 1 cm ꢁ 1 cm ꢁ
plasma-high resolution mass spectrometry (ICP-HRMS). Efforts
to prepare polymer 2 with a greater degree of DTDTCA attach-
ment results in polymers that were insoluble, likely due to excessive
ditelluride links between PAH chains. As shown Fig. 1(A),
polymer 2 exhibits the characteristic IR bands of DTDTCA such
5
0 mm)
Polymer 3
ꢀ1
ꢀ1
ꢀ1
as at 2924 cm (–C–H), 2864 cm (–C–H), 1635 cm (C–O),
L
1
M
3
H
ꢀ1
and 1406 cm (–C–H) as well as an additional band from the
Wt% of polymer 2 when
loaded in the cellulose
membrane ¼ theoretical
Te ratio
10
ꢀ1
newly formed amide bond, 1559 cm (–NHCO–), suggesting the
incorporation of DTDTCA into PAH. This polymer also displays
the characteristic color (light reddish brown) and UV-Vis
12
Analyzed Te ratio
in polymer 3/mg
Density of DTDTCA
1 (2.74)
1.1
3 (7.95)
3.2
8 (21.75)
8.7
absorption band (l ¼ 306 nm) of DTDTCA (1) (see Fig. 1(B)).
a
For the preparation of polymer 3, an interpenetrating polymer
network (IPN), polymer 2 was crosslinked using glutaraldehyde
within a cellulose membrane (MWCO ¼ 50 kDa, size ¼ 1 cm ꢁ
3
1/mg cm in polymer 3
b
3
Density of polymer 2/mg cm
in polymer 3
7.3
21.2
58.0
c
1
cm, thickness ꢂ 50 mm as a swollen state in water, see Experi-
a
b
mental for the details). To maximize the catalytic NO generation
by polymer 3, polymer 2 with the covalently linked organo-
tellurium compound was crosslinked within/on a small piece of
cellulose membrane (the bulk phase as well as the surface). Since
Quantitative amount of Te was obtained by ICP-HRMS. Based on
1
the molecular weight of DTDTCA 1(509 g mol ) consisting of 2
ꢀ
3
c
equivalents of Te, and the volume of polymer 3 (0.005 cm ). Based on
.5 wt% of Te in polymer 2.
7
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1784–1791 | 1787

View Article Online
Fig. 3 The profiles of RSNO decomposition mediated by a catalytic
amount of polymer (PAH or polymer 2) from a given solution (RSH/
RSNO or nitrite solution in PBS buffer, pH 7.4, containing 0.5 mM
EDTA), where the amount of detected NO at a given time was recorded
via the chemiluminescence NO analyzer (NOA), then converted into
percentage of the theoretical amount of NO that would be stored in
a given RSNO or nitrite solution; (a) polymer 2 (7. 5 wt% DTDTCA 1
in polymer 2) corresponding to 0.75 mM of the catalyst, using 10 mM
CySNO and 50 mM CySH as sample solution; (b) the same amount of
the polymer 2 added to solution containing 10 mM GSNO and 50 mM
GSH; (c) the same weight amount of PAH added to a solution of
Fig. 2 Scanning electron microscopy (SEM) images of the intersected
part of each dried sample: (A) blank, (B) L, (C) M, and (D) H film after
gold coating; the brighter islands in (B), (C), and (D) are polymer 2 which
is located in the bulk phase as well as the surface of the cellulose
membrane in proportion to its amount loaded in the film. The scale
bars are 2 mm in (A) and 5 mm in (B), (C), and (D).
1
0 mM GSNO and 50 mM GSH; (d) the same amount of polymer 2 added
to a solution of 1 mM nitrite and 50 mM GSH in PBS.
membrane has a clean surface. The brighter regions of the IPN
films were proven to possess Te (ca. 8%) as measured by energy
dispersed X-ray analysis (EDX, data not shown), where the Te
content is very close to the value of the given film (7.5%)
determined by ICP-HRMS. The sizes of polymer 2 regions
vary up to several mm, and they are found to be dispersed within
the interior as well as exterior of the IPN polymer 3.
in polymer 2 is the catalytic species responsible for NO genera-
tion (see Fig. 3).
When the IPN polymer 3 (H film) was tested for its ability to
generate NO from an RSNO/RSH solution over an extended
time period, the number of moles of NO detected by the NOA
was nearly the theoretical amount (90–100%) of NO that can
be released from a given starting concentration of RSNO
solution (see Experimental for the method and Fig. 2s in
ESI†). Fig. 4 illustrates clearly how various polymer 3 prepara-
tions (for film formulations L, M, and H) function catalytically
to generate varying NO fluxes based on the Te content of the
IPN. In the presence and absence of a blank film (cellulose
membrane) in a GSNO/GSH solution, the baseline level of NO
observed from the GSNO is not significantly changed, as shown
Fig. 4(A). Minor changes in these baseline NO levels are likely
due to the intrinsic instability of RSNOs (light and thermal
Taken together, the ICP-HRMS data and SEM images
support the fact that polymer 2, with its appended organoditel-
luride catalytic species, was immobilized within the cellulose
membrane to form an IPN material.
Catalytic function to liberate NO from S-nitrosothiols
The catalytic activity of water-soluble polymer 2 for the NO
liberation from RSNOs was tested using a chemiluminescence
NO analyzer (NOA) to quantify the amount of NO gas evolved
from the reaction mixture (see Experimental for details). When
a catalytic amount of polymer 2 in DI water (corresponding to
ca. 0.75 mM of DTDTCA, 1) was added into a mixture of
GSH (50 mM) and GSNO (10 mM) in the PBS buffer, pH 7.4,
containing 0.5 mM EDTA, NO equivalents corresponding to
the total GSNO concentration in the reaction mixture were
detected via the NOA (see Fig. 3). The results obtained with a
CySNO/CySH substrate solution under similar conditions also
demonstrate the NO generation activity of polymer 2, but with
somewhat faster kinetics. It is important to note that nitrite
15
sensitivity). However, placing a small piece of film L into the
reaction mixture under the same conditions causes the NO signal
to reach a significantly higher steady-state value, and subsequent
withdrawal of film L from the solution induces a decrease in the
observed NO signal to a near baseline value. Such a slight
increase of baseline after film immersion may originate from
the natural decomposition of GSNO or be caused by any
catalytic species leached from the membrane; e.g., any partially
linked DTDTCA species within the polymer, as discussed above.
However, continuing the insertion/removal cycles of film L yields
a similar level of NO signal at steady-state, indicating that this
film catalytically liberates NO from GSNO at physiological pH
(see Fig. 4(B)). Both films M and H display similar catalytic
GSNO decomposition to NO under the same conditions, as
shown Fig. 4(C) and (D).
(
2
NaNO , 1 mM) is not reduced to NO under the same experi-
mental conditions, implying that soluble polymer 2 is a selective
catalyst toward RSNO decomposition whereas copper ion
19
catalysts can reduce both RSNOs and nitrite. As a control
experiment, when the equivalent amount of PAH (50 mL of the
0
Additionally, the maximum NO flux at a steady-state is
gradually enhanced from film L to H (L < M < H), where the
density of polymer 2, i.e., DTDTCA, within the IPN polymer
3 also increases in the same order. Note that the increase in
ꢀ1
.1 mg mL in DI water solution) is added into a new GSNO/
GSH solution, no change in the chemiluminescence baseline
signal is observed, verifying that the organoditelluride species
1
788 | J. Mater. Chem., 2008, 18, 1784–1791
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Fig. 4 The NO generation profiles using a blank (A), film L (B), film M (C), and film H (D) (1 cm ꢁ 1 cm ꢁ 50 mm, polymer 3) from GSNO (50 mM) and
GSH (50 mM) in 2 mL of PBS buffer (10 mM), pH 7.4, containing EDTA (0.5 mM), as monitored via a chemiluminesence NOA. Arrows indicate
addition/removal of given species/films into the test solution.
NO generation does not directly correlate with the tellurium
content of polymer 3. This is likely related to the heterogeneous
nature of the catalytic process, and hence mass transfer of RSNO
and RSH to the surface may become the rate-limiting step, not
the amount of bound catalyst. Nonetheless, the data suggest
that it is possible to adjust the level of NO flux emitted in the
presence of a given amount of RSNO species merely by varying
the density of catalyst immobilized in the IPN film
Catalyst leaching study
Fig. 5 A plot of soaking time vs. the accumulated Te in sample solution
which was withdrawn daily from the solution of 10 mM GSH/GSNO with
To evaluate the leaching of catalytic species from polymer 3,
the fully immersed polymer 3, while a fresh solution of 10 mM GSH/
a soaking experiment using film H was conducted (see Experi-
GSNO was added into the original solution each day resulting in an
mental for details). Approximately 10% of Te in the film was
increasing concentration of GSH (see Experimental for details).
found in the GSH/GSNO solution after 1 d, as determined by
the ICP-HRMS analysis. Thereafter, the total amount of Te
leached from the film slowly increased to up to 20% of the
original amount, and then leveled off after 15 d (see Fig. 5).
This result can be explained by considering the proposed
mechanism of the organoditelluride-mediated RSNO decompo-
sition in which RSH added into the reaction mixture reduces
the organoditelluride species to break the ditelluride bond.
Therefore, any residual non-covalently linked halves of
DTDTCA species in which only one carboxylic acid group is
attached to the polymer backbone (as discussed above) can be
freed from the polymer in the presence of GSH/GSNO solution
to dissolve in the aqueous phase. Further, in the same manner,
even when both carboxylic groups of the DTDTCA species are
linked covalently to the PAH, the GSH-mediated disconnection
of ditelluride bonds can cause the molecular weight of polymer
polymers and/or relatively low molecular weight polymers
(<50 kDa, MWCO of cellulose membrane) which have good
water solubility may slowly dissolve into the GSH/GSNO solu-
tion phase until all of such polymeric species are removed from
the IPN (polymer 3). Indeed, the rate of Te leaching was opposite
to the concentration changes of GSH/GSNO in solution
(because of continuous addition of RSH/RSNO solution into
the original solution), excluding the possibility of direct and
continuous decomposition of the organoditelluride catalyst in
this environment. This reinforces the proposed mechanism of
organoditelluride-mediated RSNO denitrosation described in
12
a previous report. The result of this study suggests that effective
catalytic NO production activity of polymer 3 can be achieved
for extended periods of time, indicating the potential usefulness
of such polymers for biomedical applications.
2
to partially decrease; hence, some physically crosslinked
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1784–1791 | 1789

View Article Online
materials for creating a new NO generating polymer (NOGP).
NO generating catalytic activities were demonstrated using
biological RSNOs/RSHs in solution at physiological pH via
the NOA measurements. Especially exciting is the ability to the
tune NO fluxes of polymer 3 materials by varying the amount
of polymer 2 employed during the crosslinking reaction in the
cellulose membrane. The Te leaching experiments conducted
with polymer 3 indicate that most of the catalyst remains in
the polymer phase even after soaking in high concentrations of
fresh GSH/GSNO solution for two weeks, reinforcing the
proposed mechanism for organoditelluride-mediated RSNO
Fig. 6 The amperometric NO measurements in fresh sheep whole blood
12
denitrosation as well as reflecting the potential long-term utility
(30 mL; arrow indicates the time when the blood was added) diluted with
of the new polymers. Finally, spontaneous NO generation in
fresh sheep blood in vitro mediated by polymer 3 was clearly
demonstrated via amperometric NO gas sensor measurements,
suggesting the potential usefulness of the NOGP. Indeed, it
appears that such polymers can generate a surface enhanced level
of NO, which could prevent platelet activation and adhesion and
hence reduce the risk of thrombosis on medical devices possess-
ing such polymers as coatings. For example, polymer 3 can be
applied as the active membrane layer for preparation of hemo-
ꢃ
PBS buffer (70 mL, pH 7.4) at 35 C using two sensors; RSNO (a) and
NO sensor (b), where the amperometric signals were converted to the
NO and/or GSNO equivalent levels (nM) from their calibration curves.
Furthermore, the NO generating capability of polymer 3
(
film L) before/after soaking under the same conditions as the
soaking study (GSH/GSNO solution) for 7 days has been exami-
ned. A decrease in NO generating flux (approximately 21%
reduction) was observed after this period, which correlates well
with the Te leaching study where most of the Te catalyst remains
in polymer 3 (see Fig. 3s in ESI†).
dialysis fibers/membranes as a means to reduce risk of clotting
20
in such extracorporeal procedures.
Obviously, extensive
in vivo studies must be carried out to fully test the effectiveness,
life-time, and any toxicity of these materials for such biomedical
applications. It should be noted, however, that polymer 3 type
films have already been utilized successfully to prepare a new
amperometric RSNO sensor that functions effectively to detect
Spontaneous NO generation in blood
Finally, to assess whether these new polymeric materials (poly-
mer 3) can mediate spontaneous NO generation in fresh whole
blood, an amperometric RSNO sensor was assembled using
polymer 3 (H film) as an external film on the surface of a NO
selective electrochemical sensor prepared by a method described
21
the relative levels of RSNO species in whole blood.
Acknowledgements
17
elsewhere (also see Fig. 4s in ESI†). A control NO sensor was
also fabricated with a cellulose blank film instead of polymer
The authors thank Dr Ted Huston in the Department of
Geology at the University of Michigan for the analysis of Te
using ICP-HRMS, and Nathan Lafayette in the ECMO research
lab at the University of Michigan Medical School for providing
the fresh sheep blood. This research was financially supported by
NIH (EB-004527) and the U.S. Army (W81XWH-05-1-0602).
3
. Each sensor was calibrated for the intrinsic direct ampero-
metric response to NO prior to use (see Fig. 5s (A) in ESI†),
and then simultaneously used to measure NO and RSNO levels,
respectively, in fresh sheep blood diluted with PBS buffer at
ꢃ
3
5
C. As shown in Fig. 6, each sensor exhibits independent
amperometric signals which can be converted to NO equivalents
by a prior NO calibration curve.
References
The difference of NO equivalents between two sensors
indicates the spontaneous denitrosation of endogenous RSNO
species in the blood locally occurs at the outer membrane of
the sensor that has the IPN polymer 3 mounted on it (but not
the control sensor). Thus, this value should be proportional to
the total concentration of all low MW RSNO species in the
blood, which corresponds to ꢂ500 nM of GSNO equivalent
levels (based on GSNO calibration of the RSNO sensor from
a post-calibration curve continuously obtained in the same blood
sample) (see Fig. 5s (B) in ESI†). The ability of polymer 3 to
generate NO when in contact with fresh blood is clearly
demonstrated via these sensor measurements.
1
2
R. M. Palmer, A. G. Ferrige and S. Moncada, Nature, 1987, 327, 524.
B. J. Nablo, T.-Y. Chen and M. H. Schoenfisch, J. Am. Chem. Soc.,
2
001, 123, 9712.
3 L. K. Keefer, Annu. Rev. Pharmacol. Toxicol., 2003, 43, 585.
4
(a) K. Krejcy, L. Schmetterer, J. Kastner, M. Nieszpaurlos,
B. Monitzer, W. Schutz, H. G. Eichler and P. A. Kyrle,
Arterioscler. Thromb. Vasc. Biol., 1995, 15, 2063; (b)
C. M. Samama, M. Diaby, J. L. Fellahi, A. Mdhafar, D. Eyraud,
M. Arock, J. J. Guillosson, P. Coriat and J. J. Rouby,
Anesthesiology, 1995, 83, 56; (c) R. Sarkar and R. C. Webb,
J. Vasc. Res., 1998, 35, 135; (d) D. S. Bredt and S. H. Snyder,
Annu. Rev. Biochem., 1994, 63, 175.
5 (a) V. L. Dawson, T. M. Dawson, E. D. London, D. D. Bredt and
S. H. Snyder, Proc. Natl. Acad. Sci. U. S. A., 1991, 88, 6368; (b)
J. S. Stamler, Cell, 1994, 78, 931.
6
(a) J. B. Jr. Hibbs, R. R. Taintor and Z. Vavrin, Science, 1987, 235,
73; (b) M. A. Marietta, P. S. Yoon, R. Ivengar, C. D. Leaf and
4
Conclusions
J. S. Wishnok, Biochemistry, 1988, 27, 8706; (c) D. J. Stuehr,
S. S. Gross, I. Sukuma, R. Levi and C. F. Nathan, J. Exp. Med.,
In conclusion, water-soluble PAH with appended ditelluride
species and an IPN using this polymer crosslinked within a
cellulose membrane matrix were shown to be useful polymeric
1
989, 169, 1011.
7
P. G. Parzuchowski, M. C. Frost and M. E. Meyerhoff, J. Am. Chem.
Soc., 2002, 124, 12182.
1
790 | J. Mater. Chem., 2008, 18, 1784–1791
This journal is ª The Royal Society of Chemistry 2008

View Article Online
8
9
M. C. Frost and M. E. Meyerhoff, J. Am. Chem. Soc., 2004, 126,
348.
(a) M. W. Vaughn, L. Kuo and J. C. Liao, Am. J. Physiol. Heart Circ.
Physiol., 1998, 274, H2163; (b) C. Wang, L. J. Trudel, G. N. Wogan
and W. M. Deen, Chem. Res. Toxicol., 2003, 16, 1004.
14 J. S. Stamler and J. Loscalzo, Anal. Chem., 1992, 64, 779.
1
15 (a) J. S. Stamler and E. J. Toone, Curr. Opin. Chem. Biol., 2002, 6,
779; (b) K. Szacilowski and Z. Stasicka, Prog. React. Kinet. Mech.,
2001, 26, 1.
16 M. Nishimura, H. Imanaka, A. Uchiyama, C. Tashiro, D. Hess and
R. M. Kacmarek, J. Clin. Monit. Comput., 1997, 13, 241.
17 (a) Y. Lee, B. K. Oh and M. E. Meyerhoff, Anal. Chem., 2004, 76,
536; (b) W. Cha, Y. Lee, B. K. Oh and M. E. Meyerhoff, Anal.
Chem., 2005, 77, 3516; (c) W. Cha and M. E. Meyerhoff, Langmuir,
2006, 26, 10830.
1
0 (a) B. K. Oh and M. E. Meyerhoff, J. Am. Chem. Soc., 2003, 125,
552; (b) S. Hwang, W. Cha and M. E. Meyerhoff, Angew. Chem.,
9
Int. Ed., 2006, 45, 2745; (c) W. Cha and M. E. Meyerhoff,
Biomaterials, 2007, 28, 19; (d) X. B. Duan and R. S. Lewis,
Biomaterials, 2002, 23, 1197.
1
1 (a) A. P. Dicks, H. R. Swift, D. L. H. Williams, A. R. Butler,
H. H. Al-Sadoni and B. G. Cox, J. Chem. Soc., Perkin Trans. 2,
18 L. Engman and M. P. Cava, Organometallics, 1982, 1, 470.
19 J. A. Halfen, S. Mahapatra, E. C. Wilkinson, A. J. Gengenbach,
V. G. Young, Jr., L. Que, Jr. and W. B. Tolman, J. Am. Chem.
Soc., 1996, 118, 763.
20 Z. R. Zhou, G. M. Annich, Y. D. Wu and M. E. Meyerhoff,
Biomacromolecules, 2006, 7, 2565.
1996, 481; (b) Y. Hou, Z. Guo, J. Li and P. G. Wang, Biochem.
Biophys. Res. Commun., 1996, 228, 88.
2 S. Hwang and M. E. Meyerhoff, J. Mater. Chem., 2007,
1
1
4
5, 1462.
3 B. Nygard, J. Ludvik and S. Wendsjo, Electrochim. Acta, 1996, 41,
655.
21 S. Hwang, W. Cha and M. E. Meyerhoff, Electroanalysis, 2008,
20, 270.
1
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1784–1791 | 1791