Thermal and photochemical nitric oxide release from S-nitrosothiols incorporated in Pluronic F127 gel: potential uses for local and controlled nitric oxide release

Biomaterials

Thermal and photochemical nitric oxide release from S-nitrosothiolsincorporated in Pluronic F127 gel: potential uses for local and

controlled nitric oxide release

S!ılvia Mika Shishido, Amedea Barozzi Seabra, Watson Loh,Marcelo Ganzarolli de Oliveira*

Instituto de Qu!ımica, Universidade Estadual de Campinas (UNICAMP), CP 6154, Campinas, SP 13083-970, Brazil

Received 23 September 2002; accepted 9 March 2003

Abstract

The local delivery of nitric oxide (nitrogen monoxide, NO) by thermal or photochemical means to target cells or organs has a

great potential in several biomedical applications, especially if the NO donors are incorporated into non-toxic viscous matrices. In

this work, we have shown that the NO donors S-nitrosoglutathione (GSNO) and S-nitroso-N-acetylcysteine (SNAC) can be

incorporated into F127 hydrogels, from where NO can be released thermally or photochemically (with lirr > 480 nm). High

sensitivity differential scanning calorimetry (HSDSC) and a new spectrophotometric method, were used to characterize the

micellization and the reversal thermal gelation processes of the F127 hydrogels containing NO donors, and to modulate the gelation

temperatures to the range 29–32�C. Spectral monitoring of the S–NO bond cleavage showed that the initial rates of thermal and

photochemical NO release (ranging from 2 to 45 mmol l�1 min�1) are decreased in the hydrogel matrices, relative to those obtained in

aqueous solutions. This stabilization effect was assigned to a cage recombination mechanism and offers an additional advantage for

the storage and handling of S-nitrosothiols. These results indicate that F127 hydrogels might be used for the thermal and

photochemical delivery of NO from S-nitrosothiols to target areas in biomedical applications.

r 2003 Elsevier Science Ltd. All rights reserved.

Keywords: Nitric oxide release; S-nitrosothiols; PEO–PPO–PEO copolymers; Hydrogels; Photolysis

1. Introduction

Nitric oxide (NO) is an endogenous molecule thatplays an important role in the regulation of vasomotortone, platelet interaction with the vessel wall, smoothmuscle cell replication, and immune response [1–5]. It isalso implicated in numerous neuronal and non-neuronalfunctions in the central nervous system and peripherycaltissues [6]. In addition, NO has also been implicated inthe pathology of many inflammatory diseases and alarge number of pathological conditions such as cancer,diabetes, and neurodegenerative diseases. The biologicaleffects of NO have been shown to be dependent on itssite and source of production, as well as on itsconcentration [7].

Over the last years, there has been a great interest inagents that release NO by a controlled manner in livingsystems [8]. The main aim of this kind of research hasbeen the development of NO-delivery systems that couldbe used in target applications and the modulation of thekinetics of NO release [9]. Some studies suggest that thecysteine thiols (RSHs) of bioactive peptides such asglutathione (GSH), the most abundant non-protein thiolfound in vivo, and N-acetylcysteine (NAC), an endo-genous product of cysteine, can be NO carriers in tissues[10–12]. Thiols are believed to be intrinsically involvedin the metabolism and mobilization of NO [13] andGSH is especially likely to form S-nitrosoglutathione(GSNO) and NAC to form S-nitroso-N-acetylcysteine(SNAC), in vivo. GSNO and SNAC belong to a class ofcompounds named S-nitrosothiols (RSNOs), which areconsidered to be NO donors and to act as reservoirs ofNO, in vivo due to their greater stability in comparisonwith free NO (Fig. 1) [14–17]. RSNOs exhibit NO-like

ARTICLE IN PRESS

*Corresponding author. Tel.: +55-19-3788-3132; fax: +55-19-3788-

3023.

E-mail address: [email protected] (M. Ganzarolli de Oliveira).

0142-9612/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved.

doi:10.1016/S0142-9612(03)00153-4

activities such as inhibition of platelet activation,reduction of thrombosis and vasodilation [18–24] andare more likely to be the endothelium derived relaxingfactor (EDRF) than free NO [13,25,26]. They are alsoknown to decompose thermally [9,16] releasing NO andyielding the dimer RSSR according to

2RSNO ¼ RSSR þ 2NO: ð1Þ

Irradiation of RSNOs on their UV/VIS absorptionbands causes the photolysis of the S–NO bond,increasing the rates of NO release [8,11,27]. This is avery interesting property, since compounds that releaseNO photolytically could be used as photochemother-apeutic agents, provided that their thermal decomposi-tion is negligible [11,21,28].

Due to the thermal instability of most RSNOs, thereis a great interest in the preparation of non-toxicformulations that increase their thermal stability andthat are, at the same time, capable of releasingNO under irradiation with visible light. In thiscase, NO could be released in situ onto specifictissues under irradiation in controlled periods oftime acting, e.g., as a cytotoxic species againstpathogenic agents or possibly as a local vasodilator[29]. An l-arginine-dependent-NO-mediated mechanismhas already been demonstrated to be involved in themicrobicidal activity of interferon-gamma-treatedmacrophages against Trypanosoma-cruzi [30] and inthe destruction of intra-hepatic malaria parasites [31]. Ithas also been reported that NO from l-argininemediates the macrophage killing of leishmania parasitesin vivo [32,33] and that a gel containing an NO donor,increases the efficiency in the elimination of cutaneousleishmaniosis [34].

With the aim of developing a formulation based on ahydrogel capable of releasing NO from RSNOs, we haveused the triblock copolymer F127 as a matrix. Thiscopolymer is a commercially available non-ionic poly-

meric surfactant that possesses the symmetrical structurepoly(ethylene oxide)99–poly(propylene oxide)65–poly(ethyl-ene oxide)99 [35]. Due to two dissimilar moieties,hydrophilic poly(ethylene oxide) (PEO) and hydrophobicpoly(propylene oxide) (PPO) blocks, this copolymer issurface active and exhibits the unique property ofamphiphilicity, undergoing self-assembling to form mi-celles and a lyotropic liquid crystalline gel phase [35,36].Hydrogels of F127 display low toxicity [37] and are goodcandidates for drug-delivery systems. In addition, aqueoussolutions of F127 display reversible thermal gelation(gelation occurs with the increase in temperature and canbe reverted by decreasing the temperature) [35,38]. Such aproperty allows concentrated solutions of F127(B20 wt%) to be fluid at the refrigerator temperature(B5�C) and to be rigid semi-solid clear gels at bodytemperatures ( B37�C). With the approval of the Foodand Drug Administration (FDA, USA) [39], F127 hasalready been used extensively in the pharmaceuticalindustry as a vehicle for the controlled release of drugsand also as a coating for wound and protection againstmicrobes [40,41]. Medical uses of F127 as a drug carrierhave been reported in topical applications of anti-canceragents, for the covering of or burn wounds, in rectal,ophthalmic, parenteral, and nasal administration, as wellas in subcutaneous administrations [42,43]. In addition toallowing a controlled release of several drugs, F127 gels arealso widely used due to their characteristic tissue-likecompatibility, enhancement of protein stability, easymanipulation and drug permeability [39,41,44]. As thesegels are transparent, they also permit the irradiation of theincorporated drugs with visible light for phototherapeuticpurposes.

In the present work, we have incorporated GSNO andSNAC in F127 hydrogels. The endothermic heatsassociated with the micellization processes were char-acterized through high sensitivity scanning calorimetric(HSDSC) measurements and a new spectrophotometricmethod was employed to characterize the gelationprocesses. These techniques have shown that thepresence of RSNOs leads to an increase in the gelationtemperature of F127, which can be assigned to a salting-in effect. The kinetics of thermal and photochemical NOrelease from GSNO and SNAC in the gel matrices weremonitored and compared to those obtained in aqueoussolutions. Our results have shown that F127 hydrogelsreduce the rates of both thermal and photochemical NOrelease from GSNO and SNAC, compared to thosefound in aqueous solutions. This stabilization effect wasassigned to a cage recombination mechanism, which isfavored by the reduced diffusion of the RS and NOradicals in the micellar media within the gel structure.Such results indicate that RSNOs-containing-F127-hydrogels have a great potential to be used in medicaland pharmaceutical applications for the local delivery ofNO to target areas or for stabilizing and controlling the

ARTICLE IN PRESS

OHSON

O

NH

O

HO NHNH

O

OH

O

OS

NO

NH2

O

OHHS

O

NH

O

HO NHNH

O

OH

O

OSH

NH2

O

(A)

(C) (D)

(B)

Fig. 1. Structures of (a) N-acetyl-l-cysteine (NAC), (b) Glutathione

(GSH), (c) S-nitroso-N-acetylcysteine (SNAC), and (d) S-nitrosoglu-

tathione (GSNO).

S.M. Shishido et al. / Biomaterials 24 (2003) 3543-3553 3544

release of NO from NO donors in subcutaneous orintra-peritoneal applications.

2. Materials and methods

2.1. Materials

Pluronic F127 (MWB120,000 g mol�1, EO99–PO65–EO99) (ICI, USA), Glutathione (g-Glu-Cys-Glu, GSH),N-acetyl-l-cysteine (Sigma Chemical, St. Louis, MO),HCl, sodium nitrite (NaNO2), acetone and ethyl ether(Sigma-Aldrich Qu!ımica, Brazil), were used as received.Gaseous NO was obtained from a gas cylinder (WhiteMartins, SP, Brazil). Synthetic air (N2/O2, 79/21 v/v,H2O o2 ppm, THC, CO+CO2o0.3 ppm) was pur-chased from Air Liquide, SP, Brazil. All the solutionsand hydrogels were prepared using water from aMillipore Milli-Q Gradient filtration system.

2.2. Synthesis of GSNO

GSNO was synthesized at 25�C by reacting equimolar(200 mmol l�1) reduced glutathione with sodium nitritein 0.5 mol l�1 HCl. The final solution was placed in anice bath under stirring for 40 min and precipitated withacetone. The solid obtained was filtered, washed 5 timeswith 5 ml of cold water, 5 times with 5 ml of acetone and3 times with 1 ml of ethyl ether and freeze-dried for 24 h.GSNO was stored in a flask protected from light.

2.3. Synthesis of SNAC

Due to the greater solubility of SNAC in water andthe resulting difficulty in precipitating it with acetone, adifferent S-nitrosation method was employed in thiscase. S-nitrosation of NAC was achieved by bubbling amixture of NO/synthetic air through NAC solutions(61.2 mmol l�1), in a quartz spectrophotometer cuvette.Gas flows were controlled by flowmeters (Aldrich) toallow a mixture of known NO/air ratio. A gas flow-through system with a glass mixer, polyethylene tubesand Teflon connections was used to control the mixtureand delivery of gases to the solution at a pre-adjustedNO:O2 ratio of 5.6:5.6 ml min�1. To follow the course ofthe S-nitrosation reaction, the gaseous mixture wasbubbled through the solutions in short pulses, using asolenoid valve (Cole Palmer) controlled by an electroniccircuit developed for this purpose. The reaction wasfollowed spectrophotometrically at l ¼ 545 nm, whichcorresponds to the maximum of the visible absorptionband of SNAC. This and the UV band at 336 nm wereobserved to increase continuously with the total time ofbubbling until a maximum is reached (Fig. 2a and b). S-nitrosation was carried out until the achievement of thisplateau to ensure a complete reaction, avoiding an

excess of the nitrosating reactant. As a control, thisreaction was performed and monitored for four differentstarting concentrations of NAC and the maximumabsorbances achieved at the beginning of the plateauwere plotted against the starting NAC concentrations inthe range of 20–40 mmol l�1. A good linearity wasobtained (not shown). Based on this control it wasassumed that the concentration of SNAC solutions wereequal to the starting concentrations of NAC solutions.

2.4. Incorporation of GSNO and SNAC into Pluronic

F127 hydrogels

F127 hydrogels containing GSNO and SNAC wereprepared by the ‘‘cold’’ method [45].

In the case of hydrogels containing GSNO, 2.2 g ofF127 were added to 6.0 ml of cold water (5�C) withgentle mixing and the solution was transferred to therefrigerator (o10�C). After completing the dissolutionprocess overnight, 2.5 ml of a GSNO solution59.5 mmol l�1 were mixed, at 5�C, to the F127 solutionpreviously prepared. This mixture led to the preparationof 10.7 g of hydrogel, containing 20.5 wt% of F127 and13.8 mmol of GSNO g�1 of hydrogel. For the character-ization of the gelation temperature and for the kineticmeasurements, 3 ml of this solution were transferred to aquartz cuvette (see below).

In the case of hydrogels containing SNAC, 1.9 g ofF127 were added to 3.0 ml of cold water (5�C) withgentle mixing and the solution was transferred to therefrigerator (o10�C). After completing the dissolutionprocess overnight, 4.2 ml of an SNAC solution61.3 mmol l�1 were mixed, at 5�C, to the F127 solutionpreviously prepared. This mixture led to the preparation

ARTICLE IN PRESS

500 550 600 6500.0

0.2

0.4

0.6

0.8

1.0

1.2

Abs

orba

nce

Wavelength / nm

300 350 400 4500.00.51.01.52.02.5

Abs

orba

nce

Wavelength / nm

0 20 40 60 80 100 1200.0

0.2

0.4

0.6

0.8

1.0

Time / s

Abs

orba

nce

at λ

=54

5 nm

(a) (b)

Fig. 2. (a) Representative spectral changes in the visible range and in

the UV range (inset) obtained during the synthesis of SNAC by the S-

nitrosation of NAC (0.11 mol l�1). Each spectrum was obtained after a

gas (NO/air) pulse through the NAC solution (for details, see the

experimental section). The time intervals between the spectra are 3-6 s.

(b) Kinetic curve monitored at 545 nm for the synthesis of SNAC.

Error bars correspond to the standard errors of the means (SEM) of

duplicates.

S.M. Shishido et al. / Biomaterials 24 (2003) 3543-3553 3545

of 9.1 g of hydrogel, containing 20.5 wt% of F127 and28.4 mmol of SNAC g�1 of hydrogel. For the character-ization of the gelation temperature and for the kineticmeasurements, 3 ml of this solution were transferred to aquartz cuvette (see below).

2.5. Characterization of the micellization and gelation

processes

The temperature-driven micellization and gelationprocesses were characterized by a combination ofcalorimetric and spectrophotometric methods. HSDSCmeasurements were carried out with a high sensitivityDSC VP Model (MicroCal Inc., USA). Samples of0.2 ml of pure aqueous F127 solutions and of theF127 solutions containing NAC, GSH, SNAC andGSNO, were analyzed using a heating rate of 15� h�1 inthe temperature range of 3.5–45�C. Milli-Q waterwas used as a reference in all cases. The enthalpychanges (DH) associated with the observed endothermicpeaks were calculated using the software MicroCal-Origin 5.0.

A diode-array spectrophotometer (Hewlett-Packard,model 8453, USA) with a temperature-controlledsample holder was used to analyze the absorptionchanges observed in the temperature range of 20–45�Cat a heating rate of 5.5� min�1. Absorption versus timecurves for aqueous pure F127 and for the F127 solutionscontaining NAC, GSH, SNAC and GSNO weremonitored at l ¼ 700 nm. A 1 cm quartz cuvettereferenced against air was used in all cases. A holdingtime of 10 min was used before each absorbancemeasurement.

2.6. Kinetic measurements of GSNO and SNAC

decomposition with NO release

Spectral changes of GSNO and SNAC in aqueoussolutions and in F127 hydrogels in the range of 220–1100 nm were monitored in the dark and underirradiation conditions. Kinetic curves of thermal andphotochemical decomposition of GSNO and SNAC inwater and in the hydrogels were obtained from theabsorption changes at 545 nm. Kinetic data were takenat this wavelength in 10 and 5 min intervals at 37�C.Spectra of GSNO and SNAC solutions in water and inF127 hydrogels were obtained in a 1 cm quartz cuvette,referenced against air. The amount of NO released overtime was calculated directly from the amount of RSNOdecomposed. This calculation was based on the fact thatthe decay of the absorption bands of RSNOs at 338 and545 nm can be associated solely to the homolyticcleavage of the S–N bond with NO release, accordingto Eq. (1) [27]. Thus, the increase in the concentration ofNO released over time ([NO]t), was calculated from thechanges in RSNO concentration ([RSNO]0�[RSNO]t),

according to Beer’s law:

½NO�t ¼ ½RSNO�0 � ½RSNO�t¼ ðA0=eRSNOÞ � ðAt=eRSNOÞ; ð2Þ

where A0 and At are the solution absorbances at 545 nmat the beginning of the monitoring and at time t;respectively, [RSNO]0 and [RSNO]t are the concentra-tions of RSNOs at the beginning of the reaction and attime t; respectively, and eRSNO are the molar absorptioncoefficients of the RSNOs at 545 nm, which werecalculated in the hydrogels as eGSNO ¼ 17:0 mol�1 l cm�1

and eSNAC ¼ 16:4 mol�1 l cm�1.Initial rates (IR) of thermal and photochemical NO

release from the RSNOs were obtained by linear regressionof the slopes of the initial sections (less than 10% ofreaction) of the [NO] versus time plots according to

IR ¼D½NO�Dt

; ð3Þ

where D[NO] and Dt are the changes in NO concentra-tion and the corresponding time intervals, respectively.Each point in the kinetic curves represents the averageof two experiments, with the error bars between theabsorbance values expressed by their standard error ofthe mean (SEM).

2.7. Irradiation of GSNO and SNAC in aqueous solutions

and in F127 hydrogels

GSNO and SNAC in aqueous solutions and in F127hydrogel matrices were irradiated with light from a125 W mercury-arc lamp (Philips, S*ao Paulo, Brazil)equipped with a 10 cm circulating-water filter (OrielInstruments, Stratford, CT). Irradiation withl > 480 nm was achieved by using a glass filter (GG-475, Schott Optical Glass Inc., Duryea, PA). The lightillumination periods were controlled with an electro-mechanical shutter (Oriel). The same light intensitieswere used in all experiments. GSNO and SNAC inaqueous and in gel matrices were irradiated in quartzcuvettes from the top, directly inside the samplingcompartment of the spectrophotometer, by using a fiberoptic accessory with a liquid guide (Oriel). The cuvettewas covered with a quartz slide during the irradiationperiods.

3. Results

3.1. Synthesis of GSNO and SNAC

The aqueous solutions of GSNO and SNAC synthe-sized by the two methods described above displayed thecharacteristic absorption bands at 545 and 336 nm(Fig. 2a). These bands were already assigned tonN-p and p-p transitions, respectively [17,27], and

ARTICLE IN PRESSS.M. Shishido et al. / Biomaterials 24 (2003)3543-3553 3546

were used to confirm the formation of RSNOs in thiswork. Fig. 2b shows the kinetic curve corresponding tothe formation of the absorption band at 545 nm, duringthe synthesis of SNAC by bubbling NO/air through anNAC solution. It can be seen that at around 120 s aplateau starts to form, indicating the full consumptionon NAC. This monitoring allowed avoiding an excess ofnitrosating mixture when using this procedure.

3.2. Incorporation of GSNO and SNAC into F127

hydrogels

Both GSNO and SNAC solutions were easily mixedwith the aqueous F127 solutions, previously prepared bythe ‘‘cold’’ method. The final solutions obtained, stayedin the liquid phase at room temperature during the timerequired for the complete homogenization of thesolutions by stirring. Gelation could be confirmedvisually by the inversion method in both cases, whenthe solutions were heated above their gelation tempera-tures (see below).

3.3. Characterization of the gelation process of F127

solutions by spectrophotometric measurements

The gelation process of the F127 micellar solutions,on heating, was accompanied by using a spectrophoto-metric method based on the transient absorption changeobserved during the transition from the micellarsolution to the cubic liquid crystalline phase, accordingto the phase diagram of F127 obtained by Wanka et al.[46]. This transition was observed to cause a transientincrease in the baseline of the UV/VIS spectrum in allthe 200–1100 nm range. Fig. 3 shows the transient

absorption changes monitored at l ¼ 700 nm (whereno absorptions due to the RSNOs are present), duringthe heating of F127 solutions containing GSH, GSNO,NAC, and SNAC. The transient absorption change ofan F127 solution without co-solutes is shown in all casesfor comparison. It can be seen that in the absence ofsolutes, the F127 solution undergoes a transientabsorption change between 25�C and 30�C, while forthe GSH, GSNO, NAC and SNAC-containing solutionsthese changes occur between 27�C and 34�C, 28�C and38�C, 30�C and 40�C, and 31�C and 40�C, respectively.These transient absorption changes were assigned to thegelation processes (see below) and were used tocharacterize the gelation temperatures. In all cases,within these temperature ranges, gelation was alsoconfirmed visually by observing the absence of flowupon inversion of sample flask. In addition, it wasobserved that all the gels were stable for several days.

3.4. Characterization of the micellization and gelation

processes of F127 solutions by HSDSC measurements

It is usually assumed that the heat changes associatedwith the gelation process of PEO–PPO–PEO blockcopolymers are very small. Wanka et al. [46] observed,e.g., that for lyotropic liquid crystals of block copoly-mers, the heat exchange in a mesophase transitions aretwo orders of magnitude smaller than the heat ofmicellization. For this reason, we have used HSDSC foraccompanying the gelation process of F127 solutionscontaining RSHs and RSNOs. This method is moresensitive than conventional DSC, allowing measure-ments in the mW range, while conventional DSCequipment allows measurements in the mW range.Fig. 4 shows the endothermic peaks associated withthe micellization processes observed in F127 solutionswithout co-solutes and in the solutions containing RSHsand RSNOs. The transient absorption changes observedat l ¼ 700 nm in each case, and shown in Fig. 3, wereincluded in Fig. 4 to allow the comparison betweenspectrophotometric and HSDSC measurements. A weakendothermic peak in the range of 27–29�C was detectedin the heating of the F127 solution 20.5 wt% without co-solutes (inset), which is consistent with the gelationprocess. The calculated enthalpies of gelation (DgelH)associated with each endothermic peak of Fig. 4, as wellas the critical micellization temperatures (cmt) and thetemperatures of gelation (Tgel) are listed in Table 1. TheDgelH found for the weak endothermic peak is presentedas a footnote in Table 1. Fig. 4 also shows anendothermic peak obtained for an F127 solution10.0 wt% without co-solutes. Although at this concen-tration the solution is known to be above the criticalmicellar concentration (cmc) at the starting temperature(10�C), it was not observed to undergo gelation,behaving as a liquid in the temperature range of

ARTICLE IN PRESS

0.03

0.06

0.09

0.12F-127/GSH

Abs

orba

nce

at λ

= 7

00 n

m

20 25 30 35 40 45

0.03

0.06

0.09

0.12

Temperature / oC

F-127/GSNO F-127/SNAC

F-127/NAC

20 25 30 35 40 45

Fig. 3. Absorbance changes at l ¼ 700 nm observed during the

gelation process of Pluronic F127 aqueous solutions 20.5 wt%

containing 15.1 mmol of GSH; 13.8 mmol of GSNO; 28.4 mmol of

NAC and 28.4 mmol of SNAC, per gram of hydrogel, observed during

the heating from 20�C to 45�C. Curves represented with open circles

correspond to the gelation process of an F127 solution 20.5 wt%

without co-solutes and are shown in all cases for comparison.

S.M. Shishido et al. / Biomaterials 24 (2003) 3543-3553 3547

5–45�C. In this case, no transient absorption peak wasobserved either, as it can be seen in Fig. 4, confirmingthe assignment of this spectral change to the gelationprocess, and not to the micellization process. It can alsobe seen in Fig. 4 that both the onset and the peaktemperatures of the transient absorption peaks, areshifted to higher values when the solutes are incorpo-rated in the gels. In addition, it can be seen that the peaktemperatures are about 1–2�C greater for NAC andSNAC than for GSH and GSNO.

3.5. Characterization of the no release from GSNO and

SNAC in aqueous solutions and in F127 hydrogels

GSNO and SNAC in aqueous solutions or incorpo-rated in F127 hydrogels showed continuous thermaldecomposition in the dark at 37�C, associated with theS–N bond cleavage reaction which leads to the produc-tion of free NO. This reaction was confirmed in eachcase by observing the disappearance of the characteristicabsorption bands at 336 and 545 nm, associated with theS–N bond (Fig. 2). The disappearance of these bands isconcerted with the appearance of a new UV band as ashoulder at ca. 280 nm (not shown), which is consistentwith the formation of oxidized RSH (RSSR) accordingto Eq. (1).

3.6. Kinetics of thermal and photochemical no release

from GSNO and SNAC in aqueous solutions and in F127

hydrogels

Fig. 5 shows the kinetic curves of thermal andphotochemical NO release from GSNO and SNAC,accompanied at lmax ¼ 545 nm at 37�C, in F127hydrogels and in aqueous solutions. A comparisonamong the magnitude of the initial rates of NO releasecalculated from these curves is shown in the bar graphicsbelow. The initial rates of thermal and photochemicalNO releases from GSNO were found to be 2.0- and 1.5-fold decreased in the hydrogels, respectively, comparedto the rates obtained in aqueous solutions. In the case ofSNAC, the hydrogel matrix also leads to a reduction of4.3 times in the initial rate of thermal NO release, whilesimilar rates of photochemical NO release were found inboth aqueous solution and hydrogel. Table 2 allows acomparison between the behavior of GSNO and SNAConly in the hydrogel matrices. It can be seen that theinitial rates of NO released from SNAC are 3.1- and 8.4-fold higher than those obtained with GSNO in thethermal and photochemical processes, respectively. Thisgreater lability of NO from SNAC is also reflected in theirradiation of the hydrogels with l > 480 nm. The rateof NO release was 6.2-fold increased in the irradiationof SNAC and 2.3-fold increased in the irradiation ofGSNO.

ARTICLE IN PRESS

Table 1

Critical micellization temperatures (cmt) and enthalpy changes,

calculated from the endothermic peaks associated with the micelliza-

tion process (DmicH)a and gelation temperatures (Tgel)b for F127

solutions 10 and 20.5 wt%, without co-solutes and containing

15.1 mmol of GSH; 13.8 mmol of GSNO; 28.4 mmol of NAC and

28.4 mmol of SNAC, per gram of hydrogel. DH estimated for the

gelation peak in the inset of Fig. 4a=0.5 kJ mol�1.

Solution composition (wt%) cmt

(�C)cDmicH

(kJ mol�1)dTgel

(�C)c

Pluronic F127 10.0 wt% 13 408 —

Pluronic F127 20.5 wt% 10 262 26

Pluronic F127 20.5 wt%/GSH 10 225 27

Pluronic F127 20.5 wt%/GSNO 10 180 29

Pluronic F127 20.5 wt%/NAC 10 194 30

Pluronic F127 20.5 wt%/SNAC 9 182 32

a Obtained by HSDSC measurements.b Calculated from the transient absorption changes associated with

the gelation process.c cmt and Tgel were measured at the onset of curves of Fig. 4.d The uncertainty for the enthalpy measurements is estimated to be

smaller than 1%.

0

4

8

12

0.04

0.08

0.12

HSDSC

F-127 20.5 %

Temperature / oC

Abs

orba

nce

at λ

= 7

00 n

m

Cp

/ kca

l (m

ol d

egre

e)-1

0

4

8

12

HSDSC

F-127 20.5% / GSH

0.04

0.08

0.12

0

4

8

12

HSDSC

F-127 20.5 % / GSNO

0.04

0.08

0.12

0

4

8

12

HSDSC

F-127 20.5 % / NAC

0.04

0.08

0.12

0

4

8

12

HSDSC

F-127 20.5 % / SNAC

0.04

0.08

0.12

0 5 10 15 20 25 30 35 40 45 50

0

4

8

12

HSDSC

F-127 10 %

0.04

0.08

0.12

27 28 29125

150

175

200

Cp

/ kc

al(m

ol d

egre

e)-1

Temperature / oC

Fig. 4. Comparison between HSDSC and spectrophotometric mea-

surement of the micellization and gelation processes of F127 solutions

10 and 20.5 wt%, without co-solutes and containing 15.1 mmol of

GSH; 13.8 mmol of GSNO; 28.4 mmol of NAC and 28.4 mmol of

SNAC, per gram of hydrogel. Inset: endothermic peak obtained by

HSDSC, associated with the gelation process of an F127 solution

20.5 wt% without co-solutes.

S.M. Shishido et al. / Biomaterials 24 (2002) 3543-3553 3548

4. Discussion

In this work, RSNOs were synthesized by either thereaction between RSH and nitrite (NO2

�) in acidic mediaor by bubbling a mixture of NO/synthetic air throughRSH solutions. Reactions of acidified RSHs solutionswith sodium nitrite are known to give RSNOs in highyields [47]. This reaction can be written as

RSH þ HNO �!Hþ

RSNO þ H2O: ð4Þ

The synthesis of RSNOs by bubbling a mixture ofNO/synthetic air through RSH solutions is known tooccur only in the presence of dioxygen [27]. Nitric oxidereacts with dioxygen in a termolecular reaction formingdNO2 according to

2NOdðgÞ þ O2ðgÞ-2dNO2ðgÞ: ð5Þ

NO can share the unpaired electron with �NO2 yieldingdinitrogen trioxide (N2O3) [48] according todNO2ðgÞ þ 2NOd

ðgÞ"N2O3ðgÞ ð6Þ

N2O3 is considered to be the nitrosating agent in severalS-nitrosation reactions [49,50]. When this gaseousmixture is bubbled in aqueous solutions, N2O3 hydro-lyses yielding nitrous acid (HONO), which is inequilibrium with its conjugate base (NO2

�):

N2O3ðaqÞ þ H2OðlÞ "2HNO2ðaqÞ

"2NO�2 ðaqÞ þ 2H þðaqÞ : ð7Þ

Nitrous acid is also an S-nitrosating species (Eq. (4)).However, at the pH around 3.0, existing in the RSNOssolutions, it can be assumed that Eq. (7) is shifted to theleft and N2O3 is the main nitrosating species. Therefore,both methods lead to similar acidic aqueous solutions ofRSNOs, where the nitrite anion (NO2

�) is present in theequilibria of Eqs. (4–7). In the GSNO synthesis, theprecipitated GSNO is separated from most of the NO2

�

ions, which stay at the supernatant solution. Thus, it canbe considered that the residual concentration of NO2

�

ions is greater in the F127 hydrogel containing SNACthan in the hydrogel containing GSNO. Except for apossible influence on the gelation temperature, thepresence of different concentrations of residual NO2

�

ions must have no other implications for pharmaceuticalpurposes, once NO2

� ions are also expected to be formedthrough the reaction between NO released and O2.

The formation and decomposition of SNAC andGSNO was followed by the spectral changes as-sociated with the band at 540–550 nm (nN-p). The

ARTICLE IN PRESS

0.0

0.2

0.4

0.6

0.8

1.0GSNO

Under irradiation(λ > 480 nm)

In the dark

(d)

(c)

(b)

(a)

[NO

] / m

mol

L-1

[NO

] / m

mol

L-1

Time / min

0.0

1.0

2.0

3.0

4.0

5.0

6.0SNAC

0

2

4

6

8

10GSNO

F-127

F-127

H2O

H2O

Init

ial r

ate

of N

O r

elea

sed

(µm

o lL-1

min

-1)

0

10

20

30

40

50

60SNAC

F-127

F-127

H2O

H2O

0 20 40 60 80 100 120Time / min

0 20 40 60 80 100 120

Under irradiation(λ > 480 nm)

In the dark

Init

ial r

ate

of N

O r

elea

sed

(d)(c)

(b)

(a)

(µm

o lL-1

min

-1)

Fig. 5. Kinetic curves of NO released from GSNO and SNAC incorporated in F127 hydrogels: (a) in the gel in the dark, (b) in H2O in the dark, (c) in

the gel under irradiation with l > 480 nm, and (d) in H2O under irradiation with l > 480 nm. The increases in the concentrations of NO were

calculated from the corresponding decomposition curves monitored at 545 nm at 37�C (for details see the experimental section). The volume in the

NO concentration refers to hydrogel volume. The traces represent the linear regressions of the curves, used to calculate the initial rates NO release in

each case. These rates are displayed in the bar graphs below.

Table 2

Initial rates of NO release from GSNO (13.8 mmol g�1) and SNAC

(28.4 mmol g�1) incorporated in F127 hydrogels 20.5 wt% in the dark

and under irradiation with visible light

Rate of NO release (mmol l�1 min�1)a

In the dark Under irradiation

(l > 480 nm)

GSNO 2.370.1 5.370.1

SNAC 7.270.9 44.770.3

Note: Rates were calculated from the kinetic curves of Fig. 5.a Volumes refer to hydrogel volumes.

S.M. Shishido et al. / Biomaterials 24 (2003) 3543-3553 3549

characteristic band in the UV at 330–350 nm (p-p)was also identified in more diluted solutions (Fig. 2a).Both bands are associated with the S–N bond. Inaccordance with other works [16,51], the disappearanceof these bands in the dark and under irradiation can beassigned to the homolytic cleavage of the S–N bond,giving rise to the corresponding disulfide (RSSR) andfree NO, in a first-order kinetics (Eq. (1)). The ac-celeration of the NO release reaction observed in bothaqueous solutions and F127 hydrogels through theirradiation of the band at 545 nm, can be assignedto the promotion of the electronic transition nN-p;which weakens the S–N bond, leading to the bondcleavage with NO ejection. The transparency of theRSNOs-containing F127 hydrogels in the visible regionof the spectrum allows their irradiation in order toincrease the rate of NO release. Although the penetra-tion of visible light through the skin is maximum atabout 600 nm, irradiation with l > 480 nm, can lead toan enhancement of the local NO release. In the case ofusing the hydrogels to cover lesions onto the skin, as inthe case of those caused by Leishmania Donovani thisstrategy could lead to a potentiation of the cytotoxiceffect of NO.

Fig. 4 shows that the HSDSC and the spectrophoto-metric techniques ‘‘see’’ different processes, once thetransient spectral changes are systematically located inthe range of 25–40�C, while the HSDSC peaks are in therange of 7–30�C. The endothermic peaks detected byHSDSC are thus consistent with the micellizationprocess and not with the gelation, once the former mustprecede the latter. The calculated enthalpies of micelli-zation of pure F127 are 408 and 262 kJ mol�1, respec-tively, for 10 and 20.5 wt% solutions (Table 1). Thesevalues are in accordance with those found in Refs.[46,52–54]. Moreover, the observed trend of decreasingenthalpy upon increasing polymer concentration is alsoreported by Wanka et al. [46]. The addition of bothRSHs and RSNOs leads to a significant decrease inthe enthalpies of micelle formation without affecting thecmt values. The assignment of these peaks to themicellization process was confirmed by the appearanceof an endothermic peak in the 10 wt% solution of F127,which does not undergoes gelation (confirmed visuallyand by the spectrophotometric method) and by thecalorimetric detection of a weak endothermic peak inthe pure F127 solution, as shown in the inset of Fig. 4.This weak peak is in the same temperature range of thetransient absorption peak of the F127 solution 20.5 wt%without co-solutes and its estimated enthalpy is0.5 kJ mol�1 (Table 1). Thus, our value is in agreementwith the experimental value reported by Wanka et al.[46]. These values are, however, smaller than anotherderived from Van’t Hoff analysis of gelation tempera-tures, for the same gel [52], but such a difference must beascribed to the different approaches used.

The monitoring of the transient absorption changes at700 nm during the heating of the micellar F127 solutionsprovided an alternative technique to characterize thegelation in all cases. The F127 solution 10 wt% is knownto be above the cmc at 20�C [55]. The absence of atransient absorption change in this solution, which doesnot undergo gelation, confirms that the absorptionchanges observed in the other cases are associated solelywith the gelation process and not with micellization. Theabsorption changes can be understood as arising from alight scattering produced by the phase transition duringgelation, according to the phase diagram presented byWanka et al. [46]. Although the micellar solutionexisting in one side of the gel-phase boundary, and themicellar cubic liquid crystalline phase existing at theother side of the boundary, are both optically isotropic(non-birrefringent) and homogeneous, the coexistenceof both phases during the phase transition constitutes amicroheterogeneous system. This system scatters lightand this scattering can be considered to be the cause ofthe transient rising of the baseline of the absorptionspectrum. It must be pointed out that such effect can beobserved only because the phase transition is notconducted at the equilibrium condition. A continuousphase-transition occurring at equilibrium would beundetected by this method. The return of the baselineto low values after about 35–38�C in the curves ofFig. 4, can thus be understood as the formation of thehomogeneous and optically isotropic micellar cubicliquid crystalline phase of the gel. This homogeneousphase is in accordance with the characterization of thegel phase of F127 by small angle neutron scattering(SANS) in a previous work [55]. The detection of thislight scattering by using a conventional spectrophot-ometer shows that this method can be a simple anduseful tool to characterize the gelation process in PEO–PPO–PEO solutions, offering a more precise alternativeto the standard tube inversion technique.

The onsets of the transient absorption changes inFig. 4 show that the presence of RSHs or RSNOs leadsto an increase in the gelation temperatures, relative tothe F127 solution without co-solutes at the sameconcentration. It is known that the phase behavior ofblock copolymers is greatly affected by salts and otherco-solutes like hydrophobic drugs, ethanol, glycerol, etc.For example, with the addition of NaCl (a typical‘‘salting-out’’ co-solute) the cmt, cmc and the cloudpoint are shifted to lower temperatures, whereas theopposite is found for NaSCN (a typical ‘‘salting-in’’ co-solute) [36,56–58]. Co-solutes that have a tendency toaccumulate at the aggregate interface like large polariz-able anions (e.g. SCN�), increase the solubility of thecopolymer, leading to higher micellization and gelationtemperatures [56]. NAC, SNAC GSH and GSNO arehydrophilic charged species and therefore it can beexpected that they will be mainly located at the

ARTICLE IN PRESSS.M. Shishido et al. / Biomaterials 24 (2003) 3543-3553 3550

hydrophilic corona of the micelles, which is consistentwith the higher gelation temperatures obtained in thepresence of both RSHs and RSNOs as co-solutes. Asrecently pointed out by Matthew et al. [54] the gel-phaseboundary can be shifted to lower or to highertemperatures due to the presence of added pharmaceu-ticals and the characterization of this shift is crucial forblock copolymer formulations to be used in the body orin tissue engineering applications.

The increased thermal and photochemical stability ofGSNO and SNAC incorporated in F127 hydrogelsrelative to aqueous solutions was assigned to the cageeffect promoted by the micellar microenvironment onthe rates of RSNO decomposition. The application ofsuch concepts to the homolytic cleavage of the S–Nbond of RSNOs, with formation of a geminate radicalpair, was already proposed in a previous work [27]. Anyprocess leading to an increase in the rate of geminateradical pair recombination, before the radicals escapefrom the solvent cage, can reduce the decomposition ofRSNOs. If the RSNOs are assumed to be mainly locatedat the corona of the micelles in the gel phase, which iscomposed by PEO blocks, the increased rate of radicalpair recombination in the gel can be assigned to anincreased microviscosity in this medium, in analogy toour previous results obtained in poly(ethylene glycol)(PEG) matrices [27]. This effect could favor boththe geminate recombination, by a solvation effect, andthe random non-geminate recombination, by reduc-ing the diffusion rates of the fragments from the coronaand into the water channels [59]. In the aqueousmicroenvironment, the radicals formed are expected toescape more easily from the solvent cage leading tohigher rates of decomposition [60,61]. The expected fateof the RS� radicals that escape the cage is the dimerizationafter encounters between themselves or with RSNOmolecules. Free NO, on the other hand, can react withO2 or escape from the solution to the gas phase. However,it must be considered also that due to the relativelipophilic properties of NO [61] it may have a slighttendency to accumulate in the micelle cores. This effectwould lead to an increase in its lifetime inside the gel whatmay constitute an additional advantage of such matricesfor pharmacological applications and local NO delivery.

Although the concentrations of RSNOs incorporatedinto the hydrogels presented here were adjusted toobtain gelation temperatures around 30�C with an F127concentration of 20.5 wt%, these formulations can bechanged in order to obtain lower or higher RSNOsconcentrations with the same gelation temperatures. Inthis case, the concentration of F127 will need to bedecreased or increased as well, respectively. The gelationtemperature of any new formulation can be determinedand modulated by using the spectrophotometric techni-que described here. Such modulation can lead toformulations that release NO at rates that are citotoxic

or at rates where NO has only a signaling effect, e.g., toactivate soluble guanylate cyclase.

To evaluate whether the NO released from thehydrogels could be cytotoxic to specific moleculartargets, several points must be considered, once itseffect on cells depends on many complex conditionssuch as: the rate of NO production and its rate ofdiffusion; the concentration of potential reactants suchas superoxide (O2

d�) and dioxygen; the levels of enzymessuch as catalase and superoxide dismutase; the levels ofantioxidants such as glutathione, and the distancesbetween generator and target cells [62]. The minimalconcentration of NO required to stimulate synthesis ofcyclic GMP in vascular smooth muscle is approximately10 nmol 1�1 [63,64]. At this level, NO has no harmfulconsequences in a range of tissues [65]. However, atconcentrations 10–100 times over normal levels, i.e. 0.1–1 mmol l�1, NO can cause both acute and chronicdamage through several different mechanisms [66–69].The measured rates of NO release from our specifichydrogel formulations vary from about 2 to45 mmol l�1 min�1 (Table 2). This implies that exposureof tissues to such hydrogels for a few minutes couldproduce substantial toxicity.

5. Conclusions

We have shown that RSNOs such as GSNO or SNACcan be incorporated into F127 hydrogels, from whereNO could be released thermally or photochemically intarget areas, especially onto the skin. It was also shownthat the RSNOs incorporated into F127 hydrogels canbe irradiated with l >480 nm in order to increase therates of NO release, which is relevant for photother-apeutic applications. Moreover, it has been shown thatthe gelation process of F127 solutions can be accom-panied by simple spectrophotometric measurements.Due to the unique reverse gelation property of thiscopolymer, the concentration of the aqueous F127solutions containing RSNOs can be adjusted to obtaingelation at body temperature. This control may allowF127 solutions containing RSNOs to be administeredsubcutaneously as a liquid, so that they will become arigid semi-solid gels at body temperature, after admin-istration. Alternatively, the gels can be used topically onthe skin. The stabilization effect obtained in the hydrogelmatrices relative to aqueous solutions is an additionaladvantage of these formulations and can allow an easierhandling and a prolonged shelf life for the RSNOs.

Acknowledgements

S.M.S. and A.B.S hold graduate fellowships fromFunda@*ao de Amparo "a Pesquisa do Estado de S*ao

ARTICLE IN PRESSS.M. Shishido et al. / Biomaterials (24 (2003) 3543-3553 3551

Paulo (FAPESP), projects 98/03738-9 and 01/7869-9,respectively. The authors wish to thank Nara L. deAlmeida and Rodrigo C. da Silva for their assistancewith the HSDSC measurements and FAPESP andCNPq for financial support.

References

[1] Moncada S, Radomski MW, Palmer RMJ. Endothelium-derived

relation factor: identification as nitric oxide donors and role in

control of vascular tone ands platelet function. Biochem

Pharmacol 1988;37:2495–501.

[2] de Graaf JC, Banga JD, Moncada S, Palmer RMJ. Nitric oxide

functions as an inhibitor of platelet adhesion under flow

conditions. Circulation 1992;85:2284–90.

[3] Groves PH, Lewis MJ, Cheadle HA, Penny WJ. SIN-1 reduces

platelet adhesion and platelet thrombus formation in a porcine

model of balloon angioplasty. Circulation 1993;87:590–7.

[4] Guo JP, Milhoan KA, Tuan RS, Lefer AM. Beneficial effect of

SPM-5185, a cysteine-containing nitric oxide donor, in rat carotid

artery intimal injury. Circ Res 1993;75:77–84.

[5] Wink DA, Hanbauer I, Krishna MC, DeGraff W, Gamson J,

Mitchell JB. Nitric oxide protects against cellular damage and

cytotoxicity from reactive oxygen species. Proc Natl Acad Sci

USA 1993;90:9813–7.

[6] Garthwait J, Boulton CL. Nitric oxide signaling in the central

nervous system. Annu Rev Physiol 1995;57:683–706.

[7] Davis KL, Marin E, Turko IV, Murad F. Novel effects of nitric

oxide. Annu Rev Pharmacol Toxicol 2001;41:203–36.

[8] Zhelyaskov VR, Gee KR, Godwin DW. Control of NO

concentration in solutions of nitrosothiol compounds by light.

Photochem Photobiol 1998;67:282–8.

[9] Ignarro LJ, Napoli C, Loscalzo J. Nitric oxide and cardiovascular

agents modulating the bioactivity of nitric oxide an overview. Circ

Res 2002;90:21–8.

[10] Gaston B. Nitric oxide and thiol groups. Biochim Biophys Acta

1999;1411:323–33.

[11] Wood PD, Mutus B, Redmond RW. The mechanism of

photochemical release of nitric oxide from S-nitrosoglutathione.

Photochem Photobiol 1996;64:518–24.

[12] Gaston B, Reilly J, Drazen JM, Fackler J, Ramdev P, Arnelle D,

Mullins ME, Sugarbarker DJ, Chee C, Singel DJ, Loscalzo J,

Stamler JS. Endogenous nitrogen oxides and bronchodilator S-

nitrosothiols in human airways. Proc Natl Acad Sci USA

1993;90:10957–61.

[13] Hogg N, Singh RJ, Kalyanaraman B. The role of glutathione in

the transport and catabolism of nitric oxide. FEBS Lett

1996;382:223–8.

[14] Singh RJ, Hogg N, Goss SPA, Antholine WE, Kalyanaraman B.

Mechanism of superoxide dismutase/H2O2-mediated nitric oxide

release from S-nitrosoglutathione-role of glutamate. Arch Bio-

chem Biophys 1999;372:8–15.

[15] Butler AR, Rhodes P. Review: chemistry, analysis and biological

roles of S-nitrosothiols. Anal Biochem 1997;249:1–9.

[16] Askew SC, Barnett DJ, McAninly J, Williams DLH. Catalysis by

Cu2+ of nitric oxide release from S-nitrosoglutathione. J Chem

Soc Perkin Trans 2 1995;4:741–5.

[17] de Oliveira MG, Shishido SM, Seabra AB, Morgon NH. Thermal

stability of primary s-nitrosothiols: roles of autocatalysis and

structural effects on the rate of nitric oxide release. J Phys Chem

A 2002;106(38):8963–70.

[18] Albert J, Daleskog M, Wall!en NH. A comparison of the

antiplatelet effect of S-nitrosoglutathione in whole blood and

platelet-rich plasma. Thromb Res 2001;102:161–5.

[19] Vodovotz Y, Waksman R, Cook JA, Kim W, Chan R, Seabron

R, Collins SD, Pierre A, Bramwell O, Wink D, Mitchell JB, Leon

MB. S-nitrosoglutathione reduces non-occlusive thrombosis rate

following balloon overstretch injury and intracoronary irradiation

of porcine coronary arteries. Int J Radiat Oncol Biol Phys

2000;48:1167–74.

[20] Al-Sa’Doni HH, Ferro A. S-nitrosothiols: a class of nitric oxide-

donor drugs. Clin Sci 2000;98:507–20.

[21] Sexton DJ, Muruganandam A, McKenney DJ, Mutus B. Visible

light photochemical release of nitric oxide from S-nitrosoglu-

tathione: potential photochemotherapeutic applications. Photo-

chem Photobiol 1994;59:463–7.

[22] Stuehr DJ, Feldaman PL, Griffith OW. The surprising life of

nitric oxide. Chem Eng News 1993;20:26–37.

[23] Santos KFR, Shishido SM, Laurindo FR, De Oliveira MG,

Krieger MH. Nitrosothiol (SNAC) recovery the vascular dysfunc-

tion in the knockout LDLR�/� mice. Hypertension 2001;

37:1039.

[24] Santos KFR, Shishido SM, Laurindo FR, De Oliveira MG,

Krieger MH. Characterization of the hypotensive effect of S-

nitroso-N-acetylcysteine in normotensive and hypertensive con-

scious rats. Nitric Oxide: Biol Chem 2002;7(1):57–66.

[25] Williams DLH. The chemistry of S-nitrosothiols. Acc Chem Res

1999;32:869–76.

[26] Meyer DJ, Kramer N, Ozer B, Ketterer B. Kinetics and equilibria

of S-niitrosothiol-thiol exchange between glutathione, cysteine,

penicillamines and serum albumin. FEBS Lett 1994;345:

177–80.

[27] Shishido SM, de Oliveira MG. Polyethylene glycol matrix reduces

the rates of photochemical and thermal release of nitric oxide

from S-nitroso-N-acetylcysteine. Photochem Photobiol 2000;

71(3):273–80.

[28] de Oliveira MG, Langley GJ, Rest AJ. Photolysis of the

[Fe(CN)5(NO)]�2 ion in water and poly(vinyl alcohol) films:

evidence for cyano radical, cyanide ion and nitric oxide loss and

redox pathways. J Chem Soc Dalton Trans 1995;12:2013–9.

[29] Bohl KS, West JL. Nitric oxide-generating polymers reduce

platelet adhesion and smooth muscle cell proliferation. Biomater-

ials 2000;21:2273–8.

[30] Gazzinelli RP, Oswald SH, Janes SJ, Sher A. The microbicidal

activity of interferon-gamma-treated macrophages against trypa-

nosoma-cruzi involves an l-arginine-dependent, nitrogen oxide-

mediated mechanism inhibitable by interleuken-10 and transform-

ing growth-factor-beta. Eur J Immunol 1992;22:2501–6.

[31] Nussler A, Drapier JC, Renia S, Pied S, Miltger M, Gentilini M,

Mazier D. l-Arginine-dependent destruction of intrahepatic

malaria parasites in response to tumor-necrosis-factor and/or

interleukin-6 stimulation. Eur J Immunol 1991;21:227–30.

[32] Adhuna A, Saltora P, Bhatnagar R. Nitric Oxide induced

expression of stress proteins in virulent and avirulent promasti-

gotes of Leishmania donovani. Immunol Lett 2000;71(3):171–6.

[33] Liew FY, Millot S, Parkinson C, Palmer RMJ, Moncada S.

Macrophage killing of leishmania parasite in vivo is mediated by

nitric-oxide from l-arginine. J Immunol 1990;144(12):4794–7.

[34] Davidson RN, Yardley V, Croft SL, Konecny P, Benjamin N. A

topical nitric oxide-generating therapy for cutaneous leshmania-

sis. Trans R Soc Trop Med Hyg 2000;94(3):319–22.

[35] Desai M, Jain NJ, Sharma R, Bahadur P. Temperature and salt-

induced micellization of some block copolymers in aqueous

solution. J Surfactants Deterg 2000;3:193–9.

[36] Ivanova R, Alexandridis P Lindman B. Interaction of polxamer

block copolymes with cosolvents and surfactants. Colloid Surf A

2001;183(Sp. Iss.):41–53.

[37] Blonder JM, Baird L, Fulfs JC, Rosenthal GJ. Dose-dependent

hyperlipidemia in rabbits following administration of poloxamer

407 gel. Life Sci 1999;65:PL261–6.

ARTICLE IN PRESSS.M. Shishido et al. / Biomaterials 24 (2003) 3543-3553 3552

[38] Scherlund M, Malmsten M, Holmqvist P, Brodin A. Thermo-

setting microemulsions and mixed micellar solutions as drug

delivery systems for periodontal anesthesia. Int J Pharm

2000;194:103–16.

[39] Liaw J, Lin Y-C. Evaluation of poly(ethylene oxide)–poly(pro-

pylene oxide)–poly(ethylene oxide) (PEO–PPO–PEO) gels as a

release vehicle for percutaneous fentanyl. J Control Release 2000;

68:273–82.

[40] Morishita M, Barichello JM, Takayama K, Chiba Y, Tokiwa S,

Nagai T. Pluronic F-127 gels incorporating highly purified

unsatured fatty acids for buccal delivery of insulin. Int J Pharm

2001;212:289–93.

[41] Wang YY, Hong CT, Fang JY. In vitro and in vivo evaluations of

topically applied capsaicin and nonivamide from hydrogels. Int J

Pharm 2001;224:89–104.

[42] Guzm!an M, Garc!ıa FF, Molpeceres J, Aberturas MR. Polyox-

yethylene-polyoxypropylene block copolymer gels as sustained

release vehicles for subcutaneous drug administration. Int J

Pharm 1992;80:119–27.

[43] Barichello JM, Morishita M, Kozo T, Nagai T. Absorption of

insulin from Pluronic F-127 gels following subcutaneous admin-

istration in rats. Int J Pharm 1999;184:189–98.

[44] Stratton LP, Dong AC, Manning MC, Carpenter JF. Drug

delivery matrix containing native protein precipitates suspended

in a poloxamer gel. J Pharm Sci 1997;86:1006–10.

[45] Schmolka IR. Artificial skin. 1. Preparation and properties of

Pluronic F-127 gels for treatment of burns. J Biomed Mater Res

1972;6:571.

[46] Wanka G, Hoffmann, H, Ulbricht W. Phase diagrams and

aggregation behavior of poly(oxyethylene)-poly(oxypropylene)-

poly(oxyethylene) triblock copolymers in aqueous solutions.

Macromolecules 1994;27:4145–59.

[47] Hogg N. Biological chemistry and clinical potential of S-nitro-

sothiols. Free Rad Biol Med 2000;28:1478–86.

[48] Guillory WA, Johnston HS. Infrared absorption by peroxy-

nitrogen trioxide free radical in the gas phase. J Chem Phys

1964;42:2457–61.

[49] Wink DA, Nims RW, Darbyshire JF, Chirstodoulou D,

Hanbauer I, Cox GW, Laval F, Laval J, Cook JA, Krishna

MC, DeGraff WG, Mitchell JB. Reactions kinetics for nitrosation

of cysteine and glutathione in aerobic nitric oxide solutions at

neutral pH. Insights into the fate and physiological effects of

intermediates generated in the NO/O2 reaction. Chem Res

Toxicol 1994;7:519–25.

[50] Stamler JS, Simon DI, Osborne JA, Mullins ME, Jaraki O,

Michel T, Single DJ, Loscalzo J. S-nitrosylation of proteins with

nitric oxide: synthesis and characterization of biologically active

compounds. Proc Natl Acad Sci USA 1992;89:444–8.

[51] Roy B, Hardemare AM, Fontecave M. New thionitrites:

synthesis, stability, and nitric oxide generation. J Org Chem

1994;59:7019–26.

[52] Beezer AE, Loh W, Mitchell JC, Royall PG, Smith DO, Tute MS,

Armstrong JK, Chowdhry BZ, Leharne SA, Eagland D, Crowter

NJ. An investigation of dilute aqueous solution behavior of

poly(oxyethylene)+poly(oxypropylene)+poly(oxyethylene) block

copolymers. Langmuir 1994;10:4001–5.

[53] da Silva RC, Olofsson G, Schillen K, Loh W. Influence of ionic

surfactants on the aggregation of poly(ethylene oxide)–

poly(propylene oxide)–poly(ethylene oxide) block copolymers

studied by differential scanning and isothermal titration calori-

metry. J Phys Chem B 2002;106:1239–46.

[54] Matthew JE, Nazario YE, Roberts CS, Bhatia SR. Effect of

mammalian cell culture medium on the gelation properties of

Pluronic F127. Biomaterials 2002;23:4615–9.

[55] Wanka G, Hoffmann H, Ulbrich W. The aggregation behavior

of poly(oxyethylene)-poly(oxypropylene)–poly(oxyethylene)-block-

copolymers in aqueous solutions. Colloid Polym Sci 1990;

268:101–17.

[56] Mortensen K. Cubic phase in a connected micellar network of

poly(propylene oxide)-poly(ethylene oxide)-poly(propylene oxide)

triblock copolymers in water. Macromolecules 1997;30:503–7.

[57] Malmsten M, Lindman B. Self-assembly in aqueous block

copolymer solutions. Macromolecules 1992;25:5440–5.

[58] da Silva RC, Loh W. Effect of additives on the cloud points of

aqueous solutions of ethylene oxide-propylene oxide-ethylene

oxide block copolymers. J Colloid Interface Sci 1998;202:385–90.

[59] Gilbert JC, Hadgraf J, Bye A, Brookes LG. Drug release from

Pluronic F-127 gels. Int J Pharm 1986;32:223–8.

[60] Wayne RP. Principles and applications of photochemistry.

Oxford: Oxford Science Publications, 1991. p. 54–6.

[61] Adachi H, Sonoki H, Hoshino M, Wakasa M, Hayashi H.

Photodissociation of nitric oxide from nitrosyl metalloporphyyr-

ins in micellar solutions. J Phys Chem 2001;105:392–8.

[62] Tamir S, Lewis RS, deRojas-Walker DWM, Wisnok JS,

Tannenbaum R A. The influence of delivery rate on the chemistry

and biological effects of nitric oxide. Chem Res Toxicol 1993;

6:895–9.

[63] Hutchinson PJA, Palmer RMJ, Moncada S. Comparative

pharmacology of EDRF and nitric oxide on vascular strips. Eur

J Pharmacol 1987;141:451–5.

[64] Liu SY, Beckman JS, Ku DD. Peroxynitrite, a product of

superoxide and nitric oxide, produces coronary vasorelaxation in

dogs. J Pharmacol Exp Ther 1994;268:1114–20.

[65] Murphy MP. Nitric oxide and cell death. Biochem Biophys Acta

1999;1411:401–14.

[66] Malinski T Taha Z. Nitric oxide release from a single cell

measured in situ by a porphirinic-based microsensor. Nature

1992;358:676–8.

[67] Lipton SA, Choi YB, Pan AH, Lei SZ, Chen HSV, Sucher NJ,

Loscalzo J, Singel DJ, Stamler JS. A redox-based mechanism for

the neuroprotective and neurodestructive effects of nitric oxide

and related nitroso-compounds. Nature 1993;364:626–32.

[68] Ionnidis I Groot H. Cytotoxicity of nitric oxide in Fu5 rat

hepatoma cells: evidence for co-operative action with hydrogen

peroxide. Biochem J 1993;296:341–5.

[69] Burney S, Tamir S, Aaron G, Tannenbaum RA. Mechanistic

analysis of nitric oxide-induced cellular toxicity. Nitric Oxide:

Biol Chem 1997;1:130–44.

ARTICLE IN PRESSS.M. Shishido et al. / Biomaterials 24 (2003) 3543-3553 3553