www.elsevier.com/locate/jconrel

Journal of Controlled Releas

Developments in the rat adjuvant arthritis model and its use in

therapeutic evaluation of novel non-invasive treatment by

SOD in Transfersomes

S.I. Simoesa,*, T.C. Delgadoa, R.M. Lopesa, S. Jesusb, A.A. Ferreirab, J.A. Moraisc,

M.E.M. Cruza, M.L. Corvoa, M.B.F. Martinsa

aUnidade de Novas Formas de Agentes Bioactivos, Departamento de Biotecnologia, Instituto Nacional de Engenharia,

Tecnologia e Inovacao, Edifıcio F, Estrada do Paco do Lumiar, 22 1649-038 Lisboa, PortugalbFaculdade de Medicina Veterinaria, Universidade Tecnica de Lisboa, Portugal

cFaculdade de Farmacia, Universidade de Lisboa, Portugal

Received 27 August 2004; accepted 13 December 2004

Available online 15 January 2005

Abstract

The aim of this study was firstly to refine a rat model of arthritis, the adjuvant arthritis (AA) model, by studying the time

course of the disease, introducing new evaluation methods such as haematological and biochemical parameters in order to

identify the main stages of the disease. An optimisation of treatment schedule and evaluation criteria was developed. This

refinement provided novel non-invasive anti-inflammatory treatment of the AA with SOD by using mixed lipid vesicles

specially developed for transdermal delivery, Transfersomes (Tfs), this being the second major aim. The time course of AA

includes a first stage: 1 day after the disease induction, the induced paw volume more than doubled and the paw circumference

increased by approx. 50%. Two weeks later, another stage occurred where the disease shifted from the local arthritis form

towards polyarthritis: an additional increase of volume and circumference of the induced and non-induced paws, occurred. The

animals also started to loose weight around day 14 after the disease induction. Radiographic observable lesions increased

correspondingly. Treatment of animals, started at day 1 after induction, by epicutaneous application of SOD–Tfs showed that 1

mg SOD/kg body weight is more efficient than 0.66 mg SOD /kg body weight. As a positive control, SOD liposomes

intravenously injected were used for comparison and confirmed the biological efficiency of epicutaneously applied SOD in Tfs.

SOD solution and empty Tfs epicutaneously applied exerted no effect. In addition, epicutaneous application of SOD–Tfs used

prophylactically was able to suppress the induced rat paw oedema. Radiographic images showed less joint lesions in SOD–Tfs

treated animals in comparison with control and placebo treated rats. It was shown for the first time that SOD incorporated into

0168-3659/$ - s

doi:10.1016/j.jco

* Correspon

E-mail addr

e 103 (2005) 419–434

ee front matter D 2004 Elsevier B.V. All rights reserved.

nrel.2004.12.008

ding author. Tel.: +351 21 0924732; fax: +351 21 7163636.

ess: sandra.simoes@ineti.pt (S.I. Simoes).

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434420

Tfs and applied onto a skin area not necessarily close to the inflamed tissue is able to promote non-invasive treatment of

induced arthritis.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Superoxide dismutase; Transfersome; Adjuvant induced arthritis; Transdermal delivery; Antioxidant enzyme therapy

1. Introduction

To date, injection is the only mode of therapeutic

enzyme administration [1–4]. This is problematic as

all such agents have a short half-life and must

therefore be administered frequently. Several strat-

egies have been studied to reduce the frequency of

administration of therapeutic enzymes, such as their

association with long circulating liposomes [1,4,5].

A major challenge would be to develop non-

injection modes of administration. Special mixed lipid

carriers in the form of ultradeformable vesicles, Trans-

fersomesR (Tfs—a trademark of IDEA AG), may

arguably deliver small molecular weight drugs trans-

cutaneously into blood circulation [6–8]. Tfs represent

a new approach in the field of systemic delivery. In our

previous work the aggregates were optimised for

enzyme loading [9]. Non-invasive transdermal protein

delivery would be a new approach especially attractive,

if the protein shows therapeutic activity, in systemic

diseases, similar to that obtained with intravenous or

subcutaneous administered protein.

Tfs vesicles when placed on the surface of the skin,

lose water that evaporates from the surface and the

vesicles start to dry out. Due to its hydrophilicity, it is

assumed that the vesicles are attracted to the areas of

higher water content in the narrow gaps between

adjoining cells in the skin. This fact associated with

the ability of Tfs to deform, allow these systems to

transiently open the pores through which water

normally evaporates between the cells. After passage,

the vesicles, possibly distributed between the cells and

bypassing the cutaneous capillary plexus, reach the

subcutaneous tissue. The vesicles finally reach the

systemic blood circulation through the lymphatic

system [8]. The resulting bioavailability reportedly

can be rather high [10] and the biodistribution similar

to that resulting from a subcutaneous injection [11].

Antioxidant enzymes, such as superoxide dismu-

tase (SOD) have been used for the treatment of

diseases and injuries where reactive oxygen species

(ROS) are implicated [1,12,13]. An example is

rheumatoid arthritis, a systemic chronic disease of

unknown aetiology, characterised by chronic hyper-

trophic synovitis, leading to joint cartilage and bone

destruction and involving systemic autoantibody

production [14].

A wide variety of inflammatory arthritis models

have been studied [15]. Among these, adjuvant

disease of rodents (AA model) is the major model

system for evaluation of anti-arthritic drugs [16–19].

Michelson and co-workers reported comparative

studies of SODs from different sources in different

rat models with pronounced differences in anti-

inflammatory properties for each model [20].

The AA model was used as a follow-up to our

previous work with liposomal SOD [21]. AA is

expressed by swelling, pain and deformity. The paw

size and volume were used in this work to evaluate

disease progression. Some blood markers of inflam-

mation were quantified in the present study to achieve

a better evaluation of the response of the animals. AA

animal model is influenced by several factors includ-

ing the animal age and strain [22]. Model variability,

with the consequent lack of easy evaluation of the

disease evolution, could affect the success of the

treatment protocol. For this reason it was important to

establish the model with both physical and biochem-

ical parameters evaluation, in order to conveniently

identify the phases of the disease for the specific

animal strain and age used in each study.

Different stages, based on the time course of

arthritis disease model, were identified by several

authors [2,23–26]. Interestingly, the studies of disease

modification by anti-inflammatory agents are usually

accompanied by a study of the development and the

time course of the experimental animal model. In the

majority of the cases, primary lesions and secondary

lesions are identified. A tertiary reaction with the

appearance of evolutive trophic lesions was also

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434 421

described [2]. For this reason some studies reported in

literature [3,21,27] started the treatments on different

days after induction. The periods for evaluation of

treatments, using AA model, also vary.

Usually, the inflammation is evaluated by non-

specific laboratory tests, such as the erythrocyte

sedimentation rate [28,29] or C reactive protein

[30,31]. More recently, the evaluation of oxidative

damage, caused by ROS produced by the phagocytes

in the joints, by the variation of antioxidant system

compounds such as thiols and ascorbic acid levels,

that are part of the non-enzymatic antioxidant defence

system, have been of great use. Changes in blood

antioxidant species levels are also expected, both in

rheumatoid arthritis and in AA. Biological thiols have

a key role in the coordination of antioxidant defences

and can be used as markers in AA [32]. Ascorbic acid

levels in rat sera were also studied as biomarkers of a

model of oxidative damage in ageing [33] and for

evaluation of adjuvant arthritis [34]. Other reports

[32] emphasised the effect of vitamin C administration

on the development of the acute and the chronic phase

of rat AA and on the decrease in leukocyte infiltration

in AA animals [35].

Although the schedule of treatments and the

evaluation criteria of the regression of the disease

vary depending on the authors [3,23,36], bovine

copper SODs were shown to be fully active during

secondary and tertiary arthritic reactions and to delay

the appearance of bone damage [20]. The pharmaco-

logical efficacy of SOD is dependent on its bioavail-

ability, rate of cellular transport and plasma half-life

[37]. According to some authors [2], SOD is active

when intravenously administered at low dose (0.033

mg/kg) from days 7 to 17 after induction. Other

authors report responses to intra-muscular adminis-

tration of higher dose (10 mg/kg) at alternating days

for day 3 to day 21 [37]. This points to an anti-

inflammatory and anti-arthritic activity dependency

on the scheme, dose regimen, type of formulation and

the stage of inflammation [38].

One of the major goals of this work is to evaluate

the therapeutic activity of SOD-loaded Tfs, in

epicutaneous administration in the treatment of

rheumatoid arthritis, in order to achieve a novel

non-invasive antioxidant treatment strategy. Conse-

quently, the careful definition of the model for

evaluation of antioxidant activity is the other main

goal of this work. This would allow us to identify the

main stages of the disease and optimise either the

selection of treatment schedules or the evaluation

criteria of the therapeutic efficacy.

To evaluate the performances of the carrier

mediated transdermally transported SOD, different

applications schemes were tested within the frame-

work of established AA. The treatment of inflamma-

tory process, before the establishment of severe

ostheo-articular disease, was given particular empha-

sis. Additionally, prophylactic effect, in such model,

will be studied.

The evaluation of the performances of the treat-

ment by the oedema regression (OR), a parameter that

correlates measurements of the volume or circum-

ference of the paw at different stages of the disease

[21], was used in the present work when therapy was

administered after the induction of the disease. The

evaluation of therapeutic activity was also performed

using biological markers.

In our previous biodistribution studies, SOD-

loaded Tfs epicutaneously applied had a profile

similar to those of subcutaneously administered one

[39]. Blood contained approximately ~6% of the

epicutaneously applied SOD by means of Tfs. This

was the basis for the selection of the SOD-loaded Tfs

doses used in this work, in order to apply on the skin a

sufficient SOD amount to obtain effective therapeutic

enzyme concentration in the bloodstream. This will

allow us to compare the therapeutic efficacy of the

non-invasive approach with the results obtained with

SOD-long circulating liposome injection tested as a

positive control.

2. Materials and methods

2.1. Materials

Soybean phosphatidylcholine—SPC-(S100) and

egg phosphatidylcholine were obtained from Lipoid

KG (Ludwigshafen, Germany). Distearoylphosphati-

dylethanolamine-poly(ethylene glycol)200 (DSPE-

PEG) was obtained from Avanti Polar Lipids

(Alabaster, AI, USA). Cholesterol, sodium cholate,

Cu,Zn-superoxide dismutase (SOD, from bovine

erythrocytes, 98% protein, 2500–7000 U/mg protein),

5,5V-dithio-bis-(2V-nitrobenzoic) acid (DTNB), gluta-

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434422

thione (GSH), ascorbic acid and 1,2-phenylene

diamine were purchased from Sigma Chemical (St.

Louis, MO, USA). Dehydroascorbic acid was pur-

chased from Aldrich (Steinheim, Germany). All other

chemicals were of reagent grade. Mycobacterium

butyricum (killed and dried) and the Incomplete

Freund Adjuvant were purchased from Difco Labo-

ratories (Detroit, MI, USA). Track-etched polycar-

bonate membranes were obtained from Poretics

(Livermore, USA). Millex GV13mm filters were

obtained from Millipore (Bedford, USA). SPC con-

centration was determined with an enzymatic–colori-

metric test from Spinreact (Girona, Spain). SOD

activity was measured with BIOXYTECHR bSOD-525k kitQ (Oxys, Portland, USA).

2.2. Preparation of SOD-Transfersomes

Enzyme-loaded vesicles comprised a mixture of

SPC and a surfactant in the ratio that ensured adequate

vesicle adaptability, protein loading, and stability. In

brief, to prepare highly adaptable vesicles suspension,

SPC was mixed with the bio-surfactant sodium

cholate in the molar ratio 3.75:1. The mixture was

taken up in sodium phosphate buffer solution (50

mM, pH=7.4) to yield a 10% total lipid suspension.

SOD was added to the resulting blend of heteroge-

neous vesicles, which were brought to final size of

approximately 150F50 nm by sequential filtration

through the track-etched polycarbonate membranes.

2.3. Preparation of long-circulating SOD–liposomes

Multilamellar liposomes were prepared by the

dehydration–rehydration method developed by Kirby

and Gregoriadis [40], followed by the sequential

extrusion as previously described [3]. Briefly, a

mixture of egg phosphatidylcholine/cholesterol/

DSPE-PEG in the molar ratio of 1.85:1:0.15 was

taken up in chloroform and then dried under a

nitrogen stream to form a homogeneous film. This

film was dispersed in a solution of 5.0 mg SOD/ml

water (lipid concentration: 32 Amol/ml), frozen in

liquid nitrogen, and lyophilised overnight. A solution

of 0.28 M mannitol/10 mM citrate buffer, pH=5.6

was then added to the lyophilised powder to obtain

1/10 of the original dispersion volume. The hydra-

tion step lasted 30 min. Subsequently, 0.145 M

NaCl/10 mM citrate buffer, pH=5.6, was added to

reach the starting, specified suspension volume.

Liposomes were extruded sequentially through the

polycarbonate filters with pore sizes of 0.05 Am.

Non-encapsulated protein was separated from lip-

osome dispersion by dilution (26 times) and ultra-

centrifugation at 300,000�g for 120 min at 4 8C in a

Beckman LM-80 ultracentrifuge. Finally, the col-

lected liposomes were dispersed in a 0.145 M NaCl/

10 mM citrate buffer, pH=5.6.

2.4. Characterization of therapeutic formulations

The mean particle size and size distribution, in

terms of polydispersity index, was deduced from the

results of photon correlation spectroscopy measured at

908 with a Malvern Zetasizer-3 (Malvern, Malvern,

UK), using 3rd order cumulant analysis method. The

total SPC concentration was determined with an

enzymatic–colorimetric test. Total protein concentra-

tion was quantified with the method of Lowry [41],

after vesicle disruption with 2% Triton X-100 and

20% sodium dodecyl sulphate.

SOD activity was measured with BIOXYTECHRbSOD-525k kitQ, which relies on a spectrophotomet-

ric assay for the enzyme.

2.5. Induction and evaluation of the adjuvant arthritis

model

bIFFA CREDO BELGIUMQ 4 month old male

Wistar rats were kept under standard laboratory

conditions. The test animals were obtained from

Charles River Laboratories (Santa Perpetua de

Mogoda, Spain).

As a rule, inflammation was induced in the test rats

by a single intradermal injection, into the subplantar

area of the right hind paws, of 0.10 mL of a 10 mg/mL

suspension of M. butyricum in Incomplete Freund

Adjuvant, homogenized by ultra-sound.

2.5.1. Measurement of paw volume

Right and left paw volumes were measured by

water displacement method, using a plethysmometer

(Ugo Basile, Comerio, Italy). The hind paw was

immersed in the measurement cell up to the hair line

to the ankle to determine the immersed organ volume

in mL.

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434 423

2.5.2. Measurement of ankle circumference

Ankle circumference was measured for both paws

with a flexible strip.

2.5.3. Measurement of body weight

Body weight of each animal was determined with a

Sartorius LP2200S balance (Sartorius AG, Gfttingen,Germany) and the daily gain/loss was calculated by

comparison with the pretreatment value. For each

measurement, 10 successive readings were used.

2.5.4. Calculation of oedema regression (OR)

The equation used to calculate oedema regression

from basic measurements is:

Cat� Cbtð Þ= Cbi� Cbtð Þ � 100

Cat=ankle circumference after treatment; Cbt=ankle

circumference before treatment; Cbi=ankle circum-

ference before the induction.

2.5.5. Calculation of increase percentage

The calculations were done considering the meas-

urement immediately before induction for paw cir-

cumference and/or paw volume as a reference.

2.5.6. Radiographic analysis

Lateral radiographs were obtained with the table-

top technique for both hind paws after animals

sacrifice, with an X-ray unit Philips CP 1000 (Philips,

Eindhoven, The Netherlands). All radiographs were

evaluated by a board-certified radiologist unaware of

the treatment group assignment. Radiographic analy-

sis was performed according to the radiographic score

used by Cuzzocrea et al. [42]. Scale: 0=no bone

damage; 1=soft tissue swelling; 2=joint erosion;

3=bone erosion and osteophyte formation.

2.5.7. White blood cells count

Leukocytes were counted in the blood smear

coloured with Giemsa (100 cells per blood smear).

Cell percentage was expressed as the percentage of

each main type of leucocytes.

2.6. Treatment and therapeutic activity evaluation

protocol

Each group contained at least 6 animals. One group

of animals remained non-induced, and is referred in

further text as naive rats. One group of induced

animals remained untreated, and is referred in further

text as control rats. These two groups of animals were

kept under the same conditions as the drug-treated

animals.

One day prior to each epicutaneous (e.c.) enzyme

application, the hair on the selected area on the rat’s

back was trimmed. Our previous studies define back

skin of the rats as the elected site of formulation

epicutaneous application (data not published). The

non-invasively applied formulation was gently spread

with a micropipette over an area such that it yielded an

average area dose of 1 mg lipid/cm2. The solutions

and suspensions were left to dry out on the skin under

non-occlusive conditions. During such time each

animal was kept in a separate cage to prevent mutual

licking.

A group of rats was treated e.c. with the empty

Tfs (data not shown) to study placebo effect. Two

groups of rats were treated e.c. with SOD solution:

one group received 0.66 mg SOD/kg body weight

(BW) and the other 1 mg SOD/kg BW (data not

shown). For all the schedules, the results based on

measurements of physical examination parameters

either for the empty Tfs or SOD solution treated

animals were not statistically different from those

observed with the control rats. As a consequence, in

all the treatment application schemes used in this

work, the studies of the placebo and free SOD effects

were omitted.

2.6.1. SOD prophylactic treatment: administration 1 h

prior to induction and evaluation during the first 24 h

The schedule used for the prophylactic application

of SOD in mixed lipid vesicles for the evaluation of

treatment of adjuvant induced arthritis in the first 24 h

is presented in Schedule 1 from Fig. 1. A single time

and 1 h prior to the disease induction, 0.66 mg of the

enzyme per kg body weight in the form of SOD-

loaded mixed lipid vesicles was applied on the upper

back skin at the area dose of 1 mg lipid/cm2. The

untreated, but adjuvant-induced animals (control rats),

were used for comparison. The dose choice was made

according to our preliminary biokinetic and biodis-

tribution data [39]. The selected dose was the one that

permits to achieve a systemic concentration of the

enzyme in the range of the therapeutic activity of

SOD. Our aim was to study the effect of a

Schedule 1

Treatment

-1 0 4 6 24

Induction X

Evaluation X

X

X X X X

Day -3 -1 0 1 3 5 8 10 12 Total nr. of

applications

X

X

-

A

X

7

X

B 5

Day 0 1 3 5 8 10 12

Induction X

SOD-Lip

X

X

X X X X X

X

X X X X X X

X X X X

X X X X

X

X X X XX X

Schedule 2

Schedule 3

Time (h)

10

Inductioncontrol

Evaluation

Induction X

Treatment XXX

XX

protocol

Evaluation X

X X X X X XX

X X X XX

Induction

Treatment

protocol

Evaluation

5

Treatment

SOD-Tfs

Evaluation

Fig. 1. Schedule 1: SOD prophylactic treatment: administration one hour prior to induction and evaluation during the first 24 h; Schedule 2:

SOD therapeutic treatment: administration starting at day 1 after induction and evaluation at day 12 after induction; Schedule 3: SOD

prophylactic treatment: administration starting at day 3 prior to induction and evaluation at day 12 after induction.

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434424

prophylactic treatment on the development of the

primary phase, specifically, during the first 24 h of the

disease. The previous application (1 h prior to

induction) of the enzyme was chosen assuming the

lag time of non-invasive delivery [43].

2.6.2. SOD therapeutic treatment: administration

starting at day 1 after induction and evaluation at

day 12 after induction

Schedule 2 from Fig. 1 was taken for treatment

and evaluation of SOD-loaded Tfs administration. To

test the treatment sensitivity to different e.c. applied

SOD–Tfs doses, the test rats received seven treat-

ments, of either nominal doses of 1 mg SOD/kg BW

or 0.66 mg SOD/kg BW on the upper back skin,

starting on the day 1 after the disease induction. The

untreated, but adjuvant-induced animals (control

rats), were used as negative control. As a positive

control, a previously [4,37] determined effective

dose of SOD–liposomes (corresponding to 0.066

mg SOD/kg BW) was injected intravenously. Lip-

osomes have been tested with success as delivery

system for SOD in experimental arthritis [44]. Our

aim in this study was to find comparable data as

already reported, i.e., by means of transdermal

delivery of SOD mediated by Tfs, to obtain

comparable therapeutic activity results as those

obtained after intravenous injection of liposomal

SOD. Our previous preliminary biokinetic and

distribution tests [39] have shown that only ~6% of

the epicutaneously applied SOD (carried by Tfs)

reached the blood. For that reason we have chosen

an intravenous dose that fits the effective circulating

SOD.

2.6.3. SOD prophylactic treatment: administration

starting at day 3 prior to induction and evaluated at

day 12 after induction

To get further insight into therapeutic efficacy of

non-invasively delivered SOD in mixed lipid vesicles

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434 425

we also studied the prophylactic effect of SOD–Tfs

on the evolution of the secondary stage of the

disease. The aim of this part of the study was to

investigate if the prophylactic administration exerts

any effect on the disease progression. For that

purpose the study was carried out according to

Schedule 3 from Fig. 1. Treatment starting on the

3rd day before arthritis induction was compared with

treatment starting in the first day after arthritis

induction. SOD-loaded mixed lipid vesicles were

used at nominal dose of 1 mg SOD/kg BW. The

untreated, but adjuvant-induced animals (control

rats), were used as negative control. In this study,

only one nominal dose of SOD was taken and it was

the dose with better results obtained in the work

described in Section 2.6.2. As the evaluation was, for

both cases, at day 12 after induction, different

treatment duration and, consequently, different num-

ber of epicutaneous applications (5 and 7) affects the

total applied dose.

2.7. Total thiols assay in plasma

Quantification of the total thiols in plasma was

performed according to Ellman, following the

adaptation by Marinho [45]. For the purpose, 70

AL of fresh plasma was mixed with 1.12 mL

methanol and 210 AL buffer (0.4 M Tris–HCl,

pH=8.9). 7 AL of 0.02 M DTNB in methanol was

then added. After 15 min, the mixture was centri-

fuged at 3000�g and 4 8C for 15 min. Absorbance

against methanol was read at 412 nm. A standard

curve was prepared with the concentrations ranging

from 0 to 800 AM GSH. For the total thiols

determination, each sample absorbance was sub-

tracted from the absorbance of a blank sample (70

AL of 0.154 M KCl instead of plasma) and the blank

reagent absorbance (DTNB was replaced with

methanol). Hb was measured in the plasma samples

to check for haemolysis. The samples with haemo-

globin values higher than 130 mg/dL were not

considered in the final data analysis.

2.8. Ascorbic acid and dehydroascorbic acid assay

Quantification of both forms of the vitamin was

performed according to Simoes et al. [34] as follows.

The test sera were extracted with the mobile phase

(serum/mobile phase, 1:2, v/v) and then centrifuged at

15,490�g within 30 min of collection. The mobile

phase was 80% acetonitrile and 20% phosphoric acid

(60 mM; pH=2.0). For ascorbic acid measurements no

further step was needed. For dehydroascorbic acid, the

extracted sample was derivatised by mixing with an

equal volume of 10 mM 1,2-phenylene diamine in

water, gassed with nitrogen, and left to react for 1 h at

4 8C before final filtration through a Millex GV13mm

filters (Millipore, Bedford, USA) and HPLC analysis.

The latter was done with a solvent module Beckman

126 (Beckman, Fullerton, USA), an injector MIDAS

(Spark Holland, AJ Emmen, The Netherlands) fitted

with a 10 AL loop, and two detectors: a molecular

absorbance detector, Beckman 166 (Beckman, Full-

erton, USA) and a fluorescence detector, Jasco 812-FP

(Jasco, Tokyo, Japan), used for ascorbic acid and

dehydroascorbic acid analysis, respectively.

Samples were injected onto a Nucleosil C18

column (Sigma, St. Louis, MO, USA) and were

eluted with 80/20, acetronitrile/phosphoric acid (60

mM; pH=2.0) v/v, at flow rate of 1 ml/min.

Ascorbic acid elution was monitored at 254 nm

by spectrophotometric analysis. For dehydroascor-

bic acid analysis, the fluorescence of quinoxaline

resulting from derivatisation procedure was detected

with excitation wavelength of 350 nm and emis-

sion wavelength of 425 nm. Calibration curves

were made for ascorbic acid and dehydroascorbic

acid.

2.9. Statistical treatment

Results are given as the mean of the measured

valuesFstandard deviation, except otherwise speci-

fied. For animal experiments, the data are represented

as the meanFstandard deviation and were tested for

significance using one-way ANOVA test.

3. Results

3.1. Time course of adjuvant-induced arthritis

The time course of AA in rats was determined by

the evolution of paw circumference and volume

measurements, as a function of time after disease

induction (Fig. 2). Swelling with erythema was

0 5 10 15 20 25

-10

-5

0

5

10

15

20

25 naive rats control rats

Bod

y w

eigh

t cha

nge

(%)

Time after arthritis induction (day)

Fig. 3. Body weight variation during AA time course in naive rats

(non-induced animals) and in control rats (induced but untreated

animals), as a function of time after bacteria injection into one paw

A B

0

50

100

150

200

250

0 5 10 15 200

50

100

150

200

250

20

40

60

80

100

0 5 10 15 200

20

40

60

80

Incr

ease

in p

aw c

ircum

fere

nce

(%)

Incr

ease

in p

aw v

olum

e (%

)

0

50

100

150

200

250

0 5 10 15 200

50

100

150

200

250

0

20

40

60

80

100right paw right paw

left paw left paw

control ratsnaive rats

control ratsnaive rats

control ratsnaive rats

control ratsnaive rats

0 5 10 15 200

20

40

60

80

100

Time after adjuvant arthritis induction (day) Time after adjuvant arthritis induction (day)

Fig. 2. (A) Circumference of the induced (upper panel) and non-induced (lower panel) paws and (B) the corresponding paws volume (induced:

upper panel) and (non-induced lower panel) as a function of time after M. butyricum injection, expressed as the percent increase of the

corresponding parameter, relative to the induction day 0. For comparison, the results measured with naive rats were plotted in all the figures.

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434426

evident within 1 day in the induced paw. Both

circumference and volume of adjuvant-injected right

paw start to increase on the 1st day after induction.

The induced right paw volume more than doubled and

the paw circumference increased in parallel by

approx. 50% whereas the untreated paw remained

unchanged. Between days 7 and 13 a quasi-plateau

was observed, at about 50% for circumference and at

about 175% for the volume. This initial increase is

followed by a step-up to another plateau, between the

days 13 and 18, at around 70% and 200%, respec-

tively. This secondary change may reflect a systemic

alteration that is also seen by the increased circum-

ference and volume of the non-induced, left paw from

practically zero to approx. 30% and 50%, respectively

(cf. Fig. 2, bottom panels).

The time course of body weight change is

illustrated in Fig. 3. For naive rats, the weight

increased with animal’s growth, whereas in control

arthritic rats a decrease in body weight was observed

after day 13, reaching a quasi-plateau around day 18

post induction.

The radiographic analysis for the induced and non-

induced paws, expressed as a radiographic score, are

.

Fig. 5. Radiographic images of a paw from a non-induced rat (naive: joi

around the joint) for comparison with the paws 21 days after arthritis indu

around the joint (oval arrows), and loss of joint space (arrow)), a rat induc

soft tissues swelling around joint (oval arrows) fluffy periosteal reaction (ar

and treated, for 6 days, with SOD (0.66 mg/kg BW) in mixed lipid vesicle

around joint (arrows)).

0 5 10 15 20 25 30 35 40 45 50

0

1

2

3 induced paw non induced paw

Rad

iogr

aphi

c sc

ore

Time after adjuvant arthritis induction (day)

Fig. 4. Radiographic score for the induced and non-induced paws

during AA development in untreated control rats, as a function of

time after the disease induction. Scale: 0=no bone damage; 1=soft

tissue swelling; 2=joint erosion; 3=bone erosion and osteophyte

formation. Results are the median of 3 rats score.

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434 427

presented in Fig. 4. Radiographic score increased to a

value of 1 (b24 h) after the disease induction. After

day 15 post-induction, an increase in the score was

observed in both paws, induced and non-induced. The

score level 3 was reached after day 40 for the induced

paw. At day 50, the score of non-induced paw reached

level 2, which is one point below the induced paw

result. The results show that it takes at least 15 days

before polyarthritis to be seen. Additional macro-

scopic evidences are the lesions in the tail and anterior

limbs. These confirm the increased severity of the

disease observed after day 13 (Figs. 2 and 3). The

disease becomes even worse by week 6 or 7 after

arthritis induction. Radiographic analysis also unveils

bone erosion in some rats after day 40.

Moreover, control rats were checked radiographi-

cally for lesions in the induced paws. Fig. 5 shows the

nt space well defined (arrow) with normal thickness of soft tissues

ction of an induced but untreated rat (control: soft tissues swelling

ed and treated with empty mixed lipid vesicles (Empty-Tfs treated:

row head) and loss of joint space (diamond arrow)) and a rat induced

s (SOD–Tfs treated: joint space well defined and normal soft tissues

control rats

Time after adjuvant arthritis induction (day)

-5 0 5 10 15 20 -5 0 5 10 15 200

10

20

30

40

50

60

70

80

90

100naive rats

lymphocytes neutrophils

lymphocytes neutrophils

Cel

l num

ber

(%)

0

10

20

30

40

50

60

70

80

90

100

Cel

l num

ber

(%)

Time (day)

Fig. 6. Leukocyte count, presented separately for lymphocytes and neutrophiles, for naive rats (left panel) and for control rats (right panel) as a

function of time. Time 0 corresponds to the day of arthritis induction on the control group.

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434428

lesions seen by radiography at day 21, in comparison

with observed signs on healthy rats (naive rats). For

control rats, oedema of soft tissues and loss of joint

space were observed.

The time course of white blood cells count for

naive and control rats was also determined, expressed

as the percentage of neutrophils and lymphocytes

relative to the starting value (Fig. 6). One day after

disease induction, the leucocytes number reached

neutrophil number. This value (50% of cells) then

remained unchanged for at least 18 days, i.e., till the

end of observation period. In contrast, the relative

count of lymphocytes and neutrophils for naive rats

was constant throughout the test period.

The time course of total thiols concentration in the

plasma of control rats is illustrated in Fig. 7. From the

average value of 447 AM on day 0 the total measured

0 2 4 6 8 10 12 14 16 18 20 220

100

200

300

400

500

[**][**][**]

[*] average

Tot

al th

iols

(µM

)

Time after adjuvant arthritis induction (day)

Fig. 7. Total thiols in the plasma of untreated, control rats during the

time course of the animal model. *pb0.05 vs. day 0, **pb0.001 vs.

day 1.

thiols concentration significantly ( pb0.05) decreased

to an average value of 377 AM on day 1. On day 14 a

strong decrease was observed, giving an average

value of 207 AM ( pb0.001). At days 16 and 21 the

averages were not significantly different from day 14.

Ascorbic acid and dehydroascorbic acid were also

quantified in rat sera during the time course of the

disease. The results show a decrease of ascorbate sera

levels followed by a comparable decrease of dehy-

droascorbate. The ratio of ascorbate/dehydroascorbate

was constant (10–13) during the time course of

inflammatory process.

According to the model time course results

presented in this work, barely 24 h after disease

induction, the primary inflammation response asso-

ciated with primary joint lesions, is established. Also,

at least 14–15 days are needed before polyarthritis is

reached. The appearance of secondary lesions,

expressed by polyarthritis, initiates a secondary stage

of the inflammatory response.

3.2. SOD prophylactic treatment: administration 1 h

prior to induction and evaluation during the first 24 h

The effect of prophylactic (1 h prior to disease

induction) e.c. administration of SOD-loaded Tfs on

the first 24 h (see Schedule 1, for details, in Fig. 1) of

the AA disease time course is illustrated in Fig. 8,

expressed in terms of paw volume change for the right

hind-paw. With the exception of the 6 h treated paw

result, the other data points for the SOD–Tfs treated

are significantly different from the respective control.

The mean increase of paw volume, 4 h after disease

0

[*][*]

Tot

al th

iols

(µM

)

control Tfs660 Tfs1000 Lip66

-25

0

25

50

75

[**][***][*]

[****][*]

OR

(%

)

100

200

300

400

500

[*][*]

Day 12 after adjuvant arthritis induction

average

-150

-125

-100

-75

-50

100

[**][***][*]

[****][*]

average

Fig. 9. Effect of SOD treatment on the total thiol concentration in ra

plasma (upper panel) and oedema regression (OR) based on paw

circumference measurement (lower panel). The animals were treated

from day 1 onwards and evaluated at day 12, in comparison with

untreated rats (control): SOD was applied epicutaneously in mixed

lipid vesicles (Tfs) at two different doses (0.66 mg SOD in mixed

lipid vesicles/kg BW (=Tfs660) or 1 mg SOD in vesicles/kg BW

(=Tfs1000). SOD-loaded liposomes, injected intravenously, were

used as the positive control (0.066 mg SOD/kg BW (=Lip66))

*pb0.01 vs. control, **pb0.05 vs. control, ***pb0.01 vs. Lip66

****pb0.05 vs. Lip66.

Control SOD-Tfs Control SOD-Tfs Control SOD-Tfs0

40

80

120

160

200

[**]

[*]

24 h6 h

Time after adjuvant arthritis injection

4 h

average

Incr

ease

in p

aw v

olum

e (%

)

Fig. 8. Increase in paw volume of right hind paw of control rats

(Control) and of SOD-loaded mixed lipid vesicles (SOD–Tfs)

prophylactically treated animals, using 0.66 mg SOD/kg BW.

*pb0.01 vs. 4 h Control, **pb0.05 vs. 24 h Control.

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434 429

induction, for the control and the treated rats was 94%

and 69%, respectively, and significantly different

( pb0.01). A day after the prophylactic administration,

the mean increase in paw volume for treated rats was

significantly lower than that observed for the control

rats ( pb0.05).

3.3. SOD therapeutic treatment: administration start-

ing at day 1 after induction and evaluation at day 12

after induction

The anti-inflammatory activity of SOD-loaded Tfs

e.c. administered on day 1 onwards post-induction of

AA, evaluated at day 12 after arthritis induction for

two SOD doses, is presented in Fig. 9. In this study, a

negative control (non-treated and induced animals)

group and a positive control (SOD-loaded long-

circulating liposomes) group were included (see

Schedule 2, for details, in Fig. 1). SOD exhibited

anti-inflammatory activity in all tested groups (Fig.

9—lower panel). The mean OR values, based on paw

circumference measurements for SOD–Tfs, was 27%

for 0.66 mg SOD/kg BW and 53% for 1 mg SOD/kg

BW. These are both significantly different ( pb0.01)

from the control (�89%). For SOD–liposomes

injected intravenously (the positive control) at a dose

of 0.066 mg SOD/kg BW, the mean OR was �26%,

and likewise significantly different from the control

( pb0.05). The result observed for the positive control

was significantly smaller than the OR observed for

epicutaneously applied SOD–Tfs at the tested doses

( pb0.01 and pb0.05 for 1 mg SOD/kg BW and 0.66

mg SOD/kg BW, respectively). The highest anti-

inflammatory effect was provided for the highest

tested SOD–Tfs dose. The highest SOD–Tfs dose

e.c. applied and the SOD-Lip (i.v. administered), also

increased the total thiol concentration in animal sera on

day 12 after arthritis induction. This is demonstrated in

Fig. 9 (upper panel). The change in thiol concentration

qualitatively parallels OR results and supports the

view that carrier delivered SOD has an anti-inflam-

matory effect in arthritis model. The observed differ-

ence is statistically significant ( pb0.01) for 1 mg SOD/

kg BW in mixed lipid vesicles on the skin and for

t

.

,

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434430

0.066 mg SOD/kg BW in injected liposomes, com-

pared to control rats.

Radiographic images (Fig. 5) of induced paws of

rats treated with SOD-loaded Tfs (SOD–Tfs treated)

showed a good preservation of the joint space and of

soft tissues around the joint. This contrasts with the

strong alterations observed on the induced paw of the

control or placebo treated (Empty-Tfs treated) animals

where soft tissues swelling around the joints, fluffy

periostal reaction and loss of joint space were

observed.

3.4. SOD prophylactic treatment: administration

starting at day 3 prior to induction and evaluation

at day 12 after induction

The anti-inflammatory activity of one selected dose

of SOD-loaded Tfs e.c. applied, starting before

disease induction (3 days before induction—protocol

A) and starting after disease induction (1 day after

induction—protocol B) were studied in parallel (for

details, see Schedule 3 in Fig. 1).

The increase in the paw circumference, at day 12,

is shown in Fig. 10. This parameter allows a

comparison of the initial and final paw circumference,

regardless of when the treatment began.

control A B0

10

20

30

40

50

[**]

[*]

Day 12 after adjuvant arthritis induction

average

Incr

ease

in p

aw c

ircum

fere

nce

(%)

Fig. 10. Increase in the induced paw circumference 12 days after

arthritis induction relative to the value measured on induction day 0.

Epicutaneous administration of 1 mg SOD in mixed lipid vesicles

per kg body weight for 7 days, starting the treatment on day �3

(=A); or epicutaneous administration of 1 mg SOD in mixed lipid

vesicles per kg body weight for 5 days, starting the treatment on day

1 (=B). Induced but untreated animals were used as control

(=control). *pb0.01 vs. control, **pb0.05 vs. control.

The average increase in paw circumference

observed for protocol A (mean 8%) is significantly

different ( pb0.01) from the control (mean 25%). For

control and protocol B (mean 15%), the averages were

also significantly different ( pb0.05). The smaller

increase in paw circumference and implicitly in paw

oedema, observed when the treatment began on day�3

showed a prophylactic effect of SOD-loaded Tfs in the

treatment of inflammatory process.

Ascorbic acid and dehydroascorbic acid were also

quantified in rat sera of treated (protocols A and B)

and of control rats, at day 12 after induction. As

referred to, in Section 3.1, the ratio of ascorbate/

dehydroascorbate for control rats was in the range of

10–13, during the time course of inflammatory

process. For treated groups this ratio was in the range

of 14–16 at day 12 after induction.

The results of anti-inflammatory effect of SOD-

loaded Tfs administered either prophylactically (Figs.

8 and 10) or therapeutically (Figs. 9 and 10), support

the view that SOD is transported from the surface of

the skin to the systemic circulation, mediated by the

ultradeformable carriers.

4. Discussion

The data presented in this work described the

adjuvant arthritis model in a refined way and

provides evidence for the anti-inflammatory effect

of SOD incorporated in a special type of colloidal

carrier, Tfs.

The studies performed in the first part of the work

demonstrated that a consistent animal model is

relevant for the definition of treatment strategies and

for the evaluation of therapeutic efficacy. Different

stages of evolution of the adjuvant arthritis disease

model were also observed in this work, as previously

reported by other authors [2]. One day after disease

induction physical, haematological and biochemical

parameters revealed the primary inflammatory

response. Based on white blood cells profile of the

animals tested (Fig. 6), the establishment of the

primary inflammatory response occurred during the

first 24 h. When Wistar rats from other source were

used, the approximation of the cell counts (~50% of

cells) was only observed at day 7 (data not shown).

These results could explain the variability of the time

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434 431

course of AA disease observed with rats with different

immunological background.

The sudden alteration of physical parameters

observed in our results at day 13, namely the sharp

increase in paw circumference and volume (Fig. 2)

and the decrease of body weight (Fig. 3), suggest that

day 13 may be the beginning of a second stage where

the secondary reactions began. Lack of mobility, as a

consequence of the joint disease, could lead to a

decrease of food intake due to deficient accessibility

to the food. Usually arthritic animals fail to gain

weight compared to naive animals [24,26]. The

decrease of body weight and the quasi-plateau

observed for paw volume and circumference, after

day 13 until day 21, suggests that the second stage is

established within that period of time. Radiographic

changes with AA model were reported by other

authors [26,46]. In the present work, the first signs

of polyarthritis, observed by radiographic evaluation,

appear after day 15. Changes in joint soft tissues and

phenomena such as loss of joint space were also

assessed by radiographic analysis. Radiographic

analysis also showed fluffy periostal reaction, at day

21, for induced empty-Tfs treated rats. The radio-

graphic score used for evaluation from days 0 to 51,

confirmed previous conclusions that the initiation of

the second stage of the disease occurs after day 13.

The radiographic findings are reinforced by the

macroscopic evidences of lesions in the tail and both

anterior limbs, which increased from days 13 to 51.

The immunological response is determinant for an

animal model, such as AA, and consequently affects

the parameters used for evaluation of the treatment

with anti-inflammatory drugs. The immunological

response and the evolution of physical and biochem-

ical parameters observed in the animals used in our

study, led to the definition of the treatment schedules,

namely the definition of day 1 after adjuvant induction

as the date for the beginning of the treatment.

The quantification of the levels of both total thiols

in rat plasma and ascorbic and dehydroascorbic acids

in rat sera, complements the physical evaluation of

disease progression (Fig. 7), and corroborates our

conclusions, namely the definition of day 1 as the

initiation of the disease and its severity increase after

day 13. Thiols consumption by the ROS produced

during the AA course can explain the decrease of this

parameter. This is expected to occur when the primary

response takes place, i.e., during the first 24 h and up

to the establishment of a chronic status, after day 13.

In our earlier studies [3], paw swelling was

described to be typically observed from day 7, which

dictated that the treatment begins at day 7 after

induction. In this study, paw swelling was seen after

the first 24 h. Other parameters evaluated during the

time course of the disease reinforced the paw swelling

measurement findings that lead to the definition of

treatment protocols beginning at day 1. Out of the

numerous possibilities for treatment schedules, the

therapeutic protocol used in this work was started

before the polyarthritis establishment.

The second part of this work demonstrated the

therapeutic efficacy of epicutaneous application of

SOD–Tfs (evaluated by physical and biochemical

parameters) with special emphasis for treatments

starting at day 1 and lasting until day 12, i.e.,

throughout the primary stage of the disease. The

ability of SOD–Tfs to interfere with the establishment

of the disease, after prophylactic application of the

formulation, was also observed. The prophylactic

effect of the treatment on the primary inflammatory

response (b24 h), was evaluated by measurement of

the paw volume, as it expressed more reliably the

changes of soft tissues that occurred during initial

stage of the disease. The highest prophylactic effect

was observed 5 h after application. According to our

results of biodistribution with radiolabelled-SOD in

Tfs [39] and other Tfs carried drugs [47], a lag time

(no systemic delivery of epicutaneous applied mate-

rial) of 4 h after application is observed, which

supports our present results.

SOD–Tfs evidenced anti-inflammatory effect dur-

ing the first stage of the disease, when the primary

response takes place, i.e., from day 1 and up to the

establishment of a chronic status, as we can conclude

from the results of treatment starting at day 1, after

induction, and lasting until day 12. The anti-inflam-

matory effect indicates good enzyme bioavailability.

The best result obtained for OR indicated in parallel

smaller thiols consumption. Pharmacological SOD

activity, evaluated by means of physical and bio-

chemical parameters, was achieved after efficient

transdermal approach.

Treatment initiated 3 days before induction was

able to inhibit the progression of the primary stage of

the disease. This prophylactic effect was demonstrated

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434432

by an even smaller increase in paw circumference

than the one obtained with the treatment initiated after

induction (1 day).

Despite lack of information on the potentialities of

prophylactic use of anti-inflammatory drugs, in differ-

ent inflammatory processes, some evidence exist for

the beneficial use in different diseases [48–50]. For

example, orthopaedic surgical procedures that can

lead to ROS mediated injuries, could benefit from the

prophylactic administration of antioxidant enzymes.

The use of SOD for clinical application still

receives great interest and attention. However the

limitation to the injection route of drug administration

has not yet been overcame. In this work, high doses of

enzyme were used, based on our previous biokinetic

and biodistribution tests [39], to achieve systemic

effective therapeutic levels after transdermal delivery.

The studies presented herein confirm the transdermal

delivery of catalytic active molecules. These findings

are mostly evidenced by the evolution of the physical

parameters discussed above and also by radiographic

evaluation that show a differential beneficial effect of

SOD–Tfs in the treatment of adjuvant induced rats.

The good preservation of the joint space and of the

soft tissues around the joint of the paws of animals

treated with SOD–Tfs, observed radiographically at

day 21, showed a systemic action of the enzyme and

consequently, the ability of ultradeformable vesicles

to mediate the transport of macromolecules across the

skin barrier.

Biochemical parameters could also help to evaluate

the evolution of the disease. A trend to a higher ratio

of ascorbate/dehydroascorbate for antioxidant enzyme

treated animals in comparison with control rats, points

to a possible evaluation of the improvement of

inflammatory status by using these biochemical

parameters. A relationship between the antioxidant

defence markers levels, particularly total thiols in the

blood, and SOD therapy was observed.

This work contributes to a novel approach for

antioxidant enzyme delivery. Based on the results

obtained, different therapeutic protocols can be

applied in further studies. The results reported show

for the first time that SOD-containing Tfs applied

non-occlusively onto the intact skin are effective in

the treatment of adjuvant arthritis. Our results also

show that the transport of SOD by means of specially

designed colloidal carriers can deliver therapeutically

active enzyme from the outer skin surface to the

systemic circulation.

5. Conclusion

It can be concluded from the first part of the

present study that a careful evaluation of the adjuvant

arthritis model in rats is relevant for the design of a

study of anti-inflammatory therapeutic activity per-

formed in a defined rat strain. Well defined progres-

sion of disease described by selected evaluation

parameters allows the establishment of treatment

protocols and to gain insight into the studies of

evaluation of the therapeutic efficacy of SOD trans-

dermal delivery using Tfs. The amelioration of disease

symptoms on animal treated with SOD–Tfs shows

that epicutaneous application of SOD in especially

developed mixed lipid vesicles can play a significant

role in the reduction of inflammation, in the adjuvant

arthritis model. The therapeutic approach used shows

practical and therapeutic advantages of the non-

invasive transdermal transport of antioxidant enzymes

in comparison with invasive administration, contribu-

ting to an innovative approach in the field of the

protein transdermal delivery.

Acknowledgements

We are grateful to Prof. G. Cevc for the helpful

discussion of the results and to Dr. M.J. Costa Ferreira

for revising the manuscript. J. Faustino is acknowl-

edged for the animal experiments technical support.

This work was financially supported by the project

POCTI/ 1999/FCB/35787 from Fundacao para a

Ciencia e para a Tecnologia.

References

[1] M.L. Corvo, O.C. Boerman, W.J. Oyen, J.C. Jorge, M.E. Cruz,

D.J. Crommelin, G. Storm, Subcutaneous administration of

superoxide dismutase entrapped in long circulating liposomes:

in vivo fate and therapeutic activity in an inflammation model,

Pharm. Res. 17 (2000) 600–606.

[2] A. Vaille, G. Jadot, A. Elizagaray, Anti-inflammatory activity

of various superoxide dismutases on polyarthritis on the Lewis

rat, Biochem. Pharmacol. 39 (1990) 247–255.

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434 433

[3] M.L. Corvo, J.C.S. Jorge, R. van’t Hof, M.E.M. Cruz, D.J.A.

Crommelin, G. Storm, Superoxide dismutase entrapped in

long-circulating liposomes: formulation design and therapeutic

activity in rat adjuvant arthritis, Biochim. Biophys. Acta 1564

(2002) 227–236.

[4] M.L. Corvo, O.C. Boerman, W.J.G. Oyen, L. Van Bloois,

M.E.M. Cruz, D.J.A. Crommelin, G. Storm, Intravenous

administration of superoxide dismutase entrapped in long

circulating liposomes: II. In vivo fate in a rat model of

adjuvant arthritis, Biochim. Biophys. Acta 1419 (1999)

325–334.

[5] M.M. Gaspar, M.B. Martins, M.L. Corvo, M.E. Cruz, Design

and characterization of enzymosomes with surface-exposed

superoxide dismutase, Biochim. Biophys. Acta 1609 (2003)

211–217.

[6] G. Cevc, G. Blume, A. Sch7tzlein, D. Gebauer, The skin: a

pathway for systemic treatment with patches and lipid-based

agent carriers, Adv. Drug Del. Rev. 18 (1996) 349–378.

[7] G. Cevc, G. Blume, A. Sch7tzlein, Transfersomes mediated

transepidermal delivery improves the regio-specificity and

biological activity of corticosteroids in vivo, J. Control.

Release 45 (1997) 211–226.

[8] G. Cevc, A. Sch7tzlein, H. Richardsen, Ultradeformable lipid

vesicles can penetrate the skin and other semi-permeable

barriers intact. Evidence from double label CLSM experiments

and direct size measurements, Biochim. Biophys. Acta 1564

(2002) 21–30.

[9] S.I. Simoes, C.M. Marques, M.E.M. Cruz, G. Cevc, M.B.F.

Martins, The effect of cholate on solubilisation and perme-

ability of simple and protein-loaded phosphatidylcholine/

sodium cholate mixed aggregates designed to mediate trans-

dermal delivery of macromolecules, Eur. J. Pharm. Biopharm.

58 (2004) 509–519.

[10] G. Cevc, Transdermal drug delivery of insulin with ultra-

deformable carriers, TransfersomesR, Clin. Pharmacokinet. 42

(2003) 461–474.

[11] G. Cevc, G. Blume, Biological activity and characteristics of

triamcinolone-acetonide formulated with the self-regulating

drug carriers, Transfersomes, Biochim. Biophys. Acta 1614

(2003) 156–164.

[12] K. Vorauer-Uhl, E. Furnschlief, A. Wagner, B. Ferko, H.

Katinger, Topically applied liposome encapsulated superoxide

dismutase reduces postburn wound size and edema formation,

Eur. J. Pharm. Sci. 17 (2001) 63–67.

[13] S. Cuzzocrea, D.P. Riley, A.P. Caputi, D. Salvemini, Anti-

oxidant therapy: a new pharmacological approach in shock,

inflammation, and ischemia/reperfusion injury, Pharmacol.

Rev. 53 (2001) 135–159.

[14] T. Langenegger, B.A. Michel, Drug treatment for rheumatoid

arthritis, Clin. Orthop. Relat. Res. 366 (1999) 22–30.

[15] E. Brahn, Animal models of rheumatoid arthritis. Clues to

etiology and treatment, Clin. Orthop. Relat. Res. 265 (1991)

42–53.

[16] C.M. Pearson, F.D. Wood, Studies of polyarthritis and other

lesions induced in rats by injection of mycobacterial adjuvant:

I. General clinical and pathologic characteristics and some

modifying factors, Arthritis Rheum. 2 (1959) 440–459.

[17] R.A. Greenwald, Animal models for evaluation of arthritis

drugs, Methods Find. Exp. Clin. Pharmacol. 13 (1991)

75–83.

[18] B.M. Weichman, Rat adjuvant arthritis: a model of chronic

inflammation, in: J.Y. Chang, A.J. Lewis (Eds.), Pharmaco-

logical Methods of in the Control of Inflammation, Alan R.

Liss, New York, 1989, pp. 363–380.

[19] A. Franch, S. Cassany, C. Castellote, M. Castell, Adjuvant

arthritis pretreatment with type II collagen and Mycobacterium

butyricum, Immunobiology 186 (1992) 351–361.

[20] A.M. Michelson, K. Puget, G. Jadot, Anti-inflammatory

activity of superoxide dismutases: comparison of enzymes

from different sources in different models in rats: mechanism

of action, Free Radic. Res. Commun. 2 (1986) 43–56.

[21] M.L. Corvo, M.B. Martins, A.P. Francisco, J.G. Morais,

M.E.M. Cruz, Liposomal formulations of Cu,Zn-superoxide

dismutase: physico-chemical characterization and activity

assessment in an inflammation model, J. Control. Release 43

(1997) 1–8.

[22] K.F. Swingle, L.W. Jaques, D.C. Kvam, Differences in the

severity of adjuvant arthritis in four strains of rats, Proc. Soc.

Exp. Biol. Med. 132 (1969) 608–612.

[23] B.B. Newbould, Chemotherapy of arthritis induced in rats by

mycobacterial adjuvant, Br. J. Pharmacol. 21 (1963) 127–136.

[24] F.R. Cochran, J. Selph, P. Sherman, Insights into the role of

nitric oxide in inflammatory arthritis, Med. Res. Rev. 16

(1996) 547–563.

[25] M.L. Graeme, E. Fabry, E.B. Sigg, Mycobacterial adjuvant

periarthritis in rodents and its modification by antiinflamma-

tory agents, J. Pharmacol. Exp. Ther. 153 (1966) 373–380.

[26] D.S. Fletcher, W.R. Widmer, S. Luell, A. Christen, C.

Orevillo, S. Shrenik, D. Visco, Therapeutic administration of

a selective inhibitor of nitric oxide synthase does not

ameliorate the chronic inflammation and tissue damage

associated with adjuvant-induced arthritis in rats, J. Pharma-

col. Exp. Ther. 284 (1998) 714–721.

[27] D. Quivy, J. Neve, J. Fontaine, W. Wasowick, J.P. Famaey, A.

Peretz, Trace elements status and inflammation parameters

during chronic indometacin treatment in adjuvant arthritic rats,

Biol. Trace Elem. Res. 47 (1995) 209–218.

[28] B.S. Bull, J.C. Westengard, M. Farr, P.A. Bacon, P.J. Meyer, J.

Stuart, Efficacy of tests used to monitor rheumatoid arthritis,

Lancet 2 (1989) 965–967.

[29] D.M.F.M. van der Heidje, Disease activity and outcome in

rheumatoid arthritis. A methodological study, PhD Thesis,

1990.

[30] I. Kushner, C-reactive protein and the acute-phase response,

Hosp. Pract. 25 (1989) 21–28.

[31] I.G. Otterness, The value of C-reactive protein measurement

in rheumatoid arthritis, Semin. Arthritis Rheum. 24 (1994)

91–104.

[32] A.A. Kheir Eldin, M.A. Hamdy, A.A. Shaheen, T.K. Motawi,

H.M. Abd el Gawad, Effect of vitamin C administration in

modulating some biochemical changes in arthritic rats, Pharm.

Res. 26 (1992) 357–366.

[33] J. Lykkesfeldt, T.M. Hagen, V. Vinarsky, B.N. Ames, Age-

associated decline in AA concentration, recycling, and biosyn-

S.I. Simoes et al. / Journal of Controlled Release 103 (2005) 419–434434

thesis in rat hepatocytes—reversal with (R)-alpha-lipoic acid

supplementation, FASEB J. 12 (1998) 1183–1189.

[34] S.I.D. Simoes, C.V. Eleuterio, M.E.M. Cruz, M.L. Corvo,

M.B. Martins, Biochemical changes in arthritic rats: dehy-

droascorbic and ascorbic acid levels, Eur. J. Pharm. Sci. 18

(2003) 185–189.

[35] R.H. Davis, K.Y. Rosenthal, L.R. Cesario, G.R. Rouw,

Vitamin C influence on localized adjuvant arthritis, J. Am.

Podiatr. Med. Assoc. 80 (1990) 414–418.

[36] E.J. Lewis, J. Bishop, S.J. Aspinall, A simple inflammation

model that distinguishes between the actions of anti-inflam-

matory and anti-rheumatic drugs, Inflamm. Res. 47 (1998)

26–35.

[37] G. Jadot, A. Vaille, J. Maldonado, P. Vanelle, Clinical

pharmacokinetics and delivery of bovine superoxide dismu-

tase, Clin. Pharmacokinet. 28 (1995) 17–25.

[38] A. Conforti, P. Caliceti, L. Sartore, A. Schiavon, F. Veronese,

G.P. Velo, Anti-inflammatory activity of monomethoxypoly-

ethylene glycol superoxide dismutase on adjuvant arthritis in

rats, Pharmacol. Res. 23 (1991) 51–56.

[39] S.I. Simoes, C.M. Marques, M.E.M. Cruz, G. Cevc, M.B.

Martins, Study of the tissue distribution of SOD-loaded

Transfersomes, Proceedings of 8th Liposome Research Days

Conference, 2002, p. 67.

[40] C. Kirby, G. Gregoriadis, Dehydration–rehydration vesicles: a

simple method for high yield drug entrapment in liposomes,

Biotechnology 2 (1984) 978–984.

[41] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein

measurement with the folin phenol reagent, J. Biol. Chem. 193

(1951) 265–275.

[42] S. Cuzzocrea, E. Mazzon, C. Bevilaqua, G. Costantino, D.

Britti, G. Mazzullo, A. De Sarro, A.P. Caputi, Cloricromene, a

coumarine derivative, protects against collagen-induced arthri-

tis in Lewis rats, Br. J. Pharmacol. 131 (2000) 1399–1407.

[43] G. Cevc, Transfersomes, liposomes and other lipid suspen-

sions on the skin: permeation enhancement, vesicle penetra-

tion, and transdermal drug delivery, Crit. Rev. Ther. Drug Carr.

Syst. 13 (1996) 257–388.

[44] M.L.T.A.R. Corvo, Liposomes as delivery systems for

superoxide dismutase in experimental arthritis, PhD thesis,

1998.

[45] H.S.P.C. Marinho, Metabolismo do glutationo no fıfado de

rato normal e no fıgado de rato hepatomizado, PhD Thesis,

1995.

[46] J.R. Connor, P.T. Manning, S.L. Settle, W.M. Moore, G.M.

Jerome, R.K. Webber, F.S. Tjoeng, M.G. Currie, Suppression

of adjuvant-induced arthritis by selective inhibition of

inducible nitric oxide synthase, Eur. J. Pharmacol. 273

(1995) 15–24.

[47] G. Cevc, D. Gebauer, J. Stieber, A. Sch7tzlein, G. Blume,

Ultraflexible vesicles, Transfersomes, have an extremely low

pore penetration resistance and transport therapeutic amounts

of insulin across the intact mammalian skin, Biochim.

Biophys. Acta 1368 (1998) 201–215.

[48] V. Strand, L.S. Simon, Low dose glucocorticoids in early

rheumatoid arthritis, Clin. Exp. Rheumatol. 21 (2003)

S186–S190.

[49] R.W. Dubois, G.Y. Melmed, J.M. Henning, L. Laine,

Guidelines for the appropriate use of non-steroidal anti-

inflammatory drugs, cyclo-oxygenase-2-specific inhibitors

and proton pump inhibitors in patients requiring chronic

anti-inflammatory therapy, Aliment. Pharmacol. Ther. 19

(2004) 1323–1324.

[50] M.P. Holzer, H.P. Sandoval, L.G. Vargas, T.J. Kasper, D.T.

Vroman, D.J. Apple, K.D. Solomon, Evaluation of preoper-

ative and postoperative prophylactic regimens for prevention

and treatment of diffuse lamellar keratitis, J. Cataract Refract.

Surg. 30 (2004) 195–199.