www.elsevier.com/locate/brainres

Brain Research 1015 (2004) 145–153

Research report

Intrathecal interleukin-1h administration induces thermal hyperalgesia

by activating inducible nitric oxide synthase expression

in the rat spinal cord

Chun-Sung Sunga,b, Zhi-Hong Wenc, Wen-Kuei Changb, Shung-Tai Hoc, Shen-Kou Tsaib,Yi-Chen Changc, Chih-Shung Wongc,*

aGraduate Institute of Medical Sciences, National Defense Medical Center, Taipei, TaiwanbDepartment of Anesthesiology, Veterans General Hospital-Taipei, Taipei, Taiwan

cDepartment of Anesthesiology, Tri-Service General Hospital and National Defense Medical Center,

#325, Section 2, Chenggung Road, Neihu 114, Taipei, Taiwan

Accepted 20 April 2004

Available online

Abstract

The effect of the pro-inflammatory cytokine interleukin-1h (IL-1h) on the inducible nitric oxide synthase-nitric oxide (iNOS-NO) cascadein nociceptive signal transduction was examined in the intact rat spinal cord. All rats were implanted with an intrathecal (i.t.) catheter; some

were also implanted with an i.t. microdialysis probe. The paw withdrawal latency to radiant heat was used to assess thermal hyperalgesia. The

iNOS protein expression in the spinal cord dorsal horn was examined by western blot analysis and NOS activity assay. NO production in the

CSF dialysate was also measured. IL-1h i.t. (100 ng) produced thermal hyperalgesia from 4 to 24 h after i.t. injection. The iNOS protein

expression was induced at 4 h after i.t. IL-1h injection, peaked at the 6th hour, and disappeared at 24 h. The iNOS activity showed a similar

time-dependent change as the iNOS protein expression. NO release increased by 1.1- to 1.9-fold between 4 and 12 h, also with a peak at the

6th hour, after i.t. IL-1h administration. Pretreatment with the iNOS inhibitor 1400W (10 Ag, i.t.) 1 h before i.t. IL-1h injection prevented all

the responses of IL-1h. Neither 1400W nor artificial CSF (aCSF) affected the thermal nociceptive threshold and NO production. These

results demonstrate that i.t. administration of IL-1h induced thermal hyperalgesia by activating the iNOS-NO cascade in the rat spinal cord.

On the basis of the present findings, we suggest that i.t. administration of iNOS inhibitors may have potential in the treatment of

inflammatory and neuropathic pain syndromes.

D 2004 Elsevier B.V. All rights reserved.

Theme: Sensory systems

Topic: Pain modulation: pharmacology; Pain: pathways

Keywords: Interleukin-1; Nitric oxide synthase; Nitric oxide; Hyperalgesia; Spinal cord

1. Introduction

Interleukin-1h (IL-1h), a pro-inflammatory cytokine, is

involved in the immune response and signal transduction

both in the periphery and the central nervous system (CNS)

[13,14,17,18,28,36]. It is released during inflammatory and

neuropathic pain conditions [8,34,39,43]. Samad et al. [34]

0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.brainres.2004.04.068

* Corresponding author. Tel.: +886-2-87927128; fax: +886-2-

87927127.

E-mail address: w82556@ndmctsgh.edu.tw (C.-S. Wong).

found that injection of complete Freund’s adjuvant (CFA)

into the rat’s hindpaw induced peripheral inflammation and

IL-1h upregulation (>10,000-fold) in the inflamed hindpaw

soon after CFA administration; this effect lasted for several

days and was correlated with the peripheral inflammation.

Moreover, in their study, the IL-1h concentration in the

CSF was increased 50- and 20-fold, at 2 and 4 h,

respectively, after CFA injection, and this was followed

by an increased cyclooxygenase-2 (COX-2) level in the

spinal cord. They also found that intrathecal (i.t.) injection

of IL-1h (50 ng) induced a 30-fold increase in COX-2

C.-S. Sung et al. / Brain Research 1015 (2004) 145–153146

mRNA levels in the rat spinal cord. Administration of IL-

1h produces hyperalgesia; in contrast, blockade of IL-1hreduces pain [10,30,32,34,35,37]. It is apparent that IL-1his an important mediator in the cross-talk between the

nervous and immune systems, and also participates in

modulation of the nociceptive threshold.

Nitric oxide (NO), a short-lived diffusible molecule of

great biological importance, plays a key role in host defense,

signal transduction, and neurotransmission. There is increas-

ing evidence that NO is also involved in nociceptive

processing after inflammation and neuropathy [26]. Three

isoforms of nitric oxide synthase (NOS) are known; two are

constitutively expressed (nNOS and eNOS), and the third is

inducible NOS (iNOS). Both nNOS (neuronal) and eNOS

(endothelial) are activated in response to physical stimuli,

and rapidly produce low levels of NO. In contrast, iNOS is

activated by stimuli such as IL-1h, and produces a high

level of NO over a long period of time [6,23]. Upregulation

of NOS expression and subsequent NO production in the

spinal cord contributes to pain following nociceptive stim-

ulation in animals [20,26,33,41,46]. i.t. administration of

NOS inhibitors significantly reduces nociceptive behavior in

response to peripheral inflammation [31,33,46]. Moreover,

IL-1h was found to increase both iNOS expression and NO

production in cultured cells [12,15]. Although both IL-1hand the NOS-NO cascade have been suggested to modulate

nociception, there is little information concerning the inter-

action between IL-1h and the iNOS-NO cascade in the

spinal cord. We hypothesized that, i.t. IL-1h causes a

markedly increased NO production via induction of iNOS

expression in the rat spinal cord. If an IL-1h induced

elevation in NO can be confirmed, it might reveal a

significant pathophysiological pathway of intrathecal IL-

1h-induced thermal hyperalgesia. The present study was

designed to examine the role of the NOS-NO cascade, and

its correspondence to nociceptive behaviors, in the spinal

cord of i.t. IL-1h-treated rats.

2. Materials and methods

2.1. Construction and implantation of the intrathecal

catheter and microdialysis probe

The i.t. catheter was constructed using a 9-cm long

polyethylene (PE5) tube and a 3.5-cm long silastic tube.

The silastic tube was inserted into the PE5 tube and sealed

with epoxy resin and silicon rubber. In addition, the spinal

microdialysis probe construction technique was modified

and adapted in previous studies [21,45]. The microdialysis

probe was constructed using two 7 cm PE5 tubes (0.008

inch inner diameter, 0.014 inch outer diameter) and a 4-cm

cuprophan hollow fiber (300 Am outer diameter, 200 Aminner diameter, 50 kDa molecular weight cut-off; DM-22,

Eicom, Kyoto, Japan). To make the probe firm enough for

implantation, a Nichrome-Formavar wire (0.0026 in.; A-M

System, Everret, WA, USA) was passed through a polycar-

bonate tube (194 Am outer diameter, 102 Am inner diameter;

0.7 cm in length) and the cuprophan hollow fiber (active

dialysis region), and connected to a PE5 catheter with epoxy

glue. The fiber was then bent in the middle section of the

cuprophan hollow fiber, forming a ‘‘U’’ shaped loop. The

two ends of the dialysis fiber, consisting of silastic tubes,

were sealed with silicon sealant. The dead space of the

dialysis probe was 8 Al. During in vitro measurements, the

recovery rate of the dialysis probe was 40% at an infusion

rate of 5 Al/min. Using this technique, it was possible to

measure levels of CSF amino acids for up to 12 days after

implantation.

2.2. Animal preparation and intrathecal drug delivery

Male Wistar rats (National Laboratory Animal Breed-

ing and Research Center, Taipei, Taiwan), weighing 350–

400 g, were implanted with an i.t. catheter either with or

without an i.t. microdialysis loop probe under chloral

hydrate anesthesia (350 mg/kg i.p.). After implantation,

all rats were allowed a 5-day recovery before use. All

rats were housed individually in cages with ad libitum

food and water, and maintained under a standard 12:12

h light–dark cycle at room temperature. For the nocicep-

tive behavioral test, western blotting and NOS activity

assay experiments, the rats were implanted with an i.t.

catheter, while, for the microdialysis study, they were also

implanted with a microdialysis probe. Rats with any

neurological deficits were excluded from the study. The

treatment and use of the animals conformed to the

guidelines of the International Association for the Study

of Pain [47], and were approved by the Animal Care and

Use Committee of our institute.

The dose-dependent effect of IL-1h, at various doses

(100, 200 and 500 ng, i.t.), on nociception and iNOS protein

expression in the rat spinal cord were examined. A reduction

of the PWL and an increase in iNOS expression in the rat

spinal cord were observed in a dose-dependent manner with

a similar time course between 4 to 24 h. Therefore, the

lowest dose of IL-1h, 100 ng, was used for the followed

experiments. The role of iNOS on IL-1h-induced thermal

hyperalgesia was examined by i.t. injection of 1400W (a

highly selective iNOS inhibitor); 50 Ag of 1400W was i.t.

injected in 10 rats, which unfortunately produced spontane-

ous nociceptive behaviors (e.g., vocalizations and flaccidity)

in three rats. The dose of 1400W was decreased to 10 Agand no spontaneous nociceptive behaviors were observed

with this dose in 10 rats; therefore, 10 Ag of 1400W was

chosen for the following experiments.

On the sixth day after i.t. catheter implantation, the rats

were assigned to one of four groups and various drugs

were injected i.t.: (1) the control aCSF group received 22

Al of artificial cerebrospinal fluid (aCSF; 151.1 mM Na+,

2.6 mM K+, 122.7 mM Cl�, 21.0 mM HCO3�, 0.9 mM

Mg2 +, 1.3 mM Ca2 +, 2.5 mM HPO42�, 3.5 mM dextrose,

C.-S. Sung et al. / Brain Research 1015 (2004) 145–153 147

bubbled with 5% CO2 in 95% O2, adjusted to a pH of

7.3); (2) the 1400W group received 10 Ag (2 Al) of 1400W(an iNOS inhibitor), followed by 20 Al of aCSF to flush

the catheter; (3) the IL group received 100 ng (2 Al) of IL-1h, followed by 20 Al of aCSF to flush the catheter; (4)

the 1400W+ IL group received 10 Ag (2 Al) of 1400W,

which was flushed with 10 Al of aCSF 1 h before i.t.

injection of IL-1h (100 ng, 2 Al), which in turn was

flushed with 10 Al of aCSF.

2.3. Behavioral assessment

The paw withdrawal latency (PWL) to radiant heat was

used to assess thermal hyperalgesia by the plantar test

(Biological Research Apparatus Type 7370, Plantar Test;

UGO Basile, Comerio, Italy). Rats were placed in plastic

cages on a glass platform and the heat source was

positioned directly beneath the right hind paw. The heat

intensity was adjusted to obtain an average PWL of

17.6F 0.4 s, and the cut-off time was set at 22F 0.4 s

to prevent tissue damage. The PWL was assessed prior to

i.t. drug delivery (baseline) and at 1, 2, 3, 4, 5, 6, 8, 10,

12, and 24 h after drug delivery.

2.4. Western blot analysis

The rats were killed at 2, 4, 6, 8, 10, 12, or 24 h after

drug injection, and the spinal cords were rapidly removed.

The dorsal part of the lumbar spinal cord was dissected and

immediately frozen and stored at � 80 jC until use. The

spinal cord samples were homogenized by sonication in a

cold lysis buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 2%

Triton X-100, 0.1 mM EDTA, 0.1 mM EGTA, 100 Ag/ml of

phenylmethylsulfonyl fluoride, 1 Ag/ml of aprotinin), then

centrifuged at 100,000� g at 4 jC for 35 min. The protein

content in the supernatant was determined by the Lowry

method [19]. An equal volume of sample buffer (2% SDS,

10% glycerol, 0.1% bromophenol blue, 2% 2-mercaptoe-

thanol, and 50 mM Tris–HCl, pH 7.2) was added into the

lysate samples (with equal protein content). The lysate

buffer, containing protein 24 Ag, was added and separated

on a 7.5% SDS polyacrylamide gel at 150 V for 75 min.

After transferring the proteins to a polyvinylidene difluoride

(PVDF) membrane (0.45 M pore size, Immobilonk-P,

Millipore, Bedford, MA. USA) in a transfer buffer (50

mM Tris–HCl, 380 mM glycine, 1% SDS and 20% meth-

anol) at 125 mA at 4 overnight, the membrane was blocked

with 5% non-fat dry milk in Tween-20 in Tris-buffered

saline (TTBS; 0.1% Tween-20, 20 mM Tris–HCl, 137 mM

NaCl, pH 7.4) for 30 min at room temperature. The

membranes were then incubated with mouse anti-iNOS

monoclonal antibodies (1:1,000 dilution; Transduction Lab-

oratories, Lexington, KY, USA) for 3 h at room temperature,

washed in TTBS for 10 min three times, blocked with 5%

non-fat dry milk in TTBS, and incubated with horseradish

peroxidase-conjugated secondary anti-mouse IgG antibody

(1:2,000 dilution; Transduction Laboratories, Lexington,

KY, USA) for another 1 h at room temperature. Finally,

bound antibody was detected by chemiluminescence (ECL;

Perkin-Elmer, Boston, MA, USA) using X-ray film

(KODAK X-OMAT LS, Kodak, Rochester, NY, USA).

The spinal cord h-actin protein was used as the control.

Densitometry was used to express the density of the band as

a relative density compared to the background. The relative

densities of the bands were calculated and standardized

using the density of the h-actin band.

2.5. NOS activity assay

NOS activity was determined by the stoichiometric

conversion of L-[3H]arginine to L-[3H]citrulline by NOS

assay kit (Cayman, Ann Arbor, MI, USA). The spinal cord

tissues were homogenized with a Dounce homogenizer

(Tenbroeck Tissue Grinder) in an ice-cold homogenization

buffer (25 mM Tris–HCl, pH 7.4, 1 mM EDTA, 1 mM

EGTA, 1 Ag/ml aprotinin, 100 Ag/ml phenylmethylsulfonyl

fluoride). The homogenates were centrifuged at 10,000� g

for 15 min at 4 jC, and the supernatants were extracted and

kept on ice. The protein concentration of the supernatant

was measured by the Lowry method with bovine serum

albumin as standard. Then, 10 Al of spinal cord extract

supernatant, corresponding to 15 Ag of total protein, was

incubated with reaction buffer provided in the kit and

supplemented with freshly prepared reduced NADPH and

L-[3H]arginine with a final volume of 50 Al for 30 min at

37 jC, either with or without NOS inhibitor NG-nitro-L-

arginine methyl ester hydrochloride (L-NAME HCl, 1

mM). The reaction was terminated by adding 400 Al of

the stop buffer (HEPES 50 mM, pH 5.5, and EDTA 5

mM), followed by 100 Al of the equilibrated resin to the

reaction mixture to bind the residual L-[3H]arginine. Then,

the mixtures were transferred to the spin cups in cup

holders, and L-[3H]arginine was separated from L-[3H]cit-

rulline by centrifugation at 12,500� g for 30 s. The

scintillation cocktail fluid was added to the eluate, contain-

ing L-[3H]citrulline, and then the radioactivity was quanti-

tated in a liquid scintillation counter (Beckman LS6500;

USA). iNOS activity was calculated as the difference in

activity between samples in the presence of L-NAME

(1 mM) or not. EGTA/EDTA (1 mM) was added to all

samples to chelate CaCl2 and calmodulin in the reaction

mixture. NOS activity was expressed as counts per minute

per Ag of total protein (cpm/Ag protein).

2.6. CSF sampling and NO analysis

One end of the microdialysis tube was connected to a

micro-syringe pump (CMA102, Acton, MA, USA) for

continuous aCSF infusion, and the other end was used to

collect the CSF dialysate. The microdialysis tube was

perfused with aCSF at a flow rate of 5 Al/min, and dialysates

were collected prior to each drug administration (baseline)

Fig. 1. Effects of various i.t. drugs treatments on the thermal nociceptive

threshold in rats. Values are normalized to the baseline (before i.t. drug

delivery) of each group. aCSF, artificial CSF (n= 6); IL, IL-1h 100 ng

(n = 8); 1400W, 1400W 10 Ag (n = 8); 1400W+ IL, 1400W 10 Agpretreatment 1 h prior to IL-1h 100 ng (n= 8). The PWL was obtained

by the plantar test. The maximal cut-off time was set at 22 s. The results are

illustrated as meanF S.E.M. *p< 0.05 indicated a significant difference

from baseline value in the same group, p< 0.05 compared to the control

group (aCSF i.t. injection) at the same time-point.

C.-S. Sung et al. / Brain Research 1015 (2004) 145–153148

and at 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 8, 10, 12, and 24

h after i.t. administration of drugs. The dialysate samples

were collected in polypropylene tubes and frozen at � 80

jC until used for NO analysis.

The total amount of NOx (NO +NO2�+NO3

�) in the

samples was used to determine the NO concentration in

the CSF. NOx in the dialysates was analyzed by using a

chemiluminescence detector (NOA 280, Sievers Instru-

ments, Boulder, CO, USA). The method was modified from

those described in previous studies [2,3]. The standard curve

was generated by using sodium nitrate and 0.1 M vanadium

(III) chloride, heated to 90 jC in a water-jacketed purge

vessel constantly bubbled with oxygen-free nitrogen. The

NOx signals from the chemiluminescence detector were

displayed and analyzed by using a computer-based data

recording and analyzing system. Background concentrations

of NOx with different agents in aCSF were measured and

subtracted. All standards and samples were analyzed in

duplicate.

2.7. Chemicals

1400W [N-(3-(aminomethyl)benzyl)acetamidine dihy-

drochloride] was purchased from Calbiochem (La Jolla,

CA, USA) and dissolved at 5 Ag/Al concentration in

aCSF. Rat recombinant IL-1h was purchased from R&D

Systems (Minneapolis, MN, USA) and dissolved at 50

ng/Al concentration in aCSF. L-[3H]arginine (1 ACi/Al)was purchased from Amersham Biosciences (Bucking-

hamshire, UK). Reduced NADPH (h-nicotinamide ade-

nine dinucleotide phosphate) was purchased from Sigma

(USA).

2.8. Data analysis

All data are presented as meanF S.E.M. Data were

analyzed by using one-way analysis of variance (ANOVA),

and followed by the post-hoc Tukey test. A p value < 0.05

was considered statistically significant.

3. Results

3.1. Effect of iNOS inhibitor 1400W on IL-1b-inducedthermal hyperalgesia

After i.t. injection of IL-1h (100 ng), thermal hyper-

algesia was observed with the PWL reduced to 50–60% of

baseline level between 4 and 24 h (Fig. 1). This hyper-

sensitivity to the heat was significantly increased at 4 h,

peaked at 6 h, and lasted for 24 h after IL-1h i.t.

administration. Either aCSF or 1400W (10 Ag) alone did

not affect the PWL to the heat, whereas i.t. pretreatment

with 1400W (10 Ag) 1 h prior to IL-1h (100 ng) injection

significantly attenuated the IL-1h-induced thermal hyper-

algesia (Fig. 1).

3.2. Effects of i.t. injection of IL-1b on iNOS protein

expression and NO production in the spinal cord

As shown in Fig. 2, i.t. injection of IL-1h (100 ng)

induced iNOS protein expression in the dorsal horn of the

rat lumbar spinal cords. The expression was first observed at

4 h, peaked at 6 h, and disappeared at 24 h after injection.

The time course of iNOS activity change was measured in

lumbar spinal cord after IL-1h 100 ng i.t. injection. As

shown in Fig. 3, there was no apparent iNOS activity in the

i.t. aCSF injected rats, while the appearance and increase of

iNOS activity were found after IL-1h i.t. injection in the rat

spinal cord. After 2 h, a marked time-dependent increase of

iNOS activity was observed between 4 and 12 h with a peak

at 6 h (5530F 856 cpm/Ag protein). The iNOS activity

returned to the baseline at 24 h after IL-1h injection. As

shown in Fig. 4, i.t. injection of IL-1h (100 ng) caused a

marked increase in NO concentration in the CSF. The NO

concentrations were increased by 1.1- to 1.9-fold of baseline

between 4 and 12 h, peaked at 6 h, then gradually fell to the

basal level 24 h after IL-1h injection.

3.3. Effect of iNOS inhibitor 1400W on iNOS protein

expression and NO production in the spinal cord

Neither aCSF nor the selective iNOS inhibitor 1400W i.t.

injection induced iNOS protein expression and activity in

the rat spinal cord (Figs. 2 and 3); however, i.t. pretreatment

with 1400W (10 Ag) 1 h before IL-1h i.t. injection signif-

icantly inhibited the IL-1h-induced iNOS activity to 1.8%

(4th hour) and 0.5% (6th hour) of that induced by IL-1hinjection (Fig. 3). Moreover, iNOS inhibitor 1400W pre-

Fig. 2. Intrathecal injection of IL-1h (100 ng) induced iNOS protein expression in the rat spinal cord. Time course of iNOS protein upregulation after i.t. IL-1hinjection was shown after western blot analysis (A) and densitometry of iNOS protein bands (B). Naive meant that rats had not received any surgical or

pharmacologic manipulation, whereas aCSF meant that rats were intrathecally injected with aCSF and were sacrificed 6 h after injection. Values of optical

density were expressed as meanF S.E.M. (n= 3) at each time-point in (B). *p< 0.05 indicated a significant difference from baseline value.

C.-S. Sung et al. / Brain Research 1015 (2004) 145–153 149

treatment also significantly inhibited IL-1h-induced NO

production, whereas i.t. injection of 1400W alone did not

affect NO release (Fig. 4). This inhibitory effect of the

1400W pretreatment indicates that iNOS is involved in the

signaling of spinal IL-1h-induced NO release.

Fig. 3. The time course of iNOS activity changes in lumbar spinal cord after i.t. in

Rats were sacrificed at baseline (prior to drug injection) and at 2, 4, 6, 8, 10, 1

determined by the stoichiometric conversion of L-[3H]arginine to L-[3H]citrulline

minute per Ag of total protein (cpm/Ag protein). All the data are presented as me

baseline value in the IL group, p< 0.05 compared to the 1400W+ IL group at th

4. Discussion

In the present study, thermal hyperalgesia was observed 4

h after IL-1h injection and lasted for 24 h. The iNOS

selective inhibitor 1400W prevented this thermal hyper-

jection of IL-1h (100 ng) with or without i.t. 1400W (10 Ag) pretreatment.

2 and 24 h, respectively after each drug injection and iNOS activity was

by using the Cayman NOS assay kit. The data are expressed as count per

anF S.E.M for 4–6 rats. *p< 0.05 indicated a significant difference from

e same time-point.

Fig. 4. Time-course of NO release into the CSF after i.t. various drugs treatment. i.t. administration of IL-1h 100 ng caused a marked increase in NO release. i.t.

injection of 1400W 10 Ag 1 h before IL-1h treatment dramatically decreases NO release. The NO concentration in the CSF dialysate was analyzed by using a

chemiluminescence detector. Values are normalized to the baseline (before i.t. drug delivery) for each group. Data are presented as meanF S.E.M for 6 rats.

*p< 0.05 compared to baseline in the same group, p< 0.05 compared to the 1400W group at the same time-point.

C.-S. Sung et al. / Brain Research 1015 (2004) 145–153150

algesia and the corresponding biochemical changes. I.t.

administration of IL-1h induced iNOS protein expression

and the subsequent NO release, in a time-dependent manner,

in the rat spinal cord. iNOS protein induction and activation

were observed at 4 h after IL-1h injection, peaked at 6 h,

then declined gradually, and disappeared at 24 h. This

increase in NO level in the CSF after IL-1h injection was

accompanied by an upregulation of iNOS protein expres-

sion. The present results show that i.t. injection of the

proinflammatory cytokine, IL-1h, induced hyperalgesia,

which was produced by iNOS protein induction and acti-

vation, and the subsequent NO production. i.t. pretreatment

with 1400W blocked the synthesis of NO and reduced the

thermal hyperalgesia which were induced by the i.t. IL-1hinjection in rats.

IL-1h was found to increase in the spinal cord in animal

models of neuropathy and peripheral inflammation

[8,34,39,40,43]. Samad et al. [34] found that injection of

CFA into the hindpaw of rats induced peripheral inflamma-

tion; IL-1h expression was increased at 2 and 4 h in the CSF

after peripheral inflammation induction, and this was fol-

lowed by an increasing level of COX-2 mRNA in the spinal

cord. i.t. injection of IL-1h (50 ng) induced an increase in

the COX-2 mRNA level in the rat spinal cord, which was

similar to the central effect of CFA-induced COX-2 upre-

gulation in the spinal cord. They also found that i.t.

administration of IL-1h receptor antagonist (IL-1ra, 6 Ag)or IL-1h-converting enzyme (ICE) inhibitor YVAD (0.5 Ag)blocked the induction of spinal COX-2 mRNA and the

central prostanoid production, and attenuated the mechani-

cal hypersensitivity produced by peripheral inflammation

[34]. They concluded that IL-1h was a major contributor in

CFA-induced inflammatory pain. Peripheral inflammation

upregulated central IL-1h expression [39]; in addition, i.t.

administration of IL-1ra and soluble tumor necrosis factor

receptor attenuated the mechanical allodynia following

peripheral neuropathy [40]. These results suggest that spinal

IL-1h may contribute to the development of nociceptor

hypersensitivity to the peripheral inflammation.

Peripheral IL-1h administration produces hyperalgesia in

rats [11,44], yet the nociceptive and anti-nociceptive effect

of central IL-1h are still debated. In the rats, researchers

have found that intracerebroventricular (i.c.v.) injection of

IL-1h exerts biphasic responses on thermal and mechanical

nociceptive thresholds depending upon the dosage. i.c.v.

injection of IL-1h causes hyperalgesia at lower doses, but in

contrast, causes analgesia at higher doses [5,24,25,29].

Different nociceptive behavioral responses to IL-1h i.t.

administration have also been reported. Falchi et al. [10]

reported thermal hyperalgesia in rats after i.t. IL-1h injec-

tion (at the dose of 50 or 500 pg) by the hot-plate test. In the

present study, we found that i.t. injection of recombinant rat

IL-1h (100 ng) produced thermal hyperalgesia in rats 4

h after injection and lasted for 24 h. However, Watkins et al.

[44] reported that i.t. injection of human recombinant IL-1h(50 ng) did not produce any effect on the nociceptive

threshold assessed by a modified tail-flick test in rats.

Moreover, Souter et al. [38] reported that i.t. administration

of murine recombinant IL-1h did not affect the nociceptive

threshold in rats, at dose of 100 ng in a 3-h observation. In

contrast, Ji et al. showed an anti-nociceptive effect after rat

recombinant IL-1h (10 and 100 ng) i.t. injection in rats

assessed by the plantar test, which appeared from 5 to 30

min after injection [16]. Our results were similar to some

C.-S. Sung et al. / Brain Research 1015 (2004) 145–153 151

studies but different from others. These controversial results

might be due to differences in experimental conditions, such

as the source of IL-1h, the methods used for thermal

threshold evaluation, the time point of observation on

thermal threshold, and the dose in i.t. delivery.

To date, evidence shows that different isoforms of NOS

may contribute to peripheral inflammatory pain. Osborne

and Coderre found that i.t. administration of either 3-bromo-

7-nitroindazole (selective nNOS inhibitor) or aminoguani-

dine (AG, selective iNOS inhibitor) effectively attenuated

carrageenan-produced thermal hyperalgesia in rats and that

iNOS contributed to the thermal hyperalgesia in the late

stage of carrageenan-induced inflammatory pain, while

nNOS played a role throughout the whole course [27].

Tao et al. found that intraplantar (ipl.) carrageenan injection

produced thermal hyperalgesia, which peaked at 4 h and

persisted for 24 h after carrageenan injection, in both wild

type and iNOS knockout mice, and the iNOS mRNA

expression in the lumbar spinal cord was markedly upregu-

lated at 24 h (late phase) in wild type mice [42]. i.t. pre-

treatment with selective iNOS inhibitor L-N-(1-iminoethyl)-

lysine had no effect in the early phase (2–6 h) but

significantly attenuated the carrageenan-produced reduction

of PWL in wild type mice in the late phase (24 h), and the

nNOS protein expression was significantly up-regulated at

24 h after carrageenan (ipl.) injection in the iNOS knockout

mice when compared to that in wild type mice [42]. i.t. pre-

treatment with the selective nNOS inhibitor, 7-nitroindazole

(7-NI), not only delayed the onset of thermal hyperalgesia in

the early phase but also significantly attenuated the thermal

hyperalgesia in the late phase after carrageenan treatment in

iNOS knockout mice. Therefore, Tao et al. [42] suggested

that iNOS might be sufficient, but not essential, for the late

phase of the carrageenan-induced thermal hyperalgesia.

Maihofner et al. [20] also found that both iNOS and nNOS

proteins were upregulated in the mouse spinal cord 168

h after injection (ipl.) of zymosan. Meller et al. [22] found

that i.t. injection of AG dose-dependently reduced the

thermal hyperalgesia produced by injection (ipl.) of zymo-

san, and they suggested that iNOS contributed to zymosan-

induced thermal hyperalgesia. In addition, Wu et al. [46]

demonstrated that intradermal injection of capsaicin pro-

duced mechanical allodynia in rats, and upregulated both

iNOS and nNOS protein expression in the rat spinal cord

from 20 to 150 min after capsaicin injection. In contrast, Tao

and Johns [41] found that intraperitoneal pretreatment of 7-

NI blocked formalin-induced long-term thermal hyperalge-

sia in rats and suppressed formalin-induced soluble guany-

late cyclase a1 subunit expression in the spinal cord as well.

This study suggested that nNOS played a role in formalin-

induced long-term thermal hyperalgesia.

IL-1h has been reported in inflammatory and neuro-

pathic pain conditions [8,34,40,43]. Our present study was

targeted at i.t. IL-1h-produced spinal iNOS activation and

nociceptive responses, and we found that i.t. iNOS inhib-

itor pretreatment significantly attenuated thermal hyper-

algesia and NO release. iNOS is not constitutively

expressed in vivo, and its expression is induced by lipo-

polysaccharide or cytokines. In contrast to nNOS and

eNOS, iNOS generates high concentrations of NO (n rather

than pmolar) and this level of synthesis is sustained for

hours, days, or even longer [6,23]. There is a lag phase of

several hours between cell activation and NO production,

reflecting the time taken for mRNA and protein synthesis

[6]. Casamenti et al. directly injected IL-1h into the right

nucleus basalis of rats and found that iNOS immunopos-

itive microglia surrounded the injection site at 7 days after

IL-1h injection; meanwhile, a transient increase in reactive

nitrogen intermediates (NO2�) was present 7 days after

injection [4]. Moreover, Maihofner et al. demonstrated an

increase of iNOS protein expression in the spinal cord at 7

days after injection (ipl.) of zymosan [20]. The latency and

time course of the induced iNOS expression in the spinal

cord may be related to the characteristics of the insults.

Upregulation of iNOS mRNA was detected in the spinal

cord 24 h after carrageenan injection (ipl.) in mice [42],

and iNOS protein upregulation in the spinal cord was

increased 20 min after intradermal injection of capsaicin

in rats [46]. Taken together, these results suggest that it

takes several hours to days for the induction of iNOS

mRNA and protein expression after stimulation. In the

present study, we intrathecally injected IL-1h in rats and

found an upregulation of iNOS protein in the rat spinal

cord. Compared with capsaicin-and carrageenan-induced

peripheral inflammatory pain models, upregulation of

iNOS protein in the rat spinal cord was detected 4 h after

i.t. administration of IL-1h in our present study, which was

later than that produced by intradermal injection of capsa-

icin, but earlier than that produced by carrageenan injection

(ipl.) [42]. The latency of iNOS protein expression in the

spinal cord induced by IL-1h injection (i.t.) might be due

to the time lag of intracellular signalling transduction,

activation and translocation of nuclear transcription factors,

gene transcription and translation, and spinal sensitization.

It is worth further investigation on the intracellular events

after IL-1h administration.

In in vivo experiments, i.t. IL-1h (10 ng) injection

upregulated the expression of iNOS mRNA in rat spinal

cords [22]; in addition, IL-1h (10 U) injected into the

nucleus basalis activated iNOS-immunopositive microglias

with the production of NO in the rats [4]. Similarly,

upregulation of iNOS protein expression and increase of

NO release from the spinal cord were also observed in

peripheral inflammatory pain models [20,46]. i.t. adminis-

tration of an iNOS inhibitor blocked the thermal hyper-

algesia and mechanical allodynia induced by peripheral

inflammation [22,27,46]. These reports suggest that the

central iNOS-NO cascade played a role in nociceptive

processing and that IL-1h activates this iNOS-NO signal

transduction pathway. In our present study, except for

thermal hyperalgesia, i.t. administered IL-1h, at doses of

100, 200 and 500 ng, did not induce any neurological deficit

C.-S. Sung et al. / Brain Research 1015 (2004) 145–153152

or abnormal behavioral changes in rats. In the present study,

by Western blot analysis and NOS enzyme activity assay,

IL-1h induced a significant expression of iNOS protein in

the spinal cord as well as NO release in the CSF. From these

results, we suggest that, in a similar temporal domain, i.t.

injection of IL-1h activated the spinal iNOS-NO cascade,

which induced the thermal hyperalgesia.

In a rat zymosan intra-articular injection-induced arthritis

pain model, pretreatment with 1400W (a highly selective

iNOS inhibitor) inhibited the zymosan-induced articular

incapacitation and nociception [7]. Beauregard et al. [1]

demonstrated a time-dependent expression of iNOS protein

and NO production in cultured rabbit lacrimal gland acinar

cells, which were significantly antagonized by co-incubation

with 1400W. Furthermore, 1400W was found to inhibit the

IL-1h-induced iNOS mRNA and protein expression, and

nitrite release in human bronchial epithelial cells [9]. Simi-

larly to our present study, i.t. injection of IL-1h activated the

iNOS-NO cascade and induced thermal hyperalgesia, and i.t.

pretreatment of 1400W 10 Ag (at this dose, no antinocicep-

tive effect was observed by itself) significantly inhibited the

IL-1h-mediated iNOS protein expression and NO release in

the rat spinal cord. The IL-1h-induced iNOS protein activity

was almost completely eliminated by the 1400W pretreat-

ment, and the IL-1h-induced thermal hyperalgesia was also

reversed.

In conclusion, we found that i.t. injection of IL-1h results

in activation of the iNOS-NO signal transduction cascade in

the rat spinal cord, which leads to thermal hyperalgesia.

Inhibition of iNOS by i.t. 1400W pretreatment significantly

reduced the IL-1h-induced iNOS-NO activation and ame-

liorated i.t. IL-1h-induced thermal hyperalgesia. On the

basis of the present study, we suggest that administration

of an i.t. iNOS inhibitor is a potential treatment for the

central iNOS-NO-mediated inflammatory and neuropathic

pain syndromes.

Acknowledgements

The authors wish to thank Dr. Yoshito Takano for his

helpful communications on the algesic effect of spinal IL-1hand Mr. Yao-Chang Chen for the statistical analysis. This

work was supported by grants from the National Science

Council (NSC 91-2314-B-075-070) and Ministry of De-

fense (DOD-93-2-08) of the Republic of China.

References

[1] C. Beauregard, P.C. Brandt, G.C. Chiou, Induction of nitric oxide

synthase and over-production of nitric oxide by interleukin-1h in cul-

tured lacrimal gland acinar cells, Exp. Eye Res. 77 (2003) 109–114.

[2] R.S. Braman, S.A. Hendrix, Nanogram nitrite and nitrate determina-

tion in environmental and biological materials by vanadium (III) re-

duction with chemiluminescence detection, Anal. Chem. 61 (1989)

2715–2718.

[3] P.A. Bush, N.E. Gonzalez, J.M. Griscavage, L.J. Ignarro, Nitric oxide

synthase from cerebellum catalyzes the formation of equimolar quan-

tities of nitric oxide and citrulline from L-arginine, Biochem. Biophys.

Res. Commun. 185 (1992) 960–966.

[4] F. Casamenti, C. Prosperi, C. Scali, L. Giovannelli, M.A. Colivicchi,

M.S. Faussone-Pellegrini, G. Pepeu, Interleukin-1h activates fore-

brain glial cells and increases nitric oxide production and cortical

glutamate and GABA release in vivo: implications for Alzheimer’s

disease, Neuroscience 91 (1999) 831–842.

[5] A. Coelho, J. Fioramonti, L. Bueno, Brain interleukin-1h and tumor

necrosis factor-alpha are involved in lipopolysaccharide-induced

delayed rectal allodynia in awake rats, Brain Res. Bull. 52 (2000)

223–228.

[6] J.W. Coleman, Nitric oxide in immunity and inflammation, Int.

Immunopharmacol. 1 (2001) 1397–1406.

[7] J.C. da S Rocha, M.E. Peixoto, S. Jancar, F. de Q Cunha, R. de A

Ribeiro, F.A. da Rocha, Dual effect of nitric oxide in articular in-

flammatory pain in zymosan-induced arthritis in rats, Br. J. Pharma-

col. 136 (2002) 588–596.

[8] J.A. DeLeo, R.W. Colburn, A.J. Ricknia,n, Cytokine and growth

factor immunohistochemical spinal profiles in two animal models of

mononeuropathy, Brain Res. 759 (1997) 50–57.

[9] L.E. Donnelly, P.J. Barnes, Expression and regulation of inducible

nitric oxide synthase from human primary airway epithelial cells,

Am. J. Respir. Cell Mol. Biol. 26 (2002) 144–151.

[10] M. Falchi, F. Ferrara, C. Gharib, B. Dib, Hyperalgesic effect of in-

trathecally administered interleukin-1 in rats, Drugs Exp. Clin. Res.

27 (2001) 97–101.

[11] S.H. Ferreira, B.B. Lorenzetti, A.F. Bristow, S. Poole, Interleukin-1has a potent hyperalgesic agent antagonized by a tripeptide analogue,

Nature 334 (1988) 698–700.

[12] J.D. Finder, J.L. Petrus, A. Hamilton, R.T. Villavicencio, B.R. Pitt,

S.M. Sebti, Signal transduction pathways of IL-1h-mediated iNOS in

pulmonary vascular smooth muscle cells, Am. J. Physiol., Lung Cell.

Mol. Physiol. 281 (2001) L816–L823.

[13] M.R. Heitmeier, M. Arnush, A.L. Scarim, J.A. Corbet, Pancreatic

beta-cell damage mediated by beta-cell production of interleukin-1.

A novel mechanism for virus-induced diabetes, J. Biol. Chem. 276

(2001) 11151–11158.

[14] D. Hildeman, D. Muller, Immunopathologic weight loss in intracra-

nial LCMV infection initiated by the anorexigenic effects of IL-1h,Viral Immunol. 13 (2000) 273–285.

[15] O.J. Igwe, J.N. Marray, A.S. Moolwaney, Interleukin 1-induced

cyclooxygenase and nitric oxide synthase gene expression in the rat

dorsal root ganglia is modulated by antioxidants, Neuroscience 105

(2001) 971–985.

[16] G.C. Ji, Y.Q. Zhang, F. Ma, G.C. Wu, Increase of nociceptive thresh-

old induced by intrathecal injection of interleukin-1h in normal and

carrageenan inflammatory rat, Cytokine 19 (2002) 31–36.

[17] F. Licastro, S. Pedrini, L. Caputo, G. Annoni, L.J. Davis, C. Ferri, V.

Casadei, L.M. Grimaldi, Increased plasma levels of interleukin-1,

interleukin-6 and a-1-antichymotrypsin in patients with Alzheimer’s

disease: peripheral inflammation or signals from the brain? J. Neuro-

immunol. 103 (2000) 97–102.

[18] T.J. Lin, R. Garduno, R.T. Boudreau, A.C. Issekutz, Pseudomonas

aeruginosa activates human mast cells to induce neutrophil transen-

dothelial migration via mast cell-derived IL-1a and h, J. Immunol.

169 (2002) 4522–4530.

[19] D.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein mea-

surement with the folin phenol reagent, J. Biol. Chem. 193 (1951)

265–275.

[20] C. Maihofner, C. Euchenhofer, I. Tegeder, K.F. Beck, J. Pfeilschifter,

G. Geisslinger, Regulation and immunohistochemical localization of

nitric oxide synthases and soluble guanylyl cyclase in mouse spinal

cord following nociceptive stimulation, Neurosci. Lett. 290 (2000)

71–75.

[21] M. Marsala, A.B. Malmberg, T.L. Yaksh, The spinal loop dialysis

C.-S. Sung et al. / Brain Research 1015 (2004) 145–153 153

catheter: characterization of use in the unanesthetized rat, J. Neurosci.

Methods 62 (1995) 43–53.

[22] S.T. Meller, C. Dykstra, D. Grzybycki, S. Murphy, G.F. Gebhart, The

possible role of glia in nociceptive processing and hyperalgesia in the

spinal cord of the rat, Neuropharmacology 33 (1994) 1471–1478.

[23] C. Nathan, Q.W. Xie, Regulation of biosynthesis of nitric oxide,

J. Biol. Chem. 269 (1994) 13725–13728.

[24] T. Oka, S. Aou, T. Hori, Intracerebroventricular injection of interleu-

kin-1h induces hyperalgesia in rats, Brain Res. 624 (1993) 61–68.

[25] T. Oka, S. Aou, T. Hori, Intracerebroventricular injection of interleu-

kin-1h enhances nociceptive neuronal responses of the trigeminal

nucleus caudalis in rats, Brain Res. 656 (1994) 236–244.

[26] K. Omote, K. Hazama, T. Kawamata, M. Kawamata, Y. Nakayaka,

M. Toriyabe, A. Namiki, Peripheral nitric oxide in carrageenan-in-

duced inflammation, Brain Res. 912 (2001) 171–175.

[27] M.G. Osborne, T.J. Coderre, Effects of intrathecal administration of

nitric oxide synthase inhibitors on carrageenan-induced thermal

hyperalgesia, Br. J. Pharmacol. 126 (1999) 1840–1846.

[28] N. Quan, E.L. Stern, M.B. Whiteside, M. Herkenham, Induction of

pro-inflammatory cytokine mRNAs in the brain after peripheral in-

jection of subseptic doses of lipopolysaccharide in the rat, J. Neuro-

immunol. 93 (1999) 72–80.

[29] J.J. Rady, J.M. Fujimoto, Confluence of antianalgesic action of di-

verse agents through brain interleukin-1h in mice, J. Pharmacol. Exp.

Ther. 299 (2001) 659–665.

[30] A.J. Reeve, S. Patel, A. Fox, K. Walker, L. Urban, Intrathecally

administered endotoxin or cytokines produce allodynia, hyperalgesia

and changes in spinal cord neuronal responses to nociceptive stimuli

in the rat, Eur. J. Pain 4 (2000) 247–257.

[31] A.K. Roche, M. Cook, G.L. Wilcox, K.C. Kajander, A nitric oxide

synthesis inhibitor (L-NAME) reduces licking behavior and Fos-la-

beling in the spinal cord of rats during formalin-induced inflamma-

tion, Pain 66 (1996) 331–341.

[32] B. Safieh-Garabedian, S. Poole, A. Allchorne, J. Winter, C.J. Woolf,

Contribution of interleukin-1h to the inflammation-induced increase

in nerve growth factor levels and inflammatory hyperalgesia, Br. J.

Pharmacol. 115 (1995) 1265–1275.

[33] T. Sakurada, A. Suhiyama, C. Sakurada, K. Tan-No, A. Yonezawa, S.

Sakurada, K. Kisara, Effect of spinal nitric oxide inhibition on cap-

saicin-induced nociceptive response, Life Sci. 59 (1996) 921–930.

[34] T.A. Samad, K.A. Moore, A. Sapirstein, S. Billet, A. Allchorne, S.

Poole, J.V. Bonventre, C.J. Woolf, Interleukin-1h-mediated induction

of COX-2 in the CNS contributes to inflammatory pain hypersensi-

tivity, Nature 410 (2001) 471–475.

[35] M. Schafers, J. Brinkhoff, S. Neukirchen, M. Marziniak, C. Sommer,

Combined epineurial therapy with neutralizing antibodies to tumor

necrosis factor-alpha and interleukin-1 receptor has an additive effect

in reducing neuropathic pain in mice, Neurosci. Lett. 310 (2001)

113–116.

[36] M.J. Schultz, A.W. Rijneveld, S. Florquin, C.K. Edwards, C.A.

Dinarello, T. van der Poll, Role of interleukin-1 in the pulmonary

immune response during Pseudomonas aeruginosa pneumonia, Am.

J. Physiol., Lung Cell. Mol. Physiol. 282 (2002) L285–L290.

[37] C. Sommer, S. Petrausch, T. Lindenlaub, K.V. Toyka, Neutralizing

antibodies to interleukin 1-receptor reduce pain associated behavior

in mice with experimental neuropathy, Neurosci. Lett. 270 (1999)

25–28.

[38] A.J. Souter, M.G. Garry, D.L. Tanelian, Spinal interleukin-1h reduces

inflammatory pain, Pain 86 (2000) 63–68.

[39] S.M. Sweitzer, R.W. Colburn, M. Rutkowski, J.A. DeLeo, Acute

peripheral inflammation induces moderate glial activation and spinal

IL-1h expression that correlates with pain behavior in the rat, Brain

Res. 829 (1999) 209–221.

[40] S. Sweitzer, D. Martin, J.A. DeLeo, Intrathecal interleukin-1 receptor

antagonist in combination with soluble tumor necrosis factor receptor

exhibits an anti-allodynic action in a rat model of neuropathic pain,

Neuroscience 103 (2001) 529–539.

[41] Y.X. Tao, R.A. Johns, Activation and up-regulation of spinal cord

nitric oxide receptor, soluble guanylate cyclase, after formalin injec-

tion into the rat hind paw, Neuroscience 112 (2002) 439–446.

[42] F. Tao, Y.X. Tao, P. Mao, C. Zhao, D. Li, W.J. Liaw, S.N. Raja, R.A.

Johns, Intact carrageenan-induced thermal hyperalgesia inmice lacking

inducible nitric oxide synthase, Neuroscience 120 (2003) 847–854.

[43] C.X. Wang, J.A. Olschowka, J.R. Wrathall, Increase of interleukin-1hmRNA and protein in the spinal cord following experimental trau-

matic injury in the rat, Brain Res. 759 (1997) 190–196.

[44] L.R. Watkins, E.P. Wiertelak, L.E. Goehler, K.P. Smith, D. Martin,

S.F. Maier, Characterization of cytokine-induced hyperalgesia, Brain

Res. 654 (1994) 15–26.

[45] Z.H. Wen, L.C. Yang, J.J. Wang, Y.C. Chang, C.H. Hsing, J.Y. Chen,

C.S. Wong, Intrathecal pertussis toxin induces thermal hyperalgesia:

involvement of excitatory and inhibitory amino acids, Neuroscience

116 (2003) 871–878.

[46] J. Wu, L. Fang, Q. Lin, W.D. Willis, Nitric oxide synthase in spinal

cord central sensitization following intradermal injection of capsaicin,

Pain 94 (2001) 47–58.

[47] M. Zimmermann, Ethical guidelines for investigations of experimen-

tal pain in conscious animals, Pain 16 (1983) 109–110.