www.elsevier.com/locate/brainres

Brain Research 1019 (2004) 255–258

Research report

Increased substance P and tumor necrosis factor-a level in the paws

following formalin injection in rat tail

Mauro Bianchia,*, Cataldo Martuccia, Gabriele Biellab, Paolo Ferrarioa, Paola Sacerdotea

aDepartment of Pharmacology, University of Milano, Via Vanvitelli 32, Milan 20129, Italyb Istituto di Bioimmagini e Fisiologia Molecolare CNR, Via Fratelli Cervi 93, Segrate 20090, Italy

Accepted 7 June 2004

Available online 4 July 2004

Abstract

We previously described a rat model where the injection of formalin in the tail induced a facilitation of the hindpaw withdrawal

reflexes (hyperalgesia). In the present work, after injecting formalin in the tail, we measured the levels of pro-nociceptive mediators

tumor necrosis factor-a (TNF) and substance P (SP) in the rat paws. A significant increase of SP levels was evident in the hindpaw,

whereas no changes in SP were observed in the forepaw. Both in the hindpaw and in the forepaw the TNF levels were higher than

normal at each stage of measurement. Our results indicate that a prolonged neuronal activation induced by formalin injection is

associated with a change in nociceptive and inflammatory mediators in distal sites of the body. The fact that SP levels are changed in

the hindpaw but not in the forepaw might point to the activation of a mechanism of retrograde signaling from central synapses to paw

afferent nerves.

D 2004 Elsevier B.V. All rights reserved.

Theme: Sensory systems

Topic: Pain modulation: anatomy and physiology

Keywords: Cytokine; Hyperalgesia; Pain mediator; Retrograde signaling

1. Introduction

We have previously shown that the injection of

formalin in the rat tail produces a facilitation of the

responses to noxious thermal stimuli applied to the

hindpaw [2]. It has been demonstrated that this threshold

reduction (hyperalgesia) is due to the sensitization of the

postsynaptic neurons in the spinal segments, induced by

‘inappropriate’ projections from the tail on sciatic pro-

jection fields [6]. This model of centrally mediated

hyperalgesia has been used for evaluating the analgesic

effects of several drugs [3,4,1]. We showed also that,

after the formalin injection in the tail, the forepaws did

not show any threshold change to noxious stimuli, thus

confirming the somatotopic hypothesis [2]. Additional to

these evaluations, we now decided to explore possible

retrograde signaling mechanisms involved in the periph-

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

doi:10.1016/j.brainres.2004.06.007

* Corresponding author. Tel.: +39-02-50316930; fax: +39-02-

50316949.

E-mail address: mauro.bianchi@unimi.it (M. Bianchi).

eral sensitization. Retrograde signaling in mature adult

neurons is a critical factor in the enhancement and

stabilization of synaptic activity generating LTP–LTD

phenomena [16]. These mechanisms are deeply involved

in the development of persistent pain. It has recently

been provided evidence of a widespread role of tumor

necrosis factor-a (TNF) in mediating hyperalgesia at

different levels [14], both facilitating neuronal excitability

and triggering the release of other pro-inflammatory

substances [15,17].

Substance P (SP) is another mediator that plays a key

role in the nociceptive input transmission in the spinal

cord [12]. It is also the major mediator of neurogenic

inflammation released by primary sensory afferent fibers

activated by noxious stimuli in the skin and in the joints

[19,22].

In most of the studies conducted on the involvement of

cytokines and SP in hyperalgesia, this latter is achieved by

the injection of inflammatory substances in the same site

where the hyperalgesia is to be evaluated. With this

approach, however, it is difficult to identify all the mech-

M. Bianchi et al. / Brain Research 1019 (2004) 255–258256

anisms involved in this phenomenon, and differentiate

between peripheral and central events.

In the rat, it has been demonstrated that the unilateral

stimulation of one paw can induce neurogenically mediated

changes also in the contralateral side [7,8].

In light of all the above cited observations, we thought

it might prove interesting to evaluate the levels of SP and

TNF in the paws following the injection of formalin in the

rat tail. We wanted to investigate with a designed bio-

chemical approach whether or not the persistent activation

of the nociceptive fibers in the tail was able to activate a

mechanism of retrograde signaling from the central syn-

apses to the peripheral afferent nerves in the paw, produc-

ing a modification of hyperalgesic mediators within the

paw.

2. Materials and methods

All the experiments were carried out on adult male

Sprague–Dawley rats (Charles River, Calco Italy), weigh-

ing 200–250 g, and housed in groups of four per cage, at

22F 2 jC with a light–dark cycle of 12:12 h and free

access to water and food. Eight rats were used in each

experimental group.

The ethical guidelines of the International Association

for the Study of Pain (IASP) were adhered to in the study

[23]. To induce a state of hindpaw hyperalgesia, a solution

of formalin (10% formaldehyde, 100 Al) was injected

intradermally in the distal part of the tail; 50 Al of this

solution was injected in the right side of the tail, and 50 Al inthe left side [2]. We chose to inject the solution in both sides

of the tail simply to obtain a more homogeneous distribution

of the formalin in the tail. The method of Hargreaves [9]

was used to assess the paw nociceptive thresholds to thermal

stimuli and hyperalgesia after formalin injection in the tail.

We used a Plantar Test apparatus (Ugo Basile, Comerio,

Italy). In brief, the rats were placed in a clear plastic

chamber and left to acclimatise for 5 min before testing.

Light from an 8 V–50 W halogen bulb (64607 OSRAM)

was delivered to the plantar skin of the middle part of the

rat’s hind and front paw through the base of the plastic box.

The beam was about 12 mm in diameter. The time taken for

the animal to withdraw its left paw was measured. We had

previously observed that, following formalin injection in the

tail, similar changes in nociceptive thresholds are induced in

both hindpaws. Hyperalgesic state was assessed by delta

reaction time (basal latency–test latency), positive results

thereby indicating hyperalgesia. The animals were tested

immediately before and 30 min, 1 and 24 h after formalin

injection. These points in time were chosen since we

previously demonstrated that between 30 and 60 min after

formalin injection the hyperalgesia is particularly evident

[2,1]. In order to avoid tissue damage and paw inflamma-

tion, only one measurement was taken at each time point.

Control animals did not receive any treatment.

Immediately after the evaluation of hyperalgesia, the

animals were killed by decapitation, and the entire hindpaw

and forepaw skin, both right and left, removed. The tissue

samples were weighed, frozen on dry ice and stored at � 70

jC until further processing for TNF and SP measurement.

Blood samples were collected, allowed to stand on ice for

2 h, centrifuged and serum was stored at � 70 jC.For TNF evaluation, skin samples of the right paws were

homogenized in 3 ml of phosphate-buffered saline (PBS)

containing 10 mM EDTA and 20 KIU/ml aprotinin (Sigma).

After centrifugation at 10,000� g, the supernatant were

frozen at � 70 jC for TNF assay.

For SP evaluation, skin samples of the left paws were

homogenized in 3 ml of 0.1 N acetic acid, centrifuged at

10,000� g and the supernatants frozen at � 70 for radio-

immunoassay.

The levels of TNF in blood and paw supernatants were

measured by means of an enzyme-linked immunosorbent

assay (ELISA) kit specific for rat TNF (Bender Medsystem,

Prodotti Gianni, Milano, Italy). The anti-TNF capture

monoclonal antibody (mAb) (5 Ag/ml) was absorbed on a

polystyrene 96-well plate and the TNF present in the sample

was bound to the antibody-coated wells. The biotinylated

anti-TNF detecting mAb (0.25 Ag/ml) was added to bind

TNF captured by the first antibody. After washing, strepta-

vidin-peroxidase was added to the wells to detect the

biotinylated detecting antibody and finally TMB substrate

was added. A colored product was formed in proportion to

the amount of TNF present in the sample, which was

measured at optical density 450 nm. The amount of cytokine

in each supernatant was extrapolated from the standard

curve. The standards were recombinant cytokine curves

generated in doubling dilutions from 2500 to 39 pg/ml.

SP was measured by radioimmunoassay (RIA), using

antiserum and methods previously described and validated

[5]. The antibody was raised in rabbit against synthetic SP,

and it is directed towards the C-terminal of the peptide;125I-SP was purchased from Amersham Biosciences

(Milano, Italy). Sensitivity of the radioimmunoassay is

10 pg/tube and intraassay and interassay variation coeffi-

cient are 8% and 11%.

Difference between controls and experimental groups

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

followed by Bonferroni’s t-test for multiple comparisons. An

effect was determined to be significant if the P-value was less

than 0.05.

3. Results

As expected, the formalin injection in the tail induced a

significant hyperalgesic response in the hindpaw (Table 1).

This effect was evident 30 and 60 min after formalin

treatment, but had disappeared 24 h later. The values refer

to six different groups of eight rats, since the animals were

killed immediately after undergoing the plantar test. The

Fig. 2. SP levels in the hindpaw and forepaw as measured at different time

points after formalin injection in the rat tail. Controls (CTR) were naive,

untreated, animals. Values are meansF S.D. of eight rats. *P < 0.01 vs.

controls.

Table 1

Effect of formalin injection in the tail on forepaw (left paw) and hindpaw

(left paw) withdrawal reflexes

30 min 60 min 24 h

Forepaw

Controls 0.2F 0.3 0.4F 0.5 0.04F 0.3

Formalin 0.6F 0.2 0.5F 0.4 0.1F 0.1

Hindpaw

Controls 0.1F 0.1 0.05F 0.2 0.07F 0.2

Formalin 3.2F 0.6* 3.4F 0.5* 0.3F 0.2

Values are the meansF S.E.M. of the algebraic difference between basal

and test latencies. N= 8 rats

*P< 0.05 vs. controls.

M. Bianchi et al. / Brain Research 1019 (2004) 255–258 257

table also shows that the formalin injection in the tail did not

produce any change in the forepaw withdrawal latencies at

any time considered.

A significant change of TNF paw levels was observed

after formalin injection in the tail. The levels of this

cytokine were higher compared with those of control

animals both in the hindpaw and in the forepaw (Fig. 1).

TNF levels increased 30 min and 1 h after formalin

injection, and remained elevated even 24 h later.

Neither in control nor in formalin-treated animals did

this cytokine reach detectable serum levels (data not

shown).

The SP levels in rat paws of formalin-injected and

control animals are reported in Fig. 2. In the hindpaw,

no significant changes in SP concentration were present 30

min after formalin injection, whereas a significant increase

in this peptide was evident 1 h later. The hindpaw levels of

SP were higher than in controls even 24 h later. Interest-

ingly, we did not observe any modification of SP levels in

the forepaw (Fig. 2).

Fig. 1. TNF levels in the hindpaw and forepaw as measured at different time

points after formalin injection in the rat tail. Controls (CTR) were naive,

untreated, animals. Values are meansF S.D. of eight rats. *P< 0.01 vs.

controls.

4. Discussion

In this study, we have found that formalin injection in the

tail leads to changes in paw SP and TNF levels. It is worth

noting that these two mediators are affected differently by the

formalin injection. In fact, SP levels increased only in a body

area (the hindpaw) somatotopically related to that of the

injection site (the tail), while TNF showed a diffuse, non-

somatotopic, pattern of modification (both forepaw and

hindpaw). Peripheral sensory neurons release SP both ortho-

dromically, with laminar distribution into the dorsal horn of

the spinal cord, and antidromically at the other fiber terminals

of the afferent arch, in the peripheral tissue [10,19,21,22]. We

previously demonstrated that the formalin injection in the tail

leads to the hyperactivity of the spinal lumbar neurons both in

the proper and neighboring projection fields [6]. It is likely

that this mechanism is the background of the hyperalgesic

response in our behavioural test. In the light of these obser-

vations, we can now hypothesize that the ‘‘improperly’’

responding neurons of the lumbar sciatic projection field

activated by the tail tonic stimulation can transmit signals to

primary afferent neurons from the paw. This hypothesis

implies that the antidromic release of SP from hindpaw

primary afferents may occur without a direct stimulation of

peripheral proper nociceptors, thus providing biochemical

evidence that would suggest that the informational exchange

at the central synapses is bi-directional [7,8,16].

TNF is a cytokine produced by a variety of cell types,

including inflammatory cells (i.e. neutrophils, activated

lymphocytes, and macrophages) and tissue cells (endothelial

cells, smooth muscle fibers, keratinocytes, fibroblasts). In

recent studies, it has been shown that also glial and neuronal

cells are able to produce TNF [11,17]. In our model of

hyperalgesia, a significant elevation of TNF was present in

M. Bianchi et al. / Brain Research 1019 (2004) 255–258258

both the hindpaw and forepaw, suggesting a general, non-

specific spread of the cytokine after the induction of a

peripheral damage. This observation is consistent with the

results of other studies, where the inflammation induced in

the paw of one side by the injection of complete Freund’s

Adjuvant produced an increase of TNF levels also in the

contralateral, non-inflamed paw [20]. Thus, increased TNF

in both hindpaw and forepaw might represent a general

inflammatory response to formalin injection. As the TNF

level in the serum did not change, our results suggest that

the TNF increase in the paw may be the result of a local

production related to the perception of pain.

In our experimental conditions, the TNF and SP changes

do not correlate with the development of the hyperalgesic

state. In fact, a significant increase of TNF was present also

in the forepaw, while no hyperalgesia was detectable in this

body region. In addition, the change in SP levels show some

disparity with behavioural modifications. Indeed, the in-

crease in this mediator was confined to the hindpaw, but the

time course of the hyperalgesia was different from that of SP

tissue changes. While in fact the hyperalgesic state appeared

30 min after tail formalin injection, the SP levels started to

increase only 1 h after injection. Moreover, both TNF and

SP were still higher 24 h later, when no hyperalgesia was

present. Therefore, we can assume that peripheral TNF and

SP are not sufficient to sustain the hindpaw hyperalgesic

state in our experimental model. This observation is con-

sistent with the knowledge that inflammatory hyperalgesia

is a complex phenomenon produced by a considerable

number of mediators [13,18].

In conclusion, we have showed here that a persistent

stimulus applied to a peripheral site (the tail) is translated

into biochemical changes at distal sites.

The present data contribute to the identification of the

changes in the sensory neural pathways that take place in

this model of hyperalgesia.

References

[1] M. Bianchi, M. Broggini, Anti-hyperalgesic effects of nimesulide:

studies in rats and humans, Int. J. Clin. Pract., Suppl. 128 (2002)

11–19.

[2] M. Bianchi, A.E. Panerai, Formalin injection in the tail facilitates

hindpaw withdrawal reflexes induced by thermal stimulation in the

rat: effect of paracetamol, Neurosci. Lett. 237 (1997) 89–92.

[3] M. Bianchi, A.E. Panerai, Anti-hyperalgesic effects of tramadol in the

rat, Brain Res. 797 (1998) 163–166.

[4] M. Bianchi, A.E. Panerai, Effects of lornoxicam, piroxicam and

meloxicam in a model of thermal hindpaw hyperalgesia induced by

formalin injection in rat tail, Pharmacol. Res. 45 (2002) 101–105.

[5] M. Bianchi, G. Rossoni, P. Sacerdote, A.E. Panerai, F. Berti, Carbama-

zepine exerts anti-inflammatory effects in the rat, Eur. J. Pharmacol.

294 (1995) 71–74.

[7] I. Bileviciute, C. Stenfors, E. Theodorsson, T. Lundeberg, Unilateral

injection of calcitonin gene-related peptide (CGRP) induced bilateral

oedema formation and release of CGRP-like immunoreactivity in the

rat hindpaw, Br. J. Pharmacol. 125 (1998) 1304–1312.

[6] G. Biella, M. Bianchi, M.L. Sotgiu, Facilitation of spinal sciatic neu-

ron responses to hindpaw thermal stimulation after formalin injection

in rat tail, Exp. Brain Res. 126 (1999) 501–508.

[8] E. Decaris, C. Guingamp, M. Chat, L. Philippe, J. Grillasca, A. Abid,

A. Minn, P. Gillet, P. Netter, B. Terlain, Evidence for neurogenic

transmission inducing degenerative cartilage damage distant from lo-

cal inflammation, Arthritis Rheum. 42 (1999) 1951–1960.

[9] K. Hargreaves, R. Dubner, F. Brown, C. Flores, J. Joris, A new and

sensitive method for measuring thermal nociception in cutaneous

hyperalgesia, Pain 32 (1988) 77–88.

[10] S. Kar, R.G. Rees, R. Quirion, Altered calcitonin gene-related pep-

tide, substance P and enkephalin immunoreactivities and receptor

binding sites in the spinal cord of the polyarthritic rat, Eur. J. Neuro-

sci. 6 (1994) 345–354.

[11] E.D. Milligan, C. Twining, M. Chacur, J. Biedenkapp, K. O’Connor,

S. Poole, K. Tracey, D. Martin, S.F. Maier, L.R. Watkins, Spinal glia

and proinflammatory cytokines mediate mirror-image neuropathic

pain in rats, J. Neurosci. 23 (2003) 1026–1040.

[12] R. Oku, M. Satoh, H. Takagi, Release of substance P from the spinal

dorsal horn is enhanced in polyarthritic rats, Neurosci. Lett. 74 (1987)

315–319.

[13] 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.

[14] M. Schafers, D.H. Lee, D. Brors, T.L. Yaksh, L.S. Sorkin, Increased

sensitivity of injured and adjacent uninjured rat primary sensory neu-

rons to exogenous tumor necrosis factor-alpha after spinal nerve liga-

tion, J. Neurosci. 23 (2003) 3028–3038.

[15] L.S. Sorkin, W.H. Xiao, R. Wagner, R.R. Myers, Tumor necrosis

factor alpha induces ectopic activity in nociceptive primary afferent

fibers, Neuroscience 81 (1997) 255–262.

[16] H.W. Tao, M. Poo, Retrograde signaling at central synapses, Proc.

Natl. Acad. Sci. 98 (2001) 11009–11015.

[17] L.R. Watkins, S.F. Maier, Beyond neurons: evidence that immune and

glial cells contribute to pathological pain states, Physiol. Rev. 82

(2002) 981–1011.

[18] L.R. Watkins, S.F. Maier, L.E. Goehler, Immune-activation: the role

of pro-inflammatory cytokines in inflammation, illness responses and

pathological pain states, Pain 63 (1995) 289–302.

[19] D.M. White, Release of substance P from peripheral sensory nerve

terminals, J. Peripher. Nerv. Syst. 2 (1997) 191–201.

[20] C.J. Woolf, A. Allchorne, B. Safie-Garabedian, S. Poole, Cytokines,

growth factor and inflammatory hyperalgesia: the contribution of tu-

mor necrosis factor alpha, Br. J. Pharmacol. 121 (1997) 417–427.

[21] T.L. Yaksh, Spinal systems and pain processing: development of

novel analgesic drugs with mechanistically defined models, Trends

Pharmacol. Sci. 20 (1999) 329–337.

[22] N. Yonehara, M. Yoshimura, Influence of painful chronic neuropathy

on neurogenic inflammation, Pain 92 (2001) 259–265.

[23] M. Zimmermann, Ethical guidelines for investigation of experimental

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