LA

CAa

cb

tc

Gd

le

f

g

U

AecAsrdleecb

ibp(frStpfs(atasadP

*vUEAil

Neuroscience 153 (2008) 501–506

0d

ONG-LASTING HYPERALGESIA AND SYMPATHETIC DYSREGULATION

FTER FORMALIN INJECTION INTO THE RAT HIND PAW

Kt

Whiscimpu

pswaumerAflsoare

osdtpwttcnCrcens

rut

. J. VIERCK,a,c,d,e* R. P. YEZIERSKIb,c,d,f

ND A. R. LIGHTg

Department of Neuroscience, University of Florida College of Medi-ine, Gainesville, FL 32610-0244, USA

Department of Orthodontics, University of Florida College of Den-istry, Gainesville, FL, USA

McKnight Brain Institute, University of Florida College of Medicine,ainesville, FL, USA

Comprehensive Center for Pain Research, University of Florida Col-ege of Dentistry, Gainesville, FL, USA

University of Florida College of Medicine, Gainesville, FL, USA

University of Florida College of Dentistry, Gainesville, FL, USA

Department of Anesthesiology, University of Utah, Salt Lake City, UT,SA

bstract—Subcutaneous formalin injection has been usedxtensively to evaluate acute effects (over several hours) ofhemical nociceptive stimulation on nociceptive reflexes.lso, a persistent hyperreflexia for mechanical and thermaltimulation, lasting 3 weeks after formalin injection, has beenevealed and related to microglial activation in the spinalorsal horn. The present study demonstrates more pro-

onged effects of formalin injection, lasting 6 weeks, on op-rant escape from nociceptive thermal stimulation. Operantscape requires cerebral processing of nociceptive input andan detect effects that are not limited to spinal or spinal–rain stem–spinal reflex circuits.

Compared with rats injected with saline, escape respond-ng to 44.5 °C and 47 °C stimulation was increased afterilateral s.c. injection of 5% formalin into the dorsal hindaws. The hyperalgesia outlasted visible signs of traumae.g. paw edema). Responses to 36 °C were not altered afterormalin injection, providing a control for effects of the pe-ipheral injury on activity levels or exploratory tendencies.kin temperature recordings from the forepaws and con-

ralateral hind paw during 44.5 °C stimulation of the left hindaw provided an indirect measure of cutaneous blood flow inormalin- and saline-injected animals. Normal reductions inkin temperature during thermal stimulation were attenuatednearly eliminated) at 1 and 2 weeks after formalin injectionnd partially recovered by 10 weeks. Thus, formalin-inducedissue injury produced a long-term secondary hyperalgesia,ccompanied by a reduced sympathetic responsivity. Theimilar time-course for these phenomena suggests that therere mechanistic linkages between focal injury, autonomicysregulation and enhanced pain sensitivity. © 2008 IBRO.ublished by Elsevier Ltd. All rights reserved.

Correspondence to: C. J. Vierck, Department of Neuroscience, Uni-ersity of Florida College of Medicine, Gainesville, FL 32610-0244,SA. Tel: �1-352-275-4123; fax: �1-352-371-2378.-mail address: vierck@mbi.ufl.edu (C. J. Vierck).bbreviations: ANOVA, analysis of variance; CCI, chronic constriction

rnjury of the sciatic nerve; HPA, hypothalamic–pituitary–adrenal; L/G,icking and guarding reflexes of a hind limb to nociceptive stimulation.

306-4522/08$32.00�0.00 © 2008 IBRO. Published by Elsevier Ltd. All rights reseroi:10.1016/j.neuroscience.2008.02.027

501

ey words: formalin hyperalgesia, sympathetic dysregula-ion, inflammation, operant escape, tissue injury, pain.

ithin several hours of formalin injection into the plantarind paw of rats, two distinct periods of licking and guard-

ng are associated with activation of nociceptors (Dubuis-on and Dennis, 1977) and sensitization of spinal reflexircuits (Wiertelak et al., 1994). Subsequently, axonal end-

ngs in the injected region die back, and a period of inflam-atory repair ensues. For several days after injection,rimary hyporeflexia has been observed for plantar stim-lation.

After formalin injection into the dorsal skin of a hindaw, licking and guarding responses to nociceptive heattimulation of plantar skin of the paw are increased for 3eeks (secondary hyperreflexia; Fu et al., 2001). Second-ry hyperreflexia for limb withdrawal from mechanical stim-lation is also observed. A somatotopically appropriateicroglial proliferation is observed during this period (Wut al., 2004). The proliferation of glia and secondary hyper-eflexia for segmental spinal responses (limb withdrawal;dvokat and Duke, 1999) and spinal–brain stem–spinal re-exes (licking and guarding; Woolf, 1984) suggests thatpinal sensitization results from influences of tissue injuryn pain transmission systems. The present study evalu-tes this possibility with an operant test of escape thatelies on processing of pain throughout the neuraxis (Vi-rck 2006a,b).

A previous examination of lick/guard (L/G) reflexes andperant escape after chronic constriction injury (CCI) of theciatic nerve (Vierck et al., 2005) obtained temperatureependent and temporal differences with the effect onhese two withdrawal reflexes. For this model of neuro-athic pain, responding on L/G and operant escape testsas increased after CCI for cold but not for heat stimula-

ion. These effects lasted as long as the animals wereested after CCI (60 days for L/G, 100 days for escape),orresponding to the persistence of neuropathic pain fromerve injury in humans. In contrast, numerous studies ofCI have shown that enhancement of limb withdrawal

eflexes is temporary (recovering within 40 days) for bothold and heat (reviewed in Vierck et al., 2005). Thus,ffects of peripheral injury on spinal reflex circuits shouldot be generalized to represent effects on neural sub-trates responsible for pain transmission and processing.

The time-course of behavioral changes following neu-onal and/or tissue injury is a critical consideration fornderstanding mechanisms of chronic pain. Studies in-ended to reveal substrates of abnormal pain sensitivity

elate the duration of changes in neuronal reorganizationved.

(ttaqscomropa

AFcumnn2ap

B

ALcpcia6aiaTettw4t

rsf4tarwiihdWafiirfa

Fidteda

C. J. Vierck et al. / Neuroscience 153 (2008) 501–506502

e.g. alterations of receptor distribution or sensitivity) to theime-course of behavioral changes. If there is not a match,hen the neuronal change(s) may have contributed to butre not entirely responsible for the behavioral conse-uences of injury. Also, the behavioral effects must repre-ent a change in pain sensitivity and be appropriate for thelinical condition modeled. In the case of CCI, mechanisticbservations have been related to a time-course of seg-ental reflex effects which does not match clinical expe-

ience or the duration of hyperalgesia as measured byperant escape (reviewed in Vierck et al., 2005). Theresent study compares the time-course for hyperalgesiand hyperreflexia following formalin tissue injury.

EXPERIMENTAL PROCEDURES

ll experimental procedures were approved by the University oflorida Institutional Animal Care and Utilization Committee andonformed to National Institutes of Health guidelines for care andse of experimental animals. The experiment was designed toinimize the number of animals used and their suffering. Thirty-ine adult female Long-Evans hooded rats were housed commu-ally, in groups of four, in large enclosures (32 in. high, 18 in. wide,4 in. deep) containing a hammock, shelves, an exercise wheel,nd cardboard and PVC compartments. Gnawing blocks of cy-ress and shredded paper for nesting were provided.

ehavioral testing

fter a period of acclimation to the testing apparatus, 12 femaleong-Evans rats (300–350 g) were trained to escape from noci-eptive thermal stimulation according to methods previously re-orted (Vierck et al., 2002, 2004, 2005). The testing apparatusonsisted of a dark (0.5 foot-candles) compartment (6 in. wide, 8n. long) with a thermally regulated floor (plate compartment) and

brightly lit (3200 foot-candles) escape compartment (6 in. wide,in. long) with a thermally neutral floor (at room temperature). Thenimals could ambulate freely between the compartments, choos-

ng between thermal stimulation in the dark plate compartmentnd an aversive level of bright light in the escape compartment.he apparatus was ventilated with room air to minimize differ-nces in ambient temperature between compartments and be-ween sessions with different floor temperatures. The average airemperatures recorded in the plate and escape compartmentsere 25.5 °C and 26.6 °C for test trials of 36 °C. For trials of4.5 °C the temperatures were 26.7 °C and 26.5 °C, and for testrials of 47 °C the temperatures were 28.0 °C and 27.7 °C.

Behaviorally trained animals were tested 5 days per week andeceived two 15-min trials per day in adjacent apparatuses. Theequence of temperatures presented in the two daily trials was asollows: Monday: 36–36 °C, Tuesday: 36–47 °C, Wednesday:4.5–36 °C, Thursday, 36–44.5 °C, Friday: 47–36 °C. Animalsested on such a schedule that varies temperatures between trialsnd days learn to sample the stimulus condition on each trial andespond according to its intensity. After stable pre-injection dataere obtained (eight sessions for each combination of stimulus

ntensities), each animal received a s.c. injection of saline (50 �L)n one dorsal hind paw (Monday) followed by injection in the otherind paw on Tuesday of week 1 (no behavioral testing on injectionays). The data presented in Figs. 1–3 are from the first trials onednesday through Friday (36 °C, 44.5 °C and 47 °C), which

llowed data for each stimulus intensity to be obtained during therst and subsequent weeks after saline injection. Behavioral test-ng continued for 8 weeks after saline injection. Then, each animaleceived s.c. 50 �L injections of 5% formalin (5 mL of 37%ormaldehyde in 95 mL saline) in one dorsal hind paw (Monday)

nd then the other on Tuesday. Behavioral testing continued for i

ig. 1. Escape response durations, averaged over 6 weeks followingnjection of saline (open circles) or 5% formalin (closed squares) into theorsal skin of the hind paws. Escape duration is summed from responseo response; the last point in each panel shows the total duration ofscape compartment occupancy over seven responses (1–7). Escapeuration was positively related to stimulus intensity. Formalin injection did notffect responding to 36 °C (top panel) but significantly increased respond-

ng to 44.5 °C (middle panel) and 47 °C (bottom panel) (* P�0.05).

1wigaa

T

RuhiipiPdws3pwtltaucodaamaaa

D

ST

Fr4acri

Frpie

C. J. Vierck et al. / Neuroscience 153 (2008) 501–506 503

0 weeks after formalin injection. During this period, the animalsere closely observed for signs of chronic pain. Beyond 2 h after

njection: ambulation was not visibly changed; bouts of licking anduarding were not observed; the animals ate and drank normallynd gained weight; they did not groom excessively or lose hair,nd social interactions and responses to handling were normal.

emperature recordings

ecordings of skin and body temperature were obtained to eval-ate autonomic reactivity of three groups of animals: (1) 12 be-aviorally tested animals during the 10th week after 5% formalin

njection; (2) 10 animals, 1 and 2 weeks after injection of formalinnto the dorsal skin of the left (5 animals) or right (5 animals) hindaw; and (3) control recordings from 17 animals with no previous

njections but with a history of behavioral testing and handling.rior to the recordings, each animal received a s.c. injection ofiazepam (10 mg/kg) and then was placed into an induction boxhere isoflurane (3%) was delivered. After induction of anesthe-ia, the animals were placed on a thermal blanket (maintained at7 °C) and covered. Anesthesia was maintained throughout therocedure with isoflurane (1.5%) via a nose cone, and heart rateas continuously monitored. Rectal core temperature was moni-

ored, and thermocouples were placed on the plantar skin of theeft and right forepaws and the right hind paw. The thermocoupleips were contained within wells of thermoconductive paste indhesive foam pads. Skin and core temperatures were observedntil stable (usually 15 min), and then a thermode, preheated withirculating water, was strapped in contact with the plantar surfacef the left hind paw. Skin and body temperatures were recordeduring 10 min of thermal stimulation and for an additional 10 minfter removal of the thermode from the left hind paw. All temper-ture recordings were obtained and stored using a precision ther-ocouple system (GEC Instruments, Gainesville, FL, USA). Afternesthesia was discontinued, the animals were observed untillert, and then were placed singly in a cage until fully recoverednd then were returned to their communal enclosures.

ata presentation and statistical analysis

tatistical analyses utilized Statistica software (StatSoft, Inc.,

ig. 2. Average durations over 6 weeks for the first two escapeesponses on each trial at different temperatures presented to animalsreviously injected with saline (open bars) or 5% formalin (closed bars)

nto the dorsal skin of the hind paws. The effect of formalin injection onscape duration was significant (* P�0.05).

ulsa, OK, USA). The measure of operant escape was the cumu- a

ig. 3. Time-course of changes in escape duration for the first twoesponses on each trial of 36 °C (top panel), 44.5 °C (middle panel) and7 °C (bottom panel) after 5% formalin injection. The average respondingfter saline injection is shown as open triangles (S), with brackets indi-ating one standard deviation. Weekly averages (1–10) of the two firstesponse durations after formalin injection are shown as squares. Signif-cant differences between overall means after saline and weekly values

fter formalin (t-tests) are shown as closed squares (P�0.05).

lca6ialiceadfm((p

mbjo

E

Eu1tpTpmt(p(ieoemafttnste

ftFFPmctateTcdrfi

ffistte

Ftmb1vfhtsc

C. J. Vierck et al. / Neuroscience 153 (2008) 501–506504

ative duration of successive responses (occupancies of the es-ape platform) within trials. Repeated measures analysis of vari-nce (ANOVA) was used to compare data averaged over the firstweeks after formalin injection and the first 6 weeks after saline

njection. The repeated measures were the duration of escapeccumulated over one to seven responses (Fig. 1) and the cumu-

ative duration of the first two escape responses to three stimulusntensities (Fig. 2). t-Scores were utilized to evaluate the time-ourse of formalin effects, comparing escape durations duringach of 10 weeks after formalin injection against mean durationscross 8 weeks after saline injection (Fig. 3). Skin temperaturesuring and after thermal stimulation were averaged across the twoorepaws and the contralateral (right) hind paw for successive 2.5in periods of 20 min recordings from four groups of animals

between subjects), and there were eight within-subject variablesrepeated measures during trials) for ANOVA (Fig. 4). Core tem-erature responses during thermal stimulation were computed as

ig. 4. Skin temperature measurements from three groups of animalshat received no prior s.c. injection (controls), or injection of 5% for-alin into the dorsal skin of one hind paw 1 or 2 weeks previously, orilateral injection of 5% formalin into the dorsal skin of both hind paws0 weeks previously. Skin temperatures were normalized to baselinealues and averaged across 2.5 min periods of recording from bothorepaws and the hind paw contralateral to stimulation. The thickorizontal line on the abscissa corresponds to the duration of 44.5 °Chermal stimulation of the left hind paw (10 min). Formalin injection

eignificantly attenuated the reduction of skin temperatures during no-iceptive thermal stimulation (* P�0.05).

axima (minus the resting core temperature) and were comparedetween control recordings and different times after formalin in-

ection (ANOVAs). A probability level of 0.05 was applied through-ut for significance.

RESULTS

ffects of formalin on operant escape responding

scape responses (i.e. platform durations) to thermal stim-lation after injection of saline or formalin are shown in Fig.. These graphs show response-by-response accumula-ions of escape durations during exposures to 36 °C (topanel), 44.5 °C (middle panel) and 47 °C (bottom panel).he animals spent most of 36 °C trials in the plate com-artment but did occasionally explore the escape compart-ent. Thirty-six degrees is presumed to produce sensa-

ions of warmth but can activate C nociceptors at low levelsFleischer et al., 1983). During 44.5 °C stimulation, a tem-erature likely to predominantly activate C nociceptorsYeomans and Proudfit, 1996; Vierck et al., 2002), saline-njected animals spent approximately half the trial on thescape platform. During 47 °C stimulation, a suprathresh-ld temperature for activation of C and A-delta nociceptors,scape time consumed nearly all of the trial (7/9ths),ainly distributed between responses 1–5. The bright lightnd nociceptive thermal stimuli were strong motivatingactors that induced the animals to alternatively occupy thewo compartments. In this manner, the animals distributedheir time in the escape compartment in proportion toociceptive stimulus intensity. A positive stimulus–re-ponse relationship for escape duration and stimulus in-ensity is a critical characteristic of an algesic assay (Vierckt al., 2004).

Relative to sessions after injection of saline, injection oformalin increased escape durations for 44.5 °C stimula-ion (main effect: F�5.41, df�1, P�0.02; interaction:�2.28, df�6, P�0.03) and 47 °C stimulation (main effect:�6.62, df�1, P�0.001; interaction: F�11.23, df�6,�0.001), but not for 36 °C stimulation. As shown in theiddle and bottom panels of Fig. 1, differences in es-

ape duration mainly occurred early in test trials. Par-icularly for 47 °C, the total escape duration did not differfter saline or formalin injection, but after formalin injec-

ion the animals stayed on the escape platform consid-rably longer during the first few escape responses.herefore, the maximum difference between injectiononditions can be observed as the average platformurations for the first two responses (Fig. 2). This ploteveals a more steep stimulus–response function afterormalin injection (main effect: F�32.4, df�1, P�0.001;nteraction: F�59.1, df�2, P�0.001).

Fig. 3 reveals the time-course of hyperalgesia afterormalin injection. The mean and standard deviation of therst plus second response durations are shown for ses-ions after saline injection (open triangles), and the dura-ion of responses 1 plus 2 is shown during each week ofesting after formalin injection (squares). Significant differ-nces between weekly escape durations after formalin and

scape durations averaged over 8 weeks after saline in-

jsctw

E

Ttcmdtetcsiacwsoo(ttdTcn

mFepspitd

Sut(StcTtiosstt

av

avjctet(2cattojpn

5titmpmigi

tpTapGseusaSntu2i2cf

cipis

C. J. Vierck et al. / Neuroscience 153 (2008) 501–506 505

ection are shown as closed squares (t-tests). For 44.5 °Ctimulation, escape durations early in trials were signifi-antly elevated for 5 weeks after formalin injection. Theime-course of significant hyperalgesia extended to 6eeks for 47 °C stimulation.

ffects of formalin on autonomic reactivity

wo measures of autonomic reactivity to thermal stimula-ion (paw and core temperature) are shown in Fig. 4. Whenontrol animals were stimulated on the left hind paw for 10in with 44.5 °C, the temperature of the other three pawsecreased for 5–7.5 min. Recovery to the baseline skinemperature began before the stimulus ended (upper pan-l; open triangles) and was complete within 10 min afterhe stimulus was removed. A modest and slow increase inore temperature did not fully recover by 10 min aftertimulation for control animals. One week after unilateralnjection of formalin, the phasic decrease in skin temper-ture was absent during 44.5 °C stimulation (closed cir-les). Two weeks after unilateral injection of formalin, thereas only a slight recovery of the skin temperature re-ponse to heat stimulation (closed squares). The reductionf skin temperature responses at 1 and 2 weeks wasbserved for animals injected dorsally on the stimulatedleft) hind paw or the contralateral (right) hind paw. Statis-ical comparisons revealed no significant difference be-ween left and right injection for testing at 1 week (F�2.53,f�1, P�0.12) or at 2 weeks (F�2.79, df�1, P�0.11).hus, the reduction in skin temperature responsivity oc-urred for stimulation of an injured (left) hind paw or aon-inflamed hind paw (after injection on the right).

The differences in skin temperature over time after for-alin injection were significant (F�4.73, df�3, P�0.004).ormalin injury produced a long-term attenuation of periph-ral cutaneous vasoconstriction, as measured by skin tem-erature changes that normally accompany nociceptivetimulation. The loss and then partial recovery of skin tem-erature responsivity over 10 weeks was associated with an

ncrease and then a recovery of core temperature responseso nociceptive stimulation (lower panel of Fig. 4) (F�6.00,f�3, P�0.002).

DISCUSSION

.c. injection of formalin produces a focal injury that stim-lates and then damages sensory endings. An inflamma-ory reaction to the injury is present for at least 2 weeksLin et al., 2007) and is grossly observable as edema.timulation of the plantar surface after dorsal injection of

he same paw reveals enhanced reflex responses to me-hanical or heat stimulation for 3 weeks (Fu et al., 2001).he period of secondary hyperreflexia matches a 3-week

ime-course of microglial proliferation in the dorsal hornpsilateral to formalin injection (Wu et al., 2004). However,perant escape testing of nociceptive sensitivity (presenttudy) revealed significant secondary thermal hyperalge-ia over a period of 6 weeks after formalin injection. Thus,he time-course of secondary hyperalgesia following focal

issue injury has not been fully revealed by reflex testing, n

nd it exceeded the time-course of spinal microglial acti-ation that has been described previously.

The extended period of enhanced pain sensitivity wasccompanied by evidence for abnormal sympathetic acti-ation, based on skin temperature recordings. For unin-

ected animals, nociceptive stimulation of one hind pawooled the plantar skin of the other paws but increased coreemperature, as blood was shunted away from the skin of thextremities. This spatially distributed cutaneous vasoconstric-

ion occurs in response to nociceptive heat stimulationpresent study) or cold stimulation (Cankar and Finderle,003; C. Vierck, unpublished observations). However, re-ordings 1 and 2 weeks after formalin injection revealedttenuation of phasic cutaneous vasoconstriction following

hermal stimulation (Fig. 4). The reduction in phasic sympa-hetic responsivity indicates that there was a high tonic levelf sympathetic activation in association with the formalin in-

ury. When the sympathetic nervous system is tonically hy-eractive, phasic sympathetic vasoconstriction in response toociceptive stimulation is blunted (Cooke et al., 1990).

The reduction in skin temperature developed over–7.5 min of continuous nociceptive stimulation. Also, theime elapsed after two escape responses for formalin-njected animals was 5.7 min (including plate and platformimes). Thus, the maximal difference in escape perfor-ance after saline or formalin injection occurred during theeriod when cutaneous vasoconstriction developed nor-ally in response to nociceptive stimulation but was min-

mal after formalin injection. This correspondence sug-ests that the long-term heat hyperalgesia after formalin

njection was related to sympathetic dysregulation.The sympathetic nervous system is responsive to ac-

ivation of limbic system circuits and the hypothalamic–ituitary–adrenal (HPA) axis (Herman and Cullinan, 1997).he limbic/HPA axis and sympathetic nervous system arectivated by: (1) psychological stress, (2) nociceptive in-ut, and (3) inflammation (Molina, 2005; Griffis et al., 2006;eenen et al., 2006). Each of these influences results to

ome extent from a peripheral tissue injury. However, theffects of HPA-sympathetic activation on pain are not fullynderstood. For example, an extensive literature has de-cribed effects of stress on reflex tests as “stress-inducednalgesia” (Bodner et al., 1980; Watkins et al., 1983).hortly after termination of an acute stress experience,ociceptive reflexes are attenuated. In contrast, operantesting in the same time frame and under the same stim-lus conditions reveals hyperalgesia (King et al., 2003,007). Stress-induced hyperalgesia is consistent with clin-

cal observations of chronic pain patients (Heiden et al.,005; Vierck, 2006c). Similarly, the increased operant es-ape from heat after formalin injection could represent aorm of stress-induced hyperalgesia.

Demonstration of time-locked hyperalgesia and in-reased sympathetic tone after tissue injury has clinical

mplications for development and persistence of chronicain. Inflammation, even if focally generated and seem-

ngly minor, can activate central stress circuitry. HPA andympathetic activation by inflammation is reinforced by

ociceptive input from an injured region. Enhancement of

plcwsa

AKafo

A

B

C

C

D

F

F

G

G

H

H

K

K

L

M

V

V

V

V

V

V

W

W

W

W

Y

C. J. Vierck et al. / Neuroscience 153 (2008) 501–506506

ain by tonic sympathetic activation is revealed by stimu-ation outside an injured region. Accordingly many forms ofhronic focal pain can lead to development of fibromyalgia,hich is characterized by widespread pain and hyperalge-ia and has been linked to tonic sympathetic dysregulationnd chronic stress (Vierck, 2006c).

cknowledgments—The technical assistance of Jean Kaufman,aren Murphy, Anwarul Azam and Richard Cannon is gratefullycknowledged. Support was provided by the University of Floridarom the Comprehensive Center for Pain Research and the Officef Research.

REFERENCES

dvokat C, Duke M (1999) Comparison of morphine-induced effectson thermal nociception, mechanoreception and hind limb flexion inchronic spinal rats. Exp Clin Psychopharmacol 7:219–225.

odner R, Kelly D, Brutus M, Glusman M (1980) Stress-inducedanalgesia: neural and hormonal determinants. Neurosci BiobehavRev 4:87–100.

ankar K, Finderle Z (2003) Gender differences in cutaneous vascularand autonomic nervous response to local cooling. Clin Auton Res13:214–220.

ooke J, Creager M, Osmundson P, Shepherd J (1990) Sex differ-ences in control of cutaneous blood flow. Circulation 82:1607–1615.

ubuisson D, Dennis S (1977) The formalin test: A quantitative studyof the analgesic effects of morphine, meperidine and brain stemstimulation in rats and cats. Pain 4:161–174.

leischer E, Handwerker HO, Joukhadar S (1983) Unmyelinated no-ciceptive units in two skin areas of the rat. Brain Res 267:81–92.

u K-Y, Light A, Maixner W (2001) Long-lasting inflammation andlong-term hyperalgesia after subcutaneous formalin injection intothe rat hindpaw. J Pain 2:2–11.

eenen R, van Middendorp H, Bijlsma J (2006) The impact of stres-sors on health status and hypothalamic-pituitary-adrenal axis andautonomic nervous system responsiveness in rheumatoid arthritis.Ann N Y Acad Sci 1069:77–97.

riffis C, Compton P, Doering L (2006) The effect of pain on leukocytecellular adhesion molecules. Biol Res Nurs 7:297–312.

eiden M, Barnekow-Bergkvist M, Nakata M, Lyskov E (2005) Auto-nomic activity, pain and the perceived health in patients on sickleave due to stress-related illnesses. Integr Physiol Behav Sci40:3–16.

erman J, Cullinan W (1997) Neurocircuitry of stress: central control ofthe hypothalamo-pituitary-adrenocortical axis. Trends Neurosci

20:78–84.

ing C, Devine D, Vierck C, Mauderli A, Yezierski R (2007) Opioidmodulation of reflex versus operant responses following stress inthe rat. Neurosci 147:174–182.

ing C, Rodgers J, Devine D, Vierck C, Yezierski R (2003) Differentialeffects of stress on escape versus reflex responses to nociceptivethermal stimuli in the rat. Brain Res 987:214–222.

in T, Li K, Zhang F-Y, Zhang Z-K, Light AR, Fu K-Y (2007) Dissoci-ation of spinal microglia morphological activation and peripheralinflammation in inflammatory pain models. J Neuroimmunol 192:40–48.

olina P (2005) Neurobiology of the stress response: contribution ofthe sympathetic nervous system to the neuroimmune axis in trau-matic injury. Shock 24:3–10.

ierck C (2006a) Animal studies of pain: Lessons for drug develop-ment. In: Emerging strategies for the treatment of neuropathic pain(Campbell J, Basbaum A, Dray A, Dubner R, Dworkin R, Sang C,eds), pp 475–496. Seattle: IASP Press.

ierck CJ (2006b) Animal models of pain. In: Wall and Melzack’stextbook of pain (McMahon S, Koltzenburg M, eds), pp 175–185.London: Elsevier Churchill Livingston.

ierck CJ (2006c) Mechanisms underlying development of spatiallydistributed chronic pain (fibromyalgia). Pain 124:242–263.

ierck CJ, Acosta-Rua A, Nelligan R, Tester N, Mauderli A (2002) Lowdose systemic morphine attenuates operant escape but facilitatesinnate reflex responses to thermal stimulation. J Pain 3:309–319.

ierck C, Acosta-Rua A, Johnson R (2005) Bilateral chronic constric-tion of the sciatic nerve: A model of long-term cold hyperalgesia. JPain 6:507–517.

ierck CJ, Kline R III, Wiley R (2004) Comparison of operant escapeand innate reflex responses to nociceptive skin temperatures pro-duced by heat and cold stimulation of rats. Behav Neurosci118:627–635.

atkins L, Kinscheck I, Mayer D (1983) The neural basis of footshockanalgesia: the effect of periaqueductal gray lesions and decere-bration. Brain Res 276:317–324.

iertelak E, Furness L, Horan R, Martines J, Maier S, Watkins L(1994) Subcutaneous formalin produces centrifugal hyperalgesiaat a non-injected site via and NMDA-nitric oxide cascade. BrainRes 649:19–26.

oolf CJ (1984) Long term alterations in the excitability of the flexionreflex produced by peripheral tissue injury in the chronic decere-brate rat. Pain 18:325–343.

u Y, Willkockson H, Maixner W, Light A (2004) Suramin inhibitsspinal cord microglia activation and long-term hyperalgesia in-duced by formalin injection. J Pain 5:48–55.

eomans DC, Proudfit HK (1996) Nociceptive responses to high andlow rates of noxious cutaneous heating are mediated by differentnociceptors in the rat: electrophysiological evidence. Pain 68:

141–150.

(Accepted 18 February 2008)(Available online 29 February 2008)