ORIGINAL ARTICLES: CERVICAL SPINE

ELEVATED PRODUCTION OF INFLAMMATORY MEDIATORS

INCLUDING NOCICEPTIVE CHEMOKINES IN PATIENTS WITH

NECK PAIN: A CROSS-SECTIONAL EVALUATION

Julita A. Teodorczyk-Injeyan, PhD,a John J. Triano, DC, PhD,b, c Marion McGregor, DC, PhD,d

Linda Woodhouse, PT, PhD,e and H. Stephen Injeyan, PhD, DCf

a Associate Proanadian Memorib Professor anrograms, Canadntario, Canada.c Associate Pniversity, Hamid Professor, Uhiropractic Colle Assistant Pro

versity, Hamilton

498

ABSTRACT

Objective: This study investigated whether the production of inflammatory mediators and chemotactic cytokines(chemokines) is altered in patients with chronic and recurrent neck pain (NP).Methods: Cross-sectional data evaluating blood and serum samples were obtained from 27 NP patients and 13asymptomatic (control) subjects recruited from a chiropractic outpatient clinic. Cell cultures were activated bylipopolysaccharide (LPS) and phytoheamagglutinin for 24 to 48 hours. The levels of tumor necrosis factor α (TNF-α),monocyte chemotactic protein 1, also known as CCL2 (CCL2/MCP-1), and macrophage inflammatory protein 1α orCCL3 (CCL3/MIP-1α) were determined by specific immunoassays. Serum levels of nitric oxide metabolites wereevaluated simultaneously, in vanadium III–reduced samples, by Griess reaction.Results: Low levels of constitutive (spontaneous) TNF-α production were present in 7 of the 27 cultures from patientswith NP. Both LPS-induced TNF-α production and inducer (LPS/phytoheamagglutin)-stimulated production of CCL2were significantly elevated (P = .00) in patients compared with controls. In patients, the constitutive synthesis ofCCL3 occurred significantly more frequently (P = .00) and ranged from 30 to more than 2000 pg/mL. Finally, serumlevels of nitric oxide were significantly elevated (P = .00) in NP patients.Conclusions: Production of inflammatory mediators was consistently elevated in NP patients in this study, both invitro and in vivo, and activation of inflammatory pathways was accompanied by up-regulation of CC chemokinesynthesis. This suggests that, in NP patients, CC chemokines may be involved in regulation of local inflammatoryresponse through recruitment of immune cells to the inflamed tissue and exert pronociceptive effects. (J ManipulativePhysiol Ther 2011;34:498-505)

Key Indexing Terms: Neck Pain; Inflammation; Cytokines; Chemokines

he role of inflammatory mediators in the genesis appear to correlate with pain intensity.3 However, the

Tof persistent pain and hyperalgesia has beenrecently explored in different clinical and exper-

imental models.1,2 In certain clinical conditions associatedwith chronic pain, proinflammatory cytokine levels

fessor, Graduate Education and Research Programs,al Chiropractic College, Toronto, Ontario, Canada.d Dean, Graduate Education and Researchian Memorial Chiropractic College, Toronto,

rofessor, Rehabilitation Sciences, McMasterlton, Ontario, Canada.ndergraduate Education, Canadian Memorialege, Toronto, Ontario, Canada.fessor, Rehabilitation Sciences, McMaster Uni-

f Professor and Chair, Department of Pathology and Microbi-ology, Canadian Memorial Chiropractic College, Toronto,Ontario, Canada.

Submit requests for reprints to: H. Stephen Injeyan, PhD, DC,CMCC, 6100 Leslie Street, Toronto ON, Canada M2H 3J1(e-mail: sinjeyan@cmcc.ca).

Paper submitted May 9, 2011; in revised form June 3, 2011;accepted July 5, 2011.

0161-4754/$36.00Copyright © 2011 by National University of Health Sciences.

C

PO

U

C

, Ontario, Canada.

involvement of chemical mediators of inflammation in thepathology of chronic and/or acute spinal pain remainsunclear despite localization of inflammatory cytokinessuch as tumor necrosis factor α (TNF-α), interleukin (IL)-

doi:10.1016/j.jmpt.2011.08.010

Table 1. Demographic data of subjects

Group Age (y) (range) Sex (female/male)

NP patients 40.2 ± 14.2 16/11(n = 27) (23-65)

Control 34.7 ± 9.8 7/6(n = 13) (24-54)

499Teodorczyk-Injeyan et alJournal of Manipulative and Physiological TherapeuticsInflammatory Mediators in Neck PainVolume 34, Number 8

1β, and IL-6 both in intervertebral disc specimens4-6 andin facet joint tissues of degenerating lumbar spine.7,8

The cause of neck pain (NP) is generally consideredto be multifactorial and nonspecific.9 Current medicalopinion favors a prevalence of cervical facet joint painin patients presenting with chronic NP.10,11 Evidencefrom animal studies pointing to facet involvement alsoexists. In a rodent model, Lee et al12 have observed thattensile facet injury across the C6/C7 joint producesbehavioral sensitivity suggestive of pain. Recently, acorrelation of cervical and lumbar facet joint pain hasbeen documented in a large clinical study,13 and lumbarfacet joint pain has been shown to be associated withthe synthesis of inflammatory cytokines.8 This maysuggest that analogous inflammatory mediators may alsobe involved in the etiology of NP. Immunohistologicevidence of localized production of TNF-α in herniatedand spondylotic intervertebral discs of the cervicalspine14 and increased levels of inflammatory cytokines(IL-1 and IL-6) in the cerebrospinal fluid from patientswith cervical myelopathy15 may also provide support forconsideration of the involvement of inflammatorymolecules in NP.

The initiation and continuation of the inflammatoryresponse require a sustained recruitment and migrationof immune and inflammatory cells from the circulationto inflamed tissues. This process is generated andmediated by a group of small chemotactic cytokines(chemokines), which form, based on the presence andposition of the first cysteine (C) residues, 4 distinctivefamilies.16 Members of CC family, characterized by thepresence of 2 adjacent cysteines, attract leukocytes to thesite of injury by activating specific G-protein–coupledreceptors on the leukocyte surface and act in concertwith other cytokines to induce tissue infiltration andenrich the circulating pool of a given leukocyte.17 Inpatients with inflammatory conditions, augmented pro-duction of CC chemokines has been reported to correlatewith the severity of disease18 and analgesia.19

The present study was designed to explore whether up-regulation of inflammatory mediator pathways can bedemonstrated in patients with chronic and recurrent NP. Todetermine the state of activation of circulating peripheralblood mononuclear cells and the likelihood of theirselected recruitment toward the site of inflammation, thein vitro production of a key proinflammatory cytokine,TNF-α, and the levels of synthesis of 2 chemokines,monocyte chemotactic protein 1, also known as CCL2,(MCP-1/CCL2) and macrophage inflammatory protein1α, or CCL3, (MIP-1α/CCL3), respectively, wereinvestigated. Because inflammatory and chemotacticcytokines are strong inducers of endogenous nitricoxide (NO) production and NO plays a role in mediationof peripheral and chronic nociception including spinalpain,3,20,21 serum levels of NO were also examined.

METHODS

Subject RecruitmentSubjects were recruited as part of a larger therapeutic

randomized controlled intervention trial by advertisementsposted on an electronic billboard, by announcements postedby the outpatient clinic at Canadian Memorial ChiropracticCollege, and by direct solicitation for healthy controls.

A certified research coordinator prescreened all re-spondents to postings and ensured that each was scheduledfor evaluation by 1 of 23 independent attending providerswho oversee care in the clinic system. On completion of theinitial examination, the clinical coordinator reviewed aneligibility check list with the examining physician. Thosepatients who met the qualifications of inclusion/exclusioncriteria were included in the study after completing writteninformed consent. Twenty-seven patients with NP and 13asymptomatic controls participated in the study (Table 1).

All subject-handling procedures as well as the informedconsent form were approved by the Canadian MemorialChiropractic College Research Ethics Board.

Inclusion and Exclusion CriteriaSubjects (18-65 years old) were recruited for the study

based on their presentation for treatment of chronic andrecurrent NP (defined as pain lasting longer than 3months during which pain-free episodes did not lastlonger than 45 days), with or without shoulder and upperarm referral. Symptom severity for the NP patients wasrecorded for the average severity over the preceding weekby self-report using the neck disability index22 and avisual analog scale for NP. No threshold of minimaldisability or pain was linked with inclusion into the study.Patients with history or examination findings suggestiveof herniated disc or stenosis (neural stretch signs, armpain aggravated by head movement, muscular weakness,dermatomal hypersensitivity, or change in motor reflexes)and a history of fracture or surgery to the neck orshoulder were excluded from the study. In addition,patients with a history of phlebitis, recent (within 3months) use of chemotherapy or radiation therapy,cortisone treatment (within 30 days), or Botox use (within3 months) were also excluded. Control subjects met thesame criteria except that they were asymptomatic for neckor shoulder pain for the past 3 months.

Heparinized and nonheparinized peripheral bloodsamples (5 mL) were collected by venipuncture from

500 Journal of Manipulative and Physiological TherapeuticsTeodorczyk-Injeyan et alOctober 2011Inflammatory Mediators in Neck Pain

the antecubital fossa area of the arm at the time of patientenrollment into the study and before any therapeuticintervention. Samples were transferred to the laboratoryand processed within 60 minutes of collection. Heparinizedblood, for the culture system, was diluted 10-fold withtissue culture medium (RPMI 1640, GIBCO; Invitrogen,Grand Island, NY) supplemented with 5 × 10−5 mol/L of 2-mercaptoethanol, L-glutamine, and antibiotics. Nonhepar-inized blood samples were placed in ice for 60 minutes andthen centrifuged at 4°C. Aliquoted sera samples were frozenand stored at −76°C until use.

Culture SystemDuplicate whole blood (WB) cultures from patients and

control subjects were prepared essentially as described byYaqoob et al23 and maintained for 24 or 48 hours at 37°C,in a humidified 5% CO2 incubator. At the conclusion of theincubation period, culture supernatants from each subjectwere pooled, centrifuged to remove any contaminatingcellular material, aliquoted, and frozen at −76°C untilfurther analysis.

To induce TNF-α production, cultures were stimulatedfor 24 hours with lipopolysaccharide (LPS) fromEscherichia coli serotype 055:B5 (Sigma-Aldrich, St.Louis, MO) at a concentration of 1 μg/mL. To determinelevels of constitutive (produced without external activa-tion) cytokine synthesis, a set of parallel cultures wascultivated without LPS. Constitutive and inducer-relatedproduction of monocyte chemotactic protein 1 (CCL2)was studied in intact and inducer-stimulated WB cultures.To induce CCL2 production, cultures were stimulated bythe combination of LPS (1 μg/mL) and phytohemagglu-tinin (PHA; Sigma-Aldrich) at 10 μg/mL. The capacityfor the constitutive production of macrophage inflamma-tory protein 1α (CCL3) was determined in parallel,unstimulated preparations.

Assessment of Cytokine ProductionThe levels of TNF-α and CCL2 production were

determined by specific enzyme-linked immunosorbantassay (ELISA) using DuoSet ELISA development systemfor natural and recombinant human cytokines (R&DSystems, Minneapolis, MN) essentially as describedelsewhere. 24 Briefly, Immulon 4 HBX, flat bottommicrotiter plates (Thermo Labsystems, Franklin, MA)were coated with a predetermined concentration of mouseantihuman TNF-α or CCL2 antibody. After washing andblocking the wells with a reagent diluent (R&D Systems),duplicate dilutions of standards (defined amounts of ahuman recombinant TNF-α or CCL2) and the acquiredsamples (sera or culture supernatants) were placed in thewells and incubated for 2 hours. This was followed by athorough washing of the plates and incubation with a

specific detection antibody (biotinylated goat antihumananti–TNF-α or CCL2 antibody). After 20-minute expo-sures to streptavidin-horseradish peroxidase and then asubstrate solution (mixture of H2O2 and tetramethylben-zidine), the absorbance of the developed color wasmeasured at λ = 450 nm using an automated microplatereader (EL×800; Bio-Tech Instruments, Inc, Winooski,VT). Concentrations of the tested cytokines werecalculated from the linearized (best fit) standard curvedetermined by regression analysis. Detection limit forTNF-α and CCL2 was 30 and 15 pg/mL, respectively.Each of the studied culture supernatants was tested atleast 3 times at 2 to 4 different dilutions.

The levels of spontaneous release of macrophageinflammatory protein 1 were assessed using humanCCL3/MIP-1α Quantikine kit (R&D Systems) followingstrictly the manufacturer's protocol designed for cellculture supernatants. Detection limit for this chemokinewas 30 pg/mL.

Assessment of Serum NO LevelsBecause of the short half-life and low concentrations

of NO in vivo, quantification of NO is based on themeasurement of the stable end products of NO oxidation,nitrite (NO2)

− and nitrate (NO3)−. The levels of these

metabolites in biologic fluids, including serum, are highlycorrelated with the production of endogenous NO.25 In thepresent study, simultaneous determination of nitrate andnitrite concentrations in serum samples was carried out by aspectrophotometric method developed by Miranda et al.26

This method is based on reduction of nitrate to nitrite byvanadium (III) chloride (VCl3) combined with detection bythe acidic Griess reaction. Briefly, sera samples weredeproteinized by centrifugation through 10-kd molecularweight cutoff filters (R&D Systems) and then treated withequal volumes of a saturated vanadium (III) chloride(Sigma-Aldrich) and Griess reagent. After the developmentof color, the absorbance was measured at λ = 540 nm usingmultichannel spectrophotometer (EL×800; Bio-Tech In-struments, Inc). Concentrations of (NO2)

− were calculatedfrom a linear standard curve generated from 1 to 100 μmol/L of sodium nitrite (Sigma-Aldrich) using KC-Juniorsoftware (Bio-Tech Instruments, Inc), and detection limitwas 0.5 μmol/L.

StatisticsPrimary outcomes for the study were defined as

differences in LPS-induced TNF-α, PHA/LPS-inducedCCL2, and serum NO levels, between NP patients andcontrols. Differences were tested via t test. Screening forassumptions indicated nonnormality of the TNF-α and NOdata. These were both successfully transformed using logvalues. Results of hypothesis testing for both outcomes

Culture condition

TN

Fαα

[pg

/ml]

*3000

2500

2000

1500

1000

500

0Unstimulated Stimulated

Control

NP patients

Fig 1. In vitro production of TNF-α in NP patients. Synthesis oTNF-α was determined in tissue culture (TC) supernatants fromunstimulated (inducer-free) and LPS-stimulated WB culturesfrom NP patients and controls. The bars represent the meanvalues (±SEM) of TNF-α production, after 24 hours cultureperiod, in a given study group.⁎P = .00.

Fig 2. In vivo and in vitro synthesis of MCP-1/CCL-2 in the studiedpatients. A, Levels of systemic (in vivo) CCL2 release weredetermined in serum samples from NP patients and asymptomaticcontrols. B, The capacity for in vitro synthesis of CCL2 wasexamined in TC supernatants from unstimulated and LPS/PHAstimulated WB cultures after 48 hours incubation period. ⁎P = .00

501Teodorczyk-Injeyan et alJournal of Manipulative and Physiological TherapeuticsInflammatory Mediators in Neck PainVolume 34, Number 8

f

were the same regardless of the transformation; therefore,the untransformed data are presented in this article for easeof discussion. Presence or absence of CCL3 by group wasevaluated using χ2 test. Spontaneous release of CCL2 andserum-associated CCL2 data are provided descriptively forconsideration in future investigations. All results arepresented as mean ± SEM.

RESULTS

The initial sample size for the study was 41, with 28 NPpatients and 13 asymptomatic controls. One subjectdeclined further participation after his serum sample wasobtained, and the remaining studies involved 27 patients.Depending on sample availability, NO data were obtainedfor 24 NP patients and 11 controls; CCL3, for 21 patientsand 11 controls; and TNF-α, for 27 patients and 12 controls.Patients with NP reported their symptom experience overthe week before venipuncture with neck disability indexscores of 23.1 (±10.6) and visual analog scale 45/100(±23/100). Control subjects reported no symptoms in theprior week.

Production of TNF-αSystemic (in vivo) production of TNF-α (at the level

of 218 pg/mL) was observed in serum from only 1 of27 patients and not in any of the control subjects. Figure 1depicts the magnitude of in vitro TNF-α synthesis, in WBcultures, from the same subjects. In unstimulated cultures,constitutive (spontaneous) production of TNF-α wasobserved in preparations from 7 of 27 patients and rangedfrom 32 to 1262 pg/mL with the mean level of 345 ± 163pg/mL (Fig 1). Only 1 of the control cultures contained

-.

detectable TNF-α (42 pg/mL). After 24 hours stimulationwith LPS, synthesis of TNF-α ranged from 901 to 6673pg/mL and 1123 to 1777 pg/mL in supernatants fromNP patients and controls, respectively. In cultures fromthe patients with NP, the mean level of TNF-α production(2635 ± 257 pg/mL) was significantly higher (t = 3.21,P = .00) than that in control preparations (1405 ± 61pg/mL) (Fig 1).

Production of CC-ChemokinesCCL2/MCP-1. In patients, serum content of CCL2 rangedfrom 30 to 3500 pg/mL while remaining within 20 to 145pg/mL range in control samples (Fig 2A). The mean CCL2level in patients' serum was 354 ± 138 pg/mL comparedwith 62 ± 10 pg/mL in asymptomatic controls.

The capacity for the in vitro production of CCL2 wasdetermined after 48 hours incubation of blood cultures in theabsence or presence of the inducer (PHA/LPS) (Fig 2B). Thelevel of spontaneous production of CCL2 in control culturesranged from 30 to 730 pg/mL with the mean of 251 ± 56pg/mL. Mean level of CCL2 production in correspondingcultures from patients was 939 ± 231 pg/mL (range, 30-4687pg/mL). The synthesis of CCL2 in LPS/PHA-stimulatedcultures was predictably higher both in cultures from the NPpatients and controls (Fig 2B). Nonetheless, in culturesfrom the NP group, the mean level of this chemokineproduction was 3868 ± 371 pg/mL (range, 732-8500 pg/mL) and significantly higher (t = 4.13, P = .00) than that inthe control group (mean, 1596 ± 170 pg/mL; range, 810-2830 pg/mL) (Fig 2B).

CCL3/MIP-1a. Clinical conditions linked to inflammatoryresponse are associated with spontaneous synthesis of MIP-1α (CCL3) by quiescent peripheral blood cells.34 Thus, the

B

(NO

2)- [μ

M]

A

CC

L3

pg

/ml [

pg

/ml]

*

*

Serum Unstimulated WB cultures

350

300 60

50

40

30

20

10

0

250

200

150

100

50

0Control NP

patientsControl NP

patients

Fig 3. Synthesis of CCL3 and systemic NO levels in the studiedsubjects. A, Levels of CCL3 were determined in TC supernatantsfrom unstimulated WB cultures from patients and controls after 48hours incubation. ⁎P = .00. B, Systemic levels of NO production inthe studied subjects were established through determination of thetotal nitrite (NO2)

− levels in vanadium III–reduced serum samplesobtained from NP patients and controls. ⁎P = .00.

Table 2. Inflammatory profile of the studied NP patients a

Biomarker(inflammatorymediator)

Patients vs asymptomatic control level ofproduction

In vivo In vitro

TNF-α NS Significantly elevatedCCL2/MCP-1 Increased Significantly elevatedCCL3/MIP-1α ND Significantly more frequentNO Significantly

elevatedND

a The in vivo and in vitro synthesis of the studied inflammatorymediators was assessed in serum samples and tissue culture supernatantsrespectively, as described in the “Methods” section. Levels of mediatoproduction in NP patients and asymptomatic controls were compared todetermine the significance of the observed differences. NS, not significantND, not determined.

502 Journal of Manipulative and Physiological TherapeuticsTeodorczyk-Injeyan et alOctober 2011Inflammatory Mediators in Neck Pain

capacity for production of this chemokine was assessed,after 48 hours incubation, only in unstimulated bloodcultures from patients and controls. The production andrelease of CCL3 into supernatants of cultured blood cellswere observed in 16 of 21 NP patients but not in any ofcontrol preparations (χ2 = 16.76, P = .00). The magnitudeof constitutive CCL3 synthesis in the studied patientsranged from 30 to 2310 pg/mL with the mean level of 252 ±131 (Fig 3A).Assessment of NO Production. Assessment of stable metabo-lites of NO in vanadium III–treated sera from patientsshowed that the level of the total nitrite (NO2)

− rangedfrom 12 to 106 μmol/L in patients and from 9 to 28μmol/L in controls (Fig 3B). The mean concentration ofserum (NO2)

− (39 ± 4 μmol/L) was significantly (t = 3.12,P = .00) higher in patients compared with asymptomaticsubjects (18 ± 1.9 μmol/L).

DISCUSSION

The purpose of the present study was to investigate if NPis associated with a local inflammatory response that isaccompanied by the release of quantifiable solublemediators. The results support the hypothesis that NPsyndrome is associated with activation of inflammatorycytokine pathways. Table 2 presents the summary of theseresults and depicts an inflammatory profile of the studiedNP patients. Patients with NP, enrolled in the current study,demonstrated significantly elevated levels of systemic (in

,r

;

vivo) and/or in vitro–induced production of biomarkersinherent in physiologic responses involving inflammatorycells. Of particular interest is the observation of augmentedproduction of the chemotactic cytokines, CCL2 and CCL3(Figs 2B, 3A, Table 2). The increased capacity for synthesisand release of chemokines is invariably associated withinflammation of tissues or organs,27 and the principalstimuli enticing their synthesis are inflammatory, macro-phage-derived cytokines such as TNF-α and IL-1.18

Although we were not able to observe increases in thelevel of systemic TNF-α in patients sera, the same subjectsshowed significant elevation of its production in vitro (Fig1, Table 2). This is in concordance with observations thatserum cytokine levels seldom correspond to the extent ofsynthesis and action of proinflammatory mediators pro-duced in situ and accumulate predominantly in targettissues.28,29 On the other hand, sustainment of target tissueinflammation entails continuous production of such medi-ators delivered by infiltrating inflammatory cells, whichmigrate to the inflamed tissues from the periphery, recruitedby chemokines. The investigative model used in our studiesincluded functional assessment of the capacity for in vitrosynthesis/release of inflammatory mediators by peripheralleukocytes obtained from NP patients. This approachallowed us to detect pathophysiologic changes in the stateof activation of circulating inflammatory cells. Suchchanges might have not been discerned through studiesrestricted to scrutiny of patients' sera.

Chemokine system and the G-protein–coupled chemo-kine receptors provide essential guidance for interstitialleukocyte migration and trafficking.17 Similarly, cytokinessuch as TNF-α can transmit diverse signals to differentsubsets of leukocytes and alter their adhesiveness anddirectional movement30 Recently, Xu et al31 reported thatmechanisms controlling monocyte trafficking through theperipheral blood and lymphoid tissues are altered duringlocal inflammation but the effect is systemic in maintain-ing inflammatory cells within the circulation. Thisobservation may explain increases in the level of

503Teodorczyk-Injeyan et alJournal of Manipulative and Physiological TherapeuticsInflammatory Mediators in Neck PainVolume 34, Number 8

constitutive secretion of CC chemokines in cultures ofunstimulated WB leukocytes as well as augmentedsynthesis of proinflammatory mediators (TNF-α, CCL2)in inducer-exposed preparations (Table 2).

The cellular origin of CCL-2 and CCL-3 in the studiedpatients remains unclear. As discussed above, underpathologic conditions, most chemokines are released byinflamed tissues and by infiltrating leukocytes. However,MCP-1 (CCL2) and inflammatory cytokines may bereleased also by intervertebral disc cells both in anexperimental model6 as well as in cultures of humanintervertebral disc cells.32 However, in the present study,patients with evidence of radicular pain suggesting discprotrusion or stenosis were excluded. Based on the agerange overlap between the patients and controls, it isunlikely that degenerative intervertebral disc tissue con-tributed to the observed elevation in the level of systemicCCL2. On the other hand, the recruited patients reportedlocalized neck muscle pain associated with tenderness atpalpation, stiffness, and constant muscle fatigue consistentwith mechanical NP.33 Human myoblasts and fibroblastscan spontaneously produce a variety of cytokines andchemokines; the transcription and expression of thesemediators are strongly up-regulated in inflammatoryconditions.34 Strain-induced changes of cytokine secretionin an in vitro model of cultured fibroblasts have beenreported,35 suggesting a role of fibroblasts in the etiology ofinflammatory change in conditions such as myofascial pain.Thus, neck muscles and contiguous connective tissues mayalso produce and release inflammatory and chemotacticcytokines involved in recruitment and activation ofperipheral blood cells to the site of inflammation. Increasednumbers of CC chemokine- and chemokine receptor–expressing peripheral blood mononuclear cells werereported by Kivioja et al36 in patients with whiplashinjuries. Although the capacity for chemokine release wasnot investigated in that study, their observations suggestthat a local (whiplash-associated) soft tissue injury sufficefor systemic up-regulation of chemokine and chemokinereceptor expression.

Physiologically, the production of inflammatory cyto-kines and chemokines is regulated, positively or negatively,by other mediators including prostaglandins and NO. Thepresent report demonstrates that, in NP patients, systemicNO production, assessed through the measurement of levelsof the total serum nitrogen intermediates, (NO2)

− and(NO3)

−, is significantly elevated (Fig 3B). Considering therole of NO in the induction and resolution of pain andinflammation,37 expression of high levels of its productionmay be reflective of a strong pathophysiologic responseinvolving cells of the immune, nervous, and musculoskel-etal systems. Nitric oxide is released via inducible NOsynthase, which, in contrast to other isoforms of thisenzyme, is expressed only in inflamed tissues and whenleukocyte recruitment is occurring at a rapid rate.38,39

Experimental and clinical studies are in agreement that NOplays a major role in the induction of proinflammatorymediator production as well as in modulation of nociceptivesignaling.3,40 The modulatory effect of NO on themechanical responsiveness of nociceptors depends on thebaseline level of neuronal excitability.37 Both chemokinesand inflammatory cytokines are involved in the generationof hyperexcitable sensory neurons and thereby participatein neural signaling associated with pain states.19 Further-more, chemokines, expressed by multiple cell types withinthe nervous system, may produce receptor-mediatednociceptive effects through desensitization of neural opioidreceptors.41 In this regard, CCL2 appears to be particularlyimportant because its overexpression or administrationinduces manifestation of pain sensitivity in experimentalmodels. 42,43 These observations and results of ourinvestigation suggest that the combined effects of inflam-matory cytokine(s)-CC chemokine sensitization of neuronsand enhancement of NO signaling could contribute to thegenesis of NP in the studied patients. This concept gainsfurther support from the observation of elevated levels ofnitric metabolites in sera from patients with cervicogenicheadache noted at the onset of headache but not duringpain-free periods.44

To our knowledge, this report is the first to demonstrateactivation of proinflammatory mediator pathways inpatients experiencing NP. Further elucidation of theetiologic mechanisms of mechanical NP may furnish thenecessary basic data to justify investigative studies of theeffects of various therapeutic modalities in the treatment ofthis clinical condition.

LimitationsThe limitations of this study principally pertain to the

patient population selected for the study. Patients withchronic and recurrent NP were recruited without strictlycontrolling for several key covariants that may influencethe results including the etiology of injury at onset, theunderlying pathophysiology responsible for the generationof pain, the duration of pain, and the psychological stateof the participants. In addition, the intensity of pain andfunctional limitations varied among participants. Thisstudy made no attempt to explore possible associationsbetween biomarker levels relative to pain intensity ordisability scores because this would have required a largersample size.

Future StudiesThe present pilot study was an extension derived from a

larger clinical trial investigating the effects of a physicalmodality on NP patients. The results reported hereinencourage us to initiate a prospective study to address thelimitations mentioned above and prepare groundwork for

504 Journal of Manipulative and Physiological TherapeuticsTeodorczyk-Injeyan et alOctober 2011Inflammatory Mediators in Neck Pain

clinical trials examining the effects of different treatmentinterventions on chronic NP using changes in biomarkersynthesis/secretion as outcomes.

CONCLUSION

The patients with chronic and recurrent NP in this studydemonstrated elevated levels of inflammatory biomarkersproduction both in vitro and in vivo accompanied by up-regulation of CC chemokine synthesis. Thus, in NP, CCchemokines may be involved in the mediation of the localinflammatory response through the recruitment of immunecells to the inflamed tissues. Furthermore, in concert withNO, they may exert pronociceptive effects.

Practical Applications• Production of chemotactic cytokines (chemo-kines), known to be induced by proinflammatorycytokines, is shown for the first time, to beelevated in patients with mechanical NP.

• Chemokines and inflammatory cytokines may beused as biomarkers in mechanical NP patients.

• Elevated production of inflammation-associatedchemokines and NO may contribute to the genesisof NP.

ACKNOWLEDGMENT

The authors thank Ms Maricelle Dinulos for help withthe recruitment of patients, Ms Flora Aronshtam for helpwith venipuncture, and Mr Yuri Yatsenko, BSc, fortechnical assistance.

FUNDING SOURCES AND POTENTIAL CONFLICTS OF INTERESTThis study was supported by funds from CMCC and a

grant from NRG (NeuroResource Group, Plano, TX). Noconflicts of interest were reported for this study.

REFERENCES

1. McMahon SB, Cafferty WB, Marchand F. Immune and glialcell factors as pain mediators and modulators. Exp Neurol2005;192:444-62.

2. Marchand F, Perretti M, McMahon SB. Role of the immunesystem in chronic pain. Nat Rev (Immunol) 2005;6:521-32.

3. Koch A, Zacharowski K, Boehm O, Stevens M, Lipfert P,von Giesen HJ, et al. Nitric oxide and pro-inflammatorycytokines correlate with pain intensity in chronic pain patients.Inflamm Res 2007;56:32-7.

4. Burke JG, Watson WG, McCormack D, et al. Intervertebraldiscs which cause low back pain secrete high levels of

proinflammatory mediators. J Bone Joint Surg 2002;84-B:196-201.

5. Freemont AJ, Watkins A, Le Maitre C, et al. Currentunderstanding of cellular and molecular events in interverteb-ral disc degeneration: implication for therapy. J Pathol 2002;196:374-9.

6. Yoshida M, Nakamura T, Sei A, Kikuchi T, Takagi K,Matsukawa A. Intervertebral disc cells produce tumornecrosis factor α, interleukin-1 β, and monocyte chemoat-tractant protein-1 immediately after herniation: an experimen-tal study using a new hernia model. Spine 2005;30:55-61.

7. Igarashi A, Kikuchi S, Konno S, Olmarker K. Inflammatorycytokines released from the facet joint tissue in degenerativelumbar spinal disorders. Spine 2004;29:2091-5.

8. Igarashi A, Kikuchi S, Konno S. Correlation betweeninflammatory cytokines released from the lumbar facet jointtissue and symptoms in degeneratve lumbar spinal disorders.J Orthop Sci 2007;12:154-60.

9. Ahn N, Ahn U, Ipsen B, An H. Mechanical neck pain andcervicogenic headache. Neurosurgery 2007;60:S1-21-7.

10. Falco F, Erhart S, Wargo B, Bryce D, Atluri S, Datta S, et al.Systematic review of diagnostic utility and therapeuticeffectiveness of cervical facet joint interventions. PainPhysician 2009;12:323-44.

11. Manchikanti L, Singh V, Rivera J, Pampati V. Prevalence ofcervical facet joint pain in chronic neck pain. Pain Physician2002;5:243-9.

12. Lee K, Thinnes J, Gokhin D, Winkelstein B. A novel rodentneck pain model of facet-mediated behavioural hypersensi-tivity: implications for persistent pain and whiplash injury.J Neurosci Meth 2004;137:151-9.

13. Manchikanti L, Singh V, Pampati V, Damron K, Beyer C,Barnhill R. Is there correlation of facet joint pain in lumbarand cervical spine? Pain Physician 2002;5:365-71.

14. Kokubo Y, Uchida K, Kobayashi S, Yayama T, Sato R,Nakajima H, et al. Herniated and spondylotic intervertebraldiscs of the human cervical spine: histological and immuno-histological findings in 500 en bloc surgical samples.Laboratory investigation. J Neurosurg Spine 2008;9:285-95.

15. Nagashima H, Morio Y, Yamane K, Nanjo Y, Teshima R.Tumor necrosis factor-α, interleukin-1β, and interleukin-6 incerebrospinal fluid of patients with cervical myelopathy andlumbar radiculopathy. Eur Spine J 2009;18:1946-50.

16. Epstein F. Chemokines—chemotactic cytokines that mediateinflammation. N Engl J Med 1998;338:436-45.

17. Sharma M. Chemokines and their receptors: orchestrating afine balance between health and disease. Clin Rev Biotechnol2010;30:1-22.

18. Gouwy M, Struyf S, Proost P, Van Damme J. Synergy incytokine and chemokine networks amplifies the inflammatoryresponse. Cytokines Growth Factor Rev 2005;16:561-80.

19. White F, Wilson N. Chemokines as pain mediators andmodulators. Curr Opin Anaesthesiol 2008;21:580-5.

20. Bogdan C. Nitric oxide and the immune response. NatImmunol 2001;2:907-16.

21. Riedel W, Neeck G. Nociception, pain, and antinociception:current concepts. J Rhematol 2001;60:404-15.

22. Vernon H. The Neck Disability Index: state-of-the-art, 1991-2008. J Manip Physiol Ther 2008;31:491-502.

23. Yaqoob P, Newsholme D, Calder PC. Comparison of cytokineproduction in cultures of whole human blood and purifiedmononuclear cells. Cytokine 1999;11:600-5.

24. Teodorczyk-Injeyan J, Injeyan S, Ruegg R. Spinal manipu-lative therapy reduces inflammatory cytokines but notsubstance P production in normal subjects. J Manip PhysiolTher 2006;29:14-21.

505Teodorczyk-Injeyan et alJournal of Manipulative and Physiological TherapeuticsInflammatory Mediators in Neck PainVolume 34, Number 8

25. Kleinbongard P,DejamA,Lauer T,Rassaf T, SchindlerA, PickerO, et al. Plasma nitrite reflects constitutive nitric oxide synthaseactivity in mammals. Free Radic Biol Med 2003;35:790-6.

26. Miranda KM, Espey MG, Wink DA. A rapid, simplespectrophotometric method for simultaneous detection ofnitrate and nitrite. Nitric Oxide 2001;5:62-71.

27. Viola A, Luster AD. Chemokines and their receptors: drugtargets in immunity and inflammation. Annu Rev PharmacolToxicol 2008;48:171-97.

28. Kubaragi Y, Shimada Y, Nagaoka T, Hasegawa M,Takehara K, Sato S. Enhanced production of CC-chemo-kines (RANTES, MCP-1, MIP-1α, and eotaxin) in patientswith atopic dermatitis. Arch Dermatol Res 2001;293:350-5.

29. Sommer C, Kress M. Recent findings on how proinflamma-tory cytokines cause pain: peripheral mechanisms in inflam-matory and neuropathic hyperalgesia. Neurosci Lett 2004;361:184-7.

30. Vaday GG, Franitza S, Schor H, Hecht I, Brill A, et al.Combinatorial signals by inflammatory cytokines and che-mokines mediate leukocyte interactions with extracellularmatrix. L Leukocyte Biol 2001;60:885-92.

31. Xu H, Manivannan A, Crane I, Dawson R, Liversidge J. Criticalbut divergent roles for CD62L and CD44 in directing monocytetrafficking in vivo during inflammation. Blood 2008;112:1166-74.

32. Burke JG, Watson WG, McCormack D, Dowling FE, WalshMG, Fitzpatrick JM. Spontaneous production of monocytechemoattractant protein-1 and interleukin-8 by the humanlumbar intervertebral disc. Spine 2002;27:1402-7.

33. Andersen LL, Kjaer M, Sogaard K, Hansen L, Kryger A,Sjogaard G. Effect of two contrasting types of physical exerciseon chronic neck muscle pain. Arthritis Rheum 2008;59:84-91.

34. De Rossi M, Bernasconi P, Baggi F, de Waal Malefyt R,Mantegazza R. Cytokines and chemokines are both expressedby human myoblasts: possible relevance for the immune

pathogenesis of muscle inflammation. Int Immunol 2000;12:1329-55.

35. Egan TS, Meltzer KR, Standley PR. Importance of straindirection in regulating human fibroblast proliferation andcytokine secretion: a useful in vitro model for soft tissue injuryand manual medicine treatments. J Manipulative Physiol Ther2007;30:584-92.

36. Kivioja J, Rinaldi L, Ozenci V, Kouwenhoven M, Kostulas N.Chemokines and their receptors in whiplash injury: elevatedRANTES and CCR-5. J Clin Immunol 2001;21:272-7.

37. Levy D, Strassman AM. Modulation of dural nociceptorssensitivity by the nitric oxide-cyclic GMP signaling cascade.J Neurophysiol 2004;92:766-72.

38. Bogdan C. Nitrixc oxide and the immune response. NatureRev (Immunol) 2001;2:907-16.

39. Hickey MJ. Role of inducible nitric oxide synthase in theregulation of leukocyte recruitment. Clin Sci 2001;100:1-12.

40. Kobayashi Y. The regulatory role of nitric oxide inproinflammatory cytokine expression during the inductionand resolution of inflammation. J Leukoc Biol 2010;88:1157-62.

41. Zhang N, Rogers T, Caterina M, Oppenheim JJ. Proinflam-matory chemokines, such as C-C ligand 3, desensitize μ-opiodreceptors on dorsal root ganglia neurons. J Immunol 2004;173:594-9.

42. Dansereau MA, Gosselin RD, Pohl M, Pommier B, MechighelP, et al. Spinal CCL2 pronociceptove action is no longereffective in CCL2 receptor antagonist-treated rats. J Neuro-chem 2008;106:757-69.

43. Abbadie C, Bhagoo S, De Koninck Y, Malcangio M, Melik-Parsadaniantz S, et al. Chemokines and pain mechanisms.Brain Res Rev 2009;60:125-34.

44. Inan N, Yilmaz G, Surer H, Coskun O, Ucler S, et al. Is there arole for nitric oxide activity in cervicogenic headache? FunctNeurol 2007;22:155-7.