Directory UMM :Data Elmu:jurnal:B:Brain Research:Vol879.Issue1-2.2000:

(1)

www.elsevier.com / locate / bres

Interactive report

Intrathecal anti-IL-6 antibody and IgG attenuates peripheral nerve

injury-induced mechanical allodynia in the rat: possible immune

modulation in neuropathic pain

a ,

_*

b a a,b

Janice L. Arruda

, Sarah Sweitzer , Maria D. Rutkowski , Joyce A. DeLeo

Department of Anesthesiology, Dartmouth-Hitchcock Medical Center, HB 7125, 1 Medical Center Drive, Lebanon, NH 03756, USA b

Department of Pharmacology, Dartmouth-Hitchcock Medical Center, HB 7125, 1 Medical Center Drive, Lebanon, NH 03756, USA Accepted 13 August 2000

Abstract

Interleukin-6 (IL-6) is a pleiotrophic cytokine with a diverse range of actions including the modulation of the peripheral and central nervous system. We have previously shown significant IL-6 protein and messenger RNA elevation in rat spinal cord following peripheral nerve injury that results in pain behaviors suggestive of neuropathic pain. These spinal IL-6 levels correlated directly with the mechanical allodynia intensity following nerve injury. In the current study, we sought to determine whether it is possible to attenuate mechanical allodynia and / or alter spinal glial activation resulting from peripheral nerve injury by specific manipulation of IL-6 with neutralizing antibodies or by global immune modulation utilizing immunogamma-globulin (IgG). Effects of peripheral administration of normal goat IgG and intrathecal (i.t.) administration of IL-6 neutralizing antibody, normal goat or normal rat IgG on mechanical allodynia associated with L5 spinal nerve transection were compared. Spinal glial activation was assessed at day 10 post surgery by immunohistochemistry. Low dose (0.01–0.001 mg) goat anti-rat IL-6 i.t. administration (P50.025) significantly decreased allodynia and trended towards significance at the higher dose (0.08mg to 0.008mg, P50.062). Low doses (0.01–0.001mg) i.t. normal goat and rat IgG significantly attenuated mechanical allodynia, but not at higher doses (0.08–0.008mg; P50.001 for both goat and rat IgG). Peripherally administered normal goat IgG (30 or 100 mg / kg) did not attenuate mechanical allodynia. Spinal glial activation was unaltered by any treatment. These data provide further evidence for the role of central IL-6 and neuroimmune modulation in the etiology of mechanical allodynia following peripheral nerve injury.  2000 Elsevier Science B.V. All rights reserved.

Theme: Sensory systems

Topic: Pain modulation: pharmacology

Keywords: Astrocytes; Glial fibrillary acidic protein; Interleukin-6; IVIg therapy; Microglia; Neuropathic pain

1. Introduction ment, differentiation [31,35], survival [19], regeneration

and degeneration in both the peripheral and central nervous Interleukin 6 (IL-6) is a pleitrophic cytokine that is systems (CNS).

involved in a diverse range of functions within the body IL-6 has been implicated as a significant factor in the including the modulation of proliferation, differentiation central nervous system’s response to injury [2,21,23]. This and maturation of progenitor cells [28], control of cellular proinflammatory cytokine plays a key role in peripheral metabolic activities [17], the immune system cascade, and nerve injury-induced mechanical allodynia and thermal modulation within the nervous system. There is increasing hyperalgesia in both rodents and humans [13,18,37]. We evidence that supports a role of IL-6 in neuronal develop- have previously demonstrated a direct relationship between the increase in both mechanical allodynia (increased sensitivity to a non-noxious stimulus) and IL-6

immuno-1

Published on the World Wide Web on 28 August 2000.

reactive-like protein in the spinal cord in a sciatic

*Corresponding author. Tel.: 11-603-650-6205; fax: 1

1-603-650-cryoneurolysis model of mononeuropathy [13]. In addition,

4928.

E-mail address: [email protected] (J.L. Arruda). intrathecally administered human recombinant IL-6 in the

(2)

absence of nerve injury was able to produce touch evoked these therapies, neutralizing goat anti-IL-6 was adminis-allodynia in rats and produced thermal hyperalgesia in the tered intrathecally (i.t.) and normal rat or goat IgG sciatic cryoneurolysis model [13]. Furthermore, our labora- antibodies were given either i.t. or via intravenous (i.v.) tory has shown that neurons of the spinal cord produce injection. To assess alterations in this mononeuropathy IL-6 messenger RNA in response to peripheral nerve model due to the various treatments, tactile sensitivity injury resulting in neuropathic pain behaviors [3]. Recently using von Frey filaments and spinal glial activation were it has been demonstrated that in the chronic constriction assessed. Glial activation, in particular astrocytic, may injury model, cutaneous heat and pressure hypersensitivity have a principle role in nociceptive processing and in the was not elicited in mice with a null mutation of the IL-6 thermal and mechanical hyperalgesia and / or allodynia gene as compared to the wild type controls [33]. produced by peripheral nerve injury [32]. Efficacy of the Along with these direct relationships between IL-6 and pharmacological interventions was determined by their the CNS response to injury, recent literature also impli- ability to significantly attenuate one nerve-injured induced cates possible indirect effects of IL-6 through its’ ability to behavioral modality, mechanical allodynia. Our results modulate classic pain mediators in the CNS milieu. For reveal a potentially novel mechanism of modulating spinal example, in a series of related in vivo and in vitro studies, neuroinflammation that may ultimately translate into new, it was shown that the concentration of nitrate increased in non-addictive xenobiotics for the treatment of chronic pain rat hippocampus in response to IL-6 treatment, suggesting syndrome in humans.

an overproduction of nitric oxide due to the presence of IL-6 [29,30]. Recent work in our laboratory has also

revealed that in the spinal cord of L5 nerve transected 2. Materials and methods animals IL-6 is produced downstream to IL-1 beta and

TNF [47]. IL-1 beta is one of the proinflammatory 2.1. Animal subjects cytokines reported as effecting spinal sensitization possibly

through its’ influence on the induction and expression of Male Holtzman rats (Harlan, Indianapolis, IN) weighing cyclooxygenase 2, inducible nitric oxide synthase and at least 200 g at the start of the study were used. Animals substance P [22,24,44]. The exact contribution of each were housed individually with a 12 / 12 h light / dark cycle cytokine in the response cascade to nerve injury has not and free access to food and water. Procedures in this study yet been determined, but certainly there appears to be an were approved by the Institutional Animal Care and Use intimate interplay between these cytokines and their modu- Committee at Dartmouth College. Efforts were made to lation of the CNS environment. Pertinent to this study, limit distress and to use the minimum number of animals both the direct and potential indirect effects of IL-6 in necessary to achieve statistical significance as set forth by nerve injury contribute to the mounting evidence for a the International Society for the Study of Pain guidelines seminal role of IL-6 in neuropathic pain and support [8].

further investigation into IL-6 targeted interventions to

reduce or prevent peripheral nerve injury-induced pain 2.2. L spinal nerve transection5 behaviors.

In addition to the interest of elucidating the use of Animal surgery was performed under inhalation anes-specific interventions like IL-6, in the last few decades thesia using halothane in 100% O , induced at 3% and2 there has been an increased focus and use of high dose maintained at 1.5%. Each of the animals sustained an L5 intravenous immunogamma-globulin (IVIg) therapy as spinal nerve transection as previously described [7]. Briefl-global immune modulating therapies. Several immune y, the L4–L5 spinal nerves were isolated following system related diseases including rheumatoid arthritis [35], removal of the L6 tranverse processes. The L5 spinal nerve thrombocytopenic purpura [25], multiple sclerosis [42], was gently separated from the L4 spinal nerve and then Kawasaki’s disease [34,36] and Guillain–Barre syndrome was transected. The wound was closed using 3-0 polyester [49] have been treated with IVIg therapy. It has been found suture and surgical staples for fascia and skin, respectively. that IVIg therapy is able to relieve symptoms of these

diseases, such as pain and morning stiffness in rheumatoid 2.3. Intrathecal injections of anti-rat IL-6, normal goat arthritis patients [38]. The mechanism of action for IVIg is or rat IgG

currently unknown, but discovery of its beneficial

attri-butes are being expanded continually. Goat anti-rat IL-6 neutralizing antibody, normal goat In the present study, the effects of manipulating central IgG (R&D Systems, Minneapolis, MN) and normal rat IgG neuroinflammation after an L5 spinal nerve transection was (Sigma, St. Louis, MO) were each diluted in phosphate investigated by interfering with a specific proinflammatory buffered saline (PBS). Rats were randomly placed into one cytokine, IL-6, and a global immunomodulator, IgG, at the of seven groups: high dose or low dose of anti-rat IL-6 time of peripheral nerve injury and in the initial period (n55 or 8, respectively), normal goat IgG (n57 or 6, following the lesion. In order to elucidate the benefits of respectively), or normal rat IgG (n58 or 7, respectively)

(3)

and PBS vehicle control (n58). Each high dose animal 2.7. Immunohistochemical staining was given 0.08mg of treatment 1 h before surgery and on

day 1 post surgery and 0.008 mg on day 3 post surgery. IHC was performed by the avidin–biotin technique Low dose animals were given 0.01 mg of treatment 1 h (ABC Elite Kits, Vector Labs, Burlingame, CA) on free before surgery and on day 1 post surgery and 0.001mg on floating 20 mm sections. Elimination of the primary day 3 post surgery. All intrathecal injections were per- antibody was performed in each run as a negative control. formed via a direct lumbar puncture under inhalation Monoclonal OX-42 antibody to CR3 / CD11b was used as a anesthesia in a total volume of 40 ml. microglial marker at a dilution of 1:1 (Dr. W.F. Hickey, Dartmouth) and rabbit anti-glial fibrillary acidic protein 2.4. Peripheral injections of IgG (GFAP) was used as an astrocytic marker at a dilution of

1:20,000 (DAKO, Carpinteria, CA). Animals were divided into 3 groups: PBS control (n58),

low dose (30 mg / kg, n58) and high dose (100 mg / kg,

2.8. Scoring of IHC slides n57) normal goat IgG in PBS. Treatments were

adminis-tered via tail vein injection on day 1 before surgery (to

Assessment of IHC slides was performed by an ex-allow ample time to penetrate the central nervous system)

perimenter blinded to the treatment groups. Following and on days 1 and 3 post L5 transection surgery. These

staining of lumbar spinal cord sections with the specific doses were based on previously reported human doses for

marker under consideration, three or more sections from IVIg therapy [15,38,41,43] with modifications made for

each animal were surveyed under low and medium magni-the bolus injections utilized in our model.

fication to arrive at a score. Microglia and astrocytes were scored based on the following scale as previously de-2.5. Mechanical allodynia

scribed [6]: baseline staining (2), mild response (1), moderate response (11), intense response (111). The Tactile sensitivity was measured as the frequency of

activation response is based on the observed cell morphol-foot-withdrawals elicited by a defined normally

non-noxi-ogy, local cell density, and intensity of GFAP and CR3 / ous mechanical stimulus. Each rat, under non-restrained

CD11b immunoreactivity. conditions, was placed singly beneath an inverted

venti-lated Plexiglas cage upon an elevated aluminum screen surface with 1 cm mesh openings. Animals were

previous-2.9. Statistical methods ly acclimated to this environment and to the experimenter.

In each blinded testing session, rats were subjected to 10

STATAE5.0 (Stata Corp., College Station, TX) was tactile stimulations on the ipsilateral hind paw using 2 g

used to perform all statistical tests. To determine the and 12 g von Frey filaments (Stoelting Co., Wood Dale,

existence of a difference in overall behavioral outcomes IL). Baseline (pre-lesion) responsiveness was minimal (0

(mechanical sensitivities) due to surgical or pharmacologic to 1 responses out of 30 stimulations) as confirmed from

treatment, a one-way ANOVA was performed with sub-testing sessions prior to nerve lesion. Mechanical allodynia

sequent specific comparisons of behaviors between treat-was assessed by recording the total number of responses

ment groups accomplished with the Bonferroni multiple-elicited during three successive trials (10 stimulations /

comparisons test. A P value of ,0.05 was considered each filament) separated by at least 10 min for a total

significant. possible maximum response of 30. Mechanical allodynia

was measured on days 1, 3, 5, 7, and 10 post surgery.

2.6. Spinal cord tissue preparation _{3. Results}

All rats were perfusion-fixed for immunohistochemistry _{3.1. Effects of i.t. antibody administration on mechanical} (IHC). Under deep anesthesia (sodium pentobarbital, 150 _allodynia

mg / kg, intraperitoneal injection) rats were euthanized by transcardiac perfusion with 250 ml PBS followed by 250

ml of 4% paraformaldehyde in 0.1 M phosphate buffer, pH 3.1.1. IL-6 neutralizing antibody

7.4. Following perfusion and laminectomy, the lesioned Low dose anti-IL-6 intrathecal injections significantly root or nerve was verified and traced to its site of entry attenuated 2 g von Frey measurement of mechanical into the spinal cord. Appropriate (L5) spinal cord segments allodynia as compared to the PBS control group (P5

were retrieved and post-fixed for 4 h in the above fixative 0.025) while the high dose anti-IL-6 injections trended and then cryoprotected for at least 72 h in 30% sucrose / towards significance (P50.062). With the 12 g von Frey PBS at 48C. Segments were freeze-mounted in OCT filaments, low dose anti-IL-6 trended toward significance embedding medium on cork blocks for cryostat sectioning when compared to the PBS control animals (P50.085)

(4)

while the high dose did not attenuate allodynia (P50.318) (see Fig. 1).

3.1.2. Normal goat IgG

Normal goat i.t. IgG administration significantly at-tenuated mechanical allodynia at the low dose when compared to the PBS control (2 and 12 g, P50.001) while the high dose did not decrease nerve injury-induced allodynia (2 and 12 g, P51.0) (see Fig. 2).

3.1.3. Normal rat IgG

Normal rat i.t. IgG administration significantly at-tenuated mechanical allodynia at the low dose when compared to the PBS control (2 g, P50.001; 12 g, P5

0.003) while the high dose did not decrease nerve injury-induced allodynia (2 and 12 g, P51.0) (see Fig. 3). There was no significant difference between the goat and rat IgG treatments with either the 2 or 12 g von Frey filaments. This result implies that the species of IgG used is not critical for modulating the immune response to peripheral nerve-injury induced allodynia.

Fig. 2. Time course for mechanical allodynia in the ipsilateral hind paw following L5 spinal nerve transection with PBS, high dose normal goat-IgG, or low dose normal goat-IgG intrathecal treatment 1 h before and on day 1 (high dose50.08mg; low dose50.01mg) and day 3 (high dose50.008 mg; low dose50.001mg) post surgery. All animals were exposed to 30 stimulations with 12 g (A) or 2 g (B) von Frey filaments at each time point. Results are reported as the average response out of 30 stimulations per group6S.E.M. *Represents statistically significant differ-ence between treatment and PBS control at indicated time points (P,

0.05).

3.2. Effects of peripheral IgG administration on mechanical allodynia

Intravenous goat IgG administration at 30 and 100 mg / kg doses did not significantly attenuate nerve injury-induced allodynia (see Fig. 4).

3.3. Immunohistochemical staining

There were no observable differences in GFAP or OX-42 immunoreactivity between any of the treatment groups (see Fig. 5 and Tables 1 and 2). Astrocytic and microglial activation mirrored previously reported glial changes in

Fig. 1. Time course for mechanical allodynia in the ipsilateral hind paw

both the treated and untreated groups [3]. Microglial cells

following L5 spinal nerve transection with PBS, high dose goat-anti-rat

IL-6 neutralizing antibody, or low dose goat-anti-rat IL-6 neutralizing were no longer highly ramified (Fig. 5A). The processes

antibody intrathecal treatment 1 h before and on day 1 post surgery (high _{were concentrated to the cell body but did not appear} dose50.08 mg; low dose50.01 mg) and on day 3 post surgery (high _{phagocytic (Fig. 5B and C). Astrocytes demonstrated} dose50.008mg; low dose50.001mg). All animals were exposed to 30

hypertrophy in both the dorsal and ventral horn of the

stimulations with 12 g (A) or 2 g (B) von Frey filaments at each time

treated groups (Fig. 5E and F). This was an increase from

point. Results are reported as the average response out of 30 stimulations

(5)

Fig. 3. Time course for mechanical allodynia in the ipsilateral hind paw following L5 spinal nerve transection with PBS, high dose normal rat IgG, or low dose normal rat IgG intrathecal treatment 1 h before and on

day 1 post surgery (high dose50.08mg; low dose50.01mg) and on day _{Fig. 4. Time course for mechanical allodynia in the ipsilateral hind paw} 3 post surgery (high dose50.008mg; low dose50.001mg). All animals _{following L5 spinal nerve transection with PBS, 30 mg / kg normal goat} were exposed to 30 stimulations with 12 g (A) or 2 g (B) von Frey _{IgG, or 100 mg / kg normal goat IgG 1 day before and on days 1 and 3} filaments at each time point. Results are reported as the average response _{post surgery. All animals were exposed to 30 stimulations with 12 g (A)} out of 30 stimulations per group6S.E.M. *Represents statistically signifi- _{or 2 g (B) von Frey filaments at each time point. Results are reported as} cant difference between treatment and PBS control at indicated time _{the average response out of 30 stimulations per group}₆_S.E.M. points (P,0.05).

to IL-6, we have also reported spinal increases in

immuno-4. Discussion reactive-like TNF and IL-1b following nerve injury [14].

Sommer et al. [45] demonstrated dose-dependant attenua-In the present study, we demonstrated significant at- tion of both hyperalgesia and mechanical allodynia in mice tenuation of mechanical allodynia following L5 spinal with experimental neuropathy by i.t. administration of nerve transection by intrathecal injection of neutralizing neutralizing antibodies to the IL-1b receptor (10–80 mg / anti-IL-6. We also found in the same model that low doses animal). Based on these results, it is evident that there are of intrathecally administered normal goat and rat IgG were several neuroimmune factors influencing pain-related be-able to significantly attenuate enhanced tactile sensitivity. havioral responses following peripheral nerve injuries. These data suggest that both specific and global central Therefore, it is of particular interest that intrathecally immune manipulation alters nociceptive sensory process- administered normal goat and rat IgG were able to ing after a peripheral nerve injury. significantly attenuate the mechanical allodynia following Neuroimmune interactions are highly complex and an L5 spinal nerve transection. Over the past twenty years multi-faceted. Recently reported findings indicate that a normal human IVIg therapy has been used with success in variety of cytokines are involved in neuropathic pain humans for a variety of autoimmune disorders, primary including IL-6, Interleukin 1-b(IL-1b), and Tumor Necro- immunodeficiencies, and Kawasaki disease [5,16,25] to sis Factor (TNF). In addition to the previously mentioned both ameliorate disease and alleviate symptoms. For studies indicating a role for IL-6, Ramer et al. [39] have example, IVIg treatment in rheumatoid arthritis (RA) shown that IL-6 knockout mice have delayed mechanical patients has been shown to decrease plasma IL-6 levels, allodynia in response to L5 spinal nerve tight ligation and morning stiffness, and pain levels [38] associated with RA. transection when compared to the parent strain. In addition Given the nature of the illnesses treated with IVIg, it is

(6)

Fig. 5. (A–C) Representative microglial responses as visualized by anti-OX42 immunohistochemistry. (A) normal, untreated animal, (B) intrathecal saline treatment with L5 spinal transection (SPTS), and (C) intrathecal high dose normal goat IgG treatment with SPTS. (D–F) Representative astrocytic responses as visualized by anti-GFAP immunohistochemistry. (D) normal, untreated animal, (E) intrathecal saline treatment with SPTS, and (F) intrathecal high dose anti-IL-6 treatment with SPTS.

(7)

Table 1

Astrocytic responses (illuminated by GFAP ir) of individual animals at the L5 spinal cord level at day 10 post L5 spinal nerve transection with intrathecal treatment on days 0, 1, and 3

Intrathecal Ipsilateral (injured) side Contralateral (uninjured) side

treatment

DH VH DH VH

Normal, untreated 2 2 2 2

2 2 2 2

Surgery alone 1 1 1 1

111 11 11 11

11 1 2 2

Saline 11 1 11 1

1 1 11 1

11 1 1 1

LD anti-IL6 11 11 1 11

11 1 1 1

11 2 1 2

LD goat IgG 11 11 111 1

11 1 1 1

11 1 11 1

LD rat IgG 11 2 1 1

11 1 11 1

111 11 11 2

HD anti-IL6 11 1 1 1

11 1 11 1

111 1 11 11

HD goat IgG 11 1 11 1

11 1 11 2

11 1 1 2

HD rat IgG 11 1 11 1

111 1 1 1

1 1 1 1

Staining / morphology key: baseline (2), mild (1), moderate (1), or intense (111) responses. VH5ventral horn, DH5dorsal horn, LD5low dose, HD5high dose.

conceivable that the effective attenuation seen with normal study, the normal IgG donor animals also had antibodies IgG treatment in our rodent model is related to the that had direct and / or indirect effects on the cytokine therapy’s immunomodulatory actions. levels in the spinal cord milieu following nerve injury. A thorough understanding of the mechanisms of action Other proposed mechanisms for IVIg therapy include the of IVIg treatment is currently unknown, but there are neutralization of disease related autoantibodies [12,16], several proposed mechanisms of action, some of which inhibition of complement binding and the resultant mem-may apply to the reduction in allodynia we observed in the branolytic attack complex [4,10,11], modulation and present study. Of interest in our model is the ability of blocking of the antibody’s Fc (constant fragment) receptor IVIg therapy to block cytokines and growth factors. IVIg [26,27], and neutralizing effects of superantigen epitopes has been shown in human immune disorders to decrease [48].

the levels and / or production of interferon [40], TNF, Based on the many potentially diverse effects of IVIg IL-1b, transforming growth factor-b [1,46] and IL-6 [38]. therapy and our previous work showing that nerve injury The blockade could act directly from antibodies in the appears to induce a robust spinal neuroimmune response, it IVIg preparation against the cytokines or indirectly by is not surprising that intrathecal administration of IgG disruption of precursor components of the immune cas- would be able to significantly attenuate mechanical al-cade. The presence of low levels of anti-cytokine anti- lodynia in our current study. Interestingly, the higher doses bodies has been shown in normal, healthy humans [20]. of either rat or goat IgG were unable to elicit the same Based on this knowledge, it is conceivable that in our effect, demonstrating a reverse U-shaped dose–response

(8)

Table 2

Microglial responses (as illuminated by OX42 ir) of individual animals at the L5 spinal cord level at day 10 post L5 spinal nerve transection with intrathecal treatment on days 0, 1, and 3

Intrathecal Ipsilateral (injured) side Contralateral (uninjured) side

treatment

DH VH DH VH

Normal, untreated 2 2 2 2

2 2 2 2

Surgery alone 1 11 2 2

1 11 2 2

11 1 1 1

Saline 111 111 1 1

111 111 11 11

11 11 1 1

LD anti-IL6 11 11 1 1

11 11 11 11

111 111 1 1

LD goat IgG 11 11 1 1

111 111 1 1

111 11 1 1

LD rat IgG 11 11 1 1

11 1 1 1

111 11 1 1

HD anti-IL6 11 11 2 2

111 11 1 1

11 11 1 1

HD goat IgG 11 111 1 1

111 111 11 11

111 11 1 1

HD rat IgG 111 111 1 1

1 1 1 1

11 11 1 1

Staining / morphology key: baseline (2), mild (1), moderate (1), or intense (111) responses. VH5ventral horn, DH5dorsal horn, LD5low dose, HD5high dose.

curve. The reasons for this have not yet been elucidated, IgG in our model than those currently used in human but it is possible that when the system is overwhelmed disease [9]. In many of the human dosing schedules, IgG with IgG it is no longer able to exert its beneficial actions was gradually administered over an extended period of on the resultant immune cascade, i.e. may induce a time through a continuous intravenous drip and, thereby, compensatory proinflammatory cascade. Clearly, a more attained much higher doses than those employed in the complete understanding of the mechanism(s) by which the present study. High doses of IgG given in a short time immune system is being effected to reduce mechanical period are associated with adverse effects such as head-allodynia by i.t. IgG therapy is needed if this potential ache, chills, myalgia, and chest discomfort [9]. In view of method of therapy is to be utilized and / or optimized for the difficulty to accurately monitor side effects, administra-clinical treatment. tion limitations, and the animals’ overall well being, we Due to the difficulties of intrathecal administration,, chose to utilize conservative, instead of high doses of IgG peripheral therapies are preferable to intrathecal ones. For via intravenous injections. Our inability to attenuate me-this reason, we chose to explore the option of peripheral chanical allodynia, therefore, could be due to the con-IL-6 and IgG therapy to alter mechanical allodynia in our centration administered. Secondly, the lack of attenuation L5 transection model. However, unlike i.t. IgG administra- could be the result of the time course we chose to use. tion, peripheral IgG therapy was unable to attenuate Third, and perhaps of most importance, the concen-mechanical allodynia. There are several plausible reasons tration that was able to penetrate the CNS from the for this result. First, we employed lower doses of systemic periphery may not have been adequate for spinal

(9)

modula-survival and interleukin-6 protects against MPP1neurotoxicity in

tion of the cascade resulting from L5 spinal nerve injury.

cultures of fetal rat dopaminergic neurons, Exp. Neurol. 136 (1995)

Clearly, further research on the use of IgG therapy needs to

44–52.

be explored to elucidate the optimal dosing concentration / _{[3] J.L. Arruda, R.W. Colburn, A.J. Rickman, M.D. Rutkowski, J.A.} schedule, its mechanism of central action, and its potential DeLeo, Increase of interleukin-6 mRNA in the spinal cord following

role in the alleviation of neuropathic pain. peripheral nerve injury in the rat: potential role of IL-6 in neuro-pathic pain, Mol. Brain Res. 62 (1998) 228–235.

The pharmacological interventions used in this study

[4] M. Bastam, M.C. Dalakas, High dose intravenous immunoglobulin

were unable to alter robust spinal astrocytic or microglial

exerts its beneficial effect in patients with dermatomyositis by

activation that occurs following a peripheral nerve injury _{blocking endomysial deposition of activated complement fragments,} suggesting distinct mechanisms of IL-6 and IgG modula- _{J. Clin. Invest. 94 (1994) 1729–1735.}

tion and spinal glial changes. In addition, the sole use of [5] S.A. Berkman, M.L. Lee, R.P. Gale, Clinical uses of intravenous immunoglobulins, Ann. Intern. Med. 112 (1990) 278–292.

GFAP and the CR3 / CD11b antigen may be inadequate

[6] R.W. Colburn, J.A. DeLeo, A.J. Rickman, M.P. Yeager, P. Kwon,

markers to fully assess the contribution of this large glial

W.F. Hickey, Dissociation of microglial activation and neuropathic

sub-population of the CNS. At the present time, the _{pain behaviors following peripheral nerve injury in the rat, J.} complexities of the spinal neuroimmune response to a _{Neuroimmunology 79 (1997) 163–175.}

nerve injury are incompletely elucidated. It appears, how- [7] R.W. Colburn, A.J. Rickman, J.A. DeLeo, The effect of site and type of nerve injury on spinal glial activation and neuropathic pain

ever, that the major events in a peripheral inflammatory

behavior, Exp. Neurol. 157 (1999) 289–304.

response following tissue damage also occurs in the CNS

[8] B.G. Covino, R. Dubner, J. Gybels, H.W. Losterlitz, J.C. Liebeskind,

after a distant nerve lesion. This innate immunologic

R.A. Sternbach, L. Vylicky, H. Yamamura, M. Zimmerman, Ethical

reaction in the CNS is comprised of anatomic, physiologic _{standards for investigations of experimental pain in animals, Pain 9} and inflammatory processes that, if left unabated, may (1980) 141–143.

[9] M. Dalakas, Mechanism of action of intravenous immunoglobulin

facilitate spinal hypersensitization, the pathophysiologic

and therapeutic considerations in the treatment of autoimmune

correlate of persistent neruopathic pain. Future plans

neurologic diseases, Neurology 51 (1998) S1–S8.

include the investigation of centrally administered spinal

[10] M.C. Dalakas, Clinical relevance of IVIg in the modulation of the

anti-IL-6 and IgG antibodies on cellular adhesion molecule _{complement-mediated tissue damage: implication in} dermato-and Major Histocompatibility Complex expression after a myositis, Guillain Barre syndrome and myasthenia gravis, in: M.

peripheral nerve injury. Kazatchkine, A. Morel (Eds.), Intravenous Immunoglobulin Re-search and Therapy, Parthenon Publishing, London, 1996, pp. 89–

In conclusion, this study showed that i.t. administration

93.

of neutralizing IL-6 antibody and normal rat and goat IgG

[11] M.C. Dalakas, Experience with IVIg in the treatment of

inflamma-were effective in attenuating mechanical allodynia in a rat _{tory myopathies, in: M. Kazatchkine, A. Morel (Eds.), Intravenous} mononeuropathy model. These data are supportive of our Immunoglobulin Research and Therapy, Parthenon Publishing,

hypothesis that a complex central neuroimmune cascade is London, 1996, pp. 275–284.

[12] M.C. Dalakas, Intravenous immunoglobulin therapy for neruological

involved in the development of nerve injury-induced

diseases, Ann. Intern. Medicine 126 (1997) 721–730.

mechanical allodynia. Further research needs to be done to

[13] J.A. DeLeo, R.W. Colburn, M. Nichols, A. Malhotra,

Interleukin-6-determine specific mechanisms of action, optimally effec- _{mediated hyperalgesia / allodynia and increased spinal IL-6} expres-tive doses, dosing schedules, and administration routes of sion in a rat mononeuropathy model, J. Interferon Cytokine Res. 16

anti-IL6 or IgG therapy while minimizing side effects if (1996) 695–700.

[14] J.A. DeLeo, R.W. Colburn, A.J. Rickman, Cytokine and growth

this approach is to be considered in the treatment of

factor immunohistochemical spinal profiles in two animal models of

neuropathic pain in humans.

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

[15] I. Dibirdik, N. Durak, E. Kislauglu, T. Kutluay, C. Aytemiz, Effects of prophylactic intravenous immunoglobulin-G therapy on humoral and cellular immune components and their functions in burned Acknowledgements

patients, Burns 21 (1995) 130–135.

[16] J.M. Dwyer, Manipulating the immune system with immune

The authors would like to graciously acknowledge Dr. globulin, New England J. Med. 326 (1992) 107–116.

Nancy Birkmeyer for statistical analyses and consultation [17] J. Gauldie, C. Richard, D. Harnish, P. Lansdorp, H. Baumann, Interferon b2 / B-cell stimulating factor type 2 shares identity with

and Tracy Wynkoop for editorial assistance. The research

monocyte-derived hepatocyte-stimulating factor and regulates the

was supported by: National Institute of Drug Abuse grant

acute phase response in liver cells, Proc. Natl. Acad. Sci. USA 84

DA11276 (JAD). _{(1987) 7251–7259.}

[18] A. Geiss, E. Varadi, K. Steinbach, H. Bauer, F. Anton, Psycho-neuroimmunological correlates of persisting sciatic pain in patients who underwent discectomy, Neurosci. Lett. 237 (1997) 65–68.

References _{[19] T. Hama, Y. Kushima, M. Miyamoto, M. Kubota, N. Takei, H.}

Hatanaka, Interleukin-6 improves the survival of mesencephalic [1] Y. Abe, A. Horiuchi, M. Miyake, S. Kimura, Anti-cytokine nature of catecholaminergic and septal cholinergic neurons from postnatal, natural human immunoglobulin: one possible mechanism of the two-week-old rats in cultures, Neuroscience 40 (1991) 445–452. clinical effect of intravenous immunoglobulin therapy, Immunol. [20] M.B. Hansen, M. Svenson, M. Diamant, K. Bendtzen, Anti-inter-Rev. 139 (1994) 5–19. leukin-6 antibodies in normal human serum, Scand. J. Immunol. 33 [2] Y. Akaneya, M. Takahashi, H. Hatanaka, Interleukin-1b enhances (1991) 777–781.

(10)

[21] H. Hirohata, H. Kiyama, T. Kishimoto, T. Taga, Accelerated nerve [36] J.W. Newburger, M. Takahashi, J.C. Burns, A.S. Beiser, K.I. Chung, regeneration in mice by upregulated expression of interleukin 6 C.E. Duffy, M.P. Glode, W.H. Mason, V. Reddy, S.P. Sanders et al., (IL-6) and IL-6 receptor after trauma, J. Exp. Med. 183 (1996) The treatment of Kawasaki’s syndrome with intravenous gamma

2627–2634. globulin, New England J. Med. 315 (1986) 341–347.

[22] Z.F. Huang, J.B. Massey, D.P. Via, Differential regulation of [37] T. Oka, K. Oka, M. Hosoi, T. Hori, Intracerebroventricular injection cyclooxygenase (COX-2) mRNA stability by interleukin-1 beta and of interleukin-6 induces thermal hyperalgesia in rats, Brain Res. 692 tumor necrosis factor-alpha in human in vitro differentiated macro- (1995) 123–128.

phages, Biochem. Pharmacol. 59 (2000) 187–194. [38] T. Pap, D. Reinhold, J. Kekow, Effects of intravenous immuno-[23] K. Ikeda, Y. Iwasaki, T. Shiojima, M. Kinoshita, Neuroprotective globulins on disease activity and cytokine plasma levels in

effects of various cytokines on developing spinal motorneurons rheumatoid arthritis, Can. J. Rheumatol. 27 (1998) 157–159. following axotomy, J. Neurol. Sci. 135 (1996) 109–113. [39] M.S. Ramer, P.G. Murphy, P.M. Richardson, M.A. Bisby, Spinal [24] A. Inoue, K. Ikoma, N. Morioka, K. Kumagia, T. Hashimoto, I. nerve lesion-induced mechanoallodynia and adrenergic sprouting in Hide, Y. Nakata, Interleukin-1 beta induces substance P release from sensory ganglia are attenuated in interleukin-6 knockout mice, Pain primary afferent neurons through the cyclooygenase-2 system, J. 78 (1998) 115–121.

Neurochem. 73 (1999) 2206–2213. [40] C. Ross, M. Svenson, M.B. Hansen, G.L. Vejlsgaard, K. Bendtzen, [25] P. Imbach, S. Barandum, V. d’Apuzzo, C. Baumgartner, A. Hirt, A. High avidity IFN-neutralizing antibodies in pharmaceutically

pre-Morell, E. Rossi, M. Schoni, M. Vest, H.P. Wagner, High-dose pared human IgG, J. Clin. Invest. 95 (1995) 1974–1978. intravenous gammaglobulin for idiopathic thrombocytopenic pur- [41] C. Ross, M. Svenson, H. Nielsen, C. Lundsgaard, M.B. Hansen, K. pura in childhood, Lancet 1 (1981) 1228–1231. Bendtzen, Increased in vivo antibody activity against interferon [26] R.J. Kurlander, Reversible and irreversible loss of Fc receptor alpha, interleukin-1alpha, and interleukin-6 after high-dose Ig

function of human monocytes as a consequence of interaction with therapy, Blood 90 (1997) 2376–2380.

immunoglobulin G, J. Clin. Invest. 66 (1980) 773–781. [42] E. Schuller, A. Goverts, First results of immuno-therapy with [27] R.J. Kurlander, J. Hall, Comparison of intravenous gammaglobulin immunoglobulin G in multiple sclerosis patients, Eur. Neurol. 22

and a monoclonal anti-Fc receptor antibody as inhibitors of immune (1983) 205–212.

clearance in vivo in mice, J. Clin. Invest. 77 (1986) 2010–2018. [43] E. Sekul, E. Cupler, M. Dalakas, Aseptic meningitis associated with [28] F.D. Lee, The role of interleukin-6 in development, Dev. Biol. 151 high-dose intravenous immunoglobulin therapy: frequency and risk

(1992) 331–338. factors, Annal. Int. Med. 121 (1994) 259–262.

[29] T. Ma, X. Zhu, Interleukin-6 increases the levels of cyclic GMP and [44] M.J. Serou, M.A. DeCoster, N.G. Bazan, Interleukin-1 beta acti-nitrite in rat hippocampal slices, Eur. J. Pharmacology 321 (1997) vated expression of cyclooxygenase-2 and inducible nitric oxide 343–347. synthase in primary hippocampal neuronal culture: Platelet activat-[30] T.C. Ma, X.Z. Zhu, Effects of intrahippocampal infusion of ing factor as a preferential mediator of cyclooxegenase-2 expression,

interleukin-6 on passive avoidance and nitrite and prostaglandin J. Neurosci. Research 58 (1999) 593–598.

levels in the hippocampus in rats, Arzneimittel-Forschung 50 (2000) [45] C. Sommer, S. Petrausch, T. Lindenlaub, K. Toyka, Neutralizing 227–231. antibodies to interleukin 1-receptor reduce pain associated behavior [31] P. Marz, T. Herget, E. Lang, U. Otten, S. Rose-John, Activation of in mice with experimental neuropathy, Neurosci. Lett. 270 (1999)

gp130 by IL-6 / soluble IL-6 receptor induces neuronal differentia- 25–28.

tion, Eur.J. Neurosci. 12 (1997) 2765–2773. [46] M. Svenson, M.B. Hansen, K. Bendtzen, Binding of cytokines to [32] S.T. Meller, C. Dykstra, D. Grsybicki, S. Murphy, G.F. Gebhart, pharmaceutically prepared human immunoglobulin, J. Clin. Invest.

The possible role of glia in nociceptive processing and hyperalgesia 92 (1993) 2533–2539.

in the spinal cord of the rat, Neuropharmacology 33 (1994) 1471– [47] S. Sweitzer, D. Martin, J.A. DeLeo, Intrathecal interleukin-1

re-1478. ceptor antagonist in combination with soluble tumor necrosis factor

[33] P.G. Murphy, M.S. Ramer, L. Borthwick, J. Gauldie, P.M. Richar- receptor exhibit an anti-allodynic action in a rat model of neuropthic dson, M.A. Bisby, Endogenous interleukin-6 contributes to hyper- pain, Neuroscience, submitted.

sensitivity to cutaneous stimuli and changes in neuropeptids associ- [48] S. Takei, Y.K. Arora, S.M. Walker, Intravenous immunoglobulin ated with chronic nerve constriction in mice, Eur. J. Neurosci. 11 contains specific antibodies inhibitory of activation of T cells by (1999) 2243–2253. staphylococcal toxin superantigen, J. Clin. Invest. 91 (1993) 602– [34] M. Nagashima, M. Matsushima, H. Matsuoka, A. Ogawa, N. 607.

Okumura, High dose gammaglobulin therapy for Kawasaki disease, [49] F.G. van der Meche, R.P. Kleijweg, J. Meulster, Intravenous J. Pediatrics 110 (1987) 710–712. gammaglobulin (IVGG) versus plasmapheresis trial in patients with [35] M. Nakafuku, T. Satoh, Y. Kaziro, Differentiation factors, including acute Guillain–Barre Syndrome (GBS), Clin. Neurol. Neurosurgery

nerve growth factor, fibroblast growth factor, and interleukin-6, 89 (1987) 62–63. induce an accumulation of an active ras-GTP complex in rat

pheochromocytoma PC12 cells, J. Biol. Chem. 267 (1992) 19448– 19454.

(1)

220 J.L. Arruda et al. / Brain Research 879 (2000) 216 –225

to IL-6, we have also reported spinal increases in

immuno-4. Discussion

reactive-like TNF and IL-1

b

following nerve injury [14].

Sommer et al. [45] demonstrated dose-dependant

attenua-In the present study, we demonstrated significant at-

tion of both hyperalgesia and mechanical allodynia in mice

tenuation of mechanical allodynia following L5 spinal

with experimental neuropathy by i.t. administration of

nerve transection by intrathecal injection of neutralizing

neutralizing antibodies to the IL-1

b

receptor (10–80

m

g /

anti-IL-6. We also found in the same model that low doses

animal). Based on these results, it is evident that there are

of intrathecally administered normal goat and rat IgG were

several neuroimmune factors influencing pain-related

be-able to significantly attenuate enhanced tactile sensitivity.

havioral responses following peripheral nerve injuries.

These data suggest that both specific and global central

Therefore, it is of particular interest that intrathecally

immune manipulation alters nociceptive sensory process-

administered normal goat and rat IgG were able to

ing after a peripheral nerve injury.

significantly attenuate the mechanical allodynia following

Neuroimmune interactions are highly complex and

an L5 spinal nerve transection. Over the past twenty years

multi-faceted. Recently reported findings indicate that a

normal human IVIg therapy has been used with success in

variety of cytokines are involved in neuropathic pain

humans for a variety of autoimmune disorders, primary

including IL-6, Interleukin 1-

b

(IL-1

b

), and Tumor Necro-

immunodeficiencies, and Kawasaki disease [5,16,25] to

sis Factor (TNF). In addition to the previously mentioned

both ameliorate disease and alleviate symptoms. For

studies indicating a role for IL-6, Ramer et al. [39] have

example, IVIg treatment in rheumatoid arthritis (RA)

shown that IL-6 knockout mice have delayed mechanical

patients has been shown to decrease plasma IL-6 levels,

allodynia in response to L5 spinal nerve tight ligation and

morning stiffness, and pain levels [38] associated with RA.

transection when compared to the parent strain. In addition

Given the nature of the illnesses treated with IVIg, it is

(2)

(3)

222 J.L. Arruda et al. / Brain Research 879 (2000) 216 –225 Table 1

Astrocytic responses (illuminated by GFAP ir) of individual animals at the L5 spinal cord level at day 10 post L5 spinal nerve transection with intrathecal treatment on days 0, 1, and 3

Intrathecal Ipsilateral (injured) side Contralateral (uninjured) side

treatment

DH VH DH VH

Normal, untreated 2 2 2 2

2 2 2 2

Surgery alone 1 1 1 1

111 11 11 11

11 1 2 2

Saline 11 1 11 1

1 1 11 1

11 1 1 1

LD anti-IL6 11 11 1 11

11 1 1 1

11 2 1 2

LD goat IgG 11 11 111 1

11 1 1 1

11 1 11 1

LD rat IgG 11 2 1 1

11 1 11 1

111 11 11 2

HD anti-IL6 11 1 1 1

11 1 11 1

111 1 11 11

HD goat IgG 11 1 11 1

11 1 11 2

11 1 1 2

HD rat IgG 11 1 11 1

111 1 1 1

1 1 1 1

Staining / morphology key: baseline (2), mild (1), moderate (1), or intense (111) responses. VH5ventral horn, DH5dorsal horn, LD5low dose, HD5high dose.

conceivable that the effective attenuation seen with normal

study, the normal IgG donor animals also had antibodies

IgG treatment in our rodent model is related to the

that had direct and / or indirect effects on the cytokine

therapy’s immunomodulatory actions.

levels in the spinal cord milieu following nerve injury.

A thorough understanding of the mechanisms of action

Other proposed mechanisms for IVIg therapy include the

of IVIg treatment is currently unknown, but there are

neutralization of disease related autoantibodies [12,16],

several proposed mechanisms of action, some of which

inhibition of complement binding and the resultant

mem-may apply to the reduction in allodynia we observed in the

branolytic attack complex [4,10,11], modulation and

present study. Of interest in our model is the ability of

blocking of the antibody’s Fc (constant fragment) receptor

IVIg therapy to block cytokines and growth factors. IVIg

[26,27], and neutralizing effects of superantigen epitopes

has been shown in human immune disorders to decrease

[48].

the levels and / or production of interferon [40], TNF,

Based on the many potentially diverse effects of IVIg

IL-1

b

, transforming growth factor-

b

[1,46] and IL-6 [38].

therapy and our previous work showing that nerve injury

The blockade could act directly from antibodies in the

appears to induce a robust spinal neuroimmune response, it

IVIg preparation against the cytokines or indirectly by

is not surprising that intrathecal administration of IgG

disruption of precursor components of the immune cas-

would be able to significantly attenuate mechanical

al-cade. The presence of low levels of anti-cytokine anti-

lodynia in our current study. Interestingly, the higher doses

bodies has been shown in normal, healthy humans [20].

of either rat or goat IgG were unable to elicit the same

Based on this knowledge, it is conceivable that in our

effect, demonstrating a reverse U-shaped dose–response

(4)

Table 2

Microglial responses (as illuminated by OX42 ir) of individual animals at the L5 spinal cord level at day 10 post L5 spinal nerve transection with intrathecal treatment on days 0, 1, and 3

Intrathecal Ipsilateral (injured) side Contralateral (uninjured) side

treatment

DH VH DH VH

Normal, untreated 2 2 2 2

2 2 2 2

Surgery alone 1 11 2 2

1 11 2 2

11 1 1 1

Saline 111 111 1 1

111 111 11 11

11 11 1 1

LD anti-IL6 11 11 1 1

11 11 11 11

111 111 1 1

LD goat IgG 11 11 1 1

111 111 1 1

111 11 1 1

LD rat IgG 11 11 1 1

11 1 1 1

111 11 1 1

HD anti-IL6 11 11 2 2

111 11 1 1

11 11 1 1

HD goat IgG 11 111 1 1

111 111 11 11

111 11 1 1

HD rat IgG 111 111 1 1

1 1 1 1

11 11 1 1

Staining / morphology key: baseline (2), mild (1), moderate (1), or intense (111) responses. VH5ventral horn, DH5dorsal horn, LD5low dose, HD5high dose.

curve. The reasons for this have not yet been elucidated,

IgG in our model than those currently used in human

but it is possible that when the system is overwhelmed

disease [9]. In many of the human dosing schedules, IgG

with IgG it is no longer able to exert its beneficial actions

was gradually administered over an extended period of

on the resultant immune cascade, i.e. may induce a

time through a continuous intravenous drip and, thereby,

compensatory proinflammatory cascade. Clearly, a more

attained much higher doses than those employed in the

complete understanding of the mechanism(s) by which the

present study. High doses of IgG given in a short time

immune system is being effected to reduce mechanical

period are associated with adverse effects such as

head-allodynia by i.t. IgG therapy is needed if this potential

ache, chills, myalgia, and chest discomfort [9]. In view of

method of therapy is to be utilized and / or optimized for

the difficulty to accurately monitor side effects,

administra-clinical treatment.

tion limitations, and the animals’ overall well being, we

Due to the difficulties of intrathecal administration,,

chose to utilize conservative, instead of high doses of IgG

peripheral therapies are preferable to intrathecal ones. For

via intravenous injections. Our inability to attenuate

me-this reason, we chose to explore the option of peripheral

chanical allodynia, therefore, could be due to the

con-IL-6 and IgG therapy to alter mechanical allodynia in our

centration administered. Secondly, the lack of attenuation

L5 transection model. However, unlike i.t. IgG administra-

could be the result of the time course we chose to use.

tion, peripheral IgG therapy was unable to attenuate

Third, and perhaps of most importance, the

concen-mechanical allodynia. There are several plausible reasons

tration that was able to penetrate the CNS from the

for this result. First, we employed lower doses of systemic

periphery may not have been adequate for spinal

(5)

modula-224 J.L. Arruda et al. / Brain Research 879 (2000) 216 –225

survival and interleukin-6 protects against MPP1neurotoxicity in

tion of the cascade resulting from L5 spinal nerve injury.

cultures of fetal rat dopaminergic neurons, Exp. Neurol. 136 (1995)

Clearly, further research on the use of IgG therapy needs to

44–52.

be explored to elucidate the optimal dosing concentration /

_{[3] J.L. Arruda, R.W. Colburn, A.J. Rickman, M.D. Rutkowski, J.A.}

schedule, its mechanism of central action, and its potential

DeLeo, Increase of interleukin-6 mRNA in the spinal cord following

role in the alleviation of neuropathic pain.

peripheral nerve injury in the rat: potential role of IL-6 in neuro-pathic pain, Mol. Brain Res. 62 (1998) 228–235.

The pharmacological interventions used in this study

[4] M. Bastam, M.C. Dalakas, High dose intravenous immunoglobulin

were unable to alter robust spinal astrocytic or microglial

exerts its beneficial effect in patients with dermatomyositis by

activation that occurs following a peripheral nerve injury

_{blocking endomysial deposition of activated complement fragments,}

suggesting distinct mechanisms of IL-6 and IgG modula-

_{J. Clin. Invest. 94 (1994) 1729–1735.}

tion and spinal glial changes. In addition, the sole use of

[5] S.A. Berkman, M.L. Lee, R.P. Gale, Clinical uses of intravenous immunoglobulins, Ann. Intern. Med. 112 (1990) 278–292.

GFAP and the CR3 / CD11b antigen may be inadequate

[6] R.W. Colburn, J.A. DeLeo, A.J. Rickman, M.P. Yeager, P. Kwon,

markers to fully assess the contribution of this large glial

W.F. Hickey, Dissociation of microglial activation and neuropathic

sub-population of the CNS. At the present time, the

_{pain behaviors following peripheral nerve injury in the rat, J.}

complexities of the spinal neuroimmune response to a

_{Neuroimmunology 79 (1997) 163–175.}

nerve injury are incompletely elucidated. It appears, how-

[7] R.W. Colburn, A.J. Rickman, J.A. DeLeo, The effect of site and type of nerve injury on spinal glial activation and neuropathic pain

ever, that the major events in a peripheral inflammatory

behavior, Exp. Neurol. 157 (1999) 289–304.

response following tissue damage also occurs in the CNS

[8] B.G. Covino, R. Dubner, J. Gybels, H.W. Losterlitz, J.C. Liebeskind,

after a distant nerve lesion. This innate immunologic

R.A. Sternbach, L. Vylicky, H. Yamamura, M. Zimmerman, Ethical

reaction in the CNS is comprised of anatomic, physiologic

_{standards for investigations of experimental pain in animals, Pain 9}

and inflammatory processes that, if left unabated, may

(1980) 141–143.

[9] M. Dalakas, Mechanism of action of intravenous immunoglobulin

facilitate spinal hypersensitization, the pathophysiologic

and therapeutic considerations in the treatment of autoimmune

correlate of persistent neruopathic pain. Future plans

neurologic diseases, Neurology 51 (1998) S1–S8.

include the investigation of centrally administered spinal

[10] M.C. Dalakas, Clinical relevance of IVIg in the modulation of the

anti-IL-6 and IgG antibodies on cellular adhesion molecule

_{complement-mediated tissue damage: implication in}

dermato-and Major Histocompatibility Complex expression after a

myositis, Guillain Barre syndrome and myasthenia gravis, in: M.

peripheral nerve injury.

Kazatchkine, A. Morel (Eds.), Intravenous Immunoglobulin Re-search and Therapy, Parthenon Publishing, London, 1996, pp. 89–

In conclusion, this study showed that i.t. administration

93.

of neutralizing IL-6 antibody and normal rat and goat IgG

[11] M.C. Dalakas, Experience with IVIg in the treatment of

inflamma-were effective in attenuating mechanical allodynia in a rat

_{tory myopathies, in: M. Kazatchkine, A. Morel (Eds.), Intravenous}

mononeuropathy model. These data are supportive of our

Immunoglobulin Research and Therapy, Parthenon Publishing,

hypothesis that a complex central neuroimmune cascade is

London, 1996, pp. 275–284.

[12] M.C. Dalakas, Intravenous immunoglobulin therapy for neruological

involved in the development of nerve injury-induced

diseases, Ann. Intern. Medicine 126 (1997) 721–730.

mechanical allodynia. Further research needs to be done to

[13] J.A. DeLeo, R.W. Colburn, M. Nichols, A. Malhotra,

Interleukin-6-determine specific mechanisms of action, optimally effec-

_{mediated hyperalgesia / allodynia and increased spinal IL-6}

expres-tive doses, dosing schedules, and administration routes of

sion in a rat mononeuropathy model, J. Interferon Cytokine Res. 16

anti-IL6 or IgG therapy while minimizing side effects if

(1996) 695–700.

[14] J.A. DeLeo, R.W. Colburn, A.J. Rickman, Cytokine and growth

this approach is to be considered in the treatment of

factor immunohistochemical spinal profiles in two animal models of

neuropathic pain in humans.

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

[15] I. Dibirdik, N. Durak, E. Kislauglu, T. Kutluay, C. Aytemiz, Effects of prophylactic intravenous immunoglobulin-G therapy on humoral and cellular immune components and their functions in burned

Acknowledgements

patients, Burns 21 (1995) 130–135.

[16] J.M. Dwyer, Manipulating the immune system with immune

The authors would like to graciously acknowledge Dr.

globulin, New England J. Med. 326 (1992) 107–116.

Nancy Birkmeyer for statistical analyses and consultation

[17] J. Gauldie, C. Richard, D. Harnish, P. Lansdorp, H. Baumann, Interferon b2 / B-cell stimulating factor type 2 shares identity with

and Tracy Wynkoop for editorial assistance. The research

monocyte-derived hepatocyte-stimulating factor and regulates the

was supported by: National Institute of Drug Abuse grant

acute phase response in liver cells, Proc. Natl. Acad. Sci. USA 84

DA11276 (JAD).

_{(1987) 7251–7259.}

[18] A. Geiss, E. Varadi, K. Steinbach, H. Bauer, F. Anton, Psycho-neuroimmunological correlates of persisting sciatic pain in patients who underwent discectomy, Neurosci. Lett. 237 (1997) 65–68.

References

_{[19] T. Hama, Y. Kushima, M. Miyamoto, M. Kubota, N. Takei, H.}

Anti-inter-Rev. 139 (1994) 5–19. leukin-6 antibodies in normal human serum, Scand. J. Immunol. 33

(6)

2627–2634. globulin, New England J. Med. 315 (1986) 341–347.

function of human monocytes as a consequence of interaction with therapy, Blood 90 (1997) 2376–2380.

and a monoclonal anti-Fc receptor antibody as inhibitors of immune (1983) 205–212.

(1992) 331–338. factors, Annal. Int. Med. 121 (1994) 259–262.

343–347. synthase in primary hippocampal neuronal culture: Platelet

activat-[30] T.C. Ma, X.Z. Zhu, Effects of intrahippocampal infusion of ing factor as a preferential mediator of cyclooxegenase-2 expression, interleukin-6 on passive avoidance and nitrite and prostaglandin J. Neurosci. Research 58 (1999) 593–598.

levels in the hippocampus in rats, Arzneimittel-Forschung 50 (2000) [45] C. Sommer, S. Petrausch, T. Lindenlaub, K. Toyka, Neutralizing

227–231. antibodies to interleukin 1-receptor reduce pain associated behavior

[31] P. Marz, T. Herget, E. Lang, U. Otten, S. Rose-John, Activation of in mice with experimental neuropathy, Neurosci. Lett. 270 (1999) gp130 by IL-6 / soluble IL-6 receptor induces neuronal differentia- 25–28.

The possible role of glia in nociceptive processing and hyperalgesia 92 (1993) 2533–2539.

in the spinal cord of the rat, Neuropharmacology 33 (1994) 1471– [47] S. Sweitzer, D. Martin, J.A. DeLeo, Intrathecal interleukin-1

re-1478. ceptor antagonist in combination with soluble tumor necrosis factor

(1999) 2243–2253. staphylococcal toxin superantigen, J. Clin. Invest. 91 (1993) 602–

[34] M. Nagashima, M. Matsushima, H. Matsuoka, A. Ogawa, N. 607.

nerve growth factor, fibroblast growth factor, and interleukin-6, 89 (1987) 62–63. induce an accumulation of an active ras-GTP complex in rat

pheochromocytoma PC12 cells, J. Biol. Chem. 267 (1992) 19448– 19454.