|Home | About | Journals | Submit | Contact Us | Français|
Varicella-zoster virus (VZV) expresses immediate-early protein 62 (IE62), and zoster is associated with neuropathic pain. Brain-derived neurotrophic factor (BDNF) is involved in the neuronal mechanism underlying pain hypersensitivity. Zoster is associated with prodrome and the robust production of booster antibody to VZV. We hypothesized that the intrathecal production of antibody to IE62 cross-reacting with BDNF and the nerve injury by skin lesions may augment allodynia in zoster by enhancing BDNF activity. One of three monoclonal antibodies against the 268-556 peptide of IE62 recognized BDNF. Immunological cross-reactivity between IE62 and BDNF and the effects of anti-IE62 monoclonal antibody (anti-IE62 MAb) cross-reactivity with BDNF on BDNF activity in cultured neurons were examined. Anti-IE62 MAb and anti-BDNF MAbs recognized the 414-429 peptide of IE62 and the BDNF dimer. Anti-IE62 MAb significantly augmented BDNF-related transcription in neurons and the morphological development of spinal dorsal horn neurons. Sera from patients recognized IE62 and BDNF and enhanced BDNF activity in neurons. The effect of anti-IE62 antibody on mechanical allodynia was characterized by the threshold of allodynia using von Frey filaments in a spinal nerve injury (SNI) in mice. The administration of anti-IE62 MAb to or immunization with cross-reacting IE62 protein to mice significantly enhanced mechanical allodynia on the side with SNI but not on the uninjured side. Anti-IE62 antibody augmented BDNF activity in neurons and allodynia in mice with SNI. The intrathecal production of anti-IE62 antibody augmenting BDNF activity and peripheral nerve injury by zoster may participate in the pathogenesis of allodynia in zoster.
Zoster produces a vesicular rash in a dermatomal distribution with sensory abnormalities, and it causes neurological complications such as zoster-associated pain (ZAP) or postherpetic neuralgia (PHN) (5, 11, 14, 16, 33). Zoster is associated with robust booster antibody production to varicella-zoster virus (VZV), which makes it possible to prevent or modify varicella as varicella-zoster immune globulin. PHN is the most frequent complication of zoster, occurring in 7 to 35% of patients, and it is characterized by the combination of constant pain, lancinating pain, and allodynia. However, the pathophysiology of ZAP and PHN has not been elucidated (10, 17). VZV expresses the immediate-early protein 62 (IE62), a major transactivator of viral genes during lytic infection, and at least genes 4, 21, 29, 62, and 63 in latently infected ganglia (6, 8, 19, 20, 29, 30). Although VZV or its products such as IE62 may be involved in the pathogenesis of PHN (12, 13), VZV is ubiquitously and latently distributed in sensory ganglia after varicella and zoster, but PHN is limited only to some patients with zoster.
Brain-derived neurotrophic factor (BDNF) is a member of the neurotrophin family that plays an important role in the development and plasticity of the peripheral and central nervous systems (3, 37). BDNF is involved in the neuronal mechanism underlying clinical pain hypersensitivity, particularly neuropathic pain (7, 21-23, 27, 37, 38). BDNF from microglia has been reported to participate in the pain hypersensitivity that underlies tactile allodynia (7). We hypothesized that BDNF participates in the pathogenesis of PHN and examined the immune response to VZV and BDNF in sera from patients with zoster and PHN. We found immunological cross-reactivity between IE62 and BDNF, and this cross-reaction of IE62 and BDNF was characterized using anti-IE62 monoclonal antibody (anti-IE62 MAb) and anti-BDNF MAb. Surprisingly, the anti-IE62 MAb recognized both the linear epitope (amino acids 414 to 429, designated p414-429) of IE62 and the conformational epitope of BDNF, and it augmented the functional activity of BDNF in cultured neurons. Sera from patients with zoster and PHN recognized BDNF and augmented BDNF activity in cultured neurons. The augmentation of BDNF by anti-IE62 MAb reduced the threshold to mechanical allodynia in mice with spinal nerve injury (SNI) after the intrathecal administration of anti-IE62 MAb and immunization with glutathione S-transferase (GST)-C covering cross-reacting amino acids of IE62. Thus, this study suggests the possible participation of intrathecal anti-IE62 antibody, which recognizes BDNF, in the pathogenesis of allodynia in zoster or ZAP.
Kawaguchi strains of VZV were propagated in human embryonic lung cells or a human lung cancer cell line (A549 cells) (31). The cells were grown and maintained in Eagle's minimum essential medium supplemented with 10 and 2% fetal bovine serum, respectively.
The partial VZV IE62 proteins were divided as fragments and synthesized as GST fusion proteins to cover whole molecules with overlaps as shown in Fig. Fig.1.1. A recombinant plasmid expressing the GST-IE62 fusion protein was constructed by amplifying the IE62 gene in the VZV gene and inserting the DNA fragment into vector pGEX-4T-1 (GE Healthcare, Piscataway, NJ). After recombinant plasmids were constructed, they were transformed into Escherichia coli BL21. Isopropyl-β-D-thiogalactopyranoside (IPTG; 1 mM) was added for the induction of fusion protein in bacteria and incubated at 37°C or at room temperature for 4 h or overnight. The fusion protein was purified by glutathione Sepharose 4B (GE Healthcare) under conditions suggested by the manufacturer, and its molecular weight was identified by SDS-PAGE.
The constructs of recombinant plasmids were verified by determining their sequences. The verification of the authenticity of the IE62 fusion proteins was performed by Western blotting with zoster immune serum. Fusion proteins 1 to 5 were immunized in rabbits, and their immune sera were used for the immunofluorescent antibody (IFA) test for VZV-infected cells.
GST-IE62 fusion proteins and BDNF protein were loaded on an SDS gel, separated by electrophoresis, and transferred to a nitrocellulose membrane (Millipore, Billerica, MA). After being blocked with skim milk, the membranes were incubated with anti-IE62 MAb (1:10,000 dilution of ascites) or anti-human BDNF monoclonal antibody (0.5-μg/ml dilution; R&D Systems, Inc., Minneapolis, MN) at room temperature for 3 h. After a wash with phosphate-buffered saline (PBS), the membranes were incubated with peroxidase-conjugated goat anti-mouse immunoglobulin G (H+L) [IgG(H+L)] (1:2,000 dilution; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) for 1 h, and then they were developed by the electrogenerated chemiluminescence (ECL) method (Nacalai Tesque, Kyoto, Japan).
A549 cells cultured on the cover glass were infected with the cell-free Kawaguchi strain and were fixed with cold acetone 4 days later. After being blocked with blocking solution (10% normal goat serum, 0.1% Triton-X in PBS), the cover glass was incubated with anti-IE62 MAb (1:50,000 dilution of ascites) or anti-BDNF MAb (1:50 dilution) at 4°C overnight. After being treated with PBS, they were incubated with fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG goat serum (Jackson ImmunoResearch Laboratories, Inc.) at 37°C for 1 h.
Constructs of recombinant plasmids expressing IE62 of the Kawaguchi strain (32) were verified on the basis of their sequences. The immunogenicity of IE62 GST-1 to GST-5 was verified by Western blotting with zoster immune serum, and immune serum samples with these proteins produced in rabbits reacted positively with the nuclei of VZV-infected cells in immunofluorescent antibody tests (31, 32) and with IE62 protein of infected cells in Western blots developed by the ECL method. Thus, the IE62 proteins were validated and used for further characterization as IE62 proteins. Three anti-IE62 MAbs (2-a, 2-b, and 2-c) were produced against the peptide p268-556 of IE62, and these antibody clones were screened by enzyme-linked immunosorbent assay (ELISA) for IE62 GST protein and IFA to VZV-infected cells. Two murine anti-BDNF MAbs (R&D Systems, Inc., and Calbiochem, San Diego, CA) and rabbit anti-BDNF (130-247 peptide) polyclonal antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were used. Three anti-IE62 MAbs (2-a, 2-b, and 2-c), two anti-BDNF MAbs, and one anti-BDNF polyclonal antibody were examined for the recognition of IE62 by Western blotting. Anti-IE62 MAb 2-b and the anti-BDNF MAb from R&D Systems, Inc., showed similar staining profiles for IE62 and BDNF as determined by Western blotting. Therefore, the anti-IE62 MAb 2-b and the anti-BDNF MAb from R&D Systems, Inc., were used as the anti-IE62 and anti-BDNF MAbs in this study.
The anti-IE62 and anti-BDNF MAbs were used to characterize reactivities with IE62 and BDNF by Western blotting, and the reactivities of infected A549 lung cancer cells was determined by the IFA test. A549 cells were infected with the cell-free Kawaguchi strain and fixed with acetone-methanol 4 days after infection for the IFA test. The determination of the cross-reacting epitope was performed by Western blotting. The epitope was the overlapping area between IE62 GST-C and GST-F that corresponded to amino acids 414 to 447 of IE62. The epitope of IE62 recognized by anti-IE62 MAb was analyzed by blocking IE62 on Western blots with three synthetic peptides comprising amino acids 414 to 459. Anti-IE62 and anti-BDNF MAbs were preincubated with 180 μg/ml of the peptide overnight at 4°C and were used for Western blot analysis for IE62-GST-F and BDNF, followed by anti-mouse IgG goat serum conjugated with peroxidase (Jackson ImmunoResearch Laboratories, Inc.).
To assess the effects of IE62 on BDNF action, transgenic mice in which the expression of green fluorescent protein (GFP) is regulated by a glutamic acid decarboxylase 67 (GAD67) promoter (GAD67-GFP knock-in mice) (36) were used. GAD67 is a γ-aminobutyric acid (GABA)-synthesizing enzyme, and thus almost all of the GABAergic neurons are labeled by fluorescence from GFP. The present data were obtained from GFP-negative neurons, which most probably were excitatory neurons. The experimental procedures met the regulations of the Animal Care Committee of the Riken Brain Science Institute. Neurons were prepared from the dorsal horn of the spinal cord of transgenic mice at postnatal days 3 and 4 (18) and were cultured at low density. Recombinant human BDNF (provided by Sumitomo Pharmaceutical Co., Ltd., Osaka, Japan) was supplied to neurons through the culture medium at 100 ng/ml on the first day of culture and was replaced every 2 to 3 days. K252a (Kyowa Hakko, Co., Tokyo, Japan), an inhibitor of the activity of the BDNF receptor TrkB, was applied at 200 nM.
A piece of the dorsal horn was removed from transverse slices (500 μm thick) of the lumbosacral enlargement of the spinal cord, enzymatically dissociated with papain (20 U/ml), and triturated with a glass pipette (24). Neurons were plated on a previously prepared glial feeder layer and grown in a solution based on Neurobasal A medium (GIBCO, Rockville, MD) supplemented with 5% B27 (GIBCO). Ten days after being plated, the neurons were fixed with 4% paraformaldehyde (Sigma, St. Louis, MO) and 4% sucrose in PBS (pH 7.0) for 20 min at room temperature, incubated in PBS containing 0.2% Triton-X for 1 min, and blocked with 10% goat serum in PBS for 1 h at 37°C. Anti-microtubule-associated protein (MAP2) MAb (isotype IgG1; 1:250; Sigma) and anti-GFP rabbit polyclonal antibody (1:500; Medical & Biological Laboratories, Nagoya, Japan) were applied for 2 h at 37°C. MAP2 and GFP were visualized by being stained with an isotype-specific secondary antibody conjugated with Alexa 546 (1:2,000; Molecular Probes, Eugene, OR) and anti-rabbit secondary antibody conjugated with Alexa 488 (1:200), respectively. Fluorescent signals were observed with a 40× magnification, 0.75-numeric aperture (NA) objective (UPlan FLN; Olympus, Tokyo, Japan) attached to an inverted epifluorescence microscope (IX71; Olympus) and captured using a cooled charge-coupled-device (CCD) camera (DV43; MicroBrightField, Williston, VT). This system produced an image that comprised 1,360 by 1,036 pixels, each of which corresponded to 0.16 by 0.16 μm with the 40× objective. Filters (NIBA and WIGA; Olympus) were used for two-color immunofluorescence detection. After the immunocytochemical staining and recording of fluorescent images, neurons were incubated with anti-mouse IgG1 conjugated with biotin for 1 h at 37°C. An ABC kit (Vector Laboratories, Burlingame, CA) was used for MAP2 visualization. Neurolucida software (MicroBrightField) attached to an inverted microscope (IX71; Olympus) was used for drawing the soma and dendrites of neurons. The dendritic morphology was quantitatively assessed using the analyzing software Neuroexplore (MicroBrightField). The statistical estimation of differences was performed using one-way analysis of variance (ANOVA), followed by the Holm's test.
Primary cortical neuron cultures were prepared from the cerebral cortices of Sprague-Dawley rat embryos (Japan SLC, Shizuoka, Japan) on the 17th day of gestation (34). Total cellular RNA was extracted by the acid guanidine phenol-chloroform method with Isogen (Nippon Gene, Japan) (15). The quantitative PCR amplification of rat BDNF exons IV to IX (1), arc cDNA, and β-actin cDNA was performed after DNase treatment using arc (sense, (5′-CGCTGGAAGAAGTCCATCAA-3′; antisense, 5′-GGGCTAACAGTGTAGTCGTA-3′) and β-actin (sense, 5′-TTTGAGACCTTCAACACCCC-3′; antisense, (5′-ACGATTTCCCTCTCAGCTGT-3′) primers, and the expression level of each mRNA was normalized using the level of β-actin mRNA. The thermal profile for PCR was as follows: initial denaturation at 95°C for 10 min, followed by 45 cycles of denaturation at 95°C for 45 s, annealing at 55°C for 45 s, and extension at 72°C for 1 min. An annealing temperature of 57°C was used for the amplification of BDNF exons IV to IX. Fluorescence was acquired at the end of each 72°C extension phase. Reverse transcription-PCR (RT-PCR) was verified by comparing results with and without DNase treatment.
Patients' sera with zoster or PHN were examined for their activity in augmenting the BDNF transcription of cultured neurons. Seven patients' sera were examined for the reactivity of IE62 GST-2 and BDNF by Western blotting, and four of them were examined for their effects on the amounts of BDNF transcripts in neurons treated with BDNF. Acute- and convalescent-phase sera from three patients with zoster were evaluated for their time-course effects on the amounts of BDNF transcripts. We performed the experiments to determine the optimal conditions for this assay to evaluate the effect of patient sera on the amounts of BDNF transcripts. Potassium in serum enhanced the amounts of transcripts, and the serum K+ concentration and the dilution of serum were the two most important factors in the assay. Therefore, serum samples were dialyzed in potassium-free PBS, and the assay was performed and reproduced at the final dilution of 1:10,000.
Male C57BL/6J mice (Japan SLC, Shizuoka, Japan) were used. Experiments were conducted with the approval of the Animal Care Committee at the University of Toyama. To induce the mechanical allodynia in mice, we used the SNI model (35) with some modifications: a unilateral L5 spinal nerve of each mouse was tightly ligated with silk sutures. The mechanical allodynia was assessed by using calibrated von Frey filaments (0.02 to 2.0 g; North Coast Medical, Morgan Hill, CA), and mechanical withdrawal thresholds were determined by pressing both sides of the footpad with von Frey filaments.
Briefly, mice were acclimated in individual clear Plexiglas boxes (8 by 10 by 15 cm) with a wire mesh floor, and the von Frey filament was pressed perpendicularly against the mid-plantar surface of the hindpaw from below the mesh floor and held for 3 to 5 s with it slightly buckled. The smallest filament that caused the animal to flinch or move the paw away from the stimulus five times out of five trials at intervals of 5 s was determined to be the paw withdrawal threshold. Anti-IE62 MAb (1 μg), anti-measles MAb (1 μg), or PBS in 5 μl was administered intrathecally just before the SNI in the first series of SNI experiments.
Mice immunized three times with PBS, 10 μg of GST, or 10 μg of GST-C containing the cross-reacting p414-429 were subjected to SNI, and the antibody response was assessed by the ELISA of GST and IFA in the nuclei of infected cells in the second series of the SNI experiments.
IE62 proteins were produced as GST fusion proteins to cover whole molecules (Fig. 1A and B) and were used to characterize the reactivity of anti-IE62 and anti-BDNF MAbs. Similar amounts of the series of GST fusion proteins, as shown in Fig. Fig.1B,1B, were blotted and probed by the anti-IE62 and anti-BDNF MAbs. One anti-IE62 MAb (2-b) of three anti-IE62 MAbs (2-a, 2-b, and 2-c) to p268-556 recognized BDNF, but the other two did not, as determined by Western blotting. On the other hand, one anti-BDNF MAb (R&D Systems) among three antibodies to BDNF recognized IE62, as determined by Western blotting. Therefore, these anti-IE62 (2-b) and anti-BDNF MAb (R&D Systems) were used as the anti-IE62 and anti-BDNF MAbs for further study. Both anti-IE62 and anti-BDNF MAbs specifically recognized the same fragmented proteins, IE62 GST-2, GST-C, and GST-F (Fig. (Fig.1C),1C), indicating that the epitope common to the three fragments was amino acids 414 to 447 of IE62. Both antibodies recognized IE62 in Western blot analysis but not in immunoprecipitation (data not shown), suggesting their preference for the linear epitope to the native form. On the other hand, two BDNF products showed a 13.6-kDa monomer band in Coomassie brilliant blue staining, but the anti-IE62 and anti-BDNF MAbs similarly recognized the BDNF dimer rather than its monomer (Fig. (Fig.1D),1D), indicating that both antibodies recognized the dimer regardless of their amounts. Both MAbs stained mainly the nuclei of infected cells and did not react with their surrounding uninfected cells (Fig. (Fig.1E).1E). The anti-IE62 and anti-BDNF MAbs recognized amino acids 414 to 447 of IE62 and the BDNF dimer rather than the linear form, indicating that their recognition of BDNF was conformational.
Both MAbs recognized amino acids 414 to 447 of IE62 overlapping among GST-2, GST-C, and GST-F, and their common epitope was identified as p414-429 based on its ability to block the interaction of blotted IE62 with antibodies in Western blot analysis among the three peptides covering amino acids 414 to 459 (Fig. 2A and B). However, the amino acid homology of IE62 with BDNF was not observed in the database, and this peptide failed to block the interactions of anti-IE62 and anti-BDNF MAbs with BDNF (Fig. (Fig.2C),2C), indicating that a conformational epitope was not blocked, or that its affinity to the BDNF dimer was stronger. Taken together, anti-IE62 and anti-BDNF MAbs recognized the linear epitope (p414-429) of IE62 but may recognize the conformational epitope of BDNF, particularly that formed by dimeric BDNF.
The biological importance of this cross-recognition of anti-IE62 MAb with enhanced BDNF activity was characterized as the augmentation of BDNF action on cultured neurons (Fig. (Fig.3).3). BDNF promotes the morphological development of neurons from the spinal dorsal horn, which is located in the pain pathway. Anti-IE62 MAb significantly enhanced this promotive action of BDNF on the development of the soma and dendrites of cultured neurons (Fig. 3A to D) (18, 36). This action of anti-IE62 MAb was blocked by k252a, an inhibitor of the activity of the BDNF receptor TrkB, as shown in Fig. 3A to C, indicating that augmentation by anti-IE62 MAb was due to the BDNF-TrkB interaction and not to the other actions of the anti-IE62 MAb.
The stimulation of TrkB receptors by BDNF activates a certain set of genes, including those for Bdnf and activity-regulated cytoskeleton-associated (arc) protein (2, 4, 25). Although the transcriptions of Bdnf and arc were baseline in rat cortical neurons without BDNF, BDNF alone enhanced the transcription of Bdnf and arc. The combination of BDNF with anti-IE62 MAb significantly augmented their transcription (15, 34), whereas anti-IE62 MAb alone had no effect (Fig. (Fig.3E).3E). Thus, the effect of BDNF on the transcription of Bdnf and arc was significantly augmented by anti-IE62 MAb, and the anti-IE62 MAb morphologically and biochemically augmented BDNF action in neurons from the spinal dorsal horn of 3- to 4-day-old mice and fetal rat cortex, respectively.
Among nine patient serum samples, five reacted with the BDNF dimer and four recognized BDNF and the linear form of IE62-2 (Fig. 4A and B) in Western blot analysis. Three samples recognized the linear form of IE62-2 but not the BDNF dimer. Although the recognition of both IE62-2 and the BDNF dimer was not observed in all patients, the sera from four patients suggested the association of immune responses to IE62-2 and the BDNF dimer.
Patient sera alone had no effect on the number of BDNF transcripts in the rat cortical neurons, but four of five sera significantly augmented those induced by the addition of BDNF (Fig. (Fig.4C).4C). Thus, patient sera had activity similar to that of anti-IE62 MAb.
The paw withdrawal threshold on the ipsilateral side decreased from the presurgical level after SNI (Fig. (Fig.5A).5A). This pronounced mechanical allodynia lasted for at least 2 weeks after SNI. No changes were observed in the thresholds of the contralateral side paw. The intrathecal administration of anti-IE62 MAb (Fig. (Fig.5A)5A) significantly reduced the threshold level of mechanical allodynia in the injured side of mice with SNI (P < 0.05), while the unligated side was not affected. The thresholds of mechanical withdrawal were not influenced in mice intravenously administered anti-IE62 MAb (data not shown).
Mouse immunization with GST-C covering p414-429 significantly reduced the threshold level of mechanical allodynia in mice with SNI (P < 0.05) (Fig. (Fig.5B).5B). In both SNI systems, the pain threshold on the nonligated nerve sides was not affected by the presence of anti-IE62 antibodies. Thus, both the intrathecal administration of anti-IE62 MAb and immunization with GST-C covering p414-429 significantly reduced the threshold of mechanical allodynia on the injured side of mice with SNI, while the threshold of mechanical allodynia was not affected in the noninjured side (Fig. (Fig.55).
The injury of the sensory nerve was important for anti-IE62 MAb in enhancing the responsiveness of the neural network in the spinal pain pathway. This suggests that injury to the sensory nerve by skin lesions caused by zoster and the antibodies to IE62 produced after zoster infection participate in the induction of allodynia, as exemplified by the animal model.
The anti-IE62 (2-b) and anti-BDNF (R&D Systems) MAbs used in this study first were selected from three respective antibodies to IE62 and BDNF by their reactivity with IE62 in Western blot analysis. Among three anti-IE62 MAbs to p268-556, the anti-IE62 MAb to p414-429 did not neutralize but increased the functional activity of BDNF in neurons, indicating the specificity of the anti-IE62 MAb 2-b to this epitope of IE62. One of the possible mechanisms underlying the increase in BDNF activity is similar to that reported in the binding of antibody to CD8. The CD8 molecule of T cells has an enhancing epitope and three other blocking epitopes for antibody binding in CD8α-major histocompatibility complex class I interactions, and enhancing the antibody stabilizes a high-affinity CD8 conformation (9). Although the mechanism underlying the enhancement of BDNF action is not fully understood, bioactive BDNF is a dimer formed by strong hydrophobic interactions (3). Anti-IE62 MAb may stabilize the dimer conformation and increase the affinity to induce strong BDNF-TrkB interaction and signal transduction, resulting in the augmentation of BDNF activity in neurons. The specificity of the augmentation of BDNF activity by anti-IE62 MAb was confirmed by its blocking TrkB with k252a (Fig. (Fig.33).
Although the intravenous administration of anti-IE62 MAb without SNI failed to induce mechanical allodynia (data not shown), its intrathecal administration combined with SNI enhanced the mechanical allodynia of the injured side. Increasing evidence suggests that BDNF plays an important role in the modulation of pain transmission (7, 21-23, 27, 37, 38). The nerve injury induces an increased number of BDNF-like immunoreactive neurons in the ipsilateral dorsal root ganglia (28). In the dorsal horn of the spinal cord, BDNF from microglia contributes to the mechanical allodynia of rats with nerve injury (7). BDNF should protect neurons from SNI, but the augmentation of its activity by anti-IE62 MAb may phenotypically enhance allodynia. The antibody produced intrathecally against this cross-reacting epitope of IE62 in patients with zoster may recognize the enhancing epitope of BDNF activity, as the augmenting activity of BDNF was observed in sera of patients with zoster and PHN (Fig. (Fig.4C).4C). Thus, both the nerve injury and anti-IE62 MAb may be essential to enhance mechanical allodynia.
Zoster induces strong antibody production to viral antigens, and it is quite different from varicella. This strong antibody production made it possible to detect the antibody to IE62 in patients with zoster. Similarly, antibody to IE62 may be produced after the expression of viral proteins in the infected ganglion, and BDNF produced in the injured neurons (13) may cause strong allodynia by augmenting BDNF activity via the interaction of BDNF with anti-IE62 antibody produced intrathecally.
Herpes simplex virus (HSV) causes apparently similar vesicular skin lesions, but strong allodynia is not associated with them, possibly because HSV may not have a kind of anti-IE62 antibody augmenting BDNF activity in addition to the difference in the neural damage caused by HSV and VZV. The frequency and amount of antibody to this cross-reacting epitope of IE62 in immune response, the nature of the nerve injured by zoster skin lesions, and host BDNF response in its intrathecal production and susceptibility may be the important determinants of allodynia in patients with zoster. In this study, the augmentation of BDNF activity by anti-IE62 MAb induced significantly stronger allodynia in the injured nerve, but further study is needed to discuss its relation to the pathogenesis of PHN.
The enhancement of BDNF action by anti-IE62 antibody and the injured nerve due to zoster skin lesions may in turn enhance the responsiveness of the neural network in the spinal pain pathway so that patients become hypersensitive to stimuli that otherwise are not very painful. The BDNF activity secreted by microglia (26) may be increased by this antibody, which may render the patient more susceptible to the pain hypersensitivity that underlies tactile allodynia (7). Thus, the intrathecal immune response to IE62 and nerve injury by zoster may participate in causing allodynia augmented by BDNF in patients with zoster.
We thank Katherine Ono for editing the manuscript. We express our thanks to Y. Yanagawa and K. Obata for providing GAD67-GFP knock-in mice and to K. Murase for technical advice on the preparation of spinal cord slices. We also thank Sumitomo Pharmaceutical Co., Ltd., for the kind gift of recombinant human BDNF.
This study was supported in part by a grant for Research Promotion of Emerging and Re-emerging Infectious Diseases (H18-Shinko-013) from the Ministry of Health, Labor and Welfare of Japan, and a grant-in-aid (135508094) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
The authors have no conflicting financial interests.
Published ahead of print on 18 November 2009.