SEARCH

SEARCH BY CITATION

Keywords:

  • carrageenan;
  • cyclooxygenase-2;
  • endothelial cells;
  • hyperalgesia;
  • inflammation;
  • prostaglandin

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

Inflammation is often accompanied with hyperalgesia. This hyperalgesia is mediated partly by prostaglandin(s) produced in the CNS through the cyclooxygenase-2 (COX-2) dependent pathway. However, it remains unclear where COX-2 is induced in the CNS during inflammation, and how it is involved in hyperalgesia. We studied the precise site of COX-2 induction in the CNS, the relation between the time course of COX-2 induction and that of hyperalgesia, and the effect of COX-2-selective inhibitor by using a carrageenan model. Carrageenan injection induced expression of COX-2-like immunoreactivity in vascular endothelial cells throughout the CNS. This response became evident by 3 h, and was most prominent at 6 h after carrageenan injection. This COX-2 induction was associated with an elevation of prostaglandin E2 in the cerebrospinal fluid, being evident at 3 h, larger at 6 h, and alleviated by a COX-2-selective inhibitor. Thermal hyperalgesia became evident at 1 h, further increased thereafter, and remained elevated until 6 h. Intrathecal administration of COX-2-selective inhibitor 2 h after the carrageenan injection exerted a prominent therapeutic effect on hyperalgesia. These results demonstrate that, during carrageenan-induced inflammation, endothelial cells are the major source of prostaglandin(s) in the CNS, and this endothelial expression of COX-2 is involved in the inflammation-induced hyperalgesia.

Abbreviations used
COX-2

cyclooxygenase-2

DAB

diaminobenzidine

MPO

medial preoptic area

NGS

normal goat serum

NSAIDs

non-steroidal anti-inflammatory drugs

PGs

prostaglandins

Tissue inflammation produces spontaneous activity in otherwise silent small primary afferent axons and, consequently, evokes behavioural hyperalgesia. This peripheral hypersensitivity of the altered primary afferent can be explained in part by a local release of pro-inflammatory substances, such as bradykinin, cytokines and prostaglandins (PGs), which activate and sensitize the peripheral nerve endings (England et al. 1996; Gold et al. 1996).

Non-steroidal anti-inflammatory drugs (NSAIDs), which are widely used for therapeutic treatment of inflammatory pain, exert their action in preventing the sensitization of the peripheral afferent terminals through inhibition of cyclooxygenase (COX), an enzyme essential for the biosynthesis of PGs (Vane 1971). Owing to recent advances in molecular biology, two isozymes of COX have been cloned and crystallized, i.e. COX-1 and COX-2 (Kujubu et al. 1991; O'Banion et al. 1991; Xie et al. 1991; Smith and DeWitt 1996). COX-1 is constitutively expressed in most tissues and organs, whereas COX-2, an inducible type, is localized primarily to inflammatory cells and tissues after cellular activation by cytokines or mitogens.

Later studies following supraspinal and spinal administration of NSAIDs (Yaksh 1982) have demonstrated that there are also the sites in the CNS where NSAIDs act to exert an anti-hyperalgesic effect. This indicates that PG synthesis in the CNS is also involved in nociception and hyperalgesia. Recently, the anti-hyperalgesic effect (Seibert et al. 1994; Hay and Belleroche 1997; Hay et al. 1997; Zhang et al. 1997; Dirig et al. 1998; Smith et al. 1998) and the inhibitory effect on PGE2 synthesis (Zhang et al. 1997; Smith et al. 1998) of some highly selective COX-2 inhibitors have been demonstrated. Furthermore, up-regulation of COX-2 mRNA or COX-2 protein was reported both in the spinal cord and at the inflammation site following peripheral inflammation (Seibert et al. 1994; Anderson et al. 1996; Beiche et al. 1996; Hay and Belleroche 1997; Hay et al. 1997; Ichitani et al. 1997; Beiche et al. 1998a). Smith et al. (1998) demonstrated that COX-2 selective inhibitor, but not COX-1 selective inhibitor, suppresses hyperalgesia evoked by subcutaneous injection of carrageenan in rats, and this anti-hyperalgesic action of the COX-2 inhibitor is associated with suppression of the PGE2 level in the CSF. These results suggest the importance of the central COX-2 in hyperalgesia during inflammation. Little is known, however, about the type of cells in the CNS that express COX-2 during inflammation, and how they are involved in hyperalgesia.

In the present study, we hypothesized that COX-2 is induced in a certain type of cells in the CNS following intraplanter injection of carrageenan, and this COX-2 induction is involved in the hyperalgesic response. To test this hypothesis, we studied where COX-2 is induced in the CNS, whether the time course of its induction can explain that of behavioural hyperalgesia, and whether inhibition of CNS COX-2 activity suppresses the hyperalgesia.

Materials

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

Two commercially available anti-COX-2 antibodies were used in this study. Rabbit anti-murine COX-2 polyclonal antibody was purchased from Cayman Chemical (Ann Arbor, MI, USA). This antibody was raised against a synthetic 17-mer peptide that is unique to the C-terminus of this protein. Another one, goat anti-rat COX-2 polyclonal antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Other primary antibodies used and their sources were: OX-42 (Serotec LTD, Oxford, UK), rabbit anti-human von Willebrand (vW) factor (DAKO, Carpinteria, CA, USA), sheep anti-rat vW factor (Affinity Biologicals, Hamilton, Ontario, Canada). Cyclooxygenase inhibitors used in this study were diclofenac (Sigma, St Louis, MO, USA) and NS 398 (Cayman Chemical).

Animal model

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

All experiments were carried out according to protocols approved by the Institutional Animal Care Committee of Kyoto Prefectural University of Medicine. Male Sprague–Dawley rats (Charles River Laboratories Inc., Wilmington, MA, USA, 300–350 g) were housed in a temperature- and humidity-controlled room on a 12-h light/dark cycle with free access to food and water. The planter surface of the left paw received a subcutaneous injection of 6.0 mg carrageenan (Sigma, type IV) dissolved in 200 μL of physiological saline at time zero under halothane anaesthesia.

COX-2 immunohistochemistry

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

Under deep anaesthesia with pentobarbital, the animals were perfused through the left ventricle with 10 mm phosphate-buffered saline (PBS, pH 7.4) and their brains, spinal cords and bilateral feet were immediately removed, freshly frozen in dry-ice powder, and stored at − 80°C until use. For quantitative analysis of the time course of COX-2 expression, brains and spinal cords were cut to a thickness of 20 μm in a cryostat and thaw-mounted on silane-coated glass slides. After being air-dried at 25°C, the sections were fixed with 2% paraformaldehyde in 0.1 m PBS, pH 7.4, for 10 min at room temperature. After rinsing with PBS, they were treated with 0.3% H2O2 in PBS for 30 min followed by incubation in 3% normal goat serum (NGS) and 0.25% Triton X-100 in PBS. Endogenous biotin activity was blocked with a blocking kit (Vector Laboratories, Burlingame, CA, USA) according to the manufacturer's instructions. The sections were then incubated with rabbit anti-murine COX-2 polyclonal antibody diluted 2000 × in PBS containing 3% NGS and 0.25% Triton X-100 for 72 h at 4°C followed by incubation in biotinylated goat anti-rabbit IgG (200 × dilution) for 1 h at room temperature. COX-2-like immunoreactivity was visualized with a Vectastain Elite ABC kit using diaminobenzidine (DAB) as the chromogen. Control staining was conducted without the primary antibody or with pre-absorbed antibody, in which the diluted primary antibody and the antigen peptide (1 μg/mL) had been mixed and incubated at 4°C overnight. Four time points, i.e. 1, 3, 6 and 24 h after carrageenan injection, were chosen for animal death. Control animals received intraplantar injection of physiological saline in the left foot and were killed 3 h later.

For identification of the type of COX-2-positive cells, COX-2 staining was performed in combination with one of two cell-type markers, namely, anti-vW factor (endothelial cell marker) or OX-42 (parenchymal microglia and perivascular/meningeal macrophage marker). Double immunostaining of COX-2 and vW factor was performed with one of three combinations of anti-COX-2 antibody and anti-vW factor antibody, including (i) anti-rat COX-2 (goat IgG) and anti-human vW factor (rabbit IgG), (ii) anti-murine COX-2 (rabbit IgG) and anti-rat vW factor (rabbit IgG), and (iii) anti-murine COX-2 and anti-human vW factor. In all cases, cryostat sections were made and pre-treated in the same way as described above except for skipping H2O2 treatment in all cases and replacing 3% NGS with 3% normal donkey serum when anti-rat COX-2 (goat IgG) was used. The sections were then incubated with anti-murine COX-2 (1 : 2000 dilution) or anti-rat COX-2 (1 : 4000 dilution) overnight at room temperature. Specificity of the staining was confirmed following incubation with the antibodies pre-absorbed with the respective antigen peptide (1 μg/mL). The sections were further incubated with biotinylated anti-rabbit IgG or biotinylated anti-goat IgG. COX-2-like immunoreactivity was visualized with Cy3-conjugated streptavidin. Then, the sections were incubated with either anti-human vW factor or anti-rat vW factor for 1 h, followed by FITC-labelled anti-rabbit IgG or anti-sheep IgG. To exclude the possibility that these FITC-labelled anti-IgG, which were expected to recognize the anti-vW factor antibodies, improperly cross-bound to the anti-COX-2 antibodies, some of the sections that had been stained for COX-2 were incubated with non-immunized rabbit IgG or sheep IgG followed by FITC-labelled anti-IgG.

For double immunostaining of COX-2 and microglia/macrophage marker, the sections were incubated with anti-murine COX-2 antibody pre-mixed with OX-42 (mouse monoclonal antibody, final dilution × 800) for 72 h at 4°C followed by biotinylated anti-rabbit IgG. COX-2 immunoreactivity was visualized using Cy3-streptavidin, and then the sections were treated with an avidin-biotin blocking kit and incubated with biotinylated anti-mouse IgG. Immunoreactivity was visualized with FITC-avidin D. In some cases, the double-stained sections were further stained for nuclear DNA with TOTO-3 (Molecular Probes, Eugene, OR, USA).

Measurement of PGE2 level in the CSF

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

Three or 6 h following the carrageenan injection, CSF was collected from the cisterna magna under pentobarbital anaesthesia. The CSF samples were frozen in liquid nitrogen and stored at −80°C until analysis. PGE2 was extracted from the CSF with ethyl acetate, which was then evaporated off by vacuum centrifuge. PGE2 was measured using a PGE2 EIA kit (Cayman Chemical) according to the manufacturer's instructions. Effects of selective COX-2 inhibitor, NS398, on PGE2 level in the CSF were also examined.

Measurement of thermal hyperalgesia

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

To assess thermally evoked paw-withdrawal response, a commercially available device was used (Hargreaves et al. 1988). Paw-response latency was measured for both the injected (ipsilateral) and uninjected (contralateral) paws before and every 60 min for 360 min after carrageenan injection in separate groups of animals. Animals were assigned randomly to one of the four following treatments: intraperitoneal injection of non-selective COX inhibitor, diclofenac (30 mg/kg), selective COX-2 inhibitor, NS-398 (0.6 mg/kg, 6.0 mg/kg) or vehicle. Drugs were injected 2 h after the intraplantar carrageenan injection.

In order to assess the central effect of selective COX-2 inhibitor, another group of rats was inserted with PE10 catheters intrathecally according to the method of Yaksh and Rudy (1976). Approximately 1 week following the catheter implantation, rats were assigned to the following two groups; intrathecal injection of NS-398 (60 μg/kg) or vehicle. Drugs were injected 2 h after intraplantar carrageenan injection. These inhibitors were dissolved in dimethyl sulfoxide and then diluted 1 : 95 with 25% cyclodextrin in sterile water such that the final dose was dissolved in 0.5 mL for systemic injection and in 10 μL for intrathecal injection.

Quantitative analysis and statistics

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

All the quantitative data are presented as means ± standard errors. Statistical significance of the effect of COX-2 inhibitors was evaluated by the Student t-test for the thermal hyperalgesia and by the Mann–Whitney U-test for PGE2 level in the CSF. In the time-course study of COX-2 induction, numbers of COX-2-like immunoreactive (COX-2-LI) cells were counted in restricted areas of the brain and spinal cord in at least four coronal sections from each rat. The mean value of this number in each rat was further averaged among four or five rats at each time point. In the brain, cell counting was performed in one side of the rostral hippocampus, thalamus, and subarachnoidal space between them. This area was selected because the numbers of COX-2-LI cells were high and the variation in numbers of COX-2-LI cells among the sections was small. In the spinal cord, COX-2-LI cells were counted in the whole area of the coronal sections from L4–L6.

COX-2 immunohistochemistry

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

Figure 1 shows the results of COX-2 immunohistochemistry obtained with the rabbit anti-murine COX-2. As reported previously (Breder et al. 1995), some cortical and hippocampal neurons constitutively expressed COX-2-like immunoreactivity in both saline-treated rats and carrageenan-treated ones (Fig. 1a). This type of staining was completely eliminated when pre-absorbed antibody was used as the primary antibody (Fig. 1c). The distribution of COX-2-LI neuronal cell bodies was not influenced greatly by the carrageenan treatment. In the spinal cord, no COX-2-LI cells were observed in saline-treated group.

image

Figure 1. COX-2-like immunoreactive (COX-2-LI) cells in the brain, spinal cord and left hind paws (ipsilateral to the injection) of rats injected with normal saline (j, o, s) or carrageenan (a–i, k, l–n, p–r) 6 h before death. (a) COX-2-LI neuronal cell bodies in the hippocampus of a saline-injected rat. (b) Vascular endothelial cells (arrowheads) and neuronal cell bodies (arrows) in a carrageenan-injected rat. (c) Same area as in (a) stained with pre-absorbed antibody. (d) Same area as in (b) stained with pre-absorbed antibody. (e–g) Parenchymal blood vessels in a carrageenan-injected rat. (h) Higher magnification of the same blood vessel as in (e); (i, k) COX-2-LI cells in the subarachnoidal space of a carrageenan-injected rat. In (k), asterisk indicates the brain parenchyma and arrowheads indicate the arachnoidal membrane. Thus, COX-2-LI cells were localized to the subarachnoidal blood vessel. (j) No COX-2-LI cells were observed in parenchymal blood vessel of a saline-injected rat; (l), COX-2-LI cells in the parenchymal blood vessels in the spinal cord of a carrageenan-injected rat. (m) COX-2-LI cells in the subarachnoidal space of the spinal cord of a carrageenan-injected rat. (n) COX-2-LI cells in the basement cell layer of the foot skin in a carrageenan-injected rat. Asterisk indicates the epidermis and pound mark indicates the basement cell layer. (o) Skin of the saline- injected hindpaw. (p) Macrophage-like cells in the deeper part of carrageenan-injected footpad. (q) Unidentified cells along a blood vessel in the deeper part of carrageenan-injected footpad.; (r) Higher magnification of the same cells as in (p). (s) Same area as in (q) in a saline-injected foot pad. Scale bars: a–g, I–m, p–q, s, 50 μm; h, n–o, r; 10 μm.

Download figure to PowerPoint

Following the carrageenan injection, COX-2-LI cells of non-neuronal origin became evident, and, in almost all cases, they were associated with blood vessels in the parenchyma and subarachnoidal spaces of the brain (Figs 1b, e–i, k) and spinal cord (Figs 1l and m). COX-2 staining in these cells was most prominent at 6 h following carrageenan injection (the detailed time course is shown later). COX-2-LI structures of the non-neuronal cells were round or oval, and were situated linearly along the vessel walls.

In the hind paw of saline-injected rats, no COX-2-like immunoreactivity was found in either the injected or untreated sides (Figs 1o and s). Following the carrageenan injection, strongly stained COX-2-LI cells appeared along the skin basement cell layer only in the injected side (Fig. 1n). In addition, in the deeper part of the carrageenan-injected footpad, macrophage-like cells (Figs 1p and r) and as yet unidentified cells (Fig. 1q) were also strongly stained for COX-2. All the carrageenan-induced COX-2-LI structures described above were absent following incubation with the pre-absorbed antibody (Figs 1c and d). Another antibody, i.e. goat anti-rat COX-2, provided essentially the same results as those obtained with the rabbit anti-murine COX-2 (see the next section).

Identification of the COX-2-LI cells in blood vessels

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

Figures 2(a1) and 2(a2) show multicolour confocal laser microscopic images of COX-2 (red), vW factor (an endothelial cell marker, green) and nuclear DNA (blue) in serial spinal cord sections obtained 6 h after carrageenan injection. The former section was stained with the antibody pair of goat anti-rat COX-2 and rabbit anti-human vW factor and the latter one with that of rabbit anti-murine COX-2 and sheep anti-rat vW factor. In both cases, COX-2-LI structures were expressed surrounding the nuclear DNA, and were embedded in the vW factor-positive area. The same result was also observed in sections incubated with rabbit anti-murine COX-2 followed by rabbit anti-human vW factor (Fig. 2b1). In this case, as both primary antibodies were of rabbit origin, there was a possibility that FITC-labelled anti-rabbit IgG, which was expected to bind to anti-vW factor rabbit IgG, might also have bound to rabbit anti-COX-2, which had been added to the section first and visualized with Cy3. This possibility was, however, clearly excluded by the control experiment, in which rabbit anti-vW factor IgG was replaced with non-immunized rabbit IgG. Green fluorescence was negligible in this case (Fig. 2b2). Thus, results obtained with all three combinations of antibodies are consistent, indicating that the double immunostaining was properly done. Because vW factor is a cytosolic protein that is specifically present in endothelial cells (Jaffe et al. 1973; Wagner and Marder 1983; Theilen and Kuschinsky 1992), this staining pattern indicates that COX-2-like immunoreactivity is located in the peri-nuclear region of endothelial cells; in other words, COX-2-LI cells are endothelial cells. This was also the case for the blood vessels in brain parenchyma (Fig. 2c1–c4), and those in the subarachnoidal spaces of the brain (Fig. 2d1–d4) and spinal cord (Fig. 2e1–e4).

image

Figure 2. Multicolour laser confocal microscopic images of COX-2 (red in a–e, g), von Willebrand factor (an endothelial marker, green in a–f), OX-42 (a microglia/macrophage marker, green in g), and nuclear DNA (blue) in the brain (c, d) and spinal cord (a, e). The samples were taken 6 h following carrageenan injection. As shown in (a1), goat anti-rat COX-2 antibody recognized the same population of cells as rabbit anti-murine COX-2 shown in (a2). (b1) shows immunostaining with rabbit anti-COX-2 antibody and rabbit anti-vW factor, whereas in (b2) no immunostaining of vW factor was observed by replacing rabbit anti-vW factor to rabbit unimmunized antibody. COX-2-LI structure in blood vessels of the brain parenchyma (c1), and subarachnoidal spaces of the brain (d1) and spinal cord (e1) are shown. The same blood vessels in c1, d1 and e1 were stained with anti-vW factor (c2, d2 and e2, respectively). The COX-2 immunoreactivity (red) was overlaid with vW factor immunostaining (green) (c3, d3 and e3). The double immunostaining in c3, d3 and e3 was overlaid with TOTO-3 nuclear staining (blue) (c4, d4, e4). COX-2-LI cells were scarce (f1) in the arterial endothelial cells (f2). As shown in (g), COX-2-LI cells (red) and OX-42-positive microglia/macrophages (green) were located close to each other but remained distinct and separate. The arrowheads indicate perivascular microglia/macrophages, and the arrows indicate parenchymal microglia.

Download figure to PowerPoint

image

Figure 4. Time-dependent changes in the COX-2-LI cells in brain parenchyma (a–d) and spinal cord subarachnoidal space (e–h). Rats were killed at 1 h (a, e), 3 h (b, f), 6 h (c, g), and 24 h (d, h) following carrageenan injection. Arrows indicate faint expression of COX-2-LI vascular endothelial cells and arrowheads indicate ependymal cells with non-specific staining. Scale bar: 50 μm, *choroid plexus.

Download figure to PowerPoint

Interestingly, COX-2-positive endothelial cells were found mostly in blood vessels with a thin or little smooth muscle layer, but were scarce in vessels with a thick smooth muscle wall as shown in Fig. 2(e and f). These results indicate that COX-2-LI cells are located predominantly in veins or venules rather than in arteries or arterioles. Figure 3 shows the distribution of blood vessels in which endothelial cells expressed COX-2-like immunoreactivity following carrageenan injection. In both brain and spinal cord, the positive blood vessels were diffusely distributed with no obvious regional differences in the distribution.

image

Figure 3. Distribution of blood vessels in which endothelial cells expressed COX-2 in the brain (a) and spinal cord (b) of a carrageenan-treated rat. The rat was killed 6 h following carrageenan injection.

Download figure to PowerPoint

Perivascular microglia and meningeal macrophages may also express COX-2 in the present experimental model as reported in lipopolysaccharide-treated rats (Elmquist et al. 1997). We therefore examined whether the microglia or macrophages also expressed COX-2-like immunoreactivity in addition to the endothelial cells following carrageenan injection. However, double immunostaining with anti-COX-2 and OX-42, which recognizes the complement 3 receptor expressed in parenchymal and perivascular microglia, showed that COX-2-LI cells (red) and OX-42-positive cells (green) were distinct from each other although they were sometimes located close to one another (Fig. 2g). This result indicates that, at least under the present experimental conditions, COX-2-like immunoreactivity is induced primarily in the endothelial cells.

Timing of COX-2 induction in the CNS

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

To study the time course of COX-2 induction and compare it with that of thermal hyperalgesia, four time points, 1, 3, 6 and 24 h following carrageenan injection, were chosen. As a negative control, rats were injected with normal saline and killed 3 h later. Figure 4 shows the time-dependent appearance of COX-2 staining in the endothelial cells of the brain and spinal cord. At 1 h following carrageenan injection, there were no COX-2-LI endothelial cells detectable in the brain (Fig. 4a) and spinal cord (Fig. 4e). COX-2-LI endothelial cells were first found in the CNS at 3 h following carrageenan injection (Figs 4b and f); their number and intensity of immunostaining further increased and reached a maximum at 6 h after the injection (Figs 4c and g) and then decreased but not disappeared by 24 h following injection (Figs 4d and h). At each time point, we calculated the number of COX-2-LI cells in a restricted area of the brain (Fig. 5a) and spinal cord (Fig. 5b) as described in Materials and methods. Only the numbers of COX-2-LI endothelial cells were counted in the brain; the neuronal expression of COX-2 was not influenced by carrageenan injection. The time course of COX-2 induction in the brain and spinal cord followed a similar time course.

image

Figure 5. Time courses of the number of COX-2-LI cells in the brain (a) and spinal cord (b). The data on the numbers of COX-2-LI cells were counted in the same area as shown in Fig. 4. The data are expressed as mean ± SE.

Download figure to PowerPoint

PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

As shown in Fig. 6, carrageenan injection induced a massive increase in PGE2 in the CSF compared with saline injection. This increase was manifest at 3 h and further prominent at 6 h following carrageenan injection. NS-398 (6 mg/kg) administered 2 h following the carrageenan injection strongly suppressed the increase in PGE2 in the CSF at 6 h, whereas that administered 10 min prior to the carrageenan injection was less effective at 6 h.

image

Figure 6. Effects of carrageenan and NS-398 on prostaglandin E2 (PGE2) level in CSF. Animals were injected intraperitoneally with NS-398 (6 mg/kg) (NS) or vehicle (Veh) 10 min before (pre) or 2 h following (post) carrageenan administration (Car). Animals in the NaCl group received saline injection in their paw instead of carrageenan. CSF was sampled 6 h following carrageenan injection and used for PGE2 immunoassay. In some rats with no injection of NS-398, CSF was collected 3 h after the carrageenan injection. The data are expressed as mean ± SE from four to 15 rats in each condition. The asterisk indicates that the therapeutic treatment with NS-398 (post) produced a statistically significant decrease in the PGE2 level in the CSF (p < 0.05). Prophylactic treatment with NS-398 (pre) was less effective at this time point.

Download figure to PowerPoint

Carrageenan-induced thermal hyperalgesia

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

In all rats injected with carrageenan, prominent paw oedema was established in the left hind paw (ipsilateral to carrageenan injection) at 1 h following carrageenan injection and continued thereafter as reported in previous studies. The hyperalgesic response to thermal stimulus (thermal hyperalgesia) as evaluated by paw withdrawal latencies (PWLs) occurred by 1 h following carrageenan injection, further increased thereafter, and remained elevated during the rest of the observation period until 6 h after the injection (Fig. 7, filled square). PWLs in the contralateral side did not significantly change during the 6-h period (Fig. 7, open square).

image

Figure 7. Effects of therapeutic treatment with diclofenac or NS-398 on paw withdrawal latencies (PWLs). Mean PWLs are plotted across the 6-h experimental time course (mean ± SE, n = 15–24 rats per experimental group). All the plot series represent PWLs of the injected side (ipsi) except for one series, which represents PWLs of the contralateral (contra) side for comparison. Carrageenan was injected into the footpad of the left hind paw at time point 0 (arrow C). Drugs were administered intraperitoneally 2 h following the carrageenan injection (arrow D). Diclofenac (Diclo) was administered at the dose of 30 mg/kg, and NS-398 (NS) was administered at doses of 6.0 mg/kg and 0.6 mg/kg. Asterisks indicate significant differences (p < 0.001–0.05) of PWLs in the treatment group with the COX inhibitors from those in the group with vehicle.

Download figure to PowerPoint

Reversal of thermal hyperalgesia by COX inhibitors

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

To assess the involvement of COX-2 in the thermal hyperalgesia, diclofenac (a non-selective COX inhibitor) or NS-398 (a COX-2-selective inhibitor) was administered intraperitoneally 2 h following carrageenan injection. As shown in Fig. 7, both diclofenac and NS-398 caused a rapid suppression of thermal hyperalgesia. High dose of diclofenac or NS-398 completely reversed hyperalgesia to the baseline level and this therapeutic effect lasted throughout the entire experimental period. Low dose of NS-398 partially reversed the hyperalgesia.

To test whether COX-2 expressed in the CNS played a role in the hyperalgesia, effects of intrathecal injection of NS-398 on the hyperalgesia were also studied. As shown in Fig. 8, intrathecal administration of NS-398 (60 μg/kg) 2 h after carrageenan injection partially relieved the hyperalgesia in the period between 3 and 6 h, indicating the involvement of the central COX-2 in the hyperalgesia.

image

Figure 8. Effects of intrathecal treatment with NS-398 on paw withdrawal latencies (PWLs). Mean PWLs are plotted across the 6-h experimental time course (mean ± SE, n = 5–7 rats per experimental group). All the plot series represent PWLs of the injected side. Carrageenan was injected into the footpad of the left hind paw at time point 0 (arrow C). NS-398 (NS), at the dose of 60 μg/kg, was administered intrathecally 2 h following the carrageenan injection (arrow D). Asterisks indicate significant differences (p < 0.001–0.05) of PWLs in the treatment group with NS-398 from those in the group with vehicle.

Download figure to PowerPoint

Identification of COX-2-LI cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

The present study for the first time demonstrated that, in the brain and spinal cord, almost all COX-2-like immunoreactivity newly induced by carrageenan was confined to the perinuclear region of vascular endothelial cells, which was identified by the expression of vW-factor. vW-factor, a coagulant protein, is exclusively expressed in Weibel-Palade bodies of endothelial cells and platelets, the latter of which had been mostly removed by perfusing PBS through the heart. Thus, the remaining vW-factor-positive structure in the present study should represent endothelial cells. In particular, we would like to emphasize that we used two distinct sets of anti-COX-2 and anti-vW factor and obtained a consistent result.

One may argue that the round-shape of COX-2-like immunoreactivity is more like the shape of perivascular macrophage/microglia than that of endothelial cells. However, double immunostaining of COX-2 with OX-42, a pan-monocytic marker, clearly excluded this possibility. We have also made double immunostaining of COX-2 with ED2, which recognizes a subset of macrophages including perivascular one, and obtained a consistent result (our unpublished observation). It should be noted that the round shape of COX-2 immunoreactivity represents the shape of the nucleus but not that of the whole cell.

The staining pattern of COX-2 after carrageenan injection was very similar to that found after intraperitoneal or intracerebroventricular injection of lipopolysaccharide or cytokines (Cao et al. 1997, 1998, Matsumura et al. 1998; Cao et al. 1999; Laflamme et al. 1999). No COX-2-LI components were detected under control conditions except for a subpopulation of telencephalic neurons. These results are in agreement with those of an in situ hybridization study by Ichitani et al. (1997), in which no COX-2 mRNA signals were detected in the spinal cord under control conditions but, after carrageenan injection, mRNA was strongly expressed bilaterally in non-neuronal cells associated with blood vessels in the parenchyma and subarachnoidal space.

In contrast, Willingale et al. (1997) and Beiche et al. (1998b) demonstrated constitutive expression of COX-2 in the spinal cord under control conditions. The reason for the discrepancy between those results and ours is unclear at present. We cannot exclude the possibility that COX-2 is in fact constitutively expressed in the spinal cord in low amounts, and that the sensitivity of the present protocol was insufficient for the detection of such COX-2. Even if this is the case under the normal conditions, it is very likely that a much higher amount of COX-2 is expressed in the endothelial cells of the CNS during carrageenan-induced inflammation.

COX-2-positive endothelial cells as the source of brain PGE2

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

In the CNS, COX-2-LI endothelial cells were first found at 3 h after the carrageenan injection, most abundant at 6 h, and greatly reduced in number at 24 h after the injection. In line with this time course, elevation of PGE2 in the CSF was detected at 3 h and larger at 6 h after the carrageenan injection. In addition, although not shown in this study, our preliminary experiment indicated that, at 24 h after carrageenan injection, PGE2 level in the CSF was almost at the baseline level again, being in line with the few number of COX-2-LI cells at this time point. Furthermore, administration of NS-398, a COX-2 selective inhibitor, 2 h after the carrageenan injection potently suppressed the CSF PGE2 level, but its administration prior to the carrageenan injection was less effective. We speculate that, in the latter case, plasma NS-398 had already been dropped to the ineffective level when COX-2 was actually induced and exerting its enzymatic action. Thus, temporal relationship among COX-2 induction, PGE2 level in the CSF, and effective timing of NS-398 administration strongly indicates that COX-2-positive endothelial cells are the major source of PGE2 in the CNS during carrageenan-induced hind paw inflammation.

Role of the central COX-2 in hyperalgesia

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

Hind paw injection of carrageenan evoked thermal hyperalgesia by 1 h after the injection, and the hyperalgesic state lasted throughout the observation period of 6 h. The hyperalgesia during this time period was dose-dependently alleviated by either a non-selective COX inhibitor or a selective COX-2 inhibitor systemically administered 2 h after carrageenan injection. No essential difference was seen in the anti-hyperalgesic effects of the two COX inhibitors, indicating that COX-2 was primarily responsible for the hyperalgesia. These results are in agreement with previous studies that utilized different types of COX inhibitors in the same experimental model (Seibert et al. 1994; Hay and Belleroche 1997; Hay et al. 1997; Zhang et al. 1997; Dirig et al. 1998; Smith et al. 1998).

Intrathecal injection of COX-2 inhibitor 2 h after the carrageenan injection suppressed the thermal hyperalgesia partially but significantly. This fact strongly implicates the involvement of central COX-2 in the carrageenan-induced hyperalgesia. This is in line with a recent study by Smith et al. (1998), who emphasized the importance of the central COX-2 in carrageenan-induced hyperalgesia, at least in the later phase. In their study, 6 h after the carrageenan injection, rats that had been orally administered with a COX-2 selective inhibitor showed suppressed thermal hyperalgesia as well as PGE2 level in both the CSF and inflamed hind paw. In contrast, rats injected with a COX-1 selective inhibitor showed suppressed PGE2 levels only in the inflamed tissue with no sign of analgesia, suggesting the importance of the central COX-2 and PGE2.

There are two possible explanations why the intrathecal application of COX-2-inhibitor only partially suppressed the hyperalgesia while the systemic one almost completely did that. One is that the amount of COX-2-inhibitor applied intrathecally was not enough to suppress the entire CNS COX-2 activity. The second one is that COX-2 expressed in peripheral tissues also involved in the hyperalgesia. The second one is more likely for the following reasons.

Induction of COX-2 in the CNS endothelial cells was not evident at the first hour. However, COX-2-dependent hyperalgesia was evident from the first hour and continued thereafter. Thus, the hyperalgesic response in the early phase (at 1 h) is hardly explained by newly induced COX-2 in the CNS endothelial cells. Rather, in the early phase, it is more likely that peripheral COX-2, probably at the inflamed tissue site, contributed to the COX-2-dependent hyperalgesic response. In fact, we confirmed COX-2 expression in the inflamed hind paw at 1 h after carrageenan injection (data not shown).

Little is known about the exact CNS sites on which PGE2 or other prostaglandins act to promote hyperalgesia. These sites may reside somewhere in the spinal cord, as our present study has demonstrated an anti-hyperalgesic effect of COX inhibitors injected intrathecally. However, as shown in the present study, production of PGE2 via endothelial COX-2 most likely occurs throughout the CNS, and there are a number of supraspinal CNS regions expressing PGE2 and PGI2 receptors (Matsumura et al. 1998). One possible site involved in hyperalgesia is the medial preoptic area (MPO), in which microinjection of PGE2 evokes hyperalgesia through the EP3 subtype of the PGE2 receptor (Hosoi et al. 1997).

Mechanism of COX-2 induction

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References

As COX-2-LI cells appeared bilaterally and throughout the CNS, a humoral mechanism rather than a neuronal one is more likely involved in COX-2 induction. With respect to humoral factors, cytokines are known to mediate some of the systemic effects of inflammation such as fever or cachexia (Dinarello 1991). Intraperitoneal administration of the pro-inflammatory cytokines, i.e. interleukin-1beta (IL-1beta) result in profound expression of COX-2 in brain endothelial cells in a fashion almost identical to that seen in the present study (Cao et al. 1996). Furthermore, production of these cytokines is up-regulated at an inflammatory locus challenged with carrageenan (Cunha et al. 1992; Utsunomiya et al. 1994), and receptors for IL-1beta are constitutively expressed in brain endothelial cells (Yabuuchi et al. 1994). In combination, this evidence suggests that carrageenan activates a number of inflammatory cells such as macrophages or neutrophils and tissue cells such as fibroblasts or basal keratinocytes to produce pro-inflammatory cytokines, possibly including IL-1beta. These cytokines enter the circulation and bind to its specific receptors in the vascular endothelial cells in the CNS to accelerate the synthesis of PGs via the COX-2 dependent way. These PGs, after being released into the CSF, act on the central neurons involved in nociception control. Verification of this scenario will be attempted in future studies. Recent observations by Laflamme et al. (1999) strongly support this hypothesis. They have shown the critical role of IL-1β during the inflammatory process, which induces the expression of both IκBα (index of NF-κB activity) and COX-2 genes within the same vascular endothelial cells.

In conclusion, the present study for the first time demonstrated widespread induction of COX-2 in the vascular endothelial cells of the CNS after intraplantar injection of carrageenan. Temporal relationship between the COX-2 induction, PGE2 level in the CSF, and thermal hyperalgesia, and suppression of the hyperalgesia by intrathecal injection of COX-2 inhibitor suggested that centrally induced COX-2 is involved in the hyperalgesia during the later time period whereas peripherally induced COX-2 is essential for the initial development of the hyperalgesia.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Materials
  5. Animal model
  6. COX-2 immunohistochemistry
  7. Microscopy
  8. Measurement of PGE2 level in the CSF
  9. Measurement of thermal hyperalgesia
  10. Quantitative analysis and statistics
  11. Results
  12. COX-2 immunohistochemistry
  13. Identification of the COX-2-LI cells in blood vessels
  14. Timing of COX-2 induction in the CNS
  15. PGE2 level in the CSF during carrageenan-induced inflammation and its suppression by a COX-2 selective inhibitor
  16. Carrageenan-induced thermal hyperalgesia
  17. Reversal of thermal hyperalgesia by COX inhibitors
  18. Discussion
  19. Identification of COX-2-LI cells
  20. COX-2-positive endothelial cells as the source of brain PGE2
  21. Role of the central COX-2 in hyperalgesia
  22. Mechanism of COX-2 induction
  23. Acknowledgement
  24. References
  • Anderson G. D., Hauser S. D., McGarity K. L., Bremer M. E., Isakson P. C. and Gregory S. A. (1996) Selective inhibition of cyclooxygenase (COX)-2 reverses inflammation and expression of COX-2 and interleukin 6 in rat adjuvant arthritis. J. Clin. Invest. 97, 26722679.
  • Beiche F., Scheuerer S., Brune K., Geisslinger G. and Goppelt-Struebe M. (1996) Up-regulation of cyclooxygenase-2 mRNA in the rat spinal cord following peripheral inflammation. FEBS Lett. 390, 165169.
  • Beiche F., Burne K., Geisslinger G. and Goppelt-Struebe M. (1998a) Expression of cyclooxygenase isoforms in the rat spinal cord and their regulation during adjuvant-induced arthritis. Inflamm. Res. 47, 482487.
  • Beiche F., Klein T., Nusing R., Neuhuber W. and Goppelt-Struebe M. (1998b) Localization of cyclooxygenase-2 and prostaglandin E2 receptor EP3 in the rat lumbar spinal cord. J. Neuroimmunol. 89, 2634.
  • Breder C. D., Dewitt D. and Kraig R. P. (1995) Characterization of inducible cyclooxygenase in rat brain. J. Comp. Neurol. 355, 296315.
  • Cao C., Matsumura K., Ozaki M. and Watanabe Y. (1999) Lipopolysaccharide injected into the cerebral ventricle evokes fever through induction of cyclooxygenase-2 in brain endothelial cells. J. Neurosci. 19, 716725.
  • Cao C., Matsumura K., Yamagata K. and Watanabe Y. (1996) Endothelial cells of the rat brain vasculature express cyclooxygenase-2 mRNA in response to systemic interleukin-1β: a possible site of prostaglandin synthesis responsible for fever. Brain Res. 733, 263272.
  • Cao C., Matsumura K., Yamagata K. and Watanabe Y. (1997) Involvement of cyclooxygenase-2 in LPS-induced fever and regulation of its mRNA in the rat brain by LPS. Am. J. Physiol. 272, R1712R1725.
  • Cao C., Matsumura K., Yamagata K. and Watanabe Y. (1998) Cyclooxygenase-2 is induced in brain blood vessels during fever evoked by peripheral or central administration of tumor necrosis factor. Mol. Brain Res. 56, 4556.
  • Cunha F. Q., Poole S., Lorenzetti B. B. and Ferreira S. H. (1992) The pivotal role of tumor necrosis factor alpha in the development of inflammatory hyperalgesia. Br. J. Pharmacol. 107, 660664.
  • Dinarello C. A. (1991) Interleukin-1 and interleukin-1 antagonism. Blood 8, 16271652.
  • Dirig D., Isakson P. C. and Yaksh T. L. (1998) Effect of COX-1 and COX-2 inhibition on induction and maintenance of carrageenan-evoked thermal hyperalgesia in rats. J. Pharmacol. Exp. Ther. 285, 10311038.
  • Elmquist J. K., Breder C. D., Sherin J. E., Scammell T. E., Hickey W. F., Dewitt D. and Saper C. B. (1997) Intravenous lipopolysaccharide induces cyclooxygenase-2-like immunoreactivity in rat brain perivascular microglia and meningeal macrophages. J. Comp. Neurol. 381, 119129.
  • England S., Bevan S. and Docherty R. J. (1996) PGE2 modulates the tetrodotoxin-resistant sodium current in neonatal rat dorsal root ganglion neurons via the cyclic AMP-protein kinase A cascade. J. Physiol. 495, 429440.
  • Gold M. S., Reichling D. B., Schuster M. J. and Levine J. D. (1996) Hyperalgesic agents increase a tetrodotoxin-resistant Na+ current in nociceptors. Proc. Natl Acad. Sci. USA 93, 11081112.
  • Hargreaves K., Dubner R., Brown F., Flores C. and Joris J. (1988) A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia. Pain 32, 7788.
  • Hay C. and Belleroche J. (1997) Carrageenan-induced hyperalgesia is associated with increased cyclooxygenase-2 expression in the spinal cord. Neuroreport 8, 12491251.
  • Hay C. H., Trevethick M. A., Wheeldon A., Bowers J. S. and Belleroche J. S. D. (1997) The potential role of spinal cord cyclooxygenase-2 in the development of Freund's complete adjuvant-induced changes in hyperalgesia and allodynia. Neuroscience 78, 843850.
  • Hosoi M., Oka T. and Hori T. (1997) Prostaglandin E receptor EP3 subtype is involved in thermal hyperalgesia through its actions in the preoptic hypothalamus and the diagonal band of Broca in rats. Pain 71, 303311.
  • Ichitani Y., Shi T., Haeggstrom J. Z., Samuelsson B. and Hökfelt T. (1997) Increased levels of cyclooxygenase-2 mRNA in the rat spinal cord after peripheral inflammation: an in situ hybridization study. Neuroreport 8, 29492952.
  • Jaffe E. A., Hoyer L. W. and Nachman R. L. (1973) Synthesis of antihemophilic factor antigen by cultured human endothelial cells. J. Clin. Invest. 52, 27572764.
  • Kujubu D. A., Fletcher B. S., Varnum B. C., Lim R. W. and Herschman H. R. (1991) TIS10, a phorbol ester tumor promoter-inducible mRNA from Swiss 3T3 cell, encodes a novel prostaglandin synthase/cyclooxygenase homologue. J. Biol. Chem. 266, 1286612872.
  • Laflamme N., Lacroix S. and Rivest S. (1999) An essential role of interleukin-1β in mediating NF-κB activity and COX-2 transcription in cells of the blood–brain barrier in response to a systemic and localized inflammation but not during endotoxemia. J. Neurosci. 19, 1092310930.
  • Matsumura K., Cao C., Ozaki M., Morii H., Nakadate K. and Watanabe Y. (1998) Brain endothelial cells express cyclooxygenase-2 during lipopolysaccharide-induced fever: light and electron microscopic immunocytochemical studies. J. Neurosci. 18, 62796289.
  • O'Banion M. K., Sadowski H. B., Winn V. and Young D. A. (1991) A serum- and glucocorticoid-regulated 4-kilobase mRNA encodes a cyclooxygenase-related protein. J. Biol. Chem. 266, 2326123267.
  • Seibert K., Zhang Y., Leahy K., Hauser S., Masferrer Perkins W., Lee L. and Isakson P. (1994) Pharmacological and biochemical demonstration of the role of cyclooxygenase 2 in inflammation and pain. Proc. Natl Acad. Sci. USA 91, 1201312017.
  • Smith W. L. and DeWitt D. L. (1996) Prostaglandin endoperoxide H synthases (cyclooxygenase)-1 and -2. Adv. Immunol. 62, 167215.
  • Smith C. J., Zhang Y., Koboldt C. M., Muhammad J., Zweifel B. S., Shaffer A., Talley J. J., Masferrer J. L., Seibert K. and Isakson P. C. (1998) Pharmacological analysis of cyclooxygenase-1 in inflammation. Proc. Natl Acad. Sci. USA 95, 1331313318.
  • Theilen H. and Kuschinsky W. (1992) Fluorescence labeling of the capillary network in rat brain. J. Cereb. Blood Flow Metab. 12, 347350.
  • Utsunomiya I., Nagai S. and Ohishi S. (1994) Differential effects of indomethacin and dexamethasone on cytokine production in carrageenan-induced rat pleurisy. Eur. J. Pharmacol. 252, 213218.
  • Vane J. R. (1971) Inhibition of prostaglandin synthesis as a mechanism of action for the aspirin-like drugs. Nature (Lond.) New Biol. 231, 232235.
  • Wagner D. D. and Marder V. J. (1983) Biosynthesis of von Willebrand protein by human endothelial cells. J. Biol. Chem. 258, 20652067.
  • Willingale H. L., Gardiner N. J., McLymont N., Giblett S. and Grubb B. D. (1997) Prostanoids synthesized by cyclo-oxygenase isoforms in rat spinal cord and their contribution to the development of neuronal hyperexcitability. Br. J. Pharmacol. 122, 15931604.
  • Xie W., Chipman J. G., Robertson D. L., Erikson R. L. and Simmons D. L. (1991) Expression of a mitogen-responsive gene encoding prostaglandin synthase is regulated by mRNA splicing. Proc. Natl Acad. Sci. USA 88, 26922696.
  • Yabuuchi K., Minami M., Katsumata S. and Satoh M. (1994) Localization of type I interleukin-1 receptor mRNA in the rat brain. Brain Res. Mol. Brain Res. 27, 2736.
  • Yaksh T. L. (1982) Central and peripheral mechanism for the antialgesic action of acetylsalicylic acid, in Acetylsalicylic Acid: New Uses for an Old Drug (BarnetJ. M., HirshJ. and MustardJ. F., eds), pp. 137152. Raven Press, New York.
  • Yaksh T. L. and Rudy T. A. (1976) Chronic catheterization of the spinal subarachnoid space. Physiol. Behav. 17, 10311036.
  • Zhang Y., Shaffer A., Portanova J., Seibert K. and Isakson P. C. (1997) Inhibition of cyclooxygenase-2 rapidly reverses inflammatory hyperalgesia and prostaglandin E2 production. J. Pharmacol. Exp. Ther. 283, 10691075.