HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 15/9/1622 00-0716fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by NIEDERBERGER, E. Articles by GEISSLINGER, G. Search for Related Content PubMed PubMed Citation Articles by NIEDERBERGER, E. Articles by GEISSLINGER, G.

Celecoxib loses its anti-inflammatory efficacy at high doses through activation of NF-B^{1 ,2}

Pharmazentrum frankfurt, Klinikum der Johann Wolfgang Goethe, Universität Frankfurt, 60590 Frankfurt am Main, Germany

⁴Correspondence: Institut für klinische Pharmakologie, Klinikum der Johann Wolfgang Goethe-Universität, Frankfurt, Theodor Stern Kai 7, 60590 Frankfurt am Main, Germany. E-mail: geisslinger{at}em.uni-frankfurt.de

SPECIFIC AIMS

Celecoxib is a cyclooxygenase-2 (COX-2) selective, nonsteroidal anti-inflammatory drug (NSAID) with low gastrointestinal toxicity. In clinical studies, 800 mg/day were somewhat less effective than 200 mg/day, suggesting that celecoxib rapidly reached a ceiling effect or that it exhibited a bell-shaped dose response relationship. The aim of the present study was to find out whether the anti-inflammatory efficacy of celecoxib diminished when the dose was further increased and to evaluate possible mechanisms.

PRINCIPAL FINDINGS

1. Celecoxib reduced paw inflammation at 10 and 50 mg/kg in the zymosan-induced paw inflammation model in rats; however, at doses of 100–200 mg/kg, celecoxib had no effect
Thus, celecoxib lost its anti-inflammatory efficacy at high doses. Because gastrointestinal absorption was limited plasma concentrations in rats after oral administration of 100–200 mg/kg were only fivefold higher than those found in humans treated with a single 800 mg dose.

The anti-inflammatory efficacy of celecoxib was assessed in the zymosan-induced hindpaw inflammation model in rats. Celecoxib (10, 50, 100, or 200 mg/kg) or vehicle (lecithin-tylose slime) was administered via a gastric tube and 100 µl of a zymosan suspension (12.5 mg/ml) was injected into the plantar surface of one hindpaw. The paw volume as an indicator of the inflammatory response was measured using a plethysmometer at various times for up to 96 h. The areas under the ‘paw volume increase’ vs. ‘time’ curves are shown in Fig. 1 . In vehicle-treated rats, the paw volume increased by 125.6 ± 9.1% (mean±SE). Paw swelling was reduced in rats treated with 10 and 50 mg/kg celecoxib. At higher doses of 100 and 200 mg/kg, however, there was no difference between celecoxib- and vehicle-treated rats. Thus, celecoxib inhibited the inflammatory response at doses of 10 to 50 mg/kg. A mere doubling of the most efficacious dose, however, resulted in the complete loss of anti-inflammatory efficacy.

View larger version (22K): [in this window] [in a new window]   Figure 1. A) Time course of the anti-inflammatory effects of celecoxib in zymosan-induced hindpaw inflammation in rats (mean±SE). Male Sprague Dawley rats (6–10 in each group) received a single oral dose of 10 (), 50 (O), 100 (), and 200 mg/kg () celecoxib 15 min before an intraplantar injection of 1.25 mg zymosan. Control animals received the appropriate volume of vehicle (•). B) For statistical comparison of drug effects, the areas under the ‘paw volume increase’ vs. ‘time’ curves (AUCPW from 0 to 96 h, mean±SE) were calculated by the linear trapezoidal rule and subjected to univariate analysis of variance with subsequent Bonferroni post hoc tests. The box represents the interquartile range, the line within the box the median, the dashed line the mean; the ends of the ‘whiskers’ show the 5th and 95th percentile. Filled dots show individual values. *Statistically significant mean difference (<0.05).

Peak plasma concentrations in rats treated with the effective 50 mg/kg and the ineffective 200 mg/kg dose were 14.4 ± 1.1 µM and 29.3 ± 1.1 µM. Maximum plasma concentrations in humans after administration of a single 800 mg dose ranged from 3.2 to 5.6 µM.

2. Celecoxib inhibited PGE₂ release of stimulated cells at low and high concentrations. At high concentrations, however, it increased the expression of COX-2
Since celecoxib has been shown to inhibit the enzymatic activity of COX-2 with an IC₅₀ of 3–6 nM, we hypothesized that the observed loss of anti-inflammatory efficacy at high doses was probably not mediated through a release of prostaglandins. To address this hypothesis, we determined PGE₂ release from interleukin 1ß (IL-1ß) -stimulated rat renal mesangial cells (rMES), which were used as in vitro model. Celecoxib significantly inhibited IL-1ß-induced PGE₂ release at concentrations of 1, 10, and 50 µM. Hence, the inhibitory effect of celecoxib on PGE₂ synthesis was not reversible at high concentrations.

In parallel, the expression of COX-2 mRNA and protein was assessed. Surprisingly, celecoxib slightly reduced the IL-1ß-stimulated COX-2 mRNA and protein expression at 1 µM, had no effect at 10 µM, but increased COX-2 mRNA and protein expression at 50 µM.

3. At low concentrations (1 µM), celecoxib reduced the IL-1ß-induced activation of the transcription factor NF-B. At high celecoxib concentrations (50 µM), however, IL-1ß-induced NF-B activation was further increased
The transcription of the COX-2 gene is regulated by the transcription factor NF-B. Therefore, we were interested in whether the observed induction of COX-2 expression might be caused by activating NF-B. Electrophoretic mobility shift assays were performed to assess NF-B DNA binding capacity in rMES. Figure 2 shows the results of a representative experiment. In unstimulated control cells, DNA binding of NF-B was minimal but considerably increased after IL-1ß stimulation. Celecoxib reduced IL-1ß-induced NF-B activation at 1 µM. Pretreatment with 50 µM celecoxib however, resulted in a significant overstimulation of NF-B.

View larger version (68K): [in this window] [in a new window]   Figure 2. Electrophoretic mobility shift assay (EMSA). rMES cells were preincubated for 30 min with celecoxib at the concentrations indicated and stimulated with 1 nM IL-1ß for another 30 min. The nuclear fractions were extracted and 5 µg were subjected to EMSA using a 32P-labeled NF-B consensus oligonucleotide. A result representative of 4 repeat experiments is shown.

To evaluate whether celecoxib affected the IL-1ß-stimulated degradation of the inhibitory protein I-B and/or the nuclear translocation of NF-B, we assessed the amount of NF-B (p65 subunit) and I-B protein in nuclear and cytosolic extracts, respectively, by Western blot analysis. At 1 and 10 µM, celecoxib had no obvious effect on nuclear concentrations of NF-B but slightly reduced the degradation of I-B. At 50 µM, however, celecoxib enhanced the stimulated nuclear translocation of NF-B and led to a complete degradation of cytosolic I-B.

4. Activation of NF-B resulted in an increased transcription of NF-B-dependent genes, i.e., cyclooxygenase-2 and tumor necrosis factor (TNF-) in vitro (cell culture) and in vivo (rat)
To find out whether the observed overstimulation of NF-B caused an enhanced transcription not only of COX-2, but also of other NF-B-regulated inflammatory gene products, we assessed TNF- and iNOS expression in rMES cells and rat spinal cord. In unstimulated rMES cells, iNOS mRNA and protein were not detected but strongly increased after stimulation with IL-1ß. This was accompanied by a significant increase of NO₂^-/NO₃^- release. Treatment with 50 µM celecoxib caused a slight overexpression of iNOS mRNA whereas iNOS protein and NO₂^-/NO₃^- release remained unaffected; 1 and 10 µM celecoxib had no effect.

TNF- concentrations were determined in cell culture supernatants. IL-1ß-stimulated TNF- release was somewhat reduced in cells pretreated with 1 and 10 µM celecoxib. Pretreatment with 50 µM celecoxib, however, resulted in a further significant increase of TNF- release compared with IL-1ß stimulation alone.

Expression of COX-2, iNOS, and TNF- in rat lumbar spinal cord homogenates was assessed by Western blot analysis (COX-2, iNOS) or ELISA (TNF-) to find out whether the celecoxib-induced overexpression of the respective proteins also occurred in vivo. Rats were treated as described above and the spinal cord was excised at 96 h. The zymosan-induced up-regulation of COX-2 protein in lumbar spinal cord was reduced in rats treated with anti-inflammatory doses (10 and 50 mg/kg) of celecoxib. This effect was abolished at 100 mg/kg. At 200 mg/kg, a dose that had no anti-inflammatory activity, COX-2 protein levels exceeded those of placebo-treated rats. The zymosan-induced iNOS protein expression in lumbar spinal cord was unaffected at 10–100 mg/kg celecoxib. iNOS expression was slightly increased at 200 mg/kg. Zymosan-stimulated TNF- production in lumbar spinal cord was not significantly altered at 10–100 mg/kg celecoxib. Treatment with 200 mg/kg celecoxib, however, caused a significant TNF- increase. Thus, the effects of celecoxib on COX-2, iNOS, and TNF- expression in rat lumbar spinal cord were similar to those observed in cell culture, suggesting that the mechanisms observed in vitro also occurred in vivo.

CONCLUSIONS AND SIGNIFICANCE

Great effort has been invested in developing COX-2 selective NSAIDs as effective anti-inflammatory and analgesic drugs that spare the gastrointestinal tract. Celecoxib largely satisfies these expectations. However, celecoxib rapidly reached a ceiling effect in clinical studies, i.e., its maximum therapeutic efficacy was achieved with 100 mg b.i.d. and did not further increase despite quadrupling the dose. Using the zymosan-induced paw inflammation model in rats, we show that the anti-inflammatory efficacy of celecoxib (statistically significant at 50 mg/kg) is completely lost at doses of 100–200 mg/kg. These doses produce plasma concentrations in the range of 20–30 µM. In IL-1ß-stimulated cells, similar celecoxib concentrations evoke an overstimulation of the transcription factor NF-B that results in an enhanced production of NF-B-dependent proinflammatory proteins (cyclooxygenase-2 and TNF-). The latter effect was observed not only in cells, but also in the spinal cord of celecoxib-treated rats, and therefore probably accounts for the loss of efficacy observed in the animal experiments. In humans maximum plasma concentrations ranged from 3 to 5 µM after a single 800 mg dose. At this concentration range, we observed instead a decrease of NF-B activity and NF-B-dependent gene transcription in vitro. Thus, a complete loss of anti-inflammatory efficacy is unlikely to occur at the recommended doses. On the other hand, an inhibition of NF-B activity might contribute to the anti-inflammatory efficacy of celecoxib when low doses are administered. However, since effective and ineffective doses in rats were close together, it seems conceivable that some NF-B activation occurs in individual patients at daily doses of 400–800 mg. Concerning clinical symptoms, this effect is probably partly outweighed by the enzymatic inhibition of COX-2 activity so that the net effect is not a complete loss, but only a lack of further improvement of the anti-inflammatory efficacy. However, NF-B activation may nevertheless aggravate the underlying inflammatory process or may cause side effects independent of prostaglandin synthesis.

The finding that celecoxib evokes an activation of NF-B at high concentrations, thereby inducing the transcription of NF-B-dependent genes such as TNF-, implies that the efficacy and safety of the drug depends on the ‘correct’ dose. Thus, whereas the dose of classical nonselective NSAIDs is limited by gastrointestinal toxicity, that of celecoxib might be limited by loss of efficacy.

View larger version (36K): [in this window] [in a new window]   Figure 3. Schematic diagram illustrating the suggested mechanisms of celecoxib on NF-B and NF-B-dependent gene transcription. Activation (+), inhibition (-), reduction, no change ().

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.00-0716fje ; to cite this article, use FASEB J. (May 29, 2001) 10.1096/fj.00-0716fje

² This study was supported by the DFG (SFB 553/C6).

³ Both authors contributed equally to this work.

This Article Full Text (PDF) All Versions of this Article: 15/9/1622 00-0716fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by NIEDERBERGER, E. Articles by GEISSLINGER, G. Search for Related Content PubMed PubMed Citation Articles by NIEDERBERGER, E. Articles by GEISSLINGER, G.