Aim: To establish an in vitro injured motor neuronal model and investigate the neuroprotective effects and possible mechanism of celecoxib, a selective cyclooxygenase-2 (COX-2) inhibitor, on this model. Methods: After macrophages were stimulated with lipopolysaccharide (LPS)+interferon-γ(IFN-γ) in the pre-sence or absence of celecoxib for 24 h, the cell-free supernatant of LPS-stimulated macrophages was transferred to the culture of NSC34 cells. Viability of NSC34 cells was assessed by MTT assay after a further 24 h and 72 h incubation. After macrophages were stimulated by LPS+IFN-γ for 12 h or 24h, the release of pros-taglandin E_2 (PGE_2), nitric oxide (NO), reactive oxygen species (ROS), tumor necro-sis factor α(TNF-α) and interleukin-1β (IL-1β) from macrophages was measured by radioimmunoassay, Griess assay, fluorescence assay and enzyme-linked immunosorbent assay, respectively. The mRNA levels of COX-2, inducible nitric oxide synthase (iNOS), TNF-α and IL-1β in macrophages were determined by reverse transcription-polymerase chain reaction after macrophages were stimu-lated for 6 h and 12 h. Results: The supernatant of LPS-stimulated mouse mac-rophages induced the death of NSC34 cells and celecoxib protected the NSC34cells against this toxicity. The LPS-induced increases in the release of PGE_2, NO,TNF-α and IL-1β from macrophages were attenuated by pre-treatment with celecoxib. However, celecoxib showed no effect on the ROS levels upregulated by LPS+IFN-γin the macrophage supernatant. The mRNA levels of COX-2, iNOS, TNF-α and IL-1β were increased in LPS-activated macrophages and, except COX-2, reduced by pre-treatment with celecoxib. Conclusion: An in vitro injured motor neuronal model was established by using the toxicity of LPS-stimulated mouse macrophages toward motor neuronal NSC34 cells. In this model, celecoxib exerted neuroprotective effects on motor neurons via an inhibition of the neurotoxic secretions from activated macrophages. Aim: To establish an in vitro injured motor neuronal model and investigate the neuroprotective effects and possible mechanism of celecoxib, a selective cyclooxygenase-2 (COX-2) inhibitor, on this model. Methods: After macrophages were stimulated with lipopolysaccharide (LPS) + interferon-γ (IFN-γ) in the pre-sence or absence of celecoxib for 24 h, the cell-free supernatant of LPS-stimulated macrophages was transferred to the culture of NSC34 cells. Viability of NSC34 cells was assessed by MTT assay after macrophages were stimulated by LPS + IFN-γ for 12 h or 24h, the release of prostaglandin E_2 (PGE_2), nitric oxide (NO), reactive oxygen species (ROS), tumor necro-sis factor α (TNF-α) and interleukin-1β (IL-1β) from macrophages was measured by radioimmunoassay, Griess assay, fluorescence assay and enzyme-linked immunosorbent assay, respectively. The mRNA levels of COX-2, inducible nitric oxide synthase (iNOS), TNF-α and IL-1β in macrophage s were determined by reverse transcription-polymerase chain reaction after macrophages were stimu-lated for 6 h and 12 h. Results: The supernatant of LPS-stimulated mouse mac-rophages induced the death of NSC34 cells and celecoxib protected the NSC34 cells against this toxicity. The LPS-induced increases in the release of PGE_2, NO, TNF-α and IL-1β from macrophages were attenuated by pre-treatment with celecoxib. However, celecoxib showed no effect on the ROS levels upregulated by LPS + IFN-γin the macrophage The mRNA levels of COX-2, iNOS, TNF-α and IL-1β were increased in LPS-activated macrophages and, in addition to COX-2, reduced by pre-treatment with celecoxib. Conclusion: An in vitro injured motor neuronal model was established by using the toxicity of LPS-stimulated mouse macrophages toward motor neuronal NSC34 cells. In this model, celecoxib exerted neuroprotective effects on motor neurons via an inhibition of the neurotoxic secretions from activated macrophages.