Calcineurin-mediated BAD dephosphorylation activates the caspase-3 apoptotic cascade in traumatic spinal cord injury

Abstract

We report here that activation of the caspase-3 apoptotic cascade in spinal cord injury is regulated, in part, by calcineurin-mediated BAD dephosphorylation. BAD, a proapoptotic member of the bcl-2 gene family, is rapidly dephosphorylated after injury, dissociates from 14-3-3 in the cytosol, and translocates to the mitochondria of neurons where it binds to Bcl-x(L). Pretreatment of animals with FK506, a potent inhibitor of calcineurin activity, or MK801, an NMDA glutamate receptor antagonist, blocked BAD dephosphorylation and abolished activation of the caspase-3 apoptotic cascade. These findings extend previous in vitro observations and are the first to implicate the involvement of glutamate-mediated calcineurin activation and BAD dephosphorylation as upstream, premitochondrial signaling events leading to caspase-3 activation in traumatic spinal cord injury.

Fig. 1.

Representative immunoblots demonstrating that spinal cord injury results in rapid BAD dephosphorylation.A, Immunoblotting experiments demonstrate that spinal cord injury has no effect on overall levels of Bcl-x_L, 14-3-3, or calcineurin A over 24 hr.B, BAD levels also are not affected over this period; however, the levels of phosphorylated BAD rapidly decline as early as 30 min after injury. C, Semiquantitative analysis reveals that the levels of phosphorylated BAD are significantly reduced at all time points examined after injury. *p < 0.01 (Scheffe's post hoc analysis).

Fig. 2.

Representative immunoblots demonstrating that spinal cord injury results in the rapid dissociation of BAD from 14-3-3 and calcineurin A followed by binding to Bcl-x_L. A phosphorylation state-independent BAD antibody (New England Biolabs) was cross-linked to protein G-Sepharose beads (Sigma). Postmitochondrial supernatant fractions (14-3-3 and calcineurin) or resuspended pellet fractions (Bcl-x_L) of control or injured spinal cord were incubated overnight at 4°C with the immobilized BAD antibody. Proteins bound to BAD were eluted and separated by SDS-PAGE, and immunoblotting was used to analyze 14-3-3, calcineurin A and Bcl-x_L bound to BAD. The data are representative of experiments from four control laminectomy and four spinal cord-injured animals. CTL, Control;30′, 30 min after injury.

Fig. 3.

Representative laser-scanned confocal images demonstrating rapid translocation of BAD to the mitochondria and caspase-3 activation after spinal cord injury. Immunofluorescence histochemistry was performed on longitudinal spinal cord sections containing the lesion epicenter. A, Control section demonstrating weak diffuse immunoreactivity for BAD in uninjured gray matter neurons. B, Punctate BAD immunoreactivity in ventral horn neurons in the injury epicenter 1 hr after injury.C, HSP60 immunoreactivity double labeling of the same section as in B. D, Merged images fromB and C providing evidence that BAD immunoreactivity is associated with mitochondria. (Note the absence of colocalization in neurons denoted by arrows.) E, F, Double-labeling experiments indicate that caspase-3 activation (F) occurs in cells exhibiting BAD translocation (E). Arrows inE and F denote the absence of BAD or activated capsase-3 staining in a large α motor neuron. These photomicrographs are from 30-μm-thick spinal cord sections corresponding to 2.0 mm rostral to the injury epicenter. Scale bar, 25 μm.

Fig. 4.

FK506 and MK801 treatments inhibit BAD dephosphorylation and caspase-3 activation. Animals were pretreated with FK506 (10 mg/kg, i.p., 1 hr before injury) or MK801 (0.3 mg/kg, i.p., 30 min before injury), and spinal cords were obtained 1 hr after injury. A, Representative immunoblots demonstrating that FK506 and MK801 treatments inhibited BAD dephosphorylation, caspase-9 activation, and ICAD/DFF45 cleavage. Postmitochondrial supernatant fractions were analyzed by immunoblotting using antibodies to phosphorylation state-independent and -dependent BAD, caspase-9, and ICAD/DFF45. B, Semiquantitative analysis of bands corresponding to activated caspase-9 and caspase-3-like ICAD/DFF45 cleavage (arrows) revealed a significant effect of FK506 and MK801. *p < 0.01 for caspase-9;^#p < 0.01 for ICAD/DFF45 (Scheffe'spost hoc analysis). C, The caspase fluorogenic assay demonstrated that both FK506 and MK801 significantly reduced caspase-3 enzyme activity after spinal cord injury compared with vehicle treatments. *p < 0.01 compared with laminectomy controls; **p < 0.01 compared with spinal cord injury (Scheffe's post hoc analysis).CTL, Control; SCI, spinal cord injury.