Okadaic-acid-induced inhibition of protein phosphatase 2A produces activation of mitogen-activated protein kinases ERK1/2, MEK1/2, and p70 S6, similar to that in Alzheimer's disease

Abstract

In Alzheimer's disease (AD) brain the activity of protein phosphatase (PP)-2A is compromised and that of the extracellular signal-regulated protein kinase (ERK1/2) of the mitogen-activated protein kinase (MAPK) family, which can phosphorylate tau, is up-regulated. We investigated whether a decrease in PP-2A activity could underlie the activation of these kinases and the abnormal hyperphosphorylation of tau. Rat brain slices, 400-microm-thick, kept under metabolically active conditions in oxygenated (95% O(2), 5% CO(2)) artificial CSF were treated with 1.0 micromol/L okadaic acid (OA) for 1 hour at 33 degrees C. Under this condition, PP-2A activity was decreased to approximately 35% of the vehicle-treated control slices, and activities of PP-1 and PP-2B were not affected. In the OA-treated slices, we observed a dramatic increase in the phosphorylation/activation of ERK1/2, MEK1/2, and p70 S6 kinase both immunohistochemically and by Western blots using phosphorylation-dependent antibodies against these kinases. Treatment of 6-microm sections of the OA-treated slices with purified PP-2A reversed the phosphorylation/activation of these kinases. Hyperphosphorylation of tau at several abnormal hyperphosphorylation sites was also observed, as seen in AD brain. These results suggest 1) that PP-2A down-regulates ERK1/2, MEK1/2, and p70 S6 kinase activities through dephosphorylation at the serine/threonine residues of these kinases, and 2) that in AD brain the decrease in PP-2A activity could have caused the activation of ERK1/2, MEK1/2, and p70 S6 kinase, and the abnormal hyperphosphorylation of tau both via an increase in its phosphorylation and a decrease in its dephosphorylation.

Figure 1.

Confocal microscopy demonstrating the partial colocalization of active ERK1/2, MEK1/2, and p70 S6 kinase immunoreactivities with AT8-positive abnormally hyperphosphorylated tau in hippocampal CA1 pyramidal neurons in brains at Braak stage V/C. Formalin-fixed frozen sections, 100-μm thick, were incubated with rabbit polyclonal antibodies to active forms of ERK1/2, MEK1/2, or p70 S6 kinase, and mouse monoclonal antibody AT8 to PHF-tau, and visualized with CY3 (red)-conjugated anti-rabbit IgG and CY2 (green)-conjugated anti-mouse IgM. Active ERK1/2 (A1–3), MEK1/2 (B1–3), or p70 S6 kinase (C1–3) was found in many neurons that were AT8-positive. Scale bar, 20 μm.

Figure 2.

Inhibition of PPs, activation of protein kinases, and phosphorylation of tau in rat brain slices treated with OA or CsA. Rat brain slices were incubated in either oxygenated artificial CSF alone as controls, or in the artificial CSF plus either 1.0 μmol/L OA or 1.0 μmol/L CsA for 1 to 3 hours, followed by homogenization and centrifugation at 16,000 × g for 10 minutes. A: The resulting tissue extracts were assayed for the activities of PP-1, PP-2A, and PP2-B. The phosphatase activities of the inhibitor-treated samples were expressed as the percentage of the activities of the control samples of each set where the artificial CSF alone was used for incubation for the same period. Bars represent means ± SD of three to four independent assays. *, P < 0.01 as compared with controls by Student’s t-test. Note that under these conditions, OA selectively inhibited PP-2A activity by 65% to 70%, and CsA selectively inhibited PP-2B activity by ∼70%. B: The homogenate samples (20 μg protein per lane) with (+) or without (−) the treatment of 1.0 μmol/L OA for 1 hour or 1.0 μmol/L CsA for 3 hours were analyzed by Western blots for the levels and the activities of ERK1/2, MEK1/2, and p70 S6 kinase. Representative blots of three independent experiments with similar results are shown. Note that the levels of these kinases were not affected by the inhibitor treatments as detected with antibodies against the kinases (compare lane 2 with 1, and 7 with 6). When detected with antibodies against the phosphorylated/active form of these kinases, increased activation of all three kinases was observed in the OA-treated (compare lane 4 with 3), but not the CsA-treated (compare lane 9 with 8) brain slices. A slight up-shift of electrophoretic mobility of MEK1/2 and p70 S6 kinase is also seen in OA-treated samples (arrows). The staining of these kinases with phosphorylation-dependent antibodies disappeared after the blots were incubated with PP-2A before probing with the primary antibodies (lane 5). C: Homogenates (10 μg protein per lane) of the brain slices after incubation in the oxygenated artificial CSF in the presence (+) or absence (−) of 1.0 μmol/L OA for 1 hour were analyzed by Western blots for the phosphorylation of tau. The blots were developed with a phosphorylation-independent anti-tau antibody 92e to detect total tau, and phosphorylation-dependent anti-tau antibodies Tau-1 (which detects tau unphosphorylated at Ser198/199/202), 12E8 (which recognizes tau phosphorylated at Ser262/356), PHF-1 (which recognizes tau phosphorylated at Ser396/404), and R145 (which recognizes tau phosphorylated at Ser422), respectively. The molecular mass markers (67 and 43 kd, respectively) are shown at the left of the blots. The blots are representative results of three independent experiments with similar results. Note that the treatment of rat brain slices with 1.0 μmol/L OA for 1 hour induced phosphorylation of tau at all of the above sites studied.

Figure 3.

Phosphorylation of tau in rat brain slices after treatments with OA and U0126. A: Rat brain slices were incubated in oxygenated artificial CSF for 2 hours (lanes 1 and 2), artificial CSF for 1 hour, followed by 1.0 μmol/L OA in artificial CSF for 1 hour (lanes 3 and 4), or 50 μmol/L U0126 in artificial CSF for 1 hour, followed by 50 μmol/L U0126 plus 1.0 μmol/L OA in artificial CSF for 1 hour (lanes 5 and 6). The homogenates (10 μg protein/lane) of the treated slices were then analyzed by Western blots to assess the level of tau by using a phosphorylation-independent antibody 92e and phosphorylation levels of tau at the specific sites by using phosphorylation-dependent and site-specific anti-tau antibodies Tau-1, 12E8, PHF-1, and R145. B: The blots as shown in A were scanned and the radioimmunoreactivity was quantitated by using a phosphorimager. The radioimmunoreactivity obtained with each antibody was normalized by that of 92e in each case. The bar graphs are presented as means ± SD (n = 4). *, P < 0.05 as compared with no addition control; #, P < 0.05 as compared with that treated with OA only.

Figure 4.

Immunohistochemical staining showing topography of active ERK1/2 in rat brain slices treated with OA. Rat brain slices were incubated in either oxygenated artificial CSF alone as controls (Con) or in the artificial CSF plus 1.0 μmol/L OA for 1 hour (Exp), followed by immunohistochemical staining with antibody against the active ERK1/2. The representative staining of CA1, CA4, and dentate gyrus (DG) of the hippocampal complex, layer III of the adjacent primary sensory cortical cortex (CC), and of the whole section (E, Exp; F, Con) are shown. Arrowhead in E indicates the CA3 stratum lucidum. The bar represents 20 μm for A1 to D2 and 200 μm for E and F, respectively.

Figure 5.

Immunohistochemical staining showing topography of active MEK1/2 in rat brain slices treated with OA. Rat brain slices were incubated in either oxygenated artificial CSF alone as controls (Con), or in the artificial CSF plus 1.0 μmol/L OA for 1 hour (Exp), followed by immunohistochemical staining with antibody against the active MEK1/2. The representative staining of CA1, CA4, and DG of the hippocampal complex, layer III of the adjacent primary sensory CC, and of the whole section (E, Exp; F, Con) are shown. The magnification is indicated as scale bar of 20 μm for A1 to D2 and 200 μm for E and F.

Figure 6.

Immunohistochemical staining showing topography of active p70 S6 kinase in rat brain slices treated with OA. Rat brain slices were incubated in either oxygenated artificial CSF alone as controls (Con) or in the artificial CSF plus 1.0 μmol/L OA for 1 hour (Exp), followed by immunohistochemical staining with antibody against the active p70 S6 kinase. The representative staining of CA1, CA4, and DG of the hippocampal complex, layer III of the adjacent primary sensory CC, and of the whole section (E, Exp; F, Con) are shown. Bar represents 20 μm for A1 to D2 and 200 μm for E and F, respectively.

Figure 7.

Immunohistochemical staining showing topography of phosphorylated tau in rat brain slices treated with OA. Rat brain slices were incubated in either oxygenated artificial CSF alone as controls (Con) or in the artificial CSF plus 1.0 μmol/L OA for 1 hour (Exp), followed by immunohistochemical staining with antibody R145 against tau phosphorylated at Ser 422. The representative staining of CA1, CA2, and DG of the hippocampal complex, layer III of the adjacent primary sensory CC, and of the whole section (E, Exp; F, Con) are shown. All sections were counterstained with cresyl violet to demonstrate the cell nuclei. Bar represents 20 μm for A1 to D2 and 200 μm for E and F, respectively.

Figure 8.

Immunohistochemical staining showing the Intracellular distribution of MEK1/2 and p70 S6 kinase in layer V of the cerebral cortex of rat brain slices treated with OA. Rat brain slices were incubated in either oxygenated artificial CSF alone as controls (Co) or in the artificial CSF plus 1.0 μmol/L OA for 1 hour (Ex), followed by immunocytochemical staining with antibodies to active MEK1/2 and total MEK1/2 (A) or to active p70 S6 kinase and total p70 S6 kinase (B). The magnification is indicated as 20 μm scale bar.

Figure 9.

Immunohistochemical staining of active ERK1/2, MEK1/2, and p70 S6 kinase and of tau phosphorylated at Ser-262/356 in OA-treated tissue sections after pretreatment with PP-2A. Tissue sections from the OA-treated (same as in Figures 3 to 5 ▶ ) rat brain slices were incubated either with purified PP-2A in dephosphorylation buffer (+PP2A) or buffer alone (−PP2A), followed by immunohistochemical staining with antibodies to the active forms of ERK1/2 (A1, A2), MEK1/2 (B1, B2), or p70 S6 kinase (C1, C2), or with monoclonal antibody 12E8 to hyperphosphorylated tau (D1, D2). The sections of primary sensory CC are shown. The bar indicates 20 μm.