Knockout of striatal enriched protein tyrosine phosphatase in mice results in increased ERK1/2 phosphorylation

Abstract

STriatal Enriched protein tyrosine Phosphatase (STEP) is a brain-specific protein that is thought to play a role in synaptic plasticity. This hypothesis is based on previous findings demonstrating a role for STEP in the regulation of the extracellular signal-regulated kinase1/2 (ERK1/2). We have now generated a STEP knockout mouse and investigated the effect of knocking out STEP in the regulation of ERK1/2 activity. Here, we show that the STEP knockout mice are viable and fertile and have no detectable cytoarchitectural abnormalities in the brain. The homozygous knockout mice lack the expression of all STEP isoforms, whereas the heterozygous mice have reduced STEP protein levels when compared with the wild-type mice. The STEP knockout mice show enhanced phosphorylation of ERK1/2 in the striatum, CA2 region of the hippocampus, as well as central and lateral nuclei of the amygdala. In addition, the cultured neurons from KO mice showed significantly higher levels of pERK1/2 following synaptic stimulation when compared with wild-type controls. These data demonstrate more conclusively the role of STEP in the regulation of ERK1/2 activity.

Fig. 1

Genomic construct for the generation of STEP knockout mice. (A) Schematic diagram representing the genomic sequence of STEP along with the exons and introns. The exonic sequence deleted in the KO are shaded in black. (B) A neomycin cassette (PGK-NEO) was used to replace a 1.3 kb genomic sequence, disrupting the open reading frame. The insertion was confirmed by Southern blotting using 5′ (BamHI/BglII double digest) and 3′ (SpeI single digest) probes generated from the wild-type STEP genomic sequence, and the construct was then used to generate the STEP knockout mice (data not shown).

Fig. 2

PCR genotyping of STEP genomic DNA from WT, HT, and KO mice. (A) The genotype of the KO mice was determined by PCR amplification of the genomic DNA using primers from the STEP genomic sequence. The genomic DNA from WT mice show an expected 400 base pair (bp) band and that from the KO mice show a smaller 200 bp product. Both the bands are present in the HT mice. The table in (B) shows the number of WT, HT, and KO animals generated and the percentage of each genotype. The observed percentages are not significantly different from the expected values of 25, 50, and 25%, respectively.

Fig. 3

Expression profile of STEP protein in KO, HT, and WT mice. (A) The S2 and P2 fractions obtained from WT, HT, and KO mice total brain homogenates were analyzed by SDS-PAGE and immunoblotted with anti-STEP antibody (23E5). The levels of both STEP₆₁ and STEP₄₆ were significantly decreased in the samples obtained from HT mice, whereas both the isoforms of STEP were absent in the KO mice samples. The membranes were reprobed with anti-ERK2 antibody to ensure equal protein loading in each lane. (B) Quantification of immunoblots for STEP₆₁ and STEP₄₆ expressed in the three genotypes normalized with ERK2 loading control (n = 4). The normalized protein levels are expressed as a percentage of WT control (mean ± SE). Statistical analysis was performed using one-way ANOVA and posthoc Tukey HSD (**P < 0.01 and ***P < 0.001). (C) Coronal brain sections from the WT, HT, and KO mice were immunostained for STEP. The top panel shows representative sections through the striatum (CPu). The STEP staining was undetectable in the KO mice. STEP immunoreactivity was decreased in the lateral septum (LS) of HT when compared with WT. The middle and bottom panel show representative sections from the hippocampus and amygdala, respectively. The level of STEP expression in the HT was reduced and was not detected in the KO mice [Scale bar-200 μm].

Fig. 4

Morphological analysis of striatum, hippocampus, and amygdala from WT, HT, and KO mice. Coronal brain sections from the three genotypes were stained with cresyl violet to investigate anatomical changes in the striatum (upper panel), hippocampus (middle panel), and amygdala (lower panel). The staining did not detect any gross architectural abnormalities in the KO brain when compared with WT and HT sections at this level of analysis [Scale bar-200 μm].

Fig. 5

Increased baseline levels of phosphorylated ERK1/2 in STEP KO mice. (A) The levels of phosphorylated ERK1/2 were detected in the S2 and P2 fractions from total brain homogenates as well as the cerebellar S2 samples by probing with an antibody that recognizes ERK1/2 dually phosphorylated at the regulatory threonine and tyrosine residues (T^PEY^P-ERK1/2 or pERK1/2). The membranes were reprobed with anti-ERK2 antibody to ensure equal protein loading in each lane. STEP KO mice shows elevated levels of phosphorylated ERK1/2 when compared with WT controls in both the S2 and P2 fractions obtained from total brain homogenates. (B) Quantification of immunoblots for phosphorylated ERK2 expressed in the three genotypes normalized with total ERK2 protein loaded in the same blot (*P < 0.05 and **P < 0.01; n = 4). (C) pERK1/2 and ERK2 levels were detected in the S2 and P2 fractions from the striatum and hippocampus of WT, HT, and KO mice. The pERK1/2 levels were significantly elevated in both the fractions and regions tested. The HT showed significant increases only in the S2 fraction from the striatum. (D) The pERK1/2 levels from the immunoblots were quantified and normalized over the total ERK2 levels. The normalized pERK1/2 levels are plotted as a percentage of WT levels. The data were compared using one-way ANOVA, followed by posthoc Tukey HSD (*P < 0.05; **P < 0.01; ***P < 0.001; n = 4).

Fig. 6

Immunohistochemical analysis of phosphorylated ERK1/2 and total ERK2 in the striatum, hippocampus, and amygdala of WT, HT, and KO mice. Coronal brain sections from the three genotypes were double-labeled with antibodies that recognize the dually phosphorylated ERK1/2 (green) and total ERK2 (red). The two upper panels illustrate staining in the striatum (CPu) from the three genotypes. pERK1/2 immunoreactivity was elevated in the KO mice when compared with WT and HT. The middle and bottom panels show representative sections from the hippocampus and amygdala, respectively. Increased pERK1/2 levels were noticed in the CA2 area of the hippocampus as well as central and lateral nuclei of the amygdala [Scale bar-50 μm].

Fig. 7

Enhanced phosphorylation of ERK1/2 in the KO hippocampal cultures and synaptoneurosomes. (A) The STEP WT and KO cultures were stimulated with increasing concentration of DHPG for 5 min and stained with pERK1/2 antibody. DHPG leads to activation of pERK1/2 in both the WT and KO cultures. However, the phosphorylation of ERK1/2 is more pronounced in the KO cells when compared with WT. This suggests that in the absence of STEP, there is a higher level of pERK1/2 activation [Scale bar-50 μm]. (B) Synaptoneurosomes prepared from WT and KO mice hippocampi were stimulated with increasing concentration of DHPG and processed for Western blotting. The membranes were probed with pERK1/2 and ERK2 antibodies sequentially. (C) The pERK1/2 levels from WT and KO synaptoneurosomes were measured using ImageJ and normalized to total ERK2 levels. The normalized pERK1/2 levels are plotted as a percentage of WT control levels. The data were compared using one-way ANOVA, followed by posthoc Tukey HSD (WT 20 μm: 126% ± 2%, P > 0.1; WT 50 μm: 138% ± 7%, P < 0.02; WT 100 μm: 143% ± 12%, P < 0.01; KO Ctl: 132% ± 5%, P < 0.02; KO 20 μm: 185% ± 19%, P < 0.01; KO 50 μm: 198% ± 13%, P < 0.005; and KO 100 μm: 237% ± 37%, P < 0.001; *P < 0.05; **P < 0.01; ***P < 0.001; n = 3). (D) The integrity of the synaptoneurosomes was determined by probing the input, cytosolic and synaptosomal fractions with histone H1 (present only in the input), CaMKII (detected in all fractions), and PSD95 (present mainly in the input and synaptosomal fraction).