The pCAGGSneo expression vector was kindly provided by Professor J. Miyazaki (Osaka University, Japan). cDNAs for the CaMKIIδ2 isoform were obtained and inserted into the EcoRI site of pCAGGSneo under control of the chicken β-actin promoter (Takeuchi et al. 1999). To obtain a constitutively active mutant of CaMKII (Act KII), we replaced Thr287 of CaMKIIδ2 with Asp (Hughes et al. 2001). The C-terminal of Act KII, in which Thr287 was replaced with Asp, was obtained by amplification using PCR with a sense primer (5′-TCAACGTTCTACTGTTGCCTCCATGATGCACAGGCAGGAGGATGTAGA-3′) and an antisense primer (5′-TCTCGAGTTAGAAGACCCAAATGTGAA-3′). The sense primer contains an AclI site at the 5′ end; sequences changing Thr287 to Asp are shown by underlining. The antisense primer corresponds to nucleotides 1507–1526 and contains an EcoRI site at the 5′ end. The PCR products were cut with AclI. The N-terminal of Act KII was excised with EcoRI and AclI from cDNA encoding wild-type CaMKIIδ2. Both fragments were ligated with the AclI sites to each other and then inserted into pCAGGSneo, which was verified by sequencing (termed pActKII). To obtain constitutively active CaMKIV (Act KIV), the cDNA for the entire coding region of CaMKIVα was obtained from the rat cerebellum by RT-PCR using a sense primer (5′-ATGCTCAAAGTCACGGTGCCCTCC-3′), and an antisense primer (5′-TTAGTACTCTGGCAGAATAGCATC-3′). Act KIV was obtained by amplification of the catalytic domain of CaMKIV (1–313 amino acids) using PCR with the sense primer (5′-TTCTCGAGCTATGCTCAAAGTCACGGTG-3′) and the antisense primer (5′-TTCTCGAGAAAGCTTTTTCTGAGCGGTATC-3′), both of which contain an XhoI site 5′ end of each primer. The PCR products were cut with XhoI and inserted into the XhoI site of pEYFP-Nuc expression vector (termed pKIV-Nuc). Dominant negative JNK 1 (DN-JNK1) was constructed by replacing Tyr185 and Thr183, which had to be phosphorylated for activation, with Ala and Phe, respectively. The cDNA of the entire JNK1 coding region was obtained from RNA from NB2A cells by RT-PCR using a sense primer (5′-TTCTCGAGATGAGCAGAAGCAAACGT-3′) and an antisense primer (5′-TTCTCGAGTCATTGCTGCACCTGTGCTA-3′), both of which contain 5′XhoI sites. The C-terminal of DN-JNK1 was obtained by PCR with the antisense primer used to amplify the coding region of JNK1 and a sense primer (5′-AACAAGCTTTATGATGTTTCCTGCAGTGGT-3′), which contained an HindIII site and in which Tyr185 and Thr183 codons were replaced with Ala and Phe codons, respectively. The DN-JNK1 N-terminus was obtained by PCR with the sense primer used for amplification of the coding region of JNK1 and an antisense primer (5′-ATAAAGCTTGTTCCTGCAGTCCTC-3′), which contained a 5′HindIII site. Both amplified fragments were cut with HindIII and ligated with the HindIII sites to each other. The ligated fragments were inserted into pCAGGSneo and sequenced (termed pDN-JNK1). Similarly, the cDNA encoding all of p38MAPK was obtained from NB2A cell RNA by RT-PCR using a sense primer (5′-TTCTCGAGATGTCGCAGGAAAGGCCCACGTTC-3′) and an antisense primer (5′-TTCTCGAGTCAGGACTCCATTTCTTCTT GGTC-3′), which contain 5′XhoI sites. The C-terminus of dominant negative p38MAPK (DN-p38MAPK) was obtained by PCR with the antisense primer described above and a sense primer (5′-GTGGTACCGAGCCCCAGAGATCAT-3′). The DN-p38MAPK N-terminus was obtained by PCR with the sense primer described above and an antisense primer (5′-TCGGTACCACCTGGTAGCCACTGCGCCAAACATCTC-3′), in which Tyr182 and Thr180 codons were replaced with Ala and Phe codons, respectively. Both amplified fragments were cut with KpnI and ligated with the KpnI sites to each other. The ligated fragments were inserted into pCAGGSneo and sequenced (termed pDN-p38MAPK). The following vectors were obtained from the Path Detect System (Stratagene, La Jolla, CA, USA): pFc-MEKK (encoding amino acids 380–672 of MEKK), pFc-PKA (encoding the PKA catalytic subunit), and pSRE-Luci and pCRE-Luci. The expression vector for constitutively active MEKK encodes amino acids 380–672 of MEKK, in which the regulatory N-terminal domain is deleted. The expression vector for constitutively active PKA encodes the catalytic subunit. The cDNAs for Sp1 were obtained by amplification from NB2A cell RNA by RT-PCR with the sense primer (5′-ATGAATTCATGAGCGACCAAGATCACTCAAT-3′) and an antisense primer (5′-TTGAATTCCTCAGAAACCATTGCCACTGATA-3′), both of which contain 5′-EcoRI sites. The cDNAs for Zif268 were obtained by RT-PCR amplification from NB2A cell RNA using a sense primer (5′-ATGAATTCATGAGCGACCAAGATCACTCAAT-3′) and an antisense primer (5′-TTGAATTCCTCAGAAACCATTGCCACTGATA-3′), which also contain 5′-EcoRI sites. The cDNAs of Sp1 and Zif268 were cut with EcoRI and inserted into pCAGGSneo (termed pSp1 and pZif268, respectively). To construct DRRF-HA, we first amplified the coding region of DRRF by RT-PCR from NB2A cell RNA with a sense primer (5′-GGAATTCATGTCGGCGGCCGTGGCGTGTGT-3′), which contains a 5′-EcoRI site, and an antisense primer (5′-AACATCGTATGGGTACAAGCCAGCAGGAGCTGGGCT-3′), which contains part of the sequence of the HA-tag at the 5′ end. Using the first PCR products as a template, the second amplification was carried out with the same sense primer and an antisense primer (5′-TTGAATTCTTAAGCGTAATCTGGAACATCGTATGGGTA-3′), which consists of a 5′-EcoRI site and the sequence of HA-tag. The cDNA of DRRF-HA was cut with EcoRI, inserted into pCAGGSneo, and sequenced (termed pDRRF-HA).