Ca²⁺/calmodulin-dependent protein kinase II (CaM kinase II) is one of the most abundant protein kinases in mammalian brain and a major mediator of calcium signaling in neurons. CaM kinase II is involved in diverse functions in neurons, including neurotransmitter synthesis and neurotransmitter release, modulation of ion channel activity, synaptic plasticity, learning and memory, and gene expression (-). The α and β isoforms of CaM kinase II are expressed almost exclusively in the nervous system, but the expression of both isoforms is not uniform throughout the brain, and is regulated developmentally. In the adult rat, the α isoform is abundant in the forebrain, whereas the β isoform is abundant in the cerebellum (,). The enzyme level is increased most rapidly in postnatal brain during the most active phase of synapse formation (-). Thus, the space-temporal gene expression of CaM kinase II indicates that the enzyme is carefully regulated at the level of transcription. The transcriptional regulation of many eukaryotic genes has been extensively studied and tissue-specific enhancer and silencer elements have been identified. However, little is known about the neuronal cell type-specific expression of the CaM kinase II gene. Therefore, it is imperative that there is a method to analyze the gene expression. We have analyzed the promoter activity and a protein bound to the DNA sequence of the gene (,). Some of our procedures may be modified, although major procedures are known to provide good results.

Materials

[γ-³²P]ATP (specific activity of 110 TBq/mmol) was purchased from NEN^TM Life Science Products, Inc. (Boston, MA.). Double strand poly(dI-dC) and Sephadex G-50 were obtained from Amersham Pharmacia Japan (Tokyo, Japan). Oligonucleotide primers were synthesized by Biologica Co. (Nagoya, Japan), or by Invitrogen Life Tech. Japan (Tokyo, Japan). Luciferase assay system, luciferase reporter pGL2- and pGL3-control vectors, pGL2- and pGL3-basic vectors, and Wizard^TM Plus Minipreps DNA purification system were from Promega Corp. (Madison, WI). Geneclean II DNA purification kit was from BIO 101 Inc. (Vista, CA). FuGENE^TM6 transfection reagent and chlorophenol red β -D-galactopyranoside (CPRG) were from Roche Diagnostics Japan Corp. (Tokyo, Japan). Dulbecco's modified Eagle's medium (DMEM), Ham F12 medium, and DMEM/Ham F12 (1:1) medium were from Invitrogen Life Tech. Japan (Tokyo, Japan). Eagle's MEM amino acids and vitamins medium was from Nissui Pharmaceutical Co. Ltd. (Tokyo, Japan). Fetal bovine serum (FBS) was from JRH Bioscience (Lenexa, KS). X-OMAT film was from Eastman Kodak Company (Rochester, NY).

Construction of α and β CaM Kinase II Promoter-Luciferase Reporters

α CaM kinase II promoter-luciferase reporters were created by polymerase chain reaction (PCR) with cloned genomic DNA (CL-1 clone) () as a template. PCR fragments possessing 5' flanking sequences with various 5' sense primers and a common 3' end at +112 were inserted into a promoterless pGL3-basic vector. The 5' sense and 3' antisense primers were designed according to the sequence of the gene as shown in Table 1 (see Protocols section). To the 5' sense and 3' antisense primers were added BglII and HindIII restriction sites, respectively. PCR products were separated by agarose gel electrophoresis and excised, and then purified using the Geneclean II kit. PCR products were also separated by polyacrylamide gel electrophoresis (PAGE) and excised, and then purified as described (). Purified PCR products were digested with BglII and HindIII, and ligated into a promoterless pGL3-basic vector. The 5' flanking sequence containing 8.5 kbp of the mouse α CaM kinase II gene () was digested with SalI and inserted into pGL3-basic vector to make pGLmα8.5k.

β CaM kinase II promoter-luciferase reporters were prepared in the same way as the α CaM kinase II luciferase reporters. Genomic DNA clone (BCL-1) or the subcloned plasmid DNA of the β CaM kinase II gene () was used as a template. To the 5' sense and 3' antisense primers were added KpnI and XhoI restriction sites, respectively. PCR fragments were cloned directly into a promoterless pGL2-basic vector. A construct devoid of the transcription initiation site (pGLΔ 0.57) was prepared from the fragment of NcoI (nucleotide -577) and EheI (nucleotide -45) of the gene.

Cell Culture and Transfection

The following cell lines were used in the experiment; neuronal NG108-15, N18TG2, and CAD cells and non-neuronal BALB/c3T3, CHO-K1, and HT1080 cells. For transient transfection, cultured cells were plated on 35-mm tissue culture dishes at approximately 1-3 x 10⁵ cells/dish, and cultured 24 h. Luciferase reporter DNA (1.2 μg of pGL2- or pGL3-control vector or the molar equivalent of luciferase constructs) with β-galactosidase expression vector (pcDNA3/β -gal) DNA (0.8 μg) were introduced into the cells via FuGENE^TM6 reagent treatment (3 μl of FuGENE^TM6 reagent/dish) as instructed by the manufacturer. After 24 h, cells were harvested and extracted for assay. All transfections were performed in duplicate.

Luciferase Expression

Plated cells were washed with phosphate-buffered saline (PBS), and then lysed in 200 μl of lysis buffer. Luciferase activity was assayed in a reaction mixture containing 20 mM tricine, pH 7.8, 470 μM luciferin, 530 μM ATP, 270 μM Coenzyme A, 1.07 mM (MgCO₃)₄Mg(OH)₂·5H₂0, 2.67 mM MgSO₄, 0.1 mM EDTA, 33.3 mM dithiothreitol, and nuclear extract. Luciferase activity was measured for 20 s after the addition of reagents. The efficiency of transfection was corrected by the β -galactosidase activity. Promoter activity was compared by measuring the luciferase activity of the pGL2- or pGL3-control vector with the SV40 early gene promoter, and expressed as normalized luciferase activity. β-Galactosidase activity was assayed by a modification of the method described by Herbmel et al. () The enzymatic activity of individual samples was measured in triplicate.

Electrophoretic Mobility Shift Analysis

Nuclear extracts were prepared as described (). Sense and antisense oligonucleotide probes were annealed, and end-labeled with [γ-³²P]ATP and T4 polynucleotide kinase. The binding reaction was carried out in a reaction mixture containing 25 mM Tris buffer (pH 7.5), 75 mM KCl, 1 mM EGTA, 1 μg of poly(dI-dC), 13% glycerol, 10 μg of the nuclear extract, and 20-50 fmol ³²P-labeled probe (about 10,000 cpm) with or without non-labeled probe or competitor DNA. After incubation for 30 min at 0 °C, the reaction mixture was separated by 6.4% polyacrylamide gel electrophoresis at 200 V for 30 min in 0.5x TBE, fixed, dried, and autoradiographed.

Analysis of Promoter Activity of α CaM Kinase II Gene

We examined the promoter activity in neuronal and non-neuronal cell lines. Transient transfection was used to compare the promoter activity of the luciferase reporter gene. The structure of the 5'-flanking region of the α CaM kinase II gene and one of the luciferase reporters (pGLα2281) is shown in Fig. A. Fragments possessing a common 3' end (+112 bp) containing a sequence of 2281, 1622, 757, 275, 199, 145, and 100 bp 5' upstream from the transcription initiation site were cloned upstream of the luciferase reporter gene in a promoterless pGL3-basic vector to make respective pGLα constructs. α CaM kinase II promoter activity was compared with that of pGL3-control vector by measuring luciferase activity, and expressed as normalized luciferase activity.

In NG108-15 cells, the strongest promoter activity was obtained with pGLα 199, which yielded a 3.68-fold increase in luciferase activity relative to pGL3-control vector (Fig. B). pGLα 145 also showed high promoter activity at about 40% of that of pGLα 199, but pGLα 100 showed very little promoter activity. An internal deletion mutant (pGLα 199Δ) that had no transcription initiation site showed no significant promoter activity in NG108-15 cells, indicating that the transcription initiation site was important to the transcription of the α CaM kinase II gene. Similar results were obtained in neuronal cell lines, N18TG2 and CAD (data not shown), and normalized luciferase activity of pGLα 199 in NG108-15, N18TG2, and CAD cells was 3.68, 3.08, and 1.87, respectively. However, luciferase activity was extremely weak in non-neuronal cells - about 0.007 in BALB/c3T3 (Fig. C). In other non-neuronal cell lines, HT1080, and CHO-K1, the activity was also low and less than 0.05 (data not shown). Promoter activity in non-neuronal cells was 100- to 500-fold lower than that in neuronal cells. These results indicated that the 5' upstream region -199 to -100 bases from the transcription initiation site showed neuronal cell type-specific strong promoter activity.

The inclusion of an additional 5' flanking sequence of 275 bases (pGLα 275) decreased luciferase expression only 0.57-fold relative to the pGL3-control vector in NG108-15 cells (Fig. B). The luciferase activity of pGLα 275 was almost the same as that of pGLα 757, pGLα 1662, and pGLα 2281. Since the sequence of the 8.5 kbp 5' flanking region is known to show brain-specific expression of α CaM kinase II (), the promoter activity of pGLmα 8.5k was compared with that of the rat promoter. pGLmα 8.5k showed weak promoter activity, as did pGLα 275 in NG108-15 cells. Similar results were obtained in N18TG2 and CAD cells (data not shown). These results indicate that there is a strong silencer element within the 5' upstream region -275 to -199 bp from the transcription initiation site.

The promoter activity of all regions of the 5'-flanking sequence was much weaker in non-neuronal cells including BALB/c3T3, HT1080 and CHO-K1 than in neuronal cells (Fig. C, and data not shown).

Analysis of Promoter Activity of β CaM kinase II Gene

Since the β isoform was expressed almost exclusively in the brain, the promoter activity of the β CaM kinase II gene was also examined. The structure of the 5' flanking region of the β CaM kinase II gene and one of the luciferase-reporters (pGL3.2) is shown in Fig. D. Fragments possessing a common 3' end (+78 bp) containing sequences of 3.2, 0.57, 0.32, 0.22, 0.12, 0.06, and 0.03 kbp 5' upstream from the transcription initiation site were cloned upstream of the luciferase reporter gene in a promoterless pGL2-basic vector to make respective pGL constructs. The luciferase activity of these reporters was analyzed in neuronal cells, NG108-15, N18TG2, and CAD cells, and in non-neuronal cells, BALB/c3T3, HT1080, and CHO-K1 cells. The results indicate that the 5' upstream region -66 to -35 bp from the transcription initiation site contained neuronal cell type specific strong promoter activity, and that there are two silencing elements within the regions -222 to -123 bp and -576 to -323 bp. Representative results are shown in Fig. E and F.

The relative luciferase activity of the α isoform gene was lower than that of the β isoform, because the luciferase activity of the pGL3-control vector was about 5 to 10-fold that of the pGL2-control vector. pGL3-control vector contains more SV40 early gene promoter than pGL2-control vector as described by the manufacturer.

Investigation of Nuclear Protein Bound to DNA of the Promoter Region of β CaM kinase II

Since neuronal cell type-specific promoter activity was found in the region from -66 to -35 bp of the β CaM kinase II gene, a gel mobility shift analysis was performed using nuclear extract from various rat tissues. The 32P-labeled probe was incubated with nuclear extract of the brain, including forebrain, cerebellum, and brainstem, and non-neuronal tissues, including liver, kidney and spleen. A band specific to the brain was observed as indicated by an arrow in Fig. A, although some other bands were expressed ubiquitously. Its density was highest in the cerebellum, consistent with the expression of β CaM kinase II (,).

The radioactivity of this band detected by incubation of the extract of cerebellum was reduced almost completely in the presence of a 5-fold excess of non-radioactive probe (Fig. B). These results indicate that this band was formed by specific binding of the nuclear protein in the brain.

A protein bound to the promoter regions (nucleotides -145 to -100, and -199 to -146) of α CaM kinase II was also found in the nuclear extract of rat brain, including forebrain, brainstem and cerebellum, but not in non-neuronal tissues, including liver, kidney and spleen (data not shown). There is no homology in the sequence in the promoter region of both α and β genes. The sequence of the promoter region is 5'- CTT CTG GGC CCA CAC AGT CCT GCA GTA TTG TGT ATA TAA GGC CAG GGC AAC GGA GGA GCA GGT TTT GAA GTG AAA GGC AGG CAG GTG TTG GGG AGG CAG T -3' (-199 to -100) for the α gene and 5'- CAG CCA GCC CGG CCC CCG CCC GGC GCC GCG AG -3' (-66 to -35) for the β gene. Known transcription factor binding sites were underlined which corresponded to TATA box, v-myb, and MyoD in the α gene and Sp1 in the β gene, respectively. These sites were not specific to neuronal cells. Thus, it is suggested that there are unidentified promoter elements in these regions.

We have established a method for luciferase reporter analysis by which the number of cells expressing a specific gene can be monitored through the assay of luciferase and β -galactosidase activities. The method provides an alternative for analyzing the promoter activity of different genes. The main observations we have made with this technique are 1) neuronal cell type-specific expression of the α and β gene occurs in neuronal cell lines (silencer elements were also found in further upstream promoter regions) and 2) neuronal cell type-specific proteins bound to the promoter region are found in nuclear extracts of brain cells.