- Email: [email protected]
Identification of a negative regulatory DNA element for neuronal BC1 RNA expression by RNA polymerase III
Biochimica et Biophysica Acta 1493 (2000) 142^150
www.elsevier.com/locate/bba
Identi¢cation of a negative regulatory DNA element for neuronal BC1 RNA expression by RNA polymerase III Shunsuke Kobayashi, Sakurako Kamo, Akiko Ohmae, Keisuke Agui, Yanmei Li, Kaijiro Anzai * Department of Biochemistry, College of Pharmacy, Nihon University, Narashinodai, Funabashi, Chiba 274-8555, Japan Received 1 May 2000; received in revised form 21 June 2000; accepted 6 July 2000
Abstract BC1 RNA is a neuronal cell-specific RNA polymerase III (Pol III) transcript. The BC1 RNA gene has plural types of Pol III promoters, in addition to which an E-box sequence (E2 site) acts as a transcriptional activator, which is recognized by a brain-specific protein(s). Using an in vitro transcription system, we found that the upstream region of the BC1 RNA gene contained a sequence that interfered with the activity of the E-box element in a distance-independent manner. A tandem repeat within this sequence, which was weakly homologous with the neuron-restrictive silencer element (NRSE) found in the Pol II system, was recognized by a brain nuclear protein. Consistently, the transcriptional activity increased by deleting the tandem repeat sequence. We called this BC1 RNA-repressing element BCRE. The DNAbinding specificities of BCRE-binding protein differed from that of NRSE-binding protein (NRSF). A similar protein with an ability to bind to BCRE was also found in liver and kidney. Furthermore, the glutamate analog kainic acid increased the DNA-binding of both E2 sitebinding protein and BCRE-binding protein, and then the levels of BC1 RNA also increased transiently. Our results suggested that both positive and negative regulatory elements contribute to neuronal BC1 RNA expression. ß 2000 Elsevier Science B.V. All rights reserved. Keywords : Neuronal BC1 RNA ; Promoter; In vitro transcription system; Negative regulatory element
1. Introduction BC1 RNA is expressed preferentially in the brain by RNA polymerase III (Pol III) [1^3]. This small non-translatable RNA is distributed in neuronal dendrites as ribonucleoprotein particles, BC1 RNPs [4^7], and concentrated in synaptodendrosomes with other dendritic mRNAs [8]. It has been reported that local protein synthesis at the synapse is required for synaptic plasticity [9,10]. In a previous study, we demonstrated that translin/testis-brain RNA-binding protein, which is known to be a translational repressor [11^14], bound to BC1 RNA [15] and was distributed in the neuronal dendrites [16]. Recently, it was reported that autoantigen La/SS-B associated with BC1 RNA and suggested that La protein links
* Corresponding author. Fax: +81-47-465-5332; E-mail : [email protected]
BC1 RNP to the ribosome [17]. These observations suggest that BC1 RNP plays a role(s) in translational regulation of dendritic mRNAs. Among the many Pol III-transcribed genes (class III genes), neuronal cell speci¢city is unique to the BC1 RNA gene. The levels of this small non-translatable RNA increase during the period of synaptogenesis [18,19]. Furthermore, the somato-dendritic levels of this RNA in hippocampal neurons increased when hyperexcitability was induced with picrotoxin [20] and inhibition of the electrical activity of developing hippocampal neurons by tetrodotoxin resulted in a reduction in the level of this RNA [21]. These observations indicate that BC1 RNA transcription is regulated by neuronal activity. We are interested in determining how the unique expression pro¢le of this class III gene in the brain is achieved. Studies using an in vitro transcription system revealed that both intragenic and extragenic cis-elements are necessary for BC1 RNA transcription. The intragenic promoter elements, A- and B-boxes [22] and GCAAG/CTTGC motifs present between these two boxes, were found to be essential [23] and a 5P-£anking region containing a TATA-like sequence and a proximal sequence element
0167-4781 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 0 0 ) 0 0 1 7 5 - 5
BBAEXP 93442 25-8-00
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
(PSE) were also necessary [22]. Moreover, we demonstrated that an E-box sequence at 3248 (E2 site; CAATTG), recognized by a brain-speci¢c protein(s) (E2 site-binding protein), acted as a transcriptional activator of BC1 RNA expression by Pol III [24]. In the Pol II system, several E-box-binding basic helixloop-helix transcription factors have been identi¢ed in developing and adult neuronal tissues, and they have been suggested to play important roles in neuronal development and activity [25,26]. Recently, a neuron-restrictive silencer element (NRSE), which is recognized by a repressor/silencer factor (NRSF; also known as REST), was also identi¢ed and found to repress expression of neuronal cell-speci¢c genes in non-neuronal cells [27,28]. However, Palm et al. reported activity-dependent increases in NRSF mRNA levels in neurons and suggested that NRSF counteracts the e¡ects of positive regulators, thereby modulating target gene expression quantitatively [29]. Timmusk et al. showed that in the neurons of adult brains, kainic acid (KA)-induced expression of brain-derived neurotrophic factor (BDNF) mRNA is under the control of NRSE [30]. These observations suggest that the balance between positive and negative regulators is important for control of neuron-speci¢c gene transcription by Pol II. In this context, there is an interesting possibility that similar mechanisms may also operate the BC1 RNA expression by Pol III. In this study, we identi¢ed a tandem repeat (referred to as BC1 RNA-repressing element: BCRE), which showed weak homology with the consensus sequence of NRSE
143
[31], in the 5P-£anking region (3638V3685) of the BC1 RNA gene and found that this element interfered with the transactivation of BC1 RNA expression by E2 site. We also characterized the BCRE-binding protein and demonstrated that the DNA-binding speci¢city of the protein di¡ered from that of the NRSF. We further showed that the DNA-binding of E2 site-binding protein and BCREbinding protein and the level of BC1 RNA expression increased transiently in the brain of animals treated with KA. Our data suggest that a positive and a negative control element cooperate to regulate neuronal BC1 RNA transcription. 2. Materials and methods 2.1. DNA cloning The BC1 RNA gene was isolated from a rat genomic library using the 3P-portion of BC1 RNA as a probe. The nucleotide sequence downstream of position 3630 has been reported elsewhere [22,24]. A genomic clone, termed clone K0, containing a further region upstream of nucleotide 3740 was cloned into EcoRI and BamHI sites of the pBR 322 vector. 2.2. Plasmid construction The 5P-truncated constructs were prepared by performing polymerase chain reactions (PCRs) using the cloned
Fig. 1. Schematic representation of 5P-£anking and coding regions of the BC1 RNA gene and in vitro transcriptional analysis of 5P-deletion constructs. (A) The E2 site, the PSE, TATA-like box, A- and B-boxes and GCAAG/CTTGC motifs are shown. The arrows (K0^K4) indicate the 5P-end-points of the deletion constructs (clones K0^K4). (B) The transcriptional reactions were performed using clones K0^K4 (lanes 1^5, respectively). The arrow indicates the BC1 RNA transcript. As size markers, positions of 5.8S, 5S and 4S RNAs are also indicated.
BBAEXP 93442 25-8-00
144
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
of 3-week-old inbred rats (Wistar) or mice (ICR), essentially as described by Gorski et al. [33]. The transcription reaction was performed in a mixture (20 Wl) containing 40 Wg/ml template DNA and 3 mg/ml nuclear protein, as described previously [23]. 2.4. Electrophoretic gel mobility shift assay The sequences of the double-stranded oligonucleotide probes and competitors are shown in Figs. 3 and 5. Each 32 P-labeled probe (1U104 cpm) was incubated at room temperature for 30 min in a mixture (26 Wl) containing 10 mM Tris^HCl, pH 7.5, 50 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 5% glycerol, 1 Wg poly (dI-dC) and 3 Wg nuclear protein extract. The reaction product was subjected to electrophoresis on a 4% polyacrylamide gel containing 6.7 mM Tris^HCl, pH 7.5, 3.3 mM sodium acetate and 1 mM EDTA with electrode bu¡er recirculation. 2.5. Northern blot analysis Fig. 2. In vitro transcriptional analysis of fusion constructs. Schematic representations of the constructs are shown. The open box indicates fragment K, and clones KK2 and KK3 are fusion constructs of fragment K to clones K2 and K3, respectively. The transcription reactions were performed using clones K1, K2 and K3 (lanes 1, 3 and 5, respectively), clones K0, KK2 and KK3 (lanes 2, 4 and 6, respectively) and the pBR 322 vector (lane 7). Each BC1 RNA transcript is indicated by an arrow. As size markers, positions of 5.8S, 5S and 4S RNAs are also indicated.
BC1 RNA gene as a template, 20 nucleotides of the 5P-end primer of each site (K1^K4 in Fig. 1), which had been added to the EcoRI site, and the 3P-end primer of the BC1 RNA gene (CCTCGGGATCCTGTCATTTTGTGTAGCAATCG), which was added to the BamHI site. Each PCR product was cloned into the pBR 322 vector to produce clones K1^K4. In order to prepare fusion constructs, fragment K, which had been added to the AatII site at the 5P-end and the EcoRI site at the 3P-end, was synthesized by the PCR and ligated to the AatII and EcoRI double-digested clones K2 and K3 to produce clones KK2 and KK3, respectively. In order to produce the BCRE-deleted construct (clone dK) using fragment K as a template, a fragment consisting of a pyrimidinerich and a palindromic domain, which were added to the AatII site at the 5P-end and the EcoRI site at the 3P-end, respectively, was synthesized by the PCR and ligated to the AatII and EcoRI double-digested clone K1. All the DNAs were puri¢ed by CsCl ultracentrifugation [32]. 2.3. Preparation of nuclear protein extracts and the in vitro transcription reaction Nuclear protein extracts were prepared from the tissues
Total RNA was isolated from tissues by homogenization in guanidine isothiocyanate, followed by phenol extraction and ethanol precipitation [34]. The RNA (10 Wg) was electrophoresed in 1.5% agarose gels containing 2.2 M formaldehyde and then transferred to nitrocellulose membrane. Hybridization was performed overnight at 42³C with 32 P-labeled probes in hybridization bu¡er (5USSPE, 5UDenhardt's reagent, 50% formamide, 1% sodium dodecyl sulfate (SDS), 0.1 mg/ml salmon sperm DNA). The ¢lters were washed in 2USSC^0.1% SDS at room temperature for 30 min, followed by two 10 min washes at 52³C in 0.2USSC^0.1% SDS. 2.6. KA seizure induction Inbred male mice were injected with KA (15 Wg/kg) or physiological saline into the lateral ventricle. The animals were killed at the indicated time after injection, and then nuclear extracts or total RNAs were prepared from a region consisting of the cerebral cortex and hippocampus. 2.7. Nuclear run-on assay Nuclei were isolated from dissected cerebral cortex and hippocampal tissues of KA-treated mice by centrifugation through a dense sucrose solution [33,35]. The run-on transcription assay was performed essentially as described by Greenberg and Zi¡ [36]. The 32 P-labeled transcripts were hybridized to nitrocellulose membranes containing 1 Wg dots of denatured cDNA of BC1 RNA which was cloned into the pUC 19 vector. The vector plasmid pUC 19 was used as a control. Hybridization and washing of the ¢lters were carried out in the same way as for Northern blot analysis.
BBAEXP 93442 25-8-00
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
145
Fig. 3. Electrophoretic gel mobility shift assay of subfragments of fragment K. (A) Upper : nucleotide sequence of fragment K. The oligonucleotide probes for the gel mobility shift assay are indicated above the sequence. The pyrimidine-rich, palindromic and tandem repeat domains are indicated below the sequence. Tandem repeats I and II are boxed. Lower: comparison of tandem repeats I and II with the consensus sequence of NRSE. The boxes indicate the identical nucleotide residues in the tandem repeat and NRSE consensus sequence. (B) The gel mobility shift assay was performed using 3 Wg brain nuclear extract and 32 P-labeled subfragments of fragment K as probes. The probes (1U104 cpm) used were a, b, c, I+II, I and II (lanes 1^6, respectively).
3. Results 3.1. The upstream region of the BC1 RNA gene contains a negative regulatory sequence Fig. 1A shows the 5P-upstream and intragenic promoters of the BC1 RNA gene. The internal A- and B-boxes, GCAAG/CTTGC motifs, external TATA box and PSE are necessary for BC1 RNA transcription by Pol III [22,23]. In a previous study, we used an in vitro transcription system, an electrophoretic gel mobility shift assay and a DNase I protection assay and found that the E-box at 3248 (E2 site: CAATTG) was recognized by a
brain-speci¢c E2 site-binding protein and activated BC1 RNA transcription [24]. Transcriptional analysis of 5P-truncated constructs of the BC1 RNA gene revealed that clone K3 containing the E2 site was transcribed more e¤ciently than clone K4, which lacks the E2 site (Fig. 1B, lanes 4 and 5; see Fig. 1A for clones K3 and K4). Clones K1 and K2 were transcribed as e¤ciently as clone K3, indicating that the region between 3324 and 3630 does not contain sequences that are signi¢cant for BC1 RNA transcription (Fig. 1B, lanes 2^4). However, the presence of a further upstream region between 3631 and 3740 (clone K0) reduced the transcriptional activity (Fig. 1B, lane 1). We termed this negative regulatory region fragment K.
BBAEXP 93442 25-8-00
146
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
formed transcriptional analysis. As shown in Fig. 4, clone dK was transcribed more e¤ciently than clone K0. Therefore, it is indicated that this sequence is the BCRE consisting of a tandem repeat (BCRE-I and -II) and the BCRE-binding protein may be a negative regulator for BC1 RNA expression. 3.3. DNA-binding speci¢city of the BCRE-binding protein
Fig. 4. In vitro transcriptional analysis of the BCRE deletion construct. The hatched box indicates BCRE. The transcription reactions were performed using clones K0 (lane 1) and dK (lane 2). The BC1 RNA transcript is indicated by an arrow. Positions of 5.8S, 5S and 4S RNAs are also indicated.
In order to verify the repressor activity, we constructed templates in which fragment K was fused to regions proximal to the E2 site (clones KK2 and KK3 in Fig. 2) and performed transcriptional analysis. These fusion constructs showed reduced transcriptional activity similar to that of clone K0. These results reveal that fragment K contains a sequence that interferes with the positive e¡ect of the E2 site on BC1 RNA transcription independently of the distance from the E-box sequence. 3.2. Fragment K contains a BCRE consisting of a tandem repeat Fig. 3A shows the nucleotide sequence of fragment K, which can be subdivided into 5P-pyrimidine-rich domain, central palindromic domain and 3P-tandem repeat domain. The tandem repeat sequences I and II had limited homology (52%) with the NRSE consensus sequence [31]. In order to analyze the DNA-binding proteins, we synthesized 32 P-labeled double-stranded oligonucleotide probes corresponding to the subdomains of fragment K and performed the electrophoretic gel mobility shift assay. As shown in Fig. 3B, the tandem repeat-containing probes were recognized by a brain nuclear protein (lanes 3 and 4). This binding protein appeared to recognize either the repeat sequence I or II (Fig. 3B, lanes 5 and 6), as the mobility of each complex was identical to that of the complex consisting of the protein and the probe, containing both the repeat sequences (compare lane 4 with lanes 5 and 6). Next, we constructed a template, clone dK, which lacks the tandem repeat region of fragment K only and per-
Although BCRE-I and -II had a short stretch of nucleotides identical with those in NRSE (Fig. 3A), their overall nucleotide sequences deviated considerably from the target sequences of NRSF [31]. In order to examine the binding speci¢city of BCRE-binding protein, we performed the electrophoretic gel mobility shift assay using oligonucleotides containing the NRSEs of the SCG10 [37] and choline acetyltransferase (ChAT) [38] genes as probes and competitors (Fig. 5A). When the 32 P-labeled NRSEs of SCG10 and ChAT were used as probes, no or only weak shifts were observed under the assay conditions we employed (open arrowhead in Fig. 5B). This may be due to the low levels of NRSF in the brain nuclear extract. Furthermore, competition analysis revealed that these oligonucleotides were at least 10 times less potent than the unlabeled BCRE-I (Fig. 5C). Above observations indicate that BCRE-binding protein and NRSF are di¡erent and that BCRE-binding protein recognizes the BCRE in a sequence-speci¢c manner. Additionally, we again con¢rmed that the nuclear proteins used in these assays shown in Fig. 5B,C inhibited transcription of BCRE-containing template only (compare clone K0 with clone K1, Fig. 5D; see also Figs. 1 and 2). 3.4. Tissue distribution of BCRE-binding protein As BC1 RNA is brain-speci¢c (Fig. 6A), we next examined the tissue distribution of BCRE-binding protein. Nuclear extracts from brain and non-neuronal tissues (liver and kidney) were prepared, and an electrophoretic gel mobility shift assay was performed using BCRE-I and DNA fragment b (see Fig. 3A) as a speci¢c probe and a control probe, respectively. As shown in Fig. 6B, a protein which recognized BCRE-I, but not DNA fragment b, was also found in the nuclear extracts from liver and kidney. The mobility of each complex formed was the same as that of brain. Therefore, it is indicated that a similar nuclear protein, which binds speci¢cally to BCRE-I, is present in nonneuronal tissues. On the other hand, a transcriptional activator E2 site-binding protein, which forms one major (big arrowhead) and two minor (small arrowhead) complexes, as observed previously [24], was detected only in the brain nuclear extract (Fig. 6C). These ¢ndings suggest that tissue-speci¢c expression of BC1 RNA is controlled at the transcriptional level.
BBAEXP 93442 25-8-00
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
147
Fig. 5. Analysis of binding speci¢city of BCRE-binding protein. (A) Alignment of the oligonucleotides used as probes and competitors. BCRE-I and NRSEs of SCG10 and ChAT were underlined. (B) The gel mobility shift assay was performed using 3 Wg brain nuclear extract and 32 P-labeled (1U104 cpm) BCRE-I (lane 1), SCG10 (lane 2) and ChAT (lane 3) double-stranded oligonucleotides. (C) Competition analysis of BCRE-binding protein and 32 P-labeled BCRE-I complex formation was performed in the presence of a molar excess of unlabeled BCRE-I (lanes 1^3), SCG10 (lanes 4 and 5) or ChAT (lanes 6 and 7) double-stranded oligonucleotide. (D) In vitro transcriptional analysis was performed using the nuclear protein fraction prepared for this gel mobility shift assay (lanes 1 and 2, clones K0 and K1 in Fig. 1, respectively). The BC1 RNA transcript is indicated by an arrow. Positions of 5.8S, 5S and 4S RNAs are also indicated.
3.5. E¡ects of KA-induced seizure on BC1 RNA expression and DNA-binding activities of E2 site-binding protein and BCRE-binding protein In order to examine the neuronal activity-dependent alteration of BC1 RNA expression, we studied the e¡ect of KA-induced seizure in the mouse brain. Northern blot analysis revealed that the level of BC1 RNA in the region consisting of the cerebral cortex and hippocampus tissues increased gradually to a maximum level (3-fold) 3 h after injection of KA and then decreased, whereas no change in the level of L-actin mRNA was observed (Fig. 7A). Nuclear run-on assay was also performed on intact nuclei isolated from this brain region 2.5 h after injection of KA. As shown in Fig. 7B, the BC1 RNA gene was transcriptionally activated (2.5-fold). Using the gel mobility shift assay, we also investigated the e¡ect of KA treatment on the DNA-binding activities of E2 site-binding protein and BCRE-binding protein (Fig. 7C). The activities of both proteins increased within 0.5 h. The induction level of BCRE-binding protein activity was 4.5-fold and then
decreased gradually. Although the induction level of E2 site-binding protein activity (1.7-fold) was lower than that of BCRE-binding protein, the level of the former seemed to be maintained for a longer period than the latter. Fig. 7C also includes the electrophoretic pattern of the nuclear proteins used for analysis (bottom) to con¢rm that equal amounts of proteins were used for each experimental point. Additionally, no increases in BC1 RNA levels and DNA-binding activities of either protein were observed in control animals injected with physiological saline alone (data not shown). Taking the above results together, it is suggested that the level of BC1 RNA following KA treatment may be under the control of E2 site-binding protein and BCRE-binding protein. 4. Discussion Neuronal cell-speci¢c BC1 RNA is the only known example of a Pol III transcript that is expressed in an activity-dependent manner [21]. Unlike other Pol III tran-
BBAEXP 93442 25-8-00
148
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
tical nucleotides was present between these two elements, competition analysis revealed that BCRE-binding protein did not recognize NRSE (Fig. 5). Furthermore, Shoenherr et al. found that sequences with six or more deviations from the NRSE consensus (21 bp) were not recognized by NRSF [31], suggesting that BCRE is not a target sequence of NRSF. These ¢ndings indicate that the DNAbinding speci¢cities of BCRE-binding protein and NRSF di¡er. Moreover, NRSF was much less in amount than
Fig. 6. Tissue distribution of BC1 RNA, BCRE-binding protein and E2 site-binding protein. (A) Northern blot analysis of BC1 RNA (lanes 1^ 3, brain, liver and kidney, respectively). Positions of 28S, 18S and 4S RNAs are indicated. (B) The gel mobility shift assay was performed using brain (lanes 1 and 2), liver (lanes 3 and 4) or kidney (lanes 5 and 6) nuclear extract and the probe BCRE-I (lanes 1, 3 and 5) or DNA fragment b in Fig. 3A (lanes 2, 4 and 6). (C) The gel mobility shift assay was performed using brain (lane 1), liver (lane 2) or kidney (lane 3) nuclear extract and the double-stranded E2 site probe (5P-GAAGCTGATCAACTTGTTAGCAATTGTTGTCTAAGGAATTTACTGT-3P).
scripts, various types of Pol III promoters are essential for BC1 RNA transcription, which include intragenic A- and B-boxes and GCAAG/CTTGC motifs [22,23]. In addition to these promoter sequences, we previously demonstrated that BC1 RNA expression is also mediated by an E-box sequence (E2 site) which is present in the 5P-£anking region of the BC1 RNA gene and bound by a brain-speci¢c activator protein [24]. In this study, we have shown that the BC1 RNA gene also contains a repressor element (BCRE) in the 5P-£anking sequence and that BC1 RNA expression by Pol III is negatively regulated by a BCREbinding protein (Figs. 1 and 5). Furthermore, as this negative regulatory activity of BCRE was observed independently of the distance from the transcription initiation site of the BC1 RNA gene (Fig. 2), it is indicated that BCRE is a repressor/silencer element in the Pol III system. BCRE was found to have limited homology (52%) with NRSE that represses expression of neuronal cell-speci¢c genes by Pol II (Fig. 3). Although a short stretch of iden-
Fig. 7. E¡ects of KA-induced seizure on BC1 RNA expression and DNA-binding activities of E2 site-binding protein and BCRE-binding protein. (A) Northern blot analysis of BC1 RNA and L-actin mRNA at the indicated time after injection of KA. (B) Nuclear run-on assay of BC1 RNA. Nuclei were isolated from a region consisting of the cerebral cortex and hippocampus 2.5 h after injection of KA or physiological saline (CR). The 32 P-labeled transcripts were hybridized with the cDNA of BC1 RNA or with the pUC 19 vector. (C) The gel mobility shift assay was performed using the E2 site probe (Fig. 6C) or BCRE-I probe. The nuclear extracts were prepared from the same region as that described above (B) at the indicated time after the injection of KA. Bottom: nuclear proteins were analyzed for their integrity and quantity after electrophoresis on 10% SDS^polyacrylamide gel stained with Coomassie brilliant blue.
BBAEXP 93442 25-8-00
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
BCRE-binding protein in the nuclear extract used for in vitro transcription experiments (Fig. 5B,D). Taken together, we conclude that these two proteins are di¡erent from each other. The BC1 RNA gene is a very unique Pol III gene in that it has both an enhancer element E-box (E2 site) and a repressor element BCRE. We found a protein that was similar to BCRE-binding protein in the nuclear extracts from liver and kidney in which a transcriptional activator E2 site-binding protein was not detected (Fig. 6), suggesting that BC1 RNA transcription in non-neuronal tissues is suppressed by the absence of E2 site-binding protein and the presence of BCRE-binding protein. Thus, these observations in turn suggested that the brain speci¢city of BC1 RNA expression is ensured by both E2 site-binding protein and BCRE-binding protein. A similar example was reported in the Pol II system, where cholinergic neuronal cell-speci¢c ChAT gene expression was directed by both a cholinergic-speci¢c enhancer sequence and NRSE [38]. BCRE-binding protein was present even in the brain, where BC1 RNA is transcribed. Similar contradictory observations have also been made in the Pol II system; the presence of NRSF in the brain has been shown, and it was demonstrated that NRSE within the BDNF gene was involved in the control of the KA-induced seizure activitydependent activation of BDNF promoter in hippocampal neurons [30]. These observations suggest that BCRE-binding protein may also control the level of BC1 RNA as a negative regulator. BC1 RNA levels elevated transiently in the brain of animals treated with KA (Fig. 7A). Furthermore, di¡erences between the E2 site-binding protein and BCRE-binding protein levels in their ratio of induction and the time taken to decrease to the basal levels were also observed (Fig. 7B,C). Therefore, it is intriguing to speculate that a balance of these counteracting transcriptional regulators may be important for control of activitydependent expression of BC1 RNA which has been hypothesized to be a molecular sca¡old for the assembly of transport particle consisting of several translational regulators in dendrites [15^17]. The induction of DNA-binding activities of these two proteins was rapid (within 0.5 h after injection of KA) and transient, similar to that of the immediate early genes (IEGs) observed in Pol II system [39]. Consequently, it is also an interesting possibility that E2 site-binding protein and BCRE-binding protein may be classi¢ed as members of IEGs in Pol III system. Finally, although BC1 RNA is transcribed by Pol III, the promoter of the gene seems to be a fusion type intermediate between the Pol II and Pol III systems [22^24]. Therefore, it would be interesting to examine whether E2 site-binding protein and BCRE-binding protein are specific for the BC1 RNA gene or are also involved in the Pol II system, participating in activity-dependent transcriptional regulation of dendritic mRNAs.
149
References [1] J.G. Sutcli¡e, R.J. Milner, F.E. Bloom, R.A. Lerner, Common 82nucleotide sequence unique to brain RNA, Proc. Natl. Acad. Sci. USA 79 (1982) 4942^4946. [2] K. Anzai, S. Kobayashi, N. Kitamura, Y. Kanai, H. Nakajima, Y. Suehiro, S. Goto, Isolation and sequencing of a genomic clone for mouse brain speci¢c small RNA, J. Neurochem. 47 (1986) 673^677. [3] R.D. McKinnon, P. Danielson, M.A.D. Brow, F.E. Bloom, J.G. Sutcli¡e, Expression of small cytoplasmic transcripts of the rat identi¢er element in vivo and in cultured cells, Mol. Cell. Biol. 7 (1987) 2148^2154. [4] S. Kobayashi, S. Goto, K. Anzai, Brain-speci¢c small RNA transcript of the identi¢er sequence is present as a 10S ribonucleoprotein particle, J. Biol. Chem. 266 (1991) 4726^4730. [5] H. Tiedge, R.T. Fremeau Jr., P.H. Weinstock, O. Arancio, J. Brosius, Dendritic location of neural BC1 RNA, Proc. Natl. Acad. Sci. USA 88 (1991) 2093^2097. [6] S. Kobayashi, N. Higashi, K. Suzuki, S. Goto, K. Yumoto, K. Anzai, The 10S BC-1 ribonucleoprotein particle contains identi¢er sequence-binding proteins that interact with an array of GCAAG/ CTTGC motifs between split promoter sequences for RNA polymerase III, J. Biol. Chem. 267 (1992) 18291^18297. [7] I.A. Muslimov, E. Santi, P. Homel, S. Perini, D. Higgins, H. Tiedge, RNA transport in dendrites : A cis-acting targeting element is contained within neuronal BC1 RNA, J. Neurosci. 17 (1997) 4722^4733. [8] M.E. Chicurel, D.M. Terrian, H. Potter, mRNA at the synapse: Analysis of a synaptosomal preparation enriched in hippocampal dendritic spines, J. Neurosci. 13 (1993) 4054^4063. [9] O. Steward, C.S. Wallace, G.L. Lyford, P.F. Worley, Synaptic activation causes the mRNA for the IEG Arc to localize selectively near activated postsynaptic sites on dendrites, Neuron 21 (1998) 741^ 751. [10] H. Kang, E.M. Schuman, A requirement for local protein synthesis in neurotrophin-induced hippocampal synaptic plasticity, Science 273 (1996) 1402^1406. [11] Y.K. Kwon, N.B. Hecht, Binding of a phosphoprotein to the 3P untranslated region of the mouse protamin 2 mRNA temporally represses its translation, Mol. Cell. Biol. 13 (1993) 6547^6557. [12] J.R. Han, G.K. Yiu, N.B. Hecht, Testis/brain RNA-binding protein attaches translationally repressed and transported mRNAs to microtubules, Proc. Natl. Acad. Sci. USA 92 (1995) 9550^9554. [13] K. Aoki, K. Suzuki, T. Sugano, T. Tasaka, K. Nakahara, O. Kuge, A. Omori, M. Kasai, A novel gene, translin, encodes a recombination hotspot binding protein associated with chromosomal translocations, Nat. Genet. 10 (1995) 167^174. [14] X.-Q. Wu, W. Gu, X.H. Meng, N.B. Hecht, The RNA-binding protein, TB-RBP, is the mouse homologue of translin, a recombination protein associated with chromosomal translocations, Proc. Natl. Acad. Sci. USA 94 (1997) 5640^5645. [15] T. Muramatsu, A. Ohmae, K. Anzai, BC1 RNA protein particles in mouse brain contain two y-, h-element-binding proteins, translin and a 37 kDa protein, Biochem. Biophys. Res. Commun. 247 (1998) 7^11. [16] S. Kobayashi, A. Takashima, K. Anzai, The dendritic translocation of translin protein in the form of BC1 RNA particles in developing rat hippocampal neurons in primary culture, Biochem. Biophys. Res. Commun. 253 (1998) 448^453. [17] J. Kremerskothen, M. Nettermann, A.O.D. Bekke, M. Bachmann, J. Brosius, Identi¢cation of human autoantigen La/SS-B as BC1/BC200 RNA-binding protein, DNA Cell Biol. 17 (1998) 751^759. [18] J.G. Sutcli¡e, R.J. Milner, J.M. Gottesfeld, W. Reynolds, Control of neuronal gene expression, Science 225 (1984) 1308^1315. [19] K. Anzai, S. Goto, Brain-speci¢c small RNA during development and ageing of mice, Mech. Aging Dev. 39 (1987) 129^135.
BBAEXP 93442 25-8-00
150
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
[20] J. Brosius and H. Tiedge, Neural BC1 RNA : dendritic localization and transport, in: H.D. Lipshitz (Ed.), Localized RNAs, Molecular Biology Intelligence Unit, R.G. Landes, Austin, TX, 1995, pp. 289^ 300. [21] I.A. Muslimov, G. Banker, J. Brosius, H. Tiedge, Activity-dependent regulation of dendritic BC1 RNA in hippocampal neurons in culture, J. Cell Biol. 141 (1998) 1601^1611. [22] J.A. Martignetti, J. Brosius, BC1 RNA : Transcriptional analysis of a neural cell-speci¢c RNA polymerase III transcript, Mol. Cell. Biol. 15 (1995) 1642^1650. [23] S. Kobayashi, K. Anzai, Mutational analysis reveals that an array of GCAAG/CTTGC motifs between sprit promoter sequences for RNA polymerase III is essential for neural BC1 RNA transcription, Biochem. Biophys. Res. Commun. 239 (1997) 407^411. [24] S. Kobayashi, K. Anzai, An E-box sequence acts as a transcriptional activator for BC1 RNA expression by RNA polymerase III in the brain, Biochem. Biophys. Res. Commun. 245 (1998) 59^63. [25] J.E. Lee, Basic helix-loop-helix genes in neural development, Curr. Opin. Neurobiol. 7 (1997) 13^20. [26] W.R. Kinney, R.K. McNamara, E. Valcourt, A. Routtenberg, Prolonged alteration in E-box binding after a single systemic kainate injection: potential relation to F1/GAP-43 gene expression, Mol. Brain Res. 38 (1996) 25^36. [27] C.J. Schoenherr, D.J. Anderson, The neuron-restrictive silencer factor (NRSF): A coordinate repressor of multiple neuron-speci¢c genes, Science 267 (1995) 1360^1363. [28] J.A. Chong, J. Tapia-Ramirez, S. Kim, J.J. Toledo-Aral, Y. Zheng, M.C. Boutros, Y.M. Altshuller, M.A. Frohman, S.D. Kraner, G. Mandel, REST: A mammalian silencer protein that restricts sodium channel gene expression to neurons, Cell 80 (1995) 949^957. [29] K. Palm, N. Belluardo, M. Metsis, T. Timmusk, Neuronal expression
[30]
[31]
[32]
[33] [34]
[35]
[36] [37]
[38]
[39]
of zinc ¢nger transcription factor REST/NRSF/XBR gene, J. Neurosci. 18 (1998) 1280^1296. T. Timmusk, K. Palm, U. Lendarhl, M. Metsis, Brain-derived neurotrophic factor expression in vivo is under the control of neuron-restrictive silencer element, J. Biol. Chem. 274 (1999) 1078^1084. C.J. Schoenherr, A.J. Paquette, D.J. Anderson, Identi¢cation of potential target genes for the neuron-restrictive silencer factor, Proc. Natl. Acad. Sci. USA 93 (1996) 9881^9886. J. Shambrook, E.F. Fritsh and T. Maniatis, Molecular Cloning: A Laboratory Manual, 2nd edn., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1989. K. Gorski, M. Carneiro, U. Schibler, Tissue-speci¢c in vitro transcription from the mouse albumin promoter, Cell 47 (1986) 767^776. P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by acid guanidium thiocyanate-phenol chloroform extraction, Anal. Biochem. 162 (1987) 156^159. R.I. Scheinman, V.J. Auld, A.L. Goldin, N. Davidson, R.J. Dunn, W.A. Catterall, Developmental regulation of sodium channel expression in the rat forebrain, J. Biol. Chem. 264 (1989) 10660^10666. M.E. Greenberg, E.B. Zi¡, Stimulation of 3T3 cells induces transcription of the c-fos proto-oncogene, Nature 311 (1984) 433^438. N. Mori, C. Schoenherr, D.J. Vandenbergh, D.J. Anderson, A common silencer element in the SCG10 and type II Na channel genes binds a factor present in nonneuronal cells but not in neuronal cells, Neuron 9 (1992) 45^54. P. Lonnerberg, C.J. Schoenherr, D.J. Anderson, C.F. Ibanez, Cell type-speci¢c regulation of choline acetyltransferase gene expression, J. Biol. Chem. 271 (1996) 33358^33365. M. Sheng, M.E. Greenberg, The regulation and function of c-fos and other immediate early genes in the nervous system, Neuron 4 (1990) 477^485.
BBAEXP 93442 25-8-00
www.elsevier.com/locate/bba
Identi¢cation of a negative regulatory DNA element for neuronal BC1 RNA expression by RNA polymerase III Shunsuke Kobayashi, Sakurako Kamo, Akiko Ohmae, Keisuke Agui, Yanmei Li, Kaijiro Anzai * Department of Biochemistry, College of Pharmacy, Nihon University, Narashinodai, Funabashi, Chiba 274-8555, Japan Received 1 May 2000; received in revised form 21 June 2000; accepted 6 July 2000
Abstract BC1 RNA is a neuronal cell-specific RNA polymerase III (Pol III) transcript. The BC1 RNA gene has plural types of Pol III promoters, in addition to which an E-box sequence (E2 site) acts as a transcriptional activator, which is recognized by a brain-specific protein(s). Using an in vitro transcription system, we found that the upstream region of the BC1 RNA gene contained a sequence that interfered with the activity of the E-box element in a distance-independent manner. A tandem repeat within this sequence, which was weakly homologous with the neuron-restrictive silencer element (NRSE) found in the Pol II system, was recognized by a brain nuclear protein. Consistently, the transcriptional activity increased by deleting the tandem repeat sequence. We called this BC1 RNA-repressing element BCRE. The DNAbinding specificities of BCRE-binding protein differed from that of NRSE-binding protein (NRSF). A similar protein with an ability to bind to BCRE was also found in liver and kidney. Furthermore, the glutamate analog kainic acid increased the DNA-binding of both E2 sitebinding protein and BCRE-binding protein, and then the levels of BC1 RNA also increased transiently. Our results suggested that both positive and negative regulatory elements contribute to neuronal BC1 RNA expression. ß 2000 Elsevier Science B.V. All rights reserved. Keywords : Neuronal BC1 RNA ; Promoter; In vitro transcription system; Negative regulatory element
1. Introduction BC1 RNA is expressed preferentially in the brain by RNA polymerase III (Pol III) [1^3]. This small non-translatable RNA is distributed in neuronal dendrites as ribonucleoprotein particles, BC1 RNPs [4^7], and concentrated in synaptodendrosomes with other dendritic mRNAs [8]. It has been reported that local protein synthesis at the synapse is required for synaptic plasticity [9,10]. In a previous study, we demonstrated that translin/testis-brain RNA-binding protein, which is known to be a translational repressor [11^14], bound to BC1 RNA [15] and was distributed in the neuronal dendrites [16]. Recently, it was reported that autoantigen La/SS-B associated with BC1 RNA and suggested that La protein links
* Corresponding author. Fax: +81-47-465-5332; E-mail : [email protected]
BC1 RNP to the ribosome [17]. These observations suggest that BC1 RNP plays a role(s) in translational regulation of dendritic mRNAs. Among the many Pol III-transcribed genes (class III genes), neuronal cell speci¢city is unique to the BC1 RNA gene. The levels of this small non-translatable RNA increase during the period of synaptogenesis [18,19]. Furthermore, the somato-dendritic levels of this RNA in hippocampal neurons increased when hyperexcitability was induced with picrotoxin [20] and inhibition of the electrical activity of developing hippocampal neurons by tetrodotoxin resulted in a reduction in the level of this RNA [21]. These observations indicate that BC1 RNA transcription is regulated by neuronal activity. We are interested in determining how the unique expression pro¢le of this class III gene in the brain is achieved. Studies using an in vitro transcription system revealed that both intragenic and extragenic cis-elements are necessary for BC1 RNA transcription. The intragenic promoter elements, A- and B-boxes [22] and GCAAG/CTTGC motifs present between these two boxes, were found to be essential [23] and a 5P-£anking region containing a TATA-like sequence and a proximal sequence element
0167-4781 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 0 0 ) 0 0 1 7 5 - 5
BBAEXP 93442 25-8-00
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
(PSE) were also necessary [22]. Moreover, we demonstrated that an E-box sequence at 3248 (E2 site; CAATTG), recognized by a brain-speci¢c protein(s) (E2 site-binding protein), acted as a transcriptional activator of BC1 RNA expression by Pol III [24]. In the Pol II system, several E-box-binding basic helixloop-helix transcription factors have been identi¢ed in developing and adult neuronal tissues, and they have been suggested to play important roles in neuronal development and activity [25,26]. Recently, a neuron-restrictive silencer element (NRSE), which is recognized by a repressor/silencer factor (NRSF; also known as REST), was also identi¢ed and found to repress expression of neuronal cell-speci¢c genes in non-neuronal cells [27,28]. However, Palm et al. reported activity-dependent increases in NRSF mRNA levels in neurons and suggested that NRSF counteracts the e¡ects of positive regulators, thereby modulating target gene expression quantitatively [29]. Timmusk et al. showed that in the neurons of adult brains, kainic acid (KA)-induced expression of brain-derived neurotrophic factor (BDNF) mRNA is under the control of NRSE [30]. These observations suggest that the balance between positive and negative regulators is important for control of neuron-speci¢c gene transcription by Pol II. In this context, there is an interesting possibility that similar mechanisms may also operate the BC1 RNA expression by Pol III. In this study, we identi¢ed a tandem repeat (referred to as BC1 RNA-repressing element: BCRE), which showed weak homology with the consensus sequence of NRSE
143
[31], in the 5P-£anking region (3638V3685) of the BC1 RNA gene and found that this element interfered with the transactivation of BC1 RNA expression by E2 site. We also characterized the BCRE-binding protein and demonstrated that the DNA-binding speci¢city of the protein di¡ered from that of the NRSF. We further showed that the DNA-binding of E2 site-binding protein and BCREbinding protein and the level of BC1 RNA expression increased transiently in the brain of animals treated with KA. Our data suggest that a positive and a negative control element cooperate to regulate neuronal BC1 RNA transcription. 2. Materials and methods 2.1. DNA cloning The BC1 RNA gene was isolated from a rat genomic library using the 3P-portion of BC1 RNA as a probe. The nucleotide sequence downstream of position 3630 has been reported elsewhere [22,24]. A genomic clone, termed clone K0, containing a further region upstream of nucleotide 3740 was cloned into EcoRI and BamHI sites of the pBR 322 vector. 2.2. Plasmid construction The 5P-truncated constructs were prepared by performing polymerase chain reactions (PCRs) using the cloned
Fig. 1. Schematic representation of 5P-£anking and coding regions of the BC1 RNA gene and in vitro transcriptional analysis of 5P-deletion constructs. (A) The E2 site, the PSE, TATA-like box, A- and B-boxes and GCAAG/CTTGC motifs are shown. The arrows (K0^K4) indicate the 5P-end-points of the deletion constructs (clones K0^K4). (B) The transcriptional reactions were performed using clones K0^K4 (lanes 1^5, respectively). The arrow indicates the BC1 RNA transcript. As size markers, positions of 5.8S, 5S and 4S RNAs are also indicated.
BBAEXP 93442 25-8-00
144
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
of 3-week-old inbred rats (Wistar) or mice (ICR), essentially as described by Gorski et al. [33]. The transcription reaction was performed in a mixture (20 Wl) containing 40 Wg/ml template DNA and 3 mg/ml nuclear protein, as described previously [23]. 2.4. Electrophoretic gel mobility shift assay The sequences of the double-stranded oligonucleotide probes and competitors are shown in Figs. 3 and 5. Each 32 P-labeled probe (1U104 cpm) was incubated at room temperature for 30 min in a mixture (26 Wl) containing 10 mM Tris^HCl, pH 7.5, 50 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 5% glycerol, 1 Wg poly (dI-dC) and 3 Wg nuclear protein extract. The reaction product was subjected to electrophoresis on a 4% polyacrylamide gel containing 6.7 mM Tris^HCl, pH 7.5, 3.3 mM sodium acetate and 1 mM EDTA with electrode bu¡er recirculation. 2.5. Northern blot analysis Fig. 2. In vitro transcriptional analysis of fusion constructs. Schematic representations of the constructs are shown. The open box indicates fragment K, and clones KK2 and KK3 are fusion constructs of fragment K to clones K2 and K3, respectively. The transcription reactions were performed using clones K1, K2 and K3 (lanes 1, 3 and 5, respectively), clones K0, KK2 and KK3 (lanes 2, 4 and 6, respectively) and the pBR 322 vector (lane 7). Each BC1 RNA transcript is indicated by an arrow. As size markers, positions of 5.8S, 5S and 4S RNAs are also indicated.
BC1 RNA gene as a template, 20 nucleotides of the 5P-end primer of each site (K1^K4 in Fig. 1), which had been added to the EcoRI site, and the 3P-end primer of the BC1 RNA gene (CCTCGGGATCCTGTCATTTTGTGTAGCAATCG), which was added to the BamHI site. Each PCR product was cloned into the pBR 322 vector to produce clones K1^K4. In order to prepare fusion constructs, fragment K, which had been added to the AatII site at the 5P-end and the EcoRI site at the 3P-end, was synthesized by the PCR and ligated to the AatII and EcoRI double-digested clones K2 and K3 to produce clones KK2 and KK3, respectively. In order to produce the BCRE-deleted construct (clone dK) using fragment K as a template, a fragment consisting of a pyrimidinerich and a palindromic domain, which were added to the AatII site at the 5P-end and the EcoRI site at the 3P-end, respectively, was synthesized by the PCR and ligated to the AatII and EcoRI double-digested clone K1. All the DNAs were puri¢ed by CsCl ultracentrifugation [32]. 2.3. Preparation of nuclear protein extracts and the in vitro transcription reaction Nuclear protein extracts were prepared from the tissues
Total RNA was isolated from tissues by homogenization in guanidine isothiocyanate, followed by phenol extraction and ethanol precipitation [34]. The RNA (10 Wg) was electrophoresed in 1.5% agarose gels containing 2.2 M formaldehyde and then transferred to nitrocellulose membrane. Hybridization was performed overnight at 42³C with 32 P-labeled probes in hybridization bu¡er (5USSPE, 5UDenhardt's reagent, 50% formamide, 1% sodium dodecyl sulfate (SDS), 0.1 mg/ml salmon sperm DNA). The ¢lters were washed in 2USSC^0.1% SDS at room temperature for 30 min, followed by two 10 min washes at 52³C in 0.2USSC^0.1% SDS. 2.6. KA seizure induction Inbred male mice were injected with KA (15 Wg/kg) or physiological saline into the lateral ventricle. The animals were killed at the indicated time after injection, and then nuclear extracts or total RNAs were prepared from a region consisting of the cerebral cortex and hippocampus. 2.7. Nuclear run-on assay Nuclei were isolated from dissected cerebral cortex and hippocampal tissues of KA-treated mice by centrifugation through a dense sucrose solution [33,35]. The run-on transcription assay was performed essentially as described by Greenberg and Zi¡ [36]. The 32 P-labeled transcripts were hybridized to nitrocellulose membranes containing 1 Wg dots of denatured cDNA of BC1 RNA which was cloned into the pUC 19 vector. The vector plasmid pUC 19 was used as a control. Hybridization and washing of the ¢lters were carried out in the same way as for Northern blot analysis.
BBAEXP 93442 25-8-00
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
145
Fig. 3. Electrophoretic gel mobility shift assay of subfragments of fragment K. (A) Upper : nucleotide sequence of fragment K. The oligonucleotide probes for the gel mobility shift assay are indicated above the sequence. The pyrimidine-rich, palindromic and tandem repeat domains are indicated below the sequence. Tandem repeats I and II are boxed. Lower: comparison of tandem repeats I and II with the consensus sequence of NRSE. The boxes indicate the identical nucleotide residues in the tandem repeat and NRSE consensus sequence. (B) The gel mobility shift assay was performed using 3 Wg brain nuclear extract and 32 P-labeled subfragments of fragment K as probes. The probes (1U104 cpm) used were a, b, c, I+II, I and II (lanes 1^6, respectively).
3. Results 3.1. The upstream region of the BC1 RNA gene contains a negative regulatory sequence Fig. 1A shows the 5P-upstream and intragenic promoters of the BC1 RNA gene. The internal A- and B-boxes, GCAAG/CTTGC motifs, external TATA box and PSE are necessary for BC1 RNA transcription by Pol III [22,23]. In a previous study, we used an in vitro transcription system, an electrophoretic gel mobility shift assay and a DNase I protection assay and found that the E-box at 3248 (E2 site: CAATTG) was recognized by a
brain-speci¢c E2 site-binding protein and activated BC1 RNA transcription [24]. Transcriptional analysis of 5P-truncated constructs of the BC1 RNA gene revealed that clone K3 containing the E2 site was transcribed more e¤ciently than clone K4, which lacks the E2 site (Fig. 1B, lanes 4 and 5; see Fig. 1A for clones K3 and K4). Clones K1 and K2 were transcribed as e¤ciently as clone K3, indicating that the region between 3324 and 3630 does not contain sequences that are signi¢cant for BC1 RNA transcription (Fig. 1B, lanes 2^4). However, the presence of a further upstream region between 3631 and 3740 (clone K0) reduced the transcriptional activity (Fig. 1B, lane 1). We termed this negative regulatory region fragment K.
BBAEXP 93442 25-8-00
146
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
formed transcriptional analysis. As shown in Fig. 4, clone dK was transcribed more e¤ciently than clone K0. Therefore, it is indicated that this sequence is the BCRE consisting of a tandem repeat (BCRE-I and -II) and the BCRE-binding protein may be a negative regulator for BC1 RNA expression. 3.3. DNA-binding speci¢city of the BCRE-binding protein
Fig. 4. In vitro transcriptional analysis of the BCRE deletion construct. The hatched box indicates BCRE. The transcription reactions were performed using clones K0 (lane 1) and dK (lane 2). The BC1 RNA transcript is indicated by an arrow. Positions of 5.8S, 5S and 4S RNAs are also indicated.
In order to verify the repressor activity, we constructed templates in which fragment K was fused to regions proximal to the E2 site (clones KK2 and KK3 in Fig. 2) and performed transcriptional analysis. These fusion constructs showed reduced transcriptional activity similar to that of clone K0. These results reveal that fragment K contains a sequence that interferes with the positive e¡ect of the E2 site on BC1 RNA transcription independently of the distance from the E-box sequence. 3.2. Fragment K contains a BCRE consisting of a tandem repeat Fig. 3A shows the nucleotide sequence of fragment K, which can be subdivided into 5P-pyrimidine-rich domain, central palindromic domain and 3P-tandem repeat domain. The tandem repeat sequences I and II had limited homology (52%) with the NRSE consensus sequence [31]. In order to analyze the DNA-binding proteins, we synthesized 32 P-labeled double-stranded oligonucleotide probes corresponding to the subdomains of fragment K and performed the electrophoretic gel mobility shift assay. As shown in Fig. 3B, the tandem repeat-containing probes were recognized by a brain nuclear protein (lanes 3 and 4). This binding protein appeared to recognize either the repeat sequence I or II (Fig. 3B, lanes 5 and 6), as the mobility of each complex was identical to that of the complex consisting of the protein and the probe, containing both the repeat sequences (compare lane 4 with lanes 5 and 6). Next, we constructed a template, clone dK, which lacks the tandem repeat region of fragment K only and per-
Although BCRE-I and -II had a short stretch of nucleotides identical with those in NRSE (Fig. 3A), their overall nucleotide sequences deviated considerably from the target sequences of NRSF [31]. In order to examine the binding speci¢city of BCRE-binding protein, we performed the electrophoretic gel mobility shift assay using oligonucleotides containing the NRSEs of the SCG10 [37] and choline acetyltransferase (ChAT) [38] genes as probes and competitors (Fig. 5A). When the 32 P-labeled NRSEs of SCG10 and ChAT were used as probes, no or only weak shifts were observed under the assay conditions we employed (open arrowhead in Fig. 5B). This may be due to the low levels of NRSF in the brain nuclear extract. Furthermore, competition analysis revealed that these oligonucleotides were at least 10 times less potent than the unlabeled BCRE-I (Fig. 5C). Above observations indicate that BCRE-binding protein and NRSF are di¡erent and that BCRE-binding protein recognizes the BCRE in a sequence-speci¢c manner. Additionally, we again con¢rmed that the nuclear proteins used in these assays shown in Fig. 5B,C inhibited transcription of BCRE-containing template only (compare clone K0 with clone K1, Fig. 5D; see also Figs. 1 and 2). 3.4. Tissue distribution of BCRE-binding protein As BC1 RNA is brain-speci¢c (Fig. 6A), we next examined the tissue distribution of BCRE-binding protein. Nuclear extracts from brain and non-neuronal tissues (liver and kidney) were prepared, and an electrophoretic gel mobility shift assay was performed using BCRE-I and DNA fragment b (see Fig. 3A) as a speci¢c probe and a control probe, respectively. As shown in Fig. 6B, a protein which recognized BCRE-I, but not DNA fragment b, was also found in the nuclear extracts from liver and kidney. The mobility of each complex formed was the same as that of brain. Therefore, it is indicated that a similar nuclear protein, which binds speci¢cally to BCRE-I, is present in nonneuronal tissues. On the other hand, a transcriptional activator E2 site-binding protein, which forms one major (big arrowhead) and two minor (small arrowhead) complexes, as observed previously [24], was detected only in the brain nuclear extract (Fig. 6C). These ¢ndings suggest that tissue-speci¢c expression of BC1 RNA is controlled at the transcriptional level.
BBAEXP 93442 25-8-00
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
147
Fig. 5. Analysis of binding speci¢city of BCRE-binding protein. (A) Alignment of the oligonucleotides used as probes and competitors. BCRE-I and NRSEs of SCG10 and ChAT were underlined. (B) The gel mobility shift assay was performed using 3 Wg brain nuclear extract and 32 P-labeled (1U104 cpm) BCRE-I (lane 1), SCG10 (lane 2) and ChAT (lane 3) double-stranded oligonucleotides. (C) Competition analysis of BCRE-binding protein and 32 P-labeled BCRE-I complex formation was performed in the presence of a molar excess of unlabeled BCRE-I (lanes 1^3), SCG10 (lanes 4 and 5) or ChAT (lanes 6 and 7) double-stranded oligonucleotide. (D) In vitro transcriptional analysis was performed using the nuclear protein fraction prepared for this gel mobility shift assay (lanes 1 and 2, clones K0 and K1 in Fig. 1, respectively). The BC1 RNA transcript is indicated by an arrow. Positions of 5.8S, 5S and 4S RNAs are also indicated.
3.5. E¡ects of KA-induced seizure on BC1 RNA expression and DNA-binding activities of E2 site-binding protein and BCRE-binding protein In order to examine the neuronal activity-dependent alteration of BC1 RNA expression, we studied the e¡ect of KA-induced seizure in the mouse brain. Northern blot analysis revealed that the level of BC1 RNA in the region consisting of the cerebral cortex and hippocampus tissues increased gradually to a maximum level (3-fold) 3 h after injection of KA and then decreased, whereas no change in the level of L-actin mRNA was observed (Fig. 7A). Nuclear run-on assay was also performed on intact nuclei isolated from this brain region 2.5 h after injection of KA. As shown in Fig. 7B, the BC1 RNA gene was transcriptionally activated (2.5-fold). Using the gel mobility shift assay, we also investigated the e¡ect of KA treatment on the DNA-binding activities of E2 site-binding protein and BCRE-binding protein (Fig. 7C). The activities of both proteins increased within 0.5 h. The induction level of BCRE-binding protein activity was 4.5-fold and then
decreased gradually. Although the induction level of E2 site-binding protein activity (1.7-fold) was lower than that of BCRE-binding protein, the level of the former seemed to be maintained for a longer period than the latter. Fig. 7C also includes the electrophoretic pattern of the nuclear proteins used for analysis (bottom) to con¢rm that equal amounts of proteins were used for each experimental point. Additionally, no increases in BC1 RNA levels and DNA-binding activities of either protein were observed in control animals injected with physiological saline alone (data not shown). Taking the above results together, it is suggested that the level of BC1 RNA following KA treatment may be under the control of E2 site-binding protein and BCRE-binding protein. 4. Discussion Neuronal cell-speci¢c BC1 RNA is the only known example of a Pol III transcript that is expressed in an activity-dependent manner [21]. Unlike other Pol III tran-
BBAEXP 93442 25-8-00
148
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
tical nucleotides was present between these two elements, competition analysis revealed that BCRE-binding protein did not recognize NRSE (Fig. 5). Furthermore, Shoenherr et al. found that sequences with six or more deviations from the NRSE consensus (21 bp) were not recognized by NRSF [31], suggesting that BCRE is not a target sequence of NRSF. These ¢ndings indicate that the DNAbinding speci¢cities of BCRE-binding protein and NRSF di¡er. Moreover, NRSF was much less in amount than
Fig. 6. Tissue distribution of BC1 RNA, BCRE-binding protein and E2 site-binding protein. (A) Northern blot analysis of BC1 RNA (lanes 1^ 3, brain, liver and kidney, respectively). Positions of 28S, 18S and 4S RNAs are indicated. (B) The gel mobility shift assay was performed using brain (lanes 1 and 2), liver (lanes 3 and 4) or kidney (lanes 5 and 6) nuclear extract and the probe BCRE-I (lanes 1, 3 and 5) or DNA fragment b in Fig. 3A (lanes 2, 4 and 6). (C) The gel mobility shift assay was performed using brain (lane 1), liver (lane 2) or kidney (lane 3) nuclear extract and the double-stranded E2 site probe (5P-GAAGCTGATCAACTTGTTAGCAATTGTTGTCTAAGGAATTTACTGT-3P).
scripts, various types of Pol III promoters are essential for BC1 RNA transcription, which include intragenic A- and B-boxes and GCAAG/CTTGC motifs [22,23]. In addition to these promoter sequences, we previously demonstrated that BC1 RNA expression is also mediated by an E-box sequence (E2 site) which is present in the 5P-£anking region of the BC1 RNA gene and bound by a brain-speci¢c activator protein [24]. In this study, we have shown that the BC1 RNA gene also contains a repressor element (BCRE) in the 5P-£anking sequence and that BC1 RNA expression by Pol III is negatively regulated by a BCREbinding protein (Figs. 1 and 5). Furthermore, as this negative regulatory activity of BCRE was observed independently of the distance from the transcription initiation site of the BC1 RNA gene (Fig. 2), it is indicated that BCRE is a repressor/silencer element in the Pol III system. BCRE was found to have limited homology (52%) with NRSE that represses expression of neuronal cell-speci¢c genes by Pol II (Fig. 3). Although a short stretch of iden-
Fig. 7. E¡ects of KA-induced seizure on BC1 RNA expression and DNA-binding activities of E2 site-binding protein and BCRE-binding protein. (A) Northern blot analysis of BC1 RNA and L-actin mRNA at the indicated time after injection of KA. (B) Nuclear run-on assay of BC1 RNA. Nuclei were isolated from a region consisting of the cerebral cortex and hippocampus 2.5 h after injection of KA or physiological saline (CR). The 32 P-labeled transcripts were hybridized with the cDNA of BC1 RNA or with the pUC 19 vector. (C) The gel mobility shift assay was performed using the E2 site probe (Fig. 6C) or BCRE-I probe. The nuclear extracts were prepared from the same region as that described above (B) at the indicated time after the injection of KA. Bottom: nuclear proteins were analyzed for their integrity and quantity after electrophoresis on 10% SDS^polyacrylamide gel stained with Coomassie brilliant blue.
BBAEXP 93442 25-8-00
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
BCRE-binding protein in the nuclear extract used for in vitro transcription experiments (Fig. 5B,D). Taken together, we conclude that these two proteins are di¡erent from each other. The BC1 RNA gene is a very unique Pol III gene in that it has both an enhancer element E-box (E2 site) and a repressor element BCRE. We found a protein that was similar to BCRE-binding protein in the nuclear extracts from liver and kidney in which a transcriptional activator E2 site-binding protein was not detected (Fig. 6), suggesting that BC1 RNA transcription in non-neuronal tissues is suppressed by the absence of E2 site-binding protein and the presence of BCRE-binding protein. Thus, these observations in turn suggested that the brain speci¢city of BC1 RNA expression is ensured by both E2 site-binding protein and BCRE-binding protein. A similar example was reported in the Pol II system, where cholinergic neuronal cell-speci¢c ChAT gene expression was directed by both a cholinergic-speci¢c enhancer sequence and NRSE [38]. BCRE-binding protein was present even in the brain, where BC1 RNA is transcribed. Similar contradictory observations have also been made in the Pol II system; the presence of NRSF in the brain has been shown, and it was demonstrated that NRSE within the BDNF gene was involved in the control of the KA-induced seizure activitydependent activation of BDNF promoter in hippocampal neurons [30]. These observations suggest that BCRE-binding protein may also control the level of BC1 RNA as a negative regulator. BC1 RNA levels elevated transiently in the brain of animals treated with KA (Fig. 7A). Furthermore, di¡erences between the E2 site-binding protein and BCRE-binding protein levels in their ratio of induction and the time taken to decrease to the basal levels were also observed (Fig. 7B,C). Therefore, it is intriguing to speculate that a balance of these counteracting transcriptional regulators may be important for control of activitydependent expression of BC1 RNA which has been hypothesized to be a molecular sca¡old for the assembly of transport particle consisting of several translational regulators in dendrites [15^17]. The induction of DNA-binding activities of these two proteins was rapid (within 0.5 h after injection of KA) and transient, similar to that of the immediate early genes (IEGs) observed in Pol II system [39]. Consequently, it is also an interesting possibility that E2 site-binding protein and BCRE-binding protein may be classi¢ed as members of IEGs in Pol III system. Finally, although BC1 RNA is transcribed by Pol III, the promoter of the gene seems to be a fusion type intermediate between the Pol II and Pol III systems [22^24]. Therefore, it would be interesting to examine whether E2 site-binding protein and BCRE-binding protein are specific for the BC1 RNA gene or are also involved in the Pol II system, participating in activity-dependent transcriptional regulation of dendritic mRNAs.
149
References [1] J.G. Sutcli¡e, R.J. Milner, F.E. Bloom, R.A. Lerner, Common 82nucleotide sequence unique to brain RNA, Proc. Natl. Acad. Sci. USA 79 (1982) 4942^4946. [2] K. Anzai, S. Kobayashi, N. Kitamura, Y. Kanai, H. Nakajima, Y. Suehiro, S. Goto, Isolation and sequencing of a genomic clone for mouse brain speci¢c small RNA, J. Neurochem. 47 (1986) 673^677. [3] R.D. McKinnon, P. Danielson, M.A.D. Brow, F.E. Bloom, J.G. Sutcli¡e, Expression of small cytoplasmic transcripts of the rat identi¢er element in vivo and in cultured cells, Mol. Cell. Biol. 7 (1987) 2148^2154. [4] S. Kobayashi, S. Goto, K. Anzai, Brain-speci¢c small RNA transcript of the identi¢er sequence is present as a 10S ribonucleoprotein particle, J. Biol. Chem. 266 (1991) 4726^4730. [5] H. Tiedge, R.T. Fremeau Jr., P.H. Weinstock, O. Arancio, J. Brosius, Dendritic location of neural BC1 RNA, Proc. Natl. Acad. Sci. USA 88 (1991) 2093^2097. [6] S. Kobayashi, N. Higashi, K. Suzuki, S. Goto, K. Yumoto, K. Anzai, The 10S BC-1 ribonucleoprotein particle contains identi¢er sequence-binding proteins that interact with an array of GCAAG/ CTTGC motifs between split promoter sequences for RNA polymerase III, J. Biol. Chem. 267 (1992) 18291^18297. [7] I.A. Muslimov, E. Santi, P. Homel, S. Perini, D. Higgins, H. Tiedge, RNA transport in dendrites : A cis-acting targeting element is contained within neuronal BC1 RNA, J. Neurosci. 17 (1997) 4722^4733. [8] M.E. Chicurel, D.M. Terrian, H. Potter, mRNA at the synapse: Analysis of a synaptosomal preparation enriched in hippocampal dendritic spines, J. Neurosci. 13 (1993) 4054^4063. [9] O. Steward, C.S. Wallace, G.L. Lyford, P.F. Worley, Synaptic activation causes the mRNA for the IEG Arc to localize selectively near activated postsynaptic sites on dendrites, Neuron 21 (1998) 741^ 751. [10] H. Kang, E.M. Schuman, A requirement for local protein synthesis in neurotrophin-induced hippocampal synaptic plasticity, Science 273 (1996) 1402^1406. [11] Y.K. Kwon, N.B. Hecht, Binding of a phosphoprotein to the 3P untranslated region of the mouse protamin 2 mRNA temporally represses its translation, Mol. Cell. Biol. 13 (1993) 6547^6557. [12] J.R. Han, G.K. Yiu, N.B. Hecht, Testis/brain RNA-binding protein attaches translationally repressed and transported mRNAs to microtubules, Proc. Natl. Acad. Sci. USA 92 (1995) 9550^9554. [13] K. Aoki, K. Suzuki, T. Sugano, T. Tasaka, K. Nakahara, O. Kuge, A. Omori, M. Kasai, A novel gene, translin, encodes a recombination hotspot binding protein associated with chromosomal translocations, Nat. Genet. 10 (1995) 167^174. [14] X.-Q. Wu, W. Gu, X.H. Meng, N.B. Hecht, The RNA-binding protein, TB-RBP, is the mouse homologue of translin, a recombination protein associated with chromosomal translocations, Proc. Natl. Acad. Sci. USA 94 (1997) 5640^5645. [15] T. Muramatsu, A. Ohmae, K. Anzai, BC1 RNA protein particles in mouse brain contain two y-, h-element-binding proteins, translin and a 37 kDa protein, Biochem. Biophys. Res. Commun. 247 (1998) 7^11. [16] S. Kobayashi, A. Takashima, K. Anzai, The dendritic translocation of translin protein in the form of BC1 RNA particles in developing rat hippocampal neurons in primary culture, Biochem. Biophys. Res. Commun. 253 (1998) 448^453. [17] J. Kremerskothen, M. Nettermann, A.O.D. Bekke, M. Bachmann, J. Brosius, Identi¢cation of human autoantigen La/SS-B as BC1/BC200 RNA-binding protein, DNA Cell Biol. 17 (1998) 751^759. [18] J.G. Sutcli¡e, R.J. Milner, J.M. Gottesfeld, W. Reynolds, Control of neuronal gene expression, Science 225 (1984) 1308^1315. [19] K. Anzai, S. Goto, Brain-speci¢c small RNA during development and ageing of mice, Mech. Aging Dev. 39 (1987) 129^135.
BBAEXP 93442 25-8-00
150
S. Kobayashi et al. / Biochimica et Biophysica Acta 1493 (2000) 142^150
[20] J. Brosius and H. Tiedge, Neural BC1 RNA : dendritic localization and transport, in: H.D. Lipshitz (Ed.), Localized RNAs, Molecular Biology Intelligence Unit, R.G. Landes, Austin, TX, 1995, pp. 289^ 300. [21] I.A. Muslimov, G. Banker, J. Brosius, H. Tiedge, Activity-dependent regulation of dendritic BC1 RNA in hippocampal neurons in culture, J. Cell Biol. 141 (1998) 1601^1611. [22] J.A. Martignetti, J. Brosius, BC1 RNA : Transcriptional analysis of a neural cell-speci¢c RNA polymerase III transcript, Mol. Cell. Biol. 15 (1995) 1642^1650. [23] S. Kobayashi, K. Anzai, Mutational analysis reveals that an array of GCAAG/CTTGC motifs between sprit promoter sequences for RNA polymerase III is essential for neural BC1 RNA transcription, Biochem. Biophys. Res. Commun. 239 (1997) 407^411. [24] S. Kobayashi, K. Anzai, An E-box sequence acts as a transcriptional activator for BC1 RNA expression by RNA polymerase III in the brain, Biochem. Biophys. Res. Commun. 245 (1998) 59^63. [25] J.E. Lee, Basic helix-loop-helix genes in neural development, Curr. Opin. Neurobiol. 7 (1997) 13^20. [26] W.R. Kinney, R.K. McNamara, E. Valcourt, A. Routtenberg, Prolonged alteration in E-box binding after a single systemic kainate injection: potential relation to F1/GAP-43 gene expression, Mol. Brain Res. 38 (1996) 25^36. [27] C.J. Schoenherr, D.J. Anderson, The neuron-restrictive silencer factor (NRSF): A coordinate repressor of multiple neuron-speci¢c genes, Science 267 (1995) 1360^1363. [28] J.A. Chong, J. Tapia-Ramirez, S. Kim, J.J. Toledo-Aral, Y. Zheng, M.C. Boutros, Y.M. Altshuller, M.A. Frohman, S.D. Kraner, G. Mandel, REST: A mammalian silencer protein that restricts sodium channel gene expression to neurons, Cell 80 (1995) 949^957. [29] K. Palm, N. Belluardo, M. Metsis, T. Timmusk, Neuronal expression
[30]
[31]
[32]
[33] [34]
[35]
[36] [37]
[38]
[39]
of zinc ¢nger transcription factor REST/NRSF/XBR gene, J. Neurosci. 18 (1998) 1280^1296. T. Timmusk, K. Palm, U. Lendarhl, M. Metsis, Brain-derived neurotrophic factor expression in vivo is under the control of neuron-restrictive silencer element, J. Biol. Chem. 274 (1999) 1078^1084. C.J. Schoenherr, A.J. Paquette, D.J. Anderson, Identi¢cation of potential target genes for the neuron-restrictive silencer factor, Proc. Natl. Acad. Sci. USA 93 (1996) 9881^9886. J. Shambrook, E.F. Fritsh and T. Maniatis, Molecular Cloning: A Laboratory Manual, 2nd edn., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1989. K. Gorski, M. Carneiro, U. Schibler, Tissue-speci¢c in vitro transcription from the mouse albumin promoter, Cell 47 (1986) 767^776. P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by acid guanidium thiocyanate-phenol chloroform extraction, Anal. Biochem. 162 (1987) 156^159. R.I. Scheinman, V.J. Auld, A.L. Goldin, N. Davidson, R.J. Dunn, W.A. Catterall, Developmental regulation of sodium channel expression in the rat forebrain, J. Biol. Chem. 264 (1989) 10660^10666. M.E. Greenberg, E.B. Zi¡, Stimulation of 3T3 cells induces transcription of the c-fos proto-oncogene, Nature 311 (1984) 433^438. N. Mori, C. Schoenherr, D.J. Vandenbergh, D.J. Anderson, A common silencer element in the SCG10 and type II Na channel genes binds a factor present in nonneuronal cells but not in neuronal cells, Neuron 9 (1992) 45^54. P. Lonnerberg, C.J. Schoenherr, D.J. Anderson, C.F. Ibanez, Cell type-speci¢c regulation of choline acetyltransferase gene expression, J. Biol. Chem. 271 (1996) 33358^33365. M. Sheng, M.E. Greenberg, The regulation and function of c-fos and other immediate early genes in the nervous system, Neuron 4 (1990) 477^485.
BBAEXP 93442 25-8-00