Neuronal activity regulates the developmental expression and subcellular localization of cortical BDNF mRNA isoforms in vivo

www.elsevier.com/locate/ymcne Mol. Cell. Neurosci. 28 (2005) 556 – 570

Neuronal activity regulates the developmental expression and subcellular localization of cortical BDNF mRNA isoforms in vivo Padmanabhan Paranji Pattabiraman,a Daniela Tropea,b Cristina Chiaruttini,c Enrico Tongiorgi,c Antonino Cattaneo,a and Luciano Domenicia,d,* a

Neuroscience Program, International School for Advanced Studies (SISSA/ISAS), Trieste 34014, Italy DBCS, MIT, Cambridge, MA, USA c Department of Biology, Brain Centre for Neuroscience, University of Trieste, Trieste 34127, Italy d Institute of Neuroscience (CNR), Pisa, Italy b

Received 19 October 2004; accepted 16 November 2004 Available online 11 January 2005 Activity-dependent changes in BDNF expression have been implicated in developmental plasticity. Although its expression is widespread in visual cortex, developmental regulation of its different transcripts by visual experience has not been investigated. Here, we investigated the cellular expression of different BDNF transcripts in rat visual cortex during postnatal development. We found that transcripts I and II are expressed only in adults but III and IV are expressed from early postnatal stage. Total BDNF mRNA is expressed throughout the age groups. Transcripts III and IV show a differential intracellular localization, while former was detected only in cell bodies, latter is present both in cell bodies and dendritic processes. Inhibition of visual activity decreases the levels of exons, with exon IV transcript almost disappearing from dendrites. In vitro experiments also confirmed the above results, indicating activity-dependent regulation of different BDNF promoters with specific temporal and cellular patterns of expression in developing visual cortex. D 2004 Elsevier Inc. All rights reserved.

Introduction Brain-derived neurotrophic factor (BDNF) is a member of the neurotrophin family and is involved in neuronal survival and differentiation. Beyond its classical neurotrophic role, there are Abbreviations: BDNF, brain-derived neurotrophic factor; DAB, diaminobenzidine; DEPC, diethyl pyrocarbonate; DIG, digoxigenin; KA, Kainic acid; KCl, potassium chloride; MAP2, microtubule associated protein; MDDL, maximal distance of dendritic labeling; mRNA, messenger ribonucleic acid; PBS, phosphate buffered saline; PFA, paraformaldehyde; RMV, relative migration value; RT-PCR, reverse transcription polymerase chain reaction; SSCT, saline sodium citrate Tween-20; TTX, Tetrodotoxin. * Corresponding author. Cognitive Neuroscience Sector, International School For Advanced Studies (SISSA/ISAS), Trieste 34014, Italy. Fax: +39 40 3787243. E-mail address: [email protected] (L. Domenici). Available online on ScienceDirect (www.sciencedirect.com). 1044-7431/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.mcn.2004.11.010

evidences suggesting that BDNF plays an important role in activity-dependent plasticity and development of the visual cortex. BDNF is produced by cortical pyramidal neurons only, while its high affinity receptor TrkB, is present on both pyramidal neurons and interneurons (Cellerino et al., 1996; Gorba and Wahle, 1999). BDNF mRNA and the protein are regulated during postnatal development in the primary visual cortex and are modulated by visual experience (Bozzi et al., 1995; Capsoni et al., 1999a,b; Castren et al., 1992; Schoups et al., 1995). BDNF modulates the maturation of GABAergic cortical circuitry (Huang et al., 1999), the morphology of neurons (McAllister et al., 1995), and synaptic plasticity processes (Akaneya et al., 1997; Jiang et al., 2003; Kinoshita et al., 1999; Sermasi et al., 1999, 2000). In addition, BDNF has been shown to inhibit ocular dominance column formation (Cabelli et al., 1995). The rat BDNF gene consists of one 3V exon (exon V) encoding for the BDNF protein and for the 3V untranslated (3VUTR) tail and four different 5V exons (exons I, II, III and IV) (Timmusk et al., 1993). These 5V exons are linked each to a different promoter that directs the expression of BDNF in a tissue specific way. Alternative promoter usage, differential splicing, and the use of two different polyadenylation sites within each of the four transcription units generate eight different BDNF mRNAs (Ohara et al., 1992). Recent evidences indicate that there could be more than eight splicing forms (Zuccato et al., 2001). Therefore, probes that bind to the exon V of the BDNF gene, which contains the coding sequence, are able to recognize all the different splicing forms. Previous studies have shown that these promoters are differentially expressed in different brain regions (Timmusk et al., 1994) and that the promoters of exons I and II are neuron-specific, while the others (III and IV) are also active in non-neural tissue (Nakayama et al., 1994). Interestingly, different injury-promoting agents induce different promoters (Kokaia et al., 1994) possibly through activation of a different set of receptors (Metsis et al., 1993). It has also been shown that motor activity, hormones, and circadian rhythms differentially regulate the transcription of

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

different BDNF isoforms (Koibuchi et al., 1999; Lauterborn et al., 1998; Oliff et al., 1998; Russo-Neustadt et al., 2000). Although BDNF isoform expression has been extensively studied in many brain regions, data in primary visual cortex are still lacking. More recently, the attention has been attracted by translocation of BDNF mRNA towards neuronal processes. The first report on the translocation of BDNF mRNAs in the dendrites was shown in cultured hippocampal neurons (Tongiorgi et al., 1997). It was also reported that BDNF dendritic localization extended to the distal dendrites on depolarization. Some other mRNAs localized in dendrites of neuronal cells are MAP2, Tau, h-actin, CaMKII-a, and Arc (Burgin et al., 1990; Garner et al., 1988; Kleiman et al., 1994; Kosik et al., 1989; Link et al., 1995; Lyford et al., 1995). Since protein synthesis can occur in neuronal dendrites, the displacement of BDNF mRNA together with its activity-dependence towards distal dendrites may represent an important aspect related to activity-dependent synaptic plasticity and maturation of neuronal circuitry in the visual cortex. Indeed, we have previously shown that in rats reared in dark, BDNF mRNA disappears from the apical dendrites of cortical neurons and reappears in this compartment when animals are re-exposed to light (Capsoni et al., 1999a,b). Since it is known that BDNF plays a fundamental role in activity-dependent plasticity and development, the aims of this study were (1) to investigate the expression and subcellular localization of five BDNF isoforms in the visual cortex during the postnatal development; and (2) to know how visual/electrical activity regulates the expression and subcellular localization of the different BDNF transcript in ex vivo visual cortex and in primary cultures of cortical neurons. By the aid of reverse transcription polymerase chain reaction (RT-PCR) and non-radioactive in situ hybridization, we were able to characterize and quantify the cellular expression of different BDNF transcripts. Our analysis was carried out at different stages of visual circuitry development (P13, P23, P40, and P90) and in cultures of cortical neurons. The role of activity was further assessed using Tetrodotoxin (TTX) injection into the eyes of one set of animals to inhibit activity and intra-peritoneal injection of kainic acid to increase the activity. We found that BDNF transcripts are present in rat visual cortex with a specific temporal and cellular pattern of expression during postnatal development and that their expression levels are modulated by activity. Another important result of this study is that the subcellular localization of different BDNF transcripts is modulated by neuronal activity and is characteristic for each BDNF transcript. Results Developmental expression of different BDNF transcripts in visual cortex A semi-quantitative analysis of the developmental expression of different BDNF transcripts in the visual cortex was carried out by RT-PCR using primers specific for exons I, II, III, IV, or a pair of primers for the coding region of exon V that amplified all BDNF transcripts. Expression levels of BDNF exons I and II were very low in visual cortex and were found only at late stages of postnatal development. Exon I was detected from P40, when a faint band was visible (Fig. 1A), increasing from P40 to P90 (the increase in the band intensity was around 17%, Fig. 1B). The exon II primers produced two specific bands with size 230 and 309 bp, which are

557

designated as II short and II long, respectively. These, as described before (Zuccato et al., 2001), denote two splice variants of exon II. We observed that exon II long is more abundant than the exon II short and that these two splicing variants are differently regulated during development. The exon II short shows an up-regulation by 10% from P40 to P90, whereas the II long fragment was maintained at the same level (Fig. 1B). Exon III was present at all investigated postnatal ages. The transcript was at a very low level of expression at P13 (Fig. 1A), with a significant increase ( P b 0.001) of around 40% by P40. After P40, exon III increases twofold and thereafter maintained a stable level (Figs. 1A, B). The exon IV expression levels at the different developmental stages followed the same curve of Exon III. There was very faint expression at P13 followed by a rapid increase of expression levels from P23 to P40 (Fig. 1A). The exon V, representing all isoforms, was expressed in all age groups used in this study (Fig. 1A). BDNF exon V showed a significant increase ( P b 0.001) in expression levels from P13 onwards reaching a plateau at P40 (Fig. 1B). BDNF isoforms have a different expression pattern and subcellular localization during postnatal development We have used a high sensitive non-isotopic in situ hybridization to study and compare the cellular expression pattern of the five different transcripts of BDNF at different postnatal ages. BDNF exon I transcript was detected only at P90 and no staining was visible at earlier ages. The staining was weak even at P90, being more intense in cortical layer 4 (Figs. 2A, B, C). In all stained cells, the staining was restricted to the cell soma. BDNF exon II transcripts were detected from P40 onwards with a riboprobe that recognizes both the short and long isoforms of exon II. The staining at P40 was very faint (data not shown). At P90 (Figs. 2D, E, F), stained cells are distributed throughout the different cortical layers and both the cell bodies and proximal dendrites appeared labeled. The exon III transcript became detectable already at the earliest stage of development tested and showed increased staining from P23 onward (Figs. 2G, H), especially in the cortical layer 4. At P40, that is, when the critical period was almost over and the maturation process of cortical neurons was nearly accomplished (Figs. 2I, J), exon III transcript was heavily expressed throughout all cortical layers and a similar pattern was still present at P90 (Figs. 2K, L). At the subcellular level, the staining was restricted only in the cell body (Fig. 4J). Exon IV transcript also started to be expressed from an early developmental stage (P13) (Fig. 3A) in all cortical layers with a low staining intensity. The labeling progressively increased through P23 (Figs. 3C, D) until P40 (Figs. 3E, F), with an expression pattern similar to that of exon III transcript. Many different cell types were stained throughout all visual cortical layers. It was clear that the transcript IV was expressed in the cell soma and dendrites (Figs. 3G, H). Sense control showed no staining (Fig. 3B). As we saw not much dendritic staining in the P13 animals in our in situ hybridization experiments, we hypothesized that this could be due to an immature state of the dendrites at this stage. We then carried out an immunohistochemistry for MAP2, which could be an indicator for dendritic maturation, and found that the dendrites were well developed at P23 and later stages of development. In P13 we found lesser dendritic staining for MAP2 (data not shown).

558

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

Fig. 1. Developmental regulation of different BDNF exons in rat visual cortex using RT-PCR: RT-PCR was done after the extraction of mRNA from visual cortex of rats of different developmental stages. The PCR product was run on a 1% agarose gel. h-actin PCR was used as a control. (A) Transcripts of different BDNF Exon I, II, III and IV (lane 1, 2, 3, 4, respectively) from visual cortex of rats from different postnatal age groups: P13, P23, P40, P90. Exon II shows two bands 230 and 303 bp, called, respectively, as the exon II short and exon II long. Exon V from P13, P23, P40, P90 rats (lane 1, 2, 3, 4, respectively). Positive control h-actin from P13, P23, P40, P90 rats (lane 1, 2, 3, 4, respectively). (B) The band intensity was measured using NIH image software measuring the relative expression of different BDNF transcripts versus h-actin. Graph showing the changes in the levels of each exon in relative units. Data shown are from an average of RT-PCR from six animals in each group.

Using the riboprobe against exon V, which recognizes all BDNF mRNA isoforms, we confirmed that BDNF was already expressed at P13 in all cortical layers (Figs. 3I, J). The level of staining increased until P23 and remaining almost stable thereafter from layer II to layer VI (Figs. 3K, L, M, O). The pan-BDNF riboprobe also showed a clear labeling of cell somas and both the basal and the apical dendrites (see arrowheads in Figs. 3N, P), similarly to what was found with the riboprobe against the exon IV transcript. Each probe was checked for specificity by using its sense-transcript. Sense control showed no staining (data shown only for exon IV riboprobe). We further investigated the cellular localization of different BDNF mRNA transcripts. We focused on exons III and IV

transcripts, which were highly expressed, according to the results reported above. We performed a double in situ hybridization by processing the same slices with probes for exon III and exon IV. Fig. 4A shows higher magnification of an adult rat visual cortex (P90) processed with the DIG-labeled probe for BDNF exon IV, and the biotinylated probe for BDNF exon III. The figure clearly shows that the blue staining, corresponding to BDNF exon IV mRNA, was present in dendrites, while the red staining, corresponding to exon III was absent. At the level of cell body cytoplasm, the color appears as red-blue, meaning that the two transcripts are colocalized. To exclude the possibility that these results were due to a different sensitivity of digoxigenin versus biotinylated probes, we repeated the same

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

559

Fig. 2. Developmental expression patterns of BDNF exons I, II, and III in rat visual cortex. Non-radioactive in situ hybridization was done for the different BDNF transcripts using DIG labeled riboprobe. BDNF exon I expression pattern in the visual cortex of P90 rat. (A) 5 magnification, (B) 10 magnification, (C) 40 magnification. BDNF exon II expression pattern in various regions of visual cortex of P90 rat. (D) 5 magnification showing all the layers, (E) 10 magnification of layer IV, (F) 40 magnification of layer IV. BDNF exon III expression pattern in visual cortex at P23, P40, and P90 of rat visual cortex. (G, I, K) 5 magnification showing all the layers of P23, P40 and P90 rat visual cortex, respectively. (H, J, L) 40 magnification of layer IV. Scale bars = 100 AM.

experiment inverting the labeling of the two probes: a biotinylated probe was synthesized for exon IV mRNA and a digoxigenin probe for exon III mRNA (Fig. 4B). In this case, only the red staining was present in dendrites, while in cell bodies both BDNF exons III and/or IV were detected. These experiments shows that exon III-containing transcripts are detectable only in cell bodies, whereas mRNAs containing the exon IV are present both in cell soma and dendrites.

did not show any change in the level of expression. The decrease for exons II–V was of about 50% (Fig. 5B). Thus, for all the BDNF transcripts with the exception of exon I the blockade of activity with TTX intraocular injection determined a decrease in BDNF mRNA expression.

Visual activity regulates the expression of BDNF transcripts II-long and short, III and IV, but not I

Since the dendritic targeting of BDNF mRNA depends upon electrical activity in vitro (Tongiorgi et al., 1997) and in vivo (Capsoni et al., 1999a,b), we investigated if the different subcellular localization of exon III and IV transcripts was also under the control of neuronal activity. We chose these two transcripts since exons I and II transcripts were not very clearly detectable at P40. We then analyzed the subcellular localization of the exons III, IV transcripts in comparison with the total BDNF mRNA (exon V probe) under control conditions and in conditions of either reduced (TTX injection in both eyes) or increased neuronal activity (kainic acid intraperitoneal injection). In agreement with the RT-PCR results, we found that in animals injected with TTX, the staining levels with the exons III, IV, and V riboprobes were decreased drastically when compared with those of control. To analyze these animals for the dendritic localization of the different BDNF isoforms, we focused on the

To determine if the expression of the five BDNF isoforms in the visual cortex was under the control of visual activity, we carried out intravitreal injections of TTX in both eyes at P40. Following single TTX injection, the spiking activity of retinal ganglion cells was blocked for 24–36 h (Caleo et al., 1999), thus depriving the central targets of retinal activity. In our experimental conditions, we observed a tonic dilation of the pupil and loss of direct papillary reflex following TTX injection. Eighteen hours after TTX injections, when dilation of pupil and loss of papillary reflex were still present, the changes in expression levels of BDNF isoforms in the occipital cortex was investigated using RTPCR. The RT-PCR analysis showed a decrease in the expression levels of all the exons except for exon I with respect to the control (Fig. 5A). Although the level of transcript of exon I was low, it

Neuronal activity controls the dendritic targeting of exon IV but not the somatic localization of exon III transcripts

560

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

Fig. 3. Developmental expression pattern of BDNF exon IV and V expression in rat visual cortex. Non-radioactive in situ hybridization was done for the different BDNF transcripts using DIG labeled riboprobe. BDNF exon IV expression pattern at P13, P23, P40 and P90 rat. (A) Showing 5 magnification of all layers from P13 visual cortex. (C, D) Showing 5 magnification of all layers and 40 magnification of layer 4 from P23 visual cortex. (E) Visual cortex of P40 shown at 5 magnification of all layers and F showing layer IV at 40. (G) Visual cortex of P90 shown at 5 magnification of all layers. (H) Showing layer IV at 40 magnification. (B) Showing the 5 magnification of in situ using sense probe on P40 rat. BDNF exon V expression in visual cortex of P13, P23, P40, and P90 rat. (I, K, M, O) Showing all layers at P13, P23, P40, P90, respectively, with 5 magnification. (J, L, N, P) Showing 40 magnification of layer 4, respectively, at P13, P23, P40, P90, respectively. Scale bars = 100 AM.

cortical layer 5 due to the large size of the neurons in this layer. At the cellular level, exon III labeling was restricted to the cell soma in the control group (Fig. 6A) and in the TTX-injected

animals, although in the latter animal group the staining was weaker (Fig. 6B). A staining for the exon IV was present in both soma and dendrites, in control conditions (Fig. 6D). In the TTX-

Fig. 4. Double in situ hybridization. Double in situ hybridization was carried out using biotinylated and DIG labeled probes to show that BDNF exon IV, but not BDNF exon III is localized in dendrites. Adult visual cortical slices were processed for in situ hybridization in order to detect staining for BDNF exon III and BDNF exon IV, simultaneously. In A, digoxigenin probe was used for the detection of BDNF exon IV (blue staining), while a biotinylated probe was used for the detection of BDNF exon III (red staining); only blue staining (dark arrows) is present in dendrites. In B, on the contrary, a biotinylated probe was used for the detection of BDNF exon IV (red staining), while a digoxigenin probe was used for the detection of BDNF exon III (blue staining); in this case only the red staining (dark arrows) is present in dendrites. Together, these results suggest that that BDNF exon IV, but not exon III, is present in dendrites. Scale bars = 100 AM.

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

Fig. 5. Visual activity regulated changes of different BDNF exons in rat visual cortex using RT-PCR. To test for the activity dependent changes in the levels of different exons, one group of animals were injected with TTX into the eye and another group, the control group, were injected with citrate buffer into the eye. The brain tissue was processed after 18 h of treatment. (A) RT-PCR results from the control tissue and visual cortex from TTXinjected group. Shown on the gel picture are different BDNF transcripts I, II, III, IV, and V with h-actin control (C) and 100 bp marker (M). Exon II shows two bands 230 and 303 bp, called, respectively, as the exon II short and exon II long. (B) Graph showing the changes of the levels of each exon in relative units between the control and TTX-injected animals. Data shown are from an average of RT-PCR from six animals in each group.

injected group, the exon IV transcript was found to be almost absent in dendrites (Fig. 6E). Staining for the exon V was present in the somato-dendritic compartment in controls but it almost disappeared from dendrites in TTX-treated animals (Fig. 6H), similarly to exon IV. In the experiment with increased neuronal activity, using kainic acid-injected animals, the staining levels with each of the three riboprobes were increased in the cortex 3 h after injection, in agreement with previous studies (Castren et al., 1998; Ernfors et al., 1991; Kokaia et al., 1994; Metsis et al., 1993; Simonato et al., 1998; Timmusk et al., 1993). We also found that exon III transcript showed a somatic localization in all experimental conditions (Fig. 6C). Exon IV transcript was in the proximal dendrites in untreated animals and became very distal and intense 3 h after kainic injection, especially in large pyramidal neurons (Fig. 6F). As expected, also the exon V transcript was seen to be very dendritic after kainic acid injection (Fig. 6I). These data show that neuronal activity regulates the dendritic targeting of exon IV transcripts but does not affect exon III transcripts localization. To determine if these effects were specific, we also analyzed the expression of BDNF in areas that do not receive direct visual inputs such as the hippocampus. The hippocampus of the TTXinjected animals showed staining pattern (Fig. 6J) similar to that of the control (Fig. 6K). In vitro studies for the cellular localization of BDNF isoforms To confirm the extent of differential expression and to quantify the localization of the BDNF isoforms in response to the neuronal activity, primary cortical cultures were used. To determine if the expression of the different BDNF isoforms in cultured primary

561

cortical neurons was regulated in response to the electrical activity similarly to that observed in the visual cortex in vivo, firstly RTPCR analysis was carried out. In these experiments, KCl or TTX were added to the cultures to increase and decrease the electrical activity of the neurons, respectively. Both treatments showed changes in the levels of exons II, III, IV, and V (data not shown). In the case of KCl treatment, there was increase in levels of exons IIL, III, IV, and V. Exon I and IIS did not show any change. With TTX treatment, the changes were very well in agreement with the in vivo studies. Non-radioactive in situ hybridization was carried out on cultured cortical cells using DIG-labeled riboprobes for exons III, IV, and V. The detection procedure uses highly sensitive fluorescence signal amplification with very low background (see methods). In untreated cultures, exons IV (Fig. 7G) and V (Fig. 7J) riboprobes clearly labeled the somato-dendritic compartment, while under TTX treatment (Figs. 7I, L), mostly the cell bodies were labeled and only a few dendrites showed a very faint proximal staining. In marked contrast, in KCl-treated cultures (Figs. 7H, K) both exons IV and V probes labeled extensively the distal dendritic processes and the cell somas. In agreement with the in vivo experiments, exon III was found only in the cell body in all experimental conditions (Figs. 7D, E, F). Two different negative control experiments were carried out to assess the probe specificity and the antibody specificity. One set of experiments used sense probes against the different exons and no staining was observed (Fig. 7B). The other bamplification negativeQ control experiment was done in the absence of primary anti-DIG mouse monoclonal which also showed no staining with a clean background (Fig. 7C). In addition, anti-MAP2 immunostaining was carried out (Fig. 7A) on all the cultures in order to confirm that the in situ staining was neuronal and also to know the extent of dendritic length. As we found varied dendritic localization of different isoforms of BDNF, we did quantitative measurements on cortical cultures after in situ hybridization experiments with the respective probes. To quantify, we used the Trace function of the Image Pro software. The maximal distance of the in situ labeling detectable in dendrites was measured for the Maximal Distance of Dendritic Labeling (MDDL) value (Tongiorgi et al., 1997) and divided for the length of each dendrite to obtain the Relative Migration Value (RMV). Under the control untreated condition, exon III was found only in the cell bodies: the RMV was 10.7 F 2.1% and MDDL was 11.4 F 2.3 Am. Exon IV was found both in dendrites and cell bodies: the RMV was 65.4 F 3.6% and MDDL was 44.5 F 6.1 Am. The total BDNF mRNA, detected with the exon V probe, was also found both in dendrites and cell bodies of cultured hippocampal neurons (Tongiorgi et al., 1997) and we saw a similar dendritic labeling pattern in our cortical cultures: the RMV was 56.6 F 3.4% and MDDL was 36.3 F 5.3 Am. The experiments, which involved activation by the use of high KCl, showed that exon III did not have any change in localization after the treatment: the RMV was 10.3 F 2% and MDDL was 11.3 F 2.6 Am. There was an increase in targeting to more distal dendrites: the RMV was 70.7 F 3.2% and MDDL was 58.1 F 5.4 Am for exon IV. Exon V was also found both in dendrites and cell bodies: the RMV was 69.5 F 3.6% and MDDL was 56.2 F 5.6 Am (Figs. 8A, B). Both exon IV and exon V probes showed a significant increase in dendritic targeting ( P b 0.001) of BDNF mRNA when compared to

562

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

Fig. 6. Neuronal activity regulated changes the expression pattern of mRNAs of different BDNF exons in rat visual cortex (P40) using in situ hybridization. To test for the activity-dependent changes in the expression pattern of exon III, IV, and V, in situ hybridization (ISH) was done on cortical slices from TTX-injected group (visual activity is inhibited) and buffer-injected groups. A and B show ISH for exons III at 40 (layer 4) magnification with the animals injected with citrate buffer (control) and TTX, respectively, in both the eyes. D and E show ISH for exons IV at 40 (layer 4) magnifications for control and TTX group, respectively. G and H show exon V ISH on the control animals and TTX-injected, respectively. The cellular localization was also checked using hyperactivity induction protocol (seizure induction) using kainic acid injection. (C) Exon III remains somatic in all experimental conditions. (F) Exon IV shows the highest labeling intensity and a distal dendritic signal is present even under control conditions. The dendritic staining for exon IV becomes very distal and robust after 3 h of kainic acid-induced seizures. (I) Showing exon V also highly dendritic. TTX did not affect the regions that receive no direct inputs from the eyes like hippocampus. The staining of the different exons in these regions was normal as the controls. J, K shows control hippocampus with respect to the hippocampus of TTX-injected animal at 5 magnification. Scale bars = 100 AM.

control untreated cultures. Experiments involving blocking of spiking activity of neurons using TTX showed that the subcellular expression of exon III did not change. In contrast, there was a decrease in dendritic targeting of exon IV and V, which resulted highly significant ( P b 0.001) (Tables 1, 2).

Discussion In this study, we have shown that different BDNF transcripts are differentially regulated during postnatal development of rat visual cortex and in primary cultures of cortical neurons.

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

563

Fig. 7. In situ hybridization to check for the activity-dependent regulation of subcellular localization of different mRNAs of BDNF exons in primary cortical cultures. Primary cortical cultures were prepared and after 9 days, they were treated with either 50 mM KCl or 2 AM TTX or was left untreated for the control experiments. (A) Showing immunohistochemistry for MAP2 as an indicator of the extent of dendritic maturation on 9 days old culture. B and C showing two controls: sense control, where sense DIG-labeled riboprobe was used instead of antisense riboprobe for exon IV, and bAmplification negativeQ control where the incubation with primary mouse monoclonal antibody against DIG in the development step was omitted. These controls were used to check the probe specificity and antibody specificity. D, E, F show ISH for exons III under control, KCl- and TTX-treated conditions, respectively. G, H, I shows ISH for exon IV under control, KCl- and TTX-treated conditions, respectively. J, K, L shows ISH for exons V under control, KCl- and TTX-treated conditions, respectively. Under KCl treatment, there is an increase in dendritic staining in the case of exons IV and V and the staining is restricted to soma and to very low extent in the proximal dendrites in the case of TTX experiment. Scale bar represents 50 AM.

The developmental regulation of BDNF mRNA expression was previously studied in the cortex and hippocampus (Bozzi et al., 1995; Galuske et al., 1999; Lein et al., 2000; Pollock and Frost, 2003; Pollock et al., 2001; Tropea et al., 2003). Its relevance in various cognitive functions and synaptic plasticity has been previously elucidated (Bartoletti et al., 2002; Cotrufo et al., 2003; Ichisaka et al., 2003; Jiang et al., 2003; Sermasi et al., 1999). However, although present in the cortical areas, not any specific functional role has been so far attributed to the different BDNF mRNA transcripts. The first requirement for these isoforms to play any functional role in visual cortex maturation is that their expression needs to be regulated during postnatal development in this area. Multi-promoter system of rat BDNF gene has been used to study tissue-specific expression of different promoters (Timmusk et al., 1993) and their developmental regulation. In particular (Timmusk et al., 1994), investigations have shown the expression of different BDNF transcripts in the whole rat brain during pre- and postnatal development. Their data show that different promoters have specific regulation patterns during embry-

onic development but they are regulated coordinately in several regions of rat brain during postnatal development. However, even these studies did not address the issue of the expression of BDNF isoforms in the visual cortex. In the visual cortex, we have found that transcripts containing exons I and II are almost undetectable during postnatal development, starting to be expressed at P40 through adulthood. Since P40 coincides with the end of the critical period for monocular deprivation (Fagiolini et al., 1994), it emerges that exons I and II (short and long isoforms) are not strictly related with postnatal maturation and developmental plasticity in the primary visual cortex. In contrast to the expression profiles of exons I and II, we found that transcripts III and IV are present from P13, that is, the earliest stage investigated before eye opening, to adulthood. Moreover, the patterns of cellular expression of exons III and IV are similar and almost correspond to that of the BDNF probe recognizing all different transcripts (Capsoni et al., 1999a,b). Exons III and IV rapidly increase after eye opening suggesting that visual experience

564

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

Fig. 8. MDDL and RMV for different mRNAs of BDNF exons in primary neuron cortical cultures. After ISH using different riboprobes, the Maximal Distance of Dendritic Labeling (MDDL) in micrometers and Relative Migration Value (RMV) in relative units were measured and plotted. (A) MDDL for exon V, exon IV and III under untreated control condition, under potassium chloride-stimulated condition and in the presence of TTX, which inhibits the spiking activity of neurons. The dots in the plot refers to the data points and the vertical lines at 20 and 40 Am denotes somatic limitation and proximal dendritic limitation, respectively, and above 40 AM denotes distal localization. The mean distance of dendritic localization was measured classifying the distance of within 20 Am was considered as somatic, 20–40 Am was considered proximal dendritic targeting distance and greater than 40 Am was considered as distal dendritic targeting. (B) RMV for exon V, exons IV and III under untreated control condition, under potassium chloride-stimulated condition and in the presence of TTX. The relative migration value was calculated by measuring the MAP2 staining length from the soma, which denotes the total dendritic length and the maximal distance of the dendritic targeting of the respective exon on the same dendrite.

play a role in their developmental regulation. From P23, an age corresponding to the peak of the critical period for monocular deprivation (Fagiolini et al., 1994), to P40, when the critical period and the maturation period are almost over, exons III and IV continue to increase significantly. Thus, from the present study, it emerges a pattern of differential regulation of the various splicing forms during postnatal development of the visual cortex. The second prominent result reported in the present paper is that exon IV but not exon III is expressed in dendritic processes while both transcripts are present in the cell soma. This result was obtained using non-isotopic in situ hybridization and was particularly evident from P23 onwards, when the maturation of dendritic processes is almost accomplished. From the in vitro experiments, we showed the

localization of exon IV to the dendrites of all calibers and covering almost 65% of the total dendritic length. Thus, we suggest the possibility that exons III and IV may address the coding sequence, encoded by BDNF exon V, to different subcellular sites. Previous reports showed that neurons in the central nervous system secrete neurotrophins in activity-dependent manner (Blochl and Thoenen, 1995; Goodman et al., 1996; Wang and Poo, 1997). In the visual cortex, BDNF mRNA (Bozzi et al., 1995; Capsoni et al., 1999a,b; Castren et al., 1992; Schoups et al., 1995) and protein (Tropea et al., 2001) are regulated by visual activity. It has been already shown by in vitro studies that BDNF mRNA is targeted to proximal dendrites and that neuronal activity modulates the degree of dendritic localization of BDNF mRNA in cultured hippocampal neurons and the involvement of phosphatidylinositol-3 kinase-

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

565

Table 1 Maximum distance of dendritic labeling (MDDL)

Mean SD X5 CTRL X5 KCl X5 TTX X4 CTRL X4 KCl X4 TTX X3 CTRL X3 KCl X3 TTX

X5 CTRL (n = 75)

X5 KCl (n = 72)

X5 TTX (n = 70)

X4 CTRL (n = 76)

X4 KCl (n = 80)

X4 TTX (n = 97)

X3 CTRL (n = 74)

X3 KCl (n = 71)

X3 TTX (n = 94)

36.3 5.32

56.2 5.6 P b 0.001

11.3 2.5 P b 0.001

44.5 6.1 P b 0.001

58.1 5.4

11.2 2.8

11.4 2.3

11.3 2.6

11.3 2.4

P = 0.977

P = 0.905

P b 0.001 P b 0.001 P b 0.001

P = 0.037 P b 0.001 P = 0.037 P = 0.733

P b 0.001 P b 0.001

P = 0.733 P b 0.001 P b 0.001

P = 0.977 P = 0.905

MDDL of the different BDNF transcripts measured on primary cortical neurons after ISH. The mean distance of dendritic localization was measured classifying the distance of within 20 Am was considered as somatic, 20 and 40 Am was considered proximal dendritic targeting distance and greater than 40 Am was considered as distal dendritic localization. Statistical analysis used was ANOVA among groups and the groups averages were plotted with 95% confidence interval. SD, Standard deviation; n, number of dendrites measured.

dependent pathway (Righi et al., 2000; Tongiorgi et al., 1997). Our data provide the additional information that deprivation of visual input by intraocular injections of TTX differentially controls the subcellular localization of different BDNF transcripts. In particular, the expression of exons II short, II long, III and IV, but not exon I transcript, was reduced with TTX treatment which also induced a significant decrease in the dendritic targeting of the BDNF isoforms expressed in dendrites, particularly of the exon IV transcript. On the other hand, a strong activation of cortical activity as that elicited in animals injected with kainic acid, induced a strong increase of dendritic localization for exon IV transcripts, which was also confirmed using the riboprobe against all BDNF isoforms (exon V). In marked contrast, the exon III isoform was only present in the cell soma even after such a strong stimulus. In neuronal cultures, exon IV transcript almost disappears from dendrites with TTX treatment while it is more distally localized when neuronal activity was increased by KCl. Exon III was localized into the cytoplasm of soma in both activated and control conditions. Indeed, the KCl activation could not induce a dendritic localization for exon III. These findings are strongly in favor of a role for 5VUTR region encoded by exon III in conferring somatic restriction to

BDNF mRNA. Since the expression levels of exon III are reduced by TTX treatment we suggest that this transcript is involved in activitydependent protein synthesis of BDNF in neuronal soma. Complex machinery is presumably responsible for targeting of BDNF mRNA to dendrites. Several mechanisms for targeting mRNA to subcellular compartments have been recently described (Mohr and Richter, 2003; Van De Bor and Davis, 2004). A growing number of both cis- and trans-acting factors have been shown responsible for dendritic RNA targeting. For instance, a 21nucleotide RNA transport signal has been shown to be responsible for the Myelin Basic Protein (MBP) (Ainger et al., 1997). In addition, Cytoplasmic Polyadenylation Elements (CPE) facilitates mRNA transport to the dendrites (Wu et al., 1998). It has also been seen that it is not strictly necessary to have the localization elements in the 3VUTR, only (Chartrand et al., 1999; Long et al., 1997). For example, elements present in both 5V and 3V UTR are required for localization of gurken RNA for Drosophila oogenesis (Thio et al., 2000). Another study has shown that a 24-base sequence in the 5VUTR is required for the selective dendritic targeting of mRNAs of NR1 splice variants (Pal et al., 2003). However, which sequences are ultimately responsible of subcellular localization of the splicing

Table 2 Relative migration value (RMV) of exons in dendrites

Mean SD X5 CTRL X5 KCl X5 TTX X4 CTRL X4 KCl X4 TTX X3 CTRL X3 KCl X3 TTX

X5 CTRL (n = 75)

X5 KCl (n = 72)

X5 TTX (n = 70)

X4 CTRL (n = 76)

X4 KCl (n = 80)

X4 TTX (n = 97)

X3 CTRL (n = 74)

X3 KCl (n = 71)

X3 TTX (n = 94)

56.6 3.4

69.5 3.6 P b 0.001

9.8 2.3 P b 0.001

65.4 3.6 P b 0.001

70.7 3.2

9.9 2.1

10.7 2.1

10.3 2.0

10.5 2.1

P = 0.272

P = 0.579

P b 0.001 P b 0.001 P b 0.001

P = 0.075 P b 0.001 P = 0.075 P = 0.917

P b 0.001 P b 0.001

P = 0.917 P b 0.001 P b 0.001

P = 0.272 P = 0.579

RMV of the different BDNF transcripts calculated on primary cortical culture neurons after ISH. Here, the distance of migration of exons over dendritic length was done to calculate to RMV. The total length of the dendrite was measured by the MAP2 immunohistochemistry. Statistical analysis used was ANOVA among groups and the groups averages were plotted with 95% confidence interval. SD, Standard deviation; n, number of dendrites measured.

566

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

forms of different BDNF mRNA and which elements are regulated by neuronal activity is still unknown. Exon IV itself could be involved in aiding the distal dendritic targeting. However, the coding region of BDNF mRNA itself could contain sequence(s) responsible for its targeting to the dendrites. Interestingly, it has recently been shown that hippocampal neurons expressing human BDNF, which carries a single nucleotide polymorphism (SNP) resulting in a valine instead of methionine at codon 66 in the pro-region, are characterized by the disappearance of BDNF protein from dendrites and its restriction to the cell soma (Egan et al., 2003; Lu, 2003). This mutation leads to impairment of plasticity in hippocampal neurons and alteration of spatial memory in humans. Although the authors favor the hypothesis of a mistargeting of the protein due to an incorrect folding within the endoplasmic reticulum, the fascinating hypothesis can be put forward of the possible existence of a dendritic targeting region within the coding sequence of BDNF which may be disrupted by the V66M polymorphism. The functional importance of the different BDNF transcripts could be both temporal and spatial. First, these isoforms could be involved in activity-dependent changes of BDNF in the same or different subsets of neurons and at different times during postnatal life, for example during postnatal development versus adulthood. Activity-dependent transcription, distinct turnover properties, or translational competence of the different BDNF isoforms may account for profound difference in the availability of BDNF at specific subset of synapses located on the soma versus proximal and/ or distal dendrites. The activity-dependent changes in expression levels of different transcripts, in particular exons III and IV, may confirm the relevance with their relation to BDNF synthesis during postnatal development of visual cortex when visual experience participates in reshaping the cortical circuitry. Secondly, BDNF isoforms III and IV could aid in presenting the coding sequence to different subcellular compartments, somatic versus dendritic, for somatic protein synthesis or local and fast protein synthesis in dendrites immediately ensuing a specific synaptic input. Exon III transcript being present only in soma could be involved in functions of BDNF that are executed somatically such as the regulation of synapses with a subset of cortical inhibitory neurons, the basket cells, making synapses with the soma of the pyramidal neurons (Gonchar and Burkhalter, 1997; Gupta et al., 2000; Hendry et al., 1984; Somogyi et al., 1983; Wang et al., 2002) and expressing TrkB (Cellerino et al., 1996; Gorba and Wahle, 1999). These cells in particular, might benefit from somatic BDNF synthesis and release under the control of neuronal activity. Indeed, inhibitory cortical neurons require pre-synaptic BDNF for maturation of their processes (Kohara et al., 2003) under the control of neuronal activity (Jin et al., 2003). On the other hand, there are other neurons mainly excitatory which form synapses with dendrites, and whose synaptic maturation might be regulated. On the basis of these results, we propose a hypothetical model to represent the functional significance of these exons during cortical maturation (Fig. 9). Thus, targeting of BDNF mRNA towards different subcellular compartments could represent an essential step to translate neuronal activity into specific synaptic changes. Experimental methods Primary cortical culture and treatments Dissociated primary cortical cultures were prepared from P0 to P1 rat pups. Neurons were plated at low densities onto poly-lysine

Fig. 9. The hypothetical model showing two different modes by which the different BDNF isoform could exert their function during cortical maturation. During the cortical maturation, where there is only exon III and exon IV presenting BDNF. The exon IV transcript being localized in distal dendritic compartment, could aid in providing BDNF required for excitatory neurons (ExC) forming synapses on the distal dendrites. Exon III isoform, which is somatically localized, could be involved in functions of BDNF that is executed somatically. It could be involved in providing BDNF to the basket cells (BC), which specifically synapses on the perisomatic region (mostly on and around soma) of the pyramidal (Py) neurons.

and laminin-coated coverslips. In brief, cortex was removed after decapitation and trypsin digested in the presence of DNase, blocked with trypsin inhibitor on ice and dissociated in DNase. After centrifugation, the cell pellet was washed twice. Then the pellet was resuspended in Minimum essential medium (MEM) with Earle’s salts and Glutamax I (Life Technologies) enriched with 5– 10% fetal bovine serum, 6 mg/ml d-glucose, 3.6 mg/ml HEPES, 0.1 Ag/ml biotin, 1.5 Ag/ml vitamin B12, 30 Ag/ml insulin, and 100 Ag/ml bovine transferrin. The cells are plated on coated coverslips. Proliferation of non-neuronal cells was prevented by the addition of 5.0 AM cytosine h-d-arabinofuranoside after around 24 to 36 h. Neuronal cultures were maintained by replacing half the medium every 2 days. We used the cells after 9 days in culture at which period there were well-developed neuronal processes as determined by MAP2 immunohistochemistry. In experimental groups, the cultures were given different effective treatments, which increased or decreased the neuronal activity by adding 50 mM potassium chloride (KCl) or 2 AM TTX (Alomone Labs), respectively, to the medium and neurons were harvested after 6 h of treatment. Tissue processing Wistar rats were anaesthetized with 20% urethane, 100 Al/100 g body weights. For the RT-PCR experiments, the visual cortex was excised and frozen till further use. For in situ hybridization experiments, the rats were transcardially perfused with 4% paraformaldehyde (PFA) in 0.1 M phosphate-buffered saline (PBS). Brains were removed and cryoprotected in 20% sucrose/ PBS at 48C until sectioning. Coronal sections (40 Am) containing binocular visual cortex, OC1b (Paxinos and Watson, 1986) were cut with a freezing microtome and uniform random series (1 section every 6) were collected for the experiments. Rats of different postnatal ages, P13, P23, P40, P90 rats, were used (n = 8/age group). For the activity dependent study, six animals at P40 were subjected to intravitreal injections of TTX (3 mM) in both the eyes to induce a complete blockade of the retinal activity (Caleo et al., 1999). For the injection solution, TTX was suspended in 0.025 M

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

citrate buffer and only citrate buffer vehicle was used for the control animals (6 animals). The injections were performed under anesthesia with a Hamilton syringe. The microsyringe was inserted at ora serrata and the injection volume was slowly released into the vitreous. Silencing of the retinal activity was noted by a tonic dilation of the pupil by around 2 to 3 h. The animals were then left in the cage for 18 h and then sacrificed. Another group of animals were administered with kainic acid, an analog of excitatory amino acid glutamate, i.p. (12 mg/kg weight of animal). After 3 h, the animals were sacrificed. The experiments were performed in accordance to the European Community Council Directive for animal treatment. RT-PCR Total RNA was extracted using TriZol (Gibco BRL) from visual cortices excised. In case of primary neuronal cultures, medium was removed and cells washed once rapidly in PBS, later TriZol was added and RNA was extracted. Then, DNase treatment (DNA-free from Ambion Inc) was carried out as per the instruction manual in order to get rid of genomic DNA escaped during the RNA extraction procedure. Five micrograms of total RNA was used for first-strand synthesis using SUPERSCRIPT First strand synthesis system for RT-PCR (Life Technologies) as per the user manual. The cDNA was amplified in a total volume of 50 Al with 1.5 mM MgCl2, 0.2 mM dNTPs, 2U Taq polymerase (Life Technologies) and 0.5 AM of sense and antisense primers for all BDNF exon. The control reaction was done with h-actin primers separately. The primer sequences were as follows: BDNF Exons I, II, III, IV (Zuccato et al., 2001) and BDNF Exon V (Marmigere et al., 2001) h-actin sense: 5V-CTT TTC ACg gTT ggC CTT Agg gTT 3V Antisense: 5V-AgA TTA CTg CCC Tgg CTC CTA g 3V. The annealing temperature and the number of cycles for both hactin and the BDNF exons were verified at different levels. They were optimized within the range where there was no change in the intensity of the band both for the h-actin and various exons of BDNF with increase in cycle number. h-actin message was found to be constant after 27 cycles. Amplification was done using the primers mentioned above with the following PCR program to amplify Exon I, II, III, IV: the program consists of initial denaturating step at 948C for 4 min then followed by: 948C for 1 min, 588C for 1 min and 728C for 2 min for 30 cycles and a final step at 728C for 10 min and for Exon V and h-actin, the annealing temperature was 608C, with all the other parameters remaining the same. The PCR products are separated by electrophoresis on 1% agarose gel and visualized by staining with ethidium bromide. The densitometry analysis of the agarose gel was done as follows. Semi-quantitative assessment of the developmental expression of BDNF exons was done with the digital images of agarose gels using NIH Image freeware called Scion Image for Windows (http://www.scioncorp.com/frames/fr_download _ now.htm). The BDNF bands were standardized versus h-actin. The final equation used to the calculate the percentage expression in the form of relative units was: BDNF Exon expression=b

actin expression  100

Riboprobes The BDNF exon V sense and antisense riboprobes were synthesized from pGEM3Z vector carrying a sequence of coding

567

region (nucleotide 72–533, seq. NM _012513) for detecting BDNF mRNA. Sequences for rat BDNF alternative 5V untranslated regions were recovered by screening a neonatal rat hippocampal cDNA library (Mladinic et al., 2000). Restriction fragments of BDNF exons III and IV were subcloned in pBluescriptKS-plasmid (Stratagene) in order to act as templates for digoxigenin (DIG) labeled riboprobes synthesis. Specifically, an exon II 481 bp fragment (nucleotide 321–802, seq. S71196S2), an exon III 320 bp fragment (nucleotide 241–565, seq. S71196S3) and an exon IV BamHI 210 bp (nucleotide 387– 596, seq. S71196S4) fragment were used. For exon I, PCRamplified products after RT-PCR experiments were excised from gel and purified and subcloned into TOPO clone (Invitrogen). Five micrograms of plasmid were linearized with the appropriate restriction enzyme in a 30-Al volume, and after increasing the volume to 200 Al were extracted with an equivalent volume of phenol, followed by chloroform/phenol (1:1) and chloroform alone. Plasmids were ethanol precipitated at 208C for 2 h, centrifuged, speed-vac dried for 15 min, dissolved in 10 Al of diethylpyrocarbonate-treated (DEPC) water, and stored at 208C. The DIG-labeled riboprobes were synthesized with a SP6 or T7 or T3 DIG-RNA labeling kit (Boehringer) according to the manufacturer’s instructions. For preparation of biotinylated probes, the same procedure was used with the exception that a biotin labeled, instead of DIG-labeled, ddUTP was present in the mix. Non-radioactive in situ hybridization High sensitive in situ hybridization was performed (Tongiorgi et al., 1998) on free-floating slices. Slices were washed twice in PBS/0.1% Tween 20 (PBST), permeabilized and was digested with 1 mg/ml Proteinase K (Boehringer-Mannheim). The duration of the digestion was dependent on the age of the animals: 5 min for P13 animals, 9 min for P23 animals and 16 min for adult animals. Pre-hybridization was carried out at 558C for 1 to 6 h in the hybridization solution. Slices were then transferred into the hybridization solution, composed by the pre-hybridization solution with 10% dextran sulphate, and 50–100 ng/ml DIG-labeled riboprobes. The sense control experiments had sense DIG-labeled riboprobe instead of the antisense riboprobe in the hybridization mix. In situ hybridization was carried out overnight (at least 16 h) in multiwell plates at 558C without shaking. Post-hybridization washes were done in 2 saline sodium citrate, 0.1% Tween 20 (SSCT)/50% deionized formamide at 558C for 30 min, 20 min in 2 SSCT and twice in 0.05% SSCT at 608C for 30 min. After post-hybridization washes, the slices hybridized with riboprobes were incubated with anti-DIG F(ab)2 fragment conjugated with Alkaline phosphatase (Boehringer-Mannheim) at a dilution of 1:1000 in 10% FCS in PBS + 0.1% Tween20, over night at 48C. For primary cortical cultures, medium was removed; cells were washed and then fixed with 4% paraformaldehyde in PBS for 15 min, washed in PBS and permeabilized in 100% ethanol for 15 min at 208C. After rehydration with decreasing ethanol concentrations in PBS at room temperature, cells were prehybridized at 558C. Later, the in situ hybridization was carried out as in free-floating system. In the control experiments, sense probe was used to check for the specificity of the probe. The sense control experiments were included for in vitro experiments too. Two independent investigators carried out the experiments to confirm the results.

568

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

Colorimetric detection Probe detection was done with colorimetric development method using 4-nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-3indolyl phosphate (BCIP) in 100 mM Tris–HCl, pH 9.5, 50 mM MgCl2, 100 mM NaCl and 1 mM levamisol. Alkaline phosphatase development was done for 16 h at 48C to have reproducible results and to avoid saturation of reaction. The reaction was stopped by rinsing the sections in a stop-solution (10 mM Tris–HCl pH 8, 1 mM EDTA). Sections were then washed with PBS and mounted in water on glass slides coated with 5% gelatin (Merck) and mounted using Eukitt (Vector laboratories) after dehydration. Slices hybridized with biotinylated probes were incubated with an avidin–biotin–peroxidase complex (1:100 dilution, ABC elite, Vector Labs, Burlingame CA) for 1 h at room temperature. The labeling was revealed in 3-3V diaminobenzidine (DAB) HCl (10 mg in 25 ml Tris–HCl pH 7.5) for 20 min at room temperature. Sections were then washed and mounted as previously mentioned. Immunofluorescence detection DIG-labeled riboprobes on cortical cultures were detected in situ with the fluorescent antibody enhancer set for DIG detection (Roche Molecular Biochemicals). This detection involved an antibody cascade leading to amplification of signal intensity, with some slight changes from the instructor manual. Cells hybridized with DIG-labeled riboprobes were incubated 30 min with a suitable volume of blocking solution (10% FCS in PBS). Then, cells were incubated for 2 h at 378C with a mouse monoclonal anti-DIG diluted 1:35 in blocking solution (10% FCS in PBS). Apart from the sense control, an bamplification negativeQ control was also done in order to test for the specificity of fluorescent antibody enhancer set. In the bamplification negativeQ controls, the primary antibody incubation step was omitted. After three washes with blocking solution, cell were incubated at least for 2 h with Antimouse-Ig (FabV)2 fragments (1:35). Then, the DIG labeling was detected after washes using an anti-DIG fluorescein antibody conjugate (1:35). Double immunostaining was done on the cultured primary neuronal cells after in situ hybridization procedure described above, for microtubule-associated protein (MAP2) using a primary anti-MAP2 (Chemicon Inc) monoclonal antibody (1:500) and a detecting antibody anti-mouse TRITC conjugated (1:500). In order to avoid the cross reaction of antiMAP2 mouse monoclonal antibody on unbound anti-DIG mouse monoclonal, after the in situ detection, the coverslips were incubated with Mouse-on-Mouse kit (Vector Laboratories) to block the unspecific sites. Coverslips were then washed in PBS and mounted Vectashield (Vector laboratories). Immunohistochemistry MAP2 immunohistochemistry on tissues at various developmental stages was done using anti-MAP2 mouse monoclonal antibody. A biotinylated anti-mouse secondary antibody was used. The development was done after the incubation with avidin– biotin–peroxidase complex (Vector Labs) using DAB protocol. Image analysis Non-radioactive in situ hybridization images were captured using a Carl Zeiss Axioskop Fluorescence microscope. The

images were analyzed using Image J (freeware from NIH). The dendritic lengths were measured using the Trace program in Image J. The measurement was done from the base of the dendrite. In order to have the Relative Migration Value (RMV) of exon in dendrites, we measured the relative migration of exons to the total dendritic length; the maximal dendritic length was measured using the MAP2 immunofluorescence staining in addition to the MDDL. The statistical analysis was done using ANOVA among groups and the groups averages were plotted with 95% confidence interval. Acknowledgments We wish to thank Dr. Mauro Benedetti who kindly provided with some of the BDNF plasmids and Dr. Massimo Righi for the assistance in cell culture work. We also thank Dr. Kevin Ainger for his valuable discussion and support. This work was supported by MURST Network grant to D.T. and Intramural grant from SISSA to L.D. References Ainger, K., Avossa, D., Diana, A.S., Barry, C., Barbarese, E., Carson, J.H., 1997. Transport and localization elements in myelin basic protein mRNA. J. Cell Biol. 138, 1077 – 1087. Akaneya, Y., Tsumoto, T., Kinoshita, S., Hatanaka, H., 1997. Brain-derived neurotrophic factor enhances long-term potentiation in rat visual cortex. J. Neurosci. 17, 6707 – 6716. Bartoletti, A., Cancedda, L., Reid, S.W., Tessarollo, L., Porciatti, V., Pizzorusso, T., Maffei, L., 2002. Heterozygous knock-out mice for brain-derived neurotrophic factor show a pathway-specific impairment of long-term potentiation but normal critical period for monocular deprivation. J. Neurosci. 22, 10072 – 10077. Blochl, A., Thoenen, H., 1995. Functional postnatal development of the rat primary visual cortex and the role of visual experience: dark rearing and monocular deprivation. Eur. J. Neurosci. 7, 1220 – 1228. Bozzi, Y., Pizzorusso, T., Cremisi, F., Rossi, F.M., Barsacchi, G., Maffei, L., 1995. Monocular deprivation decreases the expression of messenger RNA for brain-derived neurotrophic factor in the rat visual cortex. Neuroscience 69, 1133 – 1144. Burgin, K.E., Waxham, M.N., Rickling, S., Westgate, S.A., Mobley, W.C., Kelly, P.T., 1990. In situ hybridization histochemistry of Ca2+/calmodulin-dependent protein kinase in developing rat brain. J. Neurosci. 10, 1788 – 1798. Cabelli, R.J., Hohn, A., Shatz, C.J., 1995. Inhibition of ocular dominance column formation by infusion of NT-4/5 or BDNF. Science 267, 1662 – 1666. Caleo, M., Lodovichi, C., Maffei, L., 1999. Effects of nerve growth factor on visual cortical plasticity require afferent electrical activity. Eur. J. Neurosci. 11, 2979 – 2984. Capsoni, S., Tongiorgi, E., Cattaneo, A., Domenici, L., 1999a. Dark rearing blocks the developmental down-regulation of brain-derived neurotrophic factor messenger RNA expression in layers IV and V of the rat visual cortex. Neuroscience 88, 393 – 403. Capsoni, S., Tongiorgi, E., Cattaneo, A., Domenici, L., 1999b. Differential regulation of brain-derived neurotrophic factor messenger RNA cellular expression in the adult rat visual cortex. Neuroscience 93, 1033 – 1040. Castren, E., Zafra, F., Thoenen, H., Lindholm, D., 1992. Light regulates expression of brain-derived neurotrophic factor mRNA in rat visual cortex. Proc. Natl. Acad. Sci. U. S. A. 89, 9444 – 9448. Castren, E., Berninger, B., Leingartner, A., Lindholm, D., 1998. Regulation of brain-derived neurotrophic factor mRNA levels in hippocampus by neuronal activity. Prog. Brain Res. 117, 57 – 64.

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570 Cellerino, A., Maffei, L., Domenici, L., 1996. The distribution of brainderived neurotrophic factor and its receptor trkB in parvalbumincontaining neurons of the rat visual cortex. Eur. J. Neurosci. 8, 1190 – 1197. Chartrand, P., Meng, X.H., Singer, R.H., Long, R.M., 1999. Structural elements required for the localization of ASH1 mRNA and of a green fluorescent protein reporter particle in vivo. Curr. Biol. 9, 333 – 336. Cotrufo, T., Viegi, A., Berardi, N., Bozzi, Y., Mascia, L., Maffei, L., 2003. Effects of neurotrophins on synaptic protein expression in the visual cortex of dark-reared rats. J. Neurosci. 23, 3566 – 3571. Egan, M.F., Kojima, M., Callicott, J.H., Goldberg, T.E., Kolachana, B.S., Bertolino, A., Zaitsev, E., Gold, B., Goldman, D., Dean, M., Lu, B., Weinberger, D.R., 2003. The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. Cell 112, 257 – 269. Ernfors, P., Bengzon, J., Kokaia, Z., Persson, H., Lindvall, O., 1991. Increased levels of messenger RNAs for neurotrophic factors in the brain during kindling epileptogenesis. Neuron 7, 165 – 176. Fagiolini, M., Pizzorusso, T., Berardi, N., Domenici, L., Maffei, L., 1994. Functional postnatal development of the rat primary visual cortex and the role of visual experience: dark rearing and monocular deprivation. Vision Res. 34, 709 – 720. Galuske, R.A., Kim, D.S., Singer, W., 1999. The role of neurotrophins in developmental cortical plasticity. Restor. Neurol. Neurosci. 15, 115 – 124. Garner, C.C., Tucker, R.P., Matus, A., 1988. Selective localization of messenger RNA for cytoskeletal protein MAP2 in dendrites. Nature 336, 674 – 677. Gonchar, Y., Burkhalter, A., 1997. Three distinct families of GABAergic neurons in rat visual cortex. Cereb. Cortex 7, 347 – 358. Goodman, L.J., Valverde, J., Lim, F., Geschwind, M.D., Federoff, H.J., Geller, A.I., Hefti, F., 1996. Regulated release and polarized localization of brain-derived neurotrophic factor in hippocampal neurons. Mol. Cell Neurosci. 7, 222 – 238. Gorba, T., Wahle, P., 1999. Expression of TrkB and TrkC but not BDNF mRNA in neurochemically identified interneurons in rat visual cortex in vivo and in organotypic cultures. Eur. J. Neurosci. 11, 1179 – 1190. Gupta, A., Wang, Y., Markram, H., 2000. Organizing principles for a diversity of GABAergic interneurons and synapses in the neocortex. Science 287, 273 – 278. Hendry, S.H., Jones, E.G., DeFelipe, J., Schmechel, D., Brandon, C., Emson, P.C., 1984. Neuropeptide-containing neurons of the cerebral cortex are also GABAergic. Proc. Natl. Acad. Sci. U. S. A. 81, 6526 – 6530. Huang, Z.J., Kirkwood, A., Pizzorusso, T., Porciatti, V., Morales, B., Bear, M.F., Maffei, L., Tonegawa, S., 1999. BDNF regulates the maturation of inhibition and the critical period of plasticity in mouse visual cortex. Cell 98, 739 – 755. Ichisaka, S., Katoh-Semba, R., Hata, Y., Ohshima, M., Kameyama, K., Tsumoto, T., 2003. Activity-dependent change in the protein level of brain-derived neurotrophic factor but no change in other neurotrophins in the visual cortex of young and adult ferrets. Neuroscience 117, 361 – 371. Jiang, B., Akaneya, Y., Hata, Y., Tsumoto, T., 2003. Long-term depression is not induced by low-frequency stimulation in rat visual cortex in vivo: a possible preventing role of endogenous brain-derived neurotrophic factor. J. Neurosci. 23, 3761 – 3770. Jin, X., Hu, H., Mathers, P.H., Agmon, A., 2003. Brain-derived neurotrophic factor mediates activity-dependent dendritic growth in nonpyramidal neocortical interneurons in developing organotypic cultures. J. Neurosci. 23, 5662 – 5673. Kinoshita, S., Yasuda, H., Taniguchi, N., Katoh-Semba, R., Hatanaka, H., Tsumoto, T., 1999. Brain-derived neurotrophic factor prevents lowfrequency inputs from inducing long-term depression in the developing visual cortex. J. Neurosci. 19, 2122 – 2130. Kleiman, R., Banker, G., Steward, O., 1994. Development of subcellular mRNA compartmentation in hippocampal neurons in culture. J. Neurosci. 14, 1130 – 1140.

569

Kohara, K., Kitamura, A., Adachi, N., Nishida, M., Itami, C., Nakamura, S., Tsumoto, T., 2003. Inhibitory but not excitatory cortical neurons require presynaptic brain-derived neurotrophic factor for dendritic development, as revealed by chimera cell culture. J. Neurosci. 23, 6123 – 6131. Koibuchi, N., Fukuda, H., Chin, W.W., 1999. Promoter-specific regulation of the brain-derived neurotropic factor gene by thyroid hormone in the developing rat cerebellum. Endocrinology 140, 3955 – 3961. Kokaia, Z., Metsis, M., Kokaia, M., Bengzon, J., Elmer, E., Smith, M.L., Timmusk, T., Slesjo, B.K., Persson, H., Lindvall, O., 1994. Brain insults in rats induce increased expression of the BDNF gene through differential use of multiple promoters. Eur. J. Neurosci. 6, 587 – 596. Kosik, K.S., Crandall, J.E., Mufson, E.J., Neve, R.L., 1989. Tau in situ hybridization in normal and Alzheimer brain: localization in the somatodendritic compartment. Ann. Neurol. 26, 352 – 361. Lauterborn, J.C., Poulsen, F.R., Stinis, C.T., Isackson, P.J., Gall, C.M., 1998. Transcript-specific effects of adrenalectomy on seizure-induced BDNF expression in rat hippocampus. Brain Res., Mol. Brain Res. 55, 81 – 91. Lein, E.S., Hohn, A., Shatz, C.J., 2000. Dynamic regulation of BDNF and NT-3 expression during visual system development. J. Comp. Neurol. 420, 1 – 18. Link, W., Konietzko, U., Kauselmann, G., Krug, M., Schwanke, B., Frey, U., Kuhl, D., 1995. Somatodendritic expression of an immediate early gene is regulated by synaptic activity. Proc. Natl. Acad. Sci. U. S. A. 92, 5734 – 5738. Long, R.M., Singer, R.H., Meng, X., Gonzalez, I., Nasmyth, K., Jansen, R.P., 1997. Mating type switching in yeast controlled by asymmetric localization of ASH1 mRNA. Science 277, 383 – 387. Lu, B., 2003. Pro-region of neurotrophins: role in synaptic modulation. Neuron 39, 735 – 738. Lyford, G.L., Yamagata, K., Kaufmann, W.E., Barnes, C.A., Sanders, L.K., Copeland, N.G., Gilbert, D.J., Jenkins, N.A., Lanahan, A.A., Worley, P.F., 1995. Arc, a growth factor and activity-regulated gene, encodes a novel cytoskeleton-associated protein that is enriched in neuronal dendrites. Neuron 14, 433 – 445. Marmigere, F., Rage, F., Tapia-Arancibia, L., 2001. Regulation of brainderived neurotrophic factor transcripts by neuronal activation in rat hypothalamic neurons. J. Neurosci. Res. 66, 377 – 389. McAllister, A.K., Lo, D.C., Katz, L.C., 1995. Neurotrophins regulate dendritic growth in developing visual cortex. Neuron 15, 791 – 803. Metsis, M., Timmusk, T., Arenas, E., Persson, H., 1993. Differential usage of multiple brain-derived neurotrophic factor promoters in the rat brain following neuronal activation. Proc. Natl. Acad. Sci. U. S. A. 90, 8802 – 8806. Mladinic, M., Didelon, F., Cherubini, E., Bradbury, A., 2000. dSpecificT oligonucleotides often recognize more than one gene: the limits of in situ hybridization applied to GABA receptors. J. Neurosci. Methods 98, 33 – 42. Mohr, E., Richter, D., 2003. Molecular determinants and physiological relevance of extrasomatic RNA localization in neurons. Front. Neuroendocrinol., 24128 – 24139. Nakayama, M., Gahara, Y., Kitamura, T., Ohara, O., 1994. Distinctive four promoters collectively direct expression of brain-derived neurotrophic factor gene. Mol. Brain Res. 21, 206 – 218Ohara, O., Gahara, Y., Teraoka, H., Kitamura, T., 1992. A rat brain-derived neurotrophic factor-encoding gene generates multiple transcripts through alternative use of 5V exons and polyadenylation sites. Gene 121, 383 – 386. Oliff, H.S., Berchtold, N.C., Isackson, P., Cotman, C.W., 1998. Exerciseinduced regulation of brain-derived neurotrophic factor (BDNF) transcripts in the rat hippocampus. Mol. Brain Res. 61, 147 – 153. Pal, R., Agbas, A., Bao, X., Hui, D., Leary, C., Hunt, J., Naniwadekar, A., Michaelis, M.L., Kumar, K.N., Michaelis, E.K., 2003. Selective dendrite-targeting of mRNAs of NR1 splice variants without exon 5: identification of a cis-acting sequence and isolation of sequence-binding proteins. Brain Res. 994, 1 – 18. Paxinos, G., Watson, C., 1986. The Rat Brain Stereotaxic Coordinates, 2nd Ed., Academic Press.

570

P.P. Pattabiraman et al. / Mol. Cell. Neurosci. 28 (2005) 556–570

Pollock, G.S., Frost, D.O., 2003. Complexity in the modulation of neurotrophic factor mRNA expression by early visual experience. Brain Res., Dev. Brain Res. 143, 225 – 232. Pollock, G.S., Vernon, E., Forbes, M.E., Yan, Q., Ma, Y.T., Hsieh, T., Robichon, R., Frost, D.O., Johnson, J.E., 2001. Effects of early visual experience and diurnal rhythms on BDNF mRNA and protein levels in the visual system, hippocampus, and cerebellum. J. Neurosci. 21, 3923 – 3931. Righi, M., Tongiorgi, E., Cattaneo, A., 2000. Brain-derived neurotrophic factor (BDNF) induces dendritic targeting of BDNF and tyrosine kinase B mRNAs in hippocampal neurons through a phosphatidylinositol-3 kinase-dependent pathway. J. Neurosci. 20, 3165 – 3174. Russo-Neustadt, A., Beard, R.C., Huang, Y.M., Cotman, C.W., 2000. Physical activity and antidepressant treatment potentiate the expression of specific brain-derived neurotrophic factor transcripts in the rat hippocampus. Neuroscience 101, 305 – 312. Schoups, A.A., Elliott, R.C., Friedman, W.J., Black, I.B., 1995. NGF and BDNF are differentially modulated by visual experience in the developing geniculocortical pathway. Brain Res. Dev. Brain Res. 86, 326 – 334. Sermasi, E., Tropea, D., Domenici, L., 1999. A new form of synaptic plasticity is transiently expressed in the developing rat visual cortex: a modulatory role for visual experience and brain-derived neurotrophic factor. Neuroscience 91, 163 – 173. Sermasi, E., Margotti, E., Cattaneo, A., Domenici, L., 2000. Trk B signalling controls LTP but not LTD expression in the developing rat visual cortex. Eur. J. Neurosci. 12, 1411 – 1419. Simonato, M., Molteni, R., Bregola, G., Muzzolini, A., Piffanelli, M., Beani, L., Racagni, G., Riva, M., 1998. Different patterns of induction of FGF-2, FGF-1 and BDNF mRNAs during kindling epileptogenesis in the rat. Eur. J. Neurosci. 10, 955 – 963. Somogyi, P., Kisvarday, Z.F., Martin, K.A., Whitteridge, D., 1983. Synaptic connections of morphologically identified and physiologically characterized large basket cells in the striate cortex of cat. Neuroscience 10, 261 – 294. Thio, G.L., Ray, R.P., Barcelo, G., Schupbach, T., 2000. Localization of gurken RNA in Drosophila oogenesis requires elements in the 5V and 3V regions of the transcript. Dev. Biol. 221, 435 – 446.

Timmusk, T., Palm, K., Metsis, M., Reintam, T., Paalme, V., Saarma, M., Persson, H., 1993. Multiple promoters direct tissue-specific expression of the rat BDNF gene. Neuron 10, 475 – 489. Timmusk, T., Belluardo, N., Persson, H., Metsis, M., 1994. Analysis of transcriptional initiation and translatability of brain-derived neurotrophic factor mRNAs in the rat brain. Neuroscience 60, 287 – 291. Tongiorgi, E., Righi, M., Cattaneo, A., 1997. Activity-dependent dendritic targeting of BDNF and TrkB mRNAs in hippocampal neurons. J. Neurosci. 17, 9492 – 9505. Tongiorgi, E., Righi, M., Cattaneo, A., 1998. A non-radioactive in situ hybridization method that does not require RNAse-free conditions. J. Neurosci. Methods 85, 129 – 139. Tropea, D., Capsoni, S., Tongiorgi, E., Giannotta, S., Cattaneo, A., Domenici, L., 2001. Mismatch between BDNF mRNA and protein expression in the developing visual cortex: the role of visual experience. Eur. J. Neurosci. 13, 709 – 721. Tropea, D., Caleo, M., Maffei, L., 2003. Synergistic effects of brain-derived neurotrophic factor and chondroitinase ABC on retinal fiber sprouting after denervation of the superior colliculus in adult rats. J. Neurosci. 23, 7034 – 7044. Van De Bor, V., Davis, I., 2004. mRNA localisation gets more complex. Curr. Opin. Cell Biol. 16, 300 – 307. Wang, X.H., Poo, M.M., 1997. Potentiation of developing synapses by postsynaptic release of neurotrophin-4. Neuron 19, 825 – 835. Wang, Y., Gupta, A., Toledo-Rodriguez, M., Wu, C.Z., Markram, H., 2002. Anatomical, physiological, molecular and circuit properties of nest basket cells in the developing somatosensory cortex. Cereb. Cortex 12, 395 – 410. Wu, L., Wells, D., Tay, J., Mendis, D., Abbott, M.A., Barnitt, A., Quinlan, E., Heynen, A., Fallon, J.R., Richter, J.D., 1998. CPEB-mediated cytoplasmic polyadenylation and the regulation of experience-dependent translation of alpha-CaMKII mRNA at synapses. Neuron 21, 1129 – 1139. Zuccato, C., Ciammola, A., Rigamonti, D., Leavitt, B.R., Goffredo, D., Conti, L., MacDonald, M.E., Friedlander, R.M., Silani, V., Hayden, M.R., Timmusk, T., Sipione, S., Cattaneo, E., 2001. Loss of huntingtinmediated BDNF gene transcription in Huntington’s disease. Science 293, 493 – 498.