[1] 9w Diazepam induces FGF-2 mRNA in the hippocampus and striatum

[1] 146w We have examined the possibility that the ␥-aminobutyric acid (GABA) system can influence the expression of basic fibroblast growth factor (FGF-2) in select regions of the brain. The GABA agonist diazepam was systemically injected in adult rats, and the expression of FGF-2 was examined between 6 h and 7 days post-injection. Results of nuclease protection assays showed significant increases in FGF-2 mRNA in the hippocampus and striatum, but not in the caudal cerebral cortical region-starting by 6 h following diazepam injection and returning to approximately control values by 24 h. In situ hybridization showed elevated FGF-2 mRNA labeling in the hippocampal formation, mostly in the pyramidal layer of the CA1 and CA2 subfields and in the dentate gyrus hilar region. These results indicate that diazepam treatment up-regulates FGF-2 expression in select regions of the brain and suggest that GABA may promote neuroplasticity in concert with FGF-2.

[1] 76w Increasing evidence indicates that the neurotransmitter ␥-aminobutyric acid (GABA) is an important modulator of structural and functional plasticity during developmental, and mature stages of the central nervous system (CNS) [35,44,54]. Recent studies indicate that GABA, similar to trophic factors, may provide neuroprotection following CNS trauma. For example, injection of the GABA full agonist diazepam [3,33,45] into the systemic circulation of rodents can have a trophic effect on ischemic damaged neurons in the hippocampus and striatum [46].

[2] 76w The action of basic fibroblast growth factor (FGF-2) can be essential for survival and repair of damaged tissue following traumatic injury. FGF-2 promotes survival and neurite growth of several neuron types [29,52], induces formation of capillary endothelial cells and vascular smooth muscle cells ([17] see review [4]). Interestingly, FGF-2 facilitates survival and growth of striatal [52,57] and hippocampal [51] neurons, suggesting the possibility that FGF-2 may be involved with the action of diazepam on these neurons.

[3] 149w It is noteworthy that the recently discovered roles of GABA in neuroplasticity makes GABA comparable to trophic factors, while the regulation of trophic factors by neural activity makes trophic factors comparable to neurotransmitters. Accordingly, we have investigated the possibility that some of the roles of GABA in neuroplasticity can be associated with the function of trophic factors. In particular, we evaluate the possibility that diazepam regulates the expression of FGF-2 in the hippocampus and striatum, because the actions of diazepam and FGF-2 have been described in these regions. Diazepam is considered a full agonist for benzodiazepine (BZ) receptors [3,33,45] which activation results in a positive modulation of GABA function. Diazepam readily cross the blood-brain barrier [32]. and is extensively prescribed for anxiety disorders. Information on the regulation of the FGF-2 system by GABA is fundamental to better understand the molecular mechanisms by which neurotransmitters and trophic factors regulate neuroplasticity.

[1] 28w Levels of FGF-2 mRNA were measured using nuclease protection assays in the hippocampus, striatum, and caudal portion of the cerebral cortex at various time periods after diazepam injection.

[2] 206w Results showed that FGF-2 mRNA in the hippocampal formation were elevated up to 136% of control values by 6 h (p Ͻ 0.05), continued to increase to 145% by 12 h, and remained at near control values between 24 h and 7 days (Fig. 1). In situ hybridization was performed in the hippocampal formation. Intact animals showed FGF-2 mRNA in situ hybridization almost exclusively in the CA2 region, as described previously [28]. There were notable increases in FGF-2 mRNA in situ hybridization in the various CA subfields and dentate gyrus after 6 h of diazepam treatment. Computer densitometry readings were taken from the various hippocampal subfields and revealed significant increases in FGF-2 mRNA expression (Fig. 2). The pyramidal layer of the CA1 region exhibited an increase in FGF-2 mRNA to 240% of control values (p Ͻ 0.05) while the CA2 region showed an increase up to 400% of controls (p Ͻ 0.01). The hilar region of the dentate gyrus exhibited an increase in FGF-2 mRNA up to 160% (p Ͻ 0.05) of control values while the stratum oriens of the CA1 region showed an increase that reached 140% of controls (p Ͻ 0.05). There were no significant changes in FGF-2 mRNA in the CA1 stratum radiatum.

[3] 80w Nuclease protection measures of FGF-2 mRNA in the striatum reached 118% of control values by 6 h, continued to increase up to 135% by 12 h (p Ͻ 0.05), and returned to approximately control levels after 24 h of diazepam injection (Fig. 3). The caudal cerebral cortex showed an increasing trend in FGF-2 mRNA to 123% of controls by 12 h, and levels of FGF-2 mRNA returned to approximately control values after 3 days of diazepam injection (data not shown).

[1] 71w Our results indicate that diazepam injected into the systemic circulation induces the expression of FGF-2 mRNA in discrete regions of the brain. A robust body of evidence supports the idea that the GABA system impacts neuroplasticity [6,31,46,48], and the present results suggest that some of these effects of GABA may be achieved in conjunction with FGF-2. Based on the trophic and growth promoting roles of FGF-2 in astrocytes and neurons, it

[1] 208w Adult male Sprague-Dawley rats (Charles River, MA, USA; 250 -350 g, approximately 3 months of age, n ϭ 60) were injected intraperitoneally (i.p.) with diazepam (RBI, Natick, MA, USA; 10 mg/kg dissolved in glycol). Control animals (n ϭ 12) were treated with an equivalent amount of injection vehicle (glycol). Rats were sacrificed 6 h, 12 h, 24 h, 3 days and 7 days after injection by rapid decapitation to provide fresh brain tissue for nuclease protection assays. The hippocampus, neocortex, and striatum were dissected out, rapidly frozen in dry ice, and stored at Ϫ70°C. Separate groups of rats to be used for in situ hybridization were injected with a lethal dose of Nembutal (75 mg/kg i.p.), then intracardially perfused with 200 ml of phosphate-buffered saline (PBS; pH 7.3) followed by 400 ml of 4% paraformaldehyde in 0.1 M Sorensen buffer (pH 7.3). The brain was removed and stored overnight in 20% sucrose solution in PBS and then rapidly frozen at Ϫ70°C. The hippocampus and striatum were selected for this study based on previous evidence showing that diazepam [46] and FGF-2 [51,52,57] have a protective action on neurons within these regions. The caudal cerebral cortex was selected because it is highly plastic to events associated with neural activity [15,40].

[2] 247w Total cellular RNA was isolated by guanidine thiocyanate extraction according to Chomczynski and Sacchi [12]. Quantifi-cation of total RNA was performed by absorption at 260 nm. Briefly, total RNA (10 g) from specific brain regions was dissolved in 30 l of hybridization buffer (80% formamide, 40 mM PIPES pH 6.4, 400 mM sodium chloride and 1 mM EDTA) containing 5 l of a 32 P-labelled cRNA probe (specific activity 10 6 cpm). After being heated at 85°C for 10 min to denature RNA, the cRNA probe was allowed to anneal the endogenous RNA at 55°C overnight. At the end of the hybridization, the solution was diluted with RNase digestion buffer containing 40 g/ml of RNase A and 2 g/ml of RNase T1, and incubated for 30 min at 37°C. Following proteinase K (240 g/ml) digestion samples were extracted with phenol/chloroform and ethanol precipitated. The pellet containing the RNA:RNA hybrid was resuspended in gel loading buffer, boiled at 85°C, and separated on a 5% polyacrylamide, 7-M urea gel unit. The gel was dried and the protected fragment was visualized by autoradiography on ␤-max-hyperfilm (Amersham, Arlington Heights, IL, USA). In some experiments, hybridization was performed with a 356-base glyceraldehyde 3-phosphate dehydrogenase (GAPDH, Ambion, Texas, USA) cRNA probe to assure that samples used for the assay contained an equivalent amount of RNA. FGF-2 mRNA contents were estimated by measuring the peak densitometry area of the autoradiogram analyzed using NIH Image software, and data was expressed as a percentage of control.

[3] 94w FGF-2 cRNA probes were prepared from cDNA templates coding the rat FGF-2 gene (RObFGF503, kindly provided by Dr. A. Baird [47]). A pBluescript SKϩ plasmid bearing 1 kb fragment was linearized with NcoI and then transcribed using T7 RNA polymerase to generate a 524-base anti-sense strand containing a 477-base rat FGF-2 cRNA (protected fragment). Transcription reactions were performed using a transcription kit (Promega, Madison, WI, USA) and ␣-32 P-CTP (Amersham; 800 Ci/mmol) for nuclease protection assays, or ␣-35 S for in situ hybridization. Quantitative changes were estimated by computer densitometry using NIH Image software.

[4] 223w We performed in situ hybridization to detect changes in FGF-2 mRNA in the cytoarchitecture of the hippocampal formation, as described previously [28]. The hippocampus was selected for this analysis because of its well-defined cytoarchitecture. Briefly, following prehybridization treatment, tissue sections were incubated for 15-20 h at 50°C in hybridization buffer (50% formamide, 0.3 M NaCl, 10 mM PIPES, 10% dextran sulfate, 0.2% ficoll, 0.2% polyvinylpyrrolidone, 0.2% bovine serum albumin, 0.1 mg/ml yeast tRNA, 0.1 mg/ml salmon sperm DNA, and 50 mM dithiothreitol, 0.2% sodium dodecyl sulfate, 10 mM EDTA) containing the 35 S-labeled cRNA probe (300,000 cpm/ml). Slices were incubated in RNAse A (Sigma Chemical Co., St. Louis, MO, USA; 2 U/ml; 10 mM Tris buffer, 1 mM EDTA, 5 mM NaCl) for 30 min at 40°C, and then rinsed at 52°C for 3 h total in decreasing concentrations of saline-sodium citrate (SSC) buffer (2ϫ SSC to end in 0.1ϫ SSC). Slides and 35 S-labeled standards were exposed on ␤-max film and the distribution of hybridization was evaluated after emulsion treatment. After development with D-19 developer (Eastman Kodak, Rochester, NY, USA) the emulsion coated slides were counter-stained with cresyl violet and observed under dark field microscopy. Controls for specificity of hybridization included RNAse A (Sigma; 4 U/ml) pre-hybridization treatment, hybridization with a FGF-2 sense probe or a probe specific for brain-derived neurotrophic factor.

[5] 37w Mean data from nuclease protection assays and in situ hybridization were computed for each group, and compared using ANOVA and Fisher's test (Statview software), as previously described [27]. The results were expressed as a percentage of controls.

[1] 153w Autoradiographic material obtained from in situ hybridization experiments was quantified using computer optical densitometry with NIH Image software. 35 S-labeled material was calibrated relative to 14 C-labeled standards (Amersham), that had previously been cross-calibrated with 35 S-labeled brain paste standards; thus, computer optical density units were converted to equivalent Ci 35 S/g tissue. Ten computer readings were taken from the pyramidal layer of the CA1 and CA2 hippocampal subfields, the stratum oriens and radiatum of the CA1 subfield, and the hilus of the dentate gyrus, at three bilateral coronal planes. Identification of the areas under study was performed directly from the cresyl violet stained sections using published atlas of the rat brain as a guideline [42]. The mean data from at least four coronal planes per animal were averaged, total means were compared across different experimental groups using analysis of variance (ANOVA) and Fisher's test, and data was expressed as percentage of control.

[2] 21w is likely that induction of FGF-2 as a result of GABA A activation can be critical for regulation of CNS plasticity.

[3] 192w Schwartz et al. [46] have recently shown that a single dose of diazepam injected into the systemic circulation results in substantial protection of striatal and hippocampal neurons from ischemic damage. Interestingly, the results of the present study indicate that an identical dose of diazepam to that used by the previous investigators induces the expression of FGF-2 in the hippocampus and striatum. The combined results from these two studies suggest the possibility that FGF-2 may mediate or supplement the neuroprotective effects of diazepam in the hippocampus and striatum. A previous report describes an action of GABA on protein synthesis in the rat brain [9], supporting the hypothesis that other molecules can complement the action of GABA on neuroplasticity. It is interesting that under some circumstances the actions of FGF-2 and GABA may be complementary. For example, FGF-2 stimulates cell proliferation and GABA A receptor expression in neocortical progenitor cells in culture [2], while GABA inhibits DNA synthesis in the same type of cells [35]. These results appear to suggest that GABA can complement the action of FGF-2 in neurogenesis by providing a feedback signal that terminates the cell division initiated by FGF-2 [2].

[4] 60w A separate line of evidence indicates that FGF-2 can also have an impact on the plasticity of GABAergic neurons. For example, it has been reported that FGF-2 affects the differentiation of GABAergic neurons [21]. Furthermore, FGF-2 enhances the survival rate [51] and neurite outgrowth [57] of striatal GABAergic neurons in vitro, and induces glutamic acid decarboxylase in septal neurons [55].

[5] 160w There is much evidence to support an involvement of FGF-2 [18,20,26] and GABA [49,50] in post-traumatic events. Abundant evidence indicates that FGF-2 has a trophic action on striatal and hippocampal neurons, which may provide clues to explain the effects of diazepam on the striatum and hippocampus as described by Schwartz et al. [46] (see above). Studies in cultured cells indicate that FGF-2 enhances survival [52] and neurite outgrowth [57] of striatal GABAergic neurons. It has been shown that FGF-2 prevents the nigrostriatal cell loss caused by MPTP lesions [10,13,34,53], which has stimulated an interest in developing FGF-2 as a therapeutic agent for the treatment of Parkinson's disease. A large amount of evidence obtained using in vitro and in vivo approaches supports a trophic action of FGF-2 on hippocampal cells. For example, FGF-2 promotes the survival of hippocampal cells in culture [11,51], facilitates synaptic transmission in hippocampal slices [1], participates in lesion-induced sprouting [16,24], and attenuates cognitive dysfunction following trauma [38].

[6] 83w The function of GABA as an inhibitory neurotransmitter is achieved by opening chloride channels and hyperpolirizing the membrane potential, thus reducing neuronal activity. After neuro- nal trauma, GABA depolirizes the membrane potential, raising cytoplasmic calcium levels and increasing neuronal activity [50], in a similar fashion to the action played by GABA during neuronal development [7]. The similarities between the actions of GABA in injured neurons and developing neurons have suggested that neuronal injury may promote a recapitulation of GABA's role during ontogeny [49].

[7] 263w It is likely that the regional selectivity displayed by diazepam on FGF-2 regulation is determined by the distribution of specific GABA A receptor subtypes [19,22,23,31,36,43]. Fifteen GABA Areceptor subunits have been described in the mammalian CNS [8,36,41], including the three ␥ subunits that confer the benzodiazepine (BZ) sensitivity. The GABA and BZ binding sites on GABA A receptors are allosterically coupled [33]. The diverse arrangement of GABA A subunits determines the pharmacological properties of the various GABA A receptor subtypes, and each GABA A receptor subtype displays a distinct regional expression. In particular, the hippocampus [5,19] and striatum [30,39] exhibits various GABA A receptors subtypes in neurons and astrocytes. It appears that the regional distribution of BZ receptors in the hippocampal formation described in the literature overlaps with those regions showing elevated FGF-2 expression after diazepam treatment in the present study. It has been reported a high density of BZ receptors in the stratum oriens of the CA1 and CA2 subfields and dentate gyrus [14]. In addition, we have shown previously that diazepam can also affect the expression of FGF-2 at the protein and mRNA levels in the spinal cord [25], where abundant BZ receptors have been described [37]. It is intriguing that the cerebral cortex did not show significant changes in FGF-2 mRNA in response to diazepam treatment, in spite of BZ receptors have been described in this structure [14]. These findings may be related to distinct cellular types involved, e.g., BZ receptors in the cerebral cortex are mainly present in neurons [14] while astrocytes are the main source of FGF-2 [26].

[8] 111w The action of diazepam on FGF-2 may be direct via activation of GABA A receptors, or indirect via activation of other intermediate molecules. A direct action is supported by the presence of GABA A receptors in the brain regions in which diazepam shows an effect on the FGF-2 system. An indirect action is supported by evidence that diazepam affects the expression of other molecules which could affect FGF-2 expression. For example, diazepam inhibits the expression of brain-derived neurotrophic factor (BDNF) [56]. Although an action of BDNF on the FGF-2 system has not been described, it is likely that maintenance of the relative levels among trophic factors is crucial for brain homeostasis.

[9] 80w The present results showing that diazepam can regulate the expression of FGF-2 in select CNS regions suggest a molecular mechanism by which neurotransmitters can impact plasticity of neurons and astrocytes. In particular, our results may stimulate a re-evaluation of the pharmacological potential of diazepam in the CNS, in terms of its possible pharmacological applications to CNS repair. It is critical to determine the interaction between additional neurotransmitter systems and trophic factors to better understand the underlying mechanisms of activity-dependent plasticity.