- Email: [email protected]
Bi-phasic change in BDNF gene expression following antidepressant drug treatment
Neuropharmacology 44 (2003) 903–910 www.elsevier.com/locate/neuropharm
Bi-phasic change in BDNF gene expression following antidepressant drug treatment A.L. Coppell a, Q. Pei a, T.S.C. Zetterstro¨m b,∗ a
University Department of Clinical Pharmacology, University of Oxford, Radcliffe Infirmary, Oxford OX2 6HE, UK b School of Pharmacy, De Montfort University, The Gateway, Leicester LE1 9BH, UK Received 7 November 2002; received in revised form 5 February 2003; accepted 6 February 2003
Abstract The gene for brain derived neurotrophic factor (BDNF) has recently received attention in relation to the therapeutic action of antidepressant treatment. This study aimed to clarify the influence of post drug interval on the effect of acute and repeated treatment with antidepressant drugs on BDNF gene expression in the rat brain. It was found that repeated administration of either the monoamine oxidase inhibitor tranylcypromine (TCP) or 5-hydroxytryptamine (5-HT) re-uptake inhibitors (fluoxetine, paroxetine and sertraline), evoke a bi-phasic and time-dependent effect on BDNF gene expression in the rat hippocampus (especially dentate gyrus). A down-regulation of the BDNF gene was detected at 4 h (TCP and fluoxetine) and an up-regulation at 24 h (TCP, paroxetine, fluoxetine, sertraline) after the last of twice daily injections for 14 days. After a single injection the down-regulation was detected at 4 h (TCP, fluoxetine, paroxetine and sertraline) but BDNF mRNA levels were not altered at 24 h post drug (TCP, fluoxetine and paroxetine). Administration of inhibitors of noradrenaline re-uptake (desipramine and maprotiline) or the atypical antidepressant mianserin had no effect on BDNF mRNA levels at either single (4 h post drug, desipramine) or repeated (24 h post drug, desipramine, maprotiline, mianserin) treatment. The gene expression for NT-3, which is distributed in a high density in the dentate gyrus, was not affected by single or repeated injections of antidepressant drugs (TCP, fluoxetine, paroxetine, sertraline, desipramine, maprotiline or mianserin) at 4 or 24 h post drug. In conclusion, these data show that the effect of antidepressant drugs on BDNF gene expression may be more complex and less widespread across treatments than previously thought. Thus, in this study drugs interacting with the central 5-HT system altered BDNF expression but the effect was bi-phasic over the 24 h post drug period. 2003 Published by Elsevier Science Ltd. Keywords: Brain-derived neurotrophic factor (BDNF); Antidepressants; Hippocampus
1. Introduction Antidepressant drugs require a course of treatment of several weeks to achieve full therapeutic effect. Although, the reason for this remains unresolved, the explanation is thought to lie in the induction of neuroadaptive changes in gene expression following an increase in brain monoamine function. Thus, antidepressant treatments (drugs and electroconvulsive shocks) increase brain monoamine function and activate second messenger signal transduction mechanisms resulting in changes of gene expression (Duman, 1998; Lesch, 2001). One gene that has recently received considerable
∗
Corresponding author. Fax: +1-44-116-2577135. E-mail address: [email protected] (T.S.C. Zetterstro¨m).
0028-3908/03/$ - see front matter 2003 Published by Elsevier Science Ltd. doi:10.1016/S0028-3908(03)00077-7
attention in relation to antidepressant action is that encoding for brain derived neurotrophic factor (BDNF; see Altar, 1999). For example, stress reduces BDNF mRNA levels in rat hippocampus (Smith et al., 1995), an effect blocked by antidepressant treatment (Nibuya et al., 1995). Moreover, repeated administration of electroconvulsive shock (ECS) as well as antidepressant drugs increase BDNF gene expression (Nibuya et al, 1995, 1996; Zetterstro¨m et al, 1998a, b; Russo-Neustadt et al., 1999, 2000). Further, administration of BDNF produces antidepressant effects in behavioural models of depression (Siuciak et al., 1997; Shirayama et al., 2002). There are however, some differences in BDNF mRNA responses after antidepressant drugs among previously published studies. Thus, some studies demonstrate, after a wide range of drugs (5-HT re-uptake inhibitors, tricyclics monoamine oxidase inhibitors and atypical
904
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
antidepressants) robust increases in BDNF mRNA levels in several parts of the brain implicated in depressive disorders, e.g. cortical and hippocampal brain regions (Nibuya et al., 1995, 1996). While, others have shown this effect to be less pronounced, anatomically wide spread and not common to all antidepressant drugs (Russo-Neustadt et al., 1999, 2000; Zetterstro¨ m et al., 1998b). Interpretation of the increase in BDNF gene expression following chronic antidepressant drug treatment is complicated by a recent report that acute administration of drugs which increase extracellular 5-HT levels, reduces BDNF gene expression in hippocampus (Zetterstro¨ m et al., 1999). This effect was observed within 2–3 h of administering: 5-HT releasing agents, a 5-HT precursor load or antidepressant drugs such as a selective 5-HT re-uptake inhibitor or a monoamine oxidase inhibitor. Whether this acute inhibitory effect on BDNF mRNA level is common to most antidepressant drugs and remains after chronic treatment needs clarification. To clarify the influence of post drug interval on BDNF gene expression, this study investigated the effect of single and repeated administration of antidepressant drugs on rat brain BDNF gene expression at 4 and 24 h after the last injection. A range of antidepressant drugs were tested, specifically inhibitors of 5-HT (fluoxetine, paroxetine, sertraline) and noradrenaline (desipramine, maprotiline) re-uptake, a monoamine oxidase inhibitor tranylcypromine (TCP) and the atypical antidepressant, mianserin. Drug effects on gene expression of neurotrophin NT-3 were also tested in some experiments. A preliminary account of these experiments was presented previously (Coppell and Zetterstro¨ m, 2000).
maprotiline (Tocris, UK) and the atypical antidepressant mianserin (Tocris, UK), were dissolved in normal saline. Paroxetine (GlaxoSmithKline, UK) and sertraline (Pfizer, UK) were dissolved in sterile water and 2% Tween-80 in sterile water, respectively. Drugs were injected at either 5 mg/kg i.p. (paroxetine, sertraline and tranylcypromine) or 10 mg/kg i.p. (fluoxetine, desipramine, maprotiline and mianserin). 2.2.2. Acute treatment For acute treatment, rats were given single injections of antidepressant drugs at the same doses as in the chronic experiments (see Section 2.2.1), or the appropriate vehicles. Rats were killed 4 h (fluoxetine, paroxetine, sertraline, TCP and desipramine) or 24 h (fluoxetine, paroxetine and TCP) after single injections. 2.3. ISHH procedure
All procedures were carried out in accordance with the UK Animals Scientific Procedures Act, (1986). Male Sprague–Dawley rats (250–280 g, Harlan-Olac Ltd, Bicester, UK) were housed in cages (4–6 rats/cage) under a 12 h light/dark cycle in a temperature controlled environment with free access to food and water.
Brain sections (12 µm) were cut on a cryostat, thawmounted onto gelatine-subbed slides and pre-treated for ISHH using a standard protocol as described previously (Pei et al., 1997). Oligonucleotides complementary to mRNAs encoding genes for all forms of BDNF (5’GGT CTC GTA GAA ATA TTG CTT CAG TTG GCC TTT TGA TAC CGG GAC 3’) or NT-3 (5’GCG TTT CCT CCG TGG TGA TGT TCT ATT GGT TAC CAC CGG GTT GCC CAC 3’) were 3’-tail labelled with 35S-dATP with terminal deoxynucleotide transferase. The labelled oligonucleotide probes (specific activity ⬎ 109 cpm/µg) was added to each section (1 × 106 cpm/section) in hybridisation buffer as previously described (Pei et al., 1997). After incubation in humid chambers (containing 50% formamide in 4 × SSC) at 35 °C (BDNF probe) or 42 °C (NT-3 probe) for 14–16 h, slides were washed in 1 × SSC (BDNF probe) or 0.5 × SSC (NT-3 probe) buffer at 55 °C for 3 × 20 min followed by 2 × 60 min at room temperature. Sections were then air-dried and exposed to autoradiography film (Hyperfilm b-max, Amersham, Buckinghamshire, UK) for 3 weeks at room temperature. Controls included the use of oligonucleotides in the sense orientation and displacement with unlabelled probes. Search of Genebank data-base using the BLAST programs revealed no significant homology of the nucleotide sequences with other previously characterised genes.
2.2. Drugs and experimental protocol
2.4. Data analysis and statistics
2.2.1. Chronic antidepressant treatment Rats were injected with antidepressant drugs or the corresponding vehicles, twice daily for 14 days and killed 4 h (fluoxetine and TCP) or 24 h (fluoxetine, paroxetine, sertraline, TCP, desipramine, maprotiline and mianserin) after the last injection. Fluoxetine (Eli. Lilly, UK) desipramine HCl (Sigma), TCP (Sigma),
The relative abundance of mRNAs was determined by densitometric quantification of autoradiograms (MCID, St Catherine, Canada). Grey density values were calibrated to 14C tissue equivalents (Amersham, UK) and corrected for 35S decay. Optical densities were converted to nCi/g tissue, and mRNA abundance in drug treated groups was expressed as a percentage of respective con-
2. Materials and methods 2.1. Animals
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
905
trols and presented as mean ± SEM values. Statistical analysis of mRNA abundance was performed using Student’s t-test for comparisons between two groups (vehicle control vs. drug treatment). Multiple group comparisons were analysed using one-way ANOVA with Dunnett’s t-test.
3. Results 3.1. Effect of antidepressant drugs on BDNF gene expression at 4 h after single or repeated injection Single injection of the 5-HT re-uptake inhibitors, fluoxetine (10 mg/kg, i.p.), paroxetine (5 mg/kg, i.p.) and sertraline (5 mg/kg, i.p.), as well as the monoamine oxidase inhibitor TCP (5 mg/kg, i.p.), reduced BDNF mRNA in the hippocampus when measured 4 h post drug. The dentate gyrus showed the most marked and consistent reduction (⫺26–30%) in BDNF gene expression (Fig. 1A). Fluoxetine and sertraline, but not paroxetine or TCP also decreased BDNF mRNA in the CA1 and CA3 subfields of the hippocampus (Table 1). In contrast, a single injection of the noradrenaline reuptake inhibitor, desipramine (10 mg/kg, i.p.), had no significant effect on BDNF mRNA expression in any of the hippocampal subfields at 4 h post-drug (Fig. 1A and Table 1). The effect of fluoxetine and TCP was also tested 4 h after the last of twice daily injections for 14 days. Both drugs reduced BDNF mRNA expression the dentate gyrus at 4 h post-drug (Fig. 1B and Fig. 2A and B). Neither single nor repeated injections of any of the drugs tested influenced BDNF mRNA in cortical regions (piriform, parietal and cingulate of frontal cortex) at 4 h after the injection (data not shown). 3.2. Effect of antidepressant drugs on BDNF mRNA levels 24 h after single or repeated injections Single injection of the 5-HT uptake inhibitor, fluoxetine (10 mg/kg, i.p.) and paroxetine (5 mg/kg, i.p.) or the monoamine oxidase inhibitor TCP (5 mg/kg, i.p.) had no effect on BDNF mRNA levels when measured 24 h post drug (Fig. 3A). In contrast, repeated administration (twice daily for 14 days) of the 5-HT uptake inhibitor, fluoxetine, paroxetine or sertraline (5 mg/kg, i.p.) and the monoamine oxidase inhibitor TCP significantly increased BDNF mRNA levels in the dentate gyrus of the hippocampus when measured 24 h after the last injection. This increase ranged from +26% for fluoxetine to +44% for sertraline, compared to the corresponding vehicle injected control groups (Fig. 3B). The fluoxetine effect is illustrated as representative autoradiograms in Fig. 2C and D. After repeated TCP there was also an increase of
Fig. 1. Effects of antidepressant drugs on BDNF gene expression in the rat dentate gyrus of the hippocampus at (A) 4 h after a single injection (fluoxetine, paroxetine, sertraline, tranylcypromine and desipramine) and at (B) 4 h after the last injection following repeated administration for 2 weeks (fluoxetine and TCP). Data are expressed as % of controls and each column represents a mean ± SEM value from 5–6 rats ∗∗∗p ⬍ 0.001; ∗∗p ⬍ 0.01; ∗p ⬍ 0.05 vs. corresponding vehicle injected controls (Student’s t-test).
BDNF mRNA in the CA1 of the hippocampus (+ 33 ± 11%, n = 6, p ⬍ 0.05 vs. controls) when measured 24 h post drug. Neither the 5-HT re-uptake inhibitors nor TCP altered BDNF mRNA in any other subfields of the hippocampus or in any cortical regions (data not shown) at 24 h after the injections. The noradrenaline reuptake inhibitors, desipramine and maprotiline (10 mg/kg, i.p.) and the atypical antidepressant, mianserin (10 mg/kg, i.p.) did not alter BDNF mRNA levels in any of the brain regions measured 24 h after the last injection (Fig. 4). 3.3. Effects of antidepressant drugs on NT-3 mRNA levels The rat brain distribution of NT-3 mRNA was in agreement with previous reports (see Smith et al., 1995).
906
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
Table 1 Effects of antidepressant drugs on BDNF mRNA expression in the CA1 and CA3 of rat hippocampus, 4 h following a single injection. Data are given as mean ± SEM (n = 5–6) and expressed as % of controls CA1
CA3
Saline Fluoxetine
100 ± 1.2 80.4 ± 2.7a
100±2.2 76.9 ± 3.1a
Vehicle Paroxetine
100 ± 4.5 76.3 ± 10.3
100 ± 4.3 84.0 ± 5.8
Vehicle Sertraline
100 ± 3.4 87.4 ± 3.0c
100 ± 2.7 79.2 ± 1.8b
Saline Desipramine
100 ± 3.4 104.3 ± 5.3
100 ± 2.7 100.3 ± 2.3
Saline Tranylcypromine
100 ± 9.8 88.3 ± 6.4
100 ± 14.6 91.0 ± 4.8
a p⬍0.001 vs. corresponding vehicle injected controls (Student’s t-test). b p⬍0.01 vs. corresponding vehicle injected controls (Student’s ttest). c p⬍0.05 vs. corresponding vehicle injected controls (Student’s ttest).
Thus, high levels of NT-3 mRNA were detected in the dentate gyrus with lower levels in the CA2 and the medial CA1 (mCA1). NT-3 gene expression was unchanged in hippocampus after either acute or chronic antidepressant treatment (fluoxetine, paroxetine, sertraline, TCP, desipramine, maprotiline or mianserin) at either 4 or 24 h (data not shown). 4. Discussion Here we report the effect of acute and repeated administration of antidepressant drugs on BDNF gene expression in rat brain. The main finding is that drugs which inhibit either the 5-HT transporter (fluoxetine, paroxetine, sertraline) or monoamine oxidase (TCP), influence BDNF gene expression in the rat hippocampus in a bi-phasic manner following repeated injections. Thus, a down-regulation in the abundance of BDNF mRNA was detected at 4 h, and an up-regulation was detected at 24 h, after the last injection of twice daily injections for two weeks. After a single injection of these drugs a decrease in BDNF mRNA was also detected at 4 h but there was no change at 24 h post drug. Repeated administration of drugs which inhibit noradrenaline reuptake (desipramine and maprotiline), or an atypical
Fig. 2. Representative autoradiogram, showing the effect of repeated fluoxetine administration (10 mg/kg i.p. twice daily for 14 days), 4 h (A, B) and 24 h (C, D) after the last injection on rat BDNF mRNA in hippocampus. Dentate gyrus (DG).
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
907
Fig. 4. Effects of desipramine, maprotiline and mianserine on BDNF gene expression in the dentate gyrus of rat hippocampus at 24 h after the last injection following a 2 weeks treatment. Data are expressed as % of controls and each column represents a mean ± SEM value from 5–6 rats.
Fig. 3. Effects of antidepressant drugs on BDNF gene expression in the rat dentate gyrus of the hippocampus at 24 h (A) after a single injection (fluoxetine, paroxetine, tranylcypromine) and at 24 h (B) after the last injection following repeated administration for 2 weeks (fluoxetine, paroxetine, sertraline and tranylcypromine). Data are expressed as % of controls and each column represents a mean ± SEM value from 5–6 rats). ∗∗∗p ⬍ 0.001; ∗∗p ⬍ 0.01; ∗p ⬍ 0.05 vs. corresponding vehicle injected controls (Student’s t-test).
antidepressant drug (mianserin) did not change BDNF gene expression at neither 24 h (repeated administration) nor 4 h (single administration, desipramine only) post drug. Gene expression for NT-3 was not affected by any of these treatments. 4.1. Effect of antidepressant drugs on BDNF gene expression 24 h after administration Repeated injection of either a monoamine oxidase inhibitor (TCP) or 5-HT re-uptake inhibitors (fluoxetine, paroxetine or sertraline) increased BDNF mRNA
expression in the hippocampus 24 h after the last injection. This effect was restricted to the dentate gyrus, with the exception of TCP, which also caused an increase in the CA1. Importantly, these increases in BDNF mRNA were not detected after a single injection, which is in agreement with previous studies and indicates the occurrence of neuroadaptive effects (Nibuya et al., 1995, 1996). In contrast, both single and repeated ECS increase hippocampal BDNF mRNA at 24 h after treatment, although the effect by repeated ECS is longer lasting (Zetterstro¨ m et al., 1998a). Repeated injection with noradrenaline uptake inhibitors (desipramine and maprotiline) or an atypical antidepressant (mianserin) did not increase BDNF mRNA levels in hippocampus 24 h after the injection. This is in contrast to a previous study by Nibuya et al. (1996), which found an up-regulation in hippocampal BDNF gene expression (18 h after the last injection) common to all antidepressant drugs tested including, a monoamine oxidase inhibitor (TCP), tricyclics (desipramine, imipramine), 5-HT re-uptake inhibitors (fluoxetine, sertraline) and an atypical antidepressant (mianserin). However, a more recent study found that repeated administration with TCP but not the tricyclic imipramine significantly increased BDNF mRNA levels in dentate gyrus (Russo-Neustadt et al., 1999). The increase in BDNF mRNA levels in the present study, 24 h after the last of repeated injections was present in the dentate gyrus, but apart from TCP not in other hippocampal subfields. In comparison, previous studies by Nibuya et al. (1995, 1996) demonstrated overall increases in BDNF gene expression in dentate gyrus,
908
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
CA1, CA3 subfields of the hippocampus, as well as in the frontal cortex. The reason for this discrepancy is not clear, but some differences in drug doses and regimes might contribute. For example, twice daily injections for 2 weeks were used here, while once daily injections for 3 weeks were used by Nibuya et al. While most drugs were given in similar doses this was however, not the case for TCP: 5 mg/kg in present study and 7.5 mg/kg for 7 days then 10 mg/kg for 14 days in studies by Nibuya et al and sertraline: 5 mg/kg and 10 mg/kg respectively. In addition, the dose of the noradrenaline reuptake inhibitor desipramine was lower in the current study (10 mg/kg) compared to previous studies by Nibuya et al. (15 mg/kg). It is unlikely that the dose of desipramine used in the present study is too low, since previous studies have shown that this dose is both pharmacologically active and achieves plasma drug levels in the therapeutic range (Ainsworth et al., 1998). The explanation for the increase in BDNF gene expression following chronic, but not acute administration of 5-HT re-uptake inhibitors or TCP is not clear. However, these drugs cause greater or more prolonged increases in extracellular levels of 5-HT following repeated compared to acute administration (Ferrer and Artigas, 1994; Hjort et al., 2000). A prolonged increase of extracellular 5-HT levels would result in long-term activation of second messenger signal transduction systems such as the cAMP/protein kinase A (PKA, 5HT4,6,7), as well as diacylglycerol (DAG)/protein kinase C (PKC, 5-HT2). Gene expression for the trancription factor cAMP response element binding protein (CREB) is up-regulated by both of these pathways and has been linked to antidepressant drug induced elevations of the BDNF gene (Duman, 1998). It is possible that increased expression of the CREB gene accounts for the increased BDNF expression seen after chronic treatment with 5HT uptake inhibitors or TCP in the present study. In this context, it is however, difficult to explain the lack of change in BDNF mRNA levels in rats treated with desipramine and maprotiline in the present study. Since, adrenergic receptors of b and a1 activate cAMP and DAG transduction systems, respectively, these drugs might also be expected to induce the CREB gene. However, chronic administration of noradrenergic antidepressants is well known to cause desensitisation of b receptors (see Pryor and Sulser, 1992). Interestingly, a recent study shows that noradrenergic antidepressant drugs (desipramine and reboxetine) decreases nuclear CREBP (regulator of CREB-CRE directed gene transcription), suggesting that noradrenergic drugs do not stimulate CRE-mediated gene expression (e.g. BDNF, Manier et al., 2002). Interestingly, there are previous reports demonstrating glutamate-dependent induction of the BDNF gene (Zafra et al., 1991; Wetmore et al., 1994; Zetterstro¨ m et al., 1998c; Mackowiak et al., 2002). In this respect, it is
possible that the enhanced BDNF expression at 24 h following repeated administration by 5-HT uptake inhibitors or TCP is due to increased activity of glutamatergic input into the dentate gyrus via the perforant path of the entorhinal cortex (Amaral and Witter, 1995). In support of this idea, 5-HT2 receptor-stimulation increases glutamate activity in the neocortex (Aghajanian and Marek, 1997). 4.2. Effect of antidepressant drugs on BDNF gene expression 4 h after administration Repeated and single administration of TCP or the 5HT re-uptake inhibitors decreased BDNF gene expression in the hippocampus at 4 h after the last injection, while the noradrenaline re-uptake inhibitor desipramine had no effect. The decrease of BDNF gene expression is likely to be mediated by different mechanisms to the increase, as the latter effect was seen only 24 h after repeated administration. Previous microdialysis studies have shown that single injections of TCP or 5-HT uptake inhibitors increase extracellular levels of 5-HT in rat hippocampus (Gartside et al., 1995) and it is possible that this causes the down regulation in BDNF gene expression. Consistent with this, we found that 5-HT precursor loading and a 5-HT releasing agent drug decrease BDNF gene expression in rat hippocampus 2-3 h after the injection (Zetterstro¨ m et al., 1999). Also, a single injection of fluoxetine does not reduce BDNF mRNA levels in 5-HT lesioned rats (Coppell and Zetterstro¨ m, 1999). Inconsistent with the present study, a previous report failed to detect a decrease in BDNF gene expression at 3 h after either single (no effect) or 3 weeks administration (an increase), (Nibuya et al., 1995). It is possible that some differences in doses and regimes contribute to these differences (see previous section). It should however, be noted that the acute dose of TCP (10 mg/kg) used in the Nibuya study is expected to increase extracellular 5-HT levels sufficiently to stimulate 5-HT receptors (Ferrer and Artigas, 1994). Including 5-HT2 receptors which have previously been shown to reduce BDNF gene expression upon activation (Vaidya et al., 1997). Alternatively the differences could be accounted for by the use of a Northern blotting technique in the Nibuya study, probing the entire hippocampus for BDNF mRNA changes, which could fail in detecting regional differences within the hippocampus. In comparison, the ISHH techniques used in the present study yields better anatomical resolution. Thus, the most consistent and pronounced decrease was detected in the dentate gyrus representing only a small part of the entire hippocampus. Electrophysiological studies indicate that excitatory 5HT receptors are localized on GABA-ergic inter-neurons in the hippocampus (Freund et al., 1990). A possible explanation of our data is that excitatory 5-HT receptors
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
(5-HT2,4,6or7) on GABA-ergic interneurons in the polymorphic layer of the hippocampus, increase the release of GABA, which inhibits glutamatergic neurons and leads to a decrease of BDNF gene expression. In support of this idea, we have recently shown that the GABAB receptor agonist baclophen inhibits hippocampal BDNF gene expression with a similar magnitude to paroxetine and TCP (Khundakar et al., 2002a). 4.3. Bi-phasic BDNF gene expression following repeated antidepressant drug treatment The present data indicate that observed changes in the expression of the BDNF gene following repeated administration of 5-HT re-uptake inhibitors (fluoxetine, paroxetine and sertraline) and a monoamine oxidase inhibitor (TCP) varies according to the elapsed time between the last injection and the BDNF mRNA measurements. The rat BDNF gene gives rise to five separate exons (I–V) which have distinct promoters and a different cellular distribution (Timmusk et al., 1993). All BDNF transcripts contain the coding region for the mature BDNF protein and therefore, should be of equal functional importance. In the present study we used a probe that binds to the coding region for the BDNF protein (exon V) that is common to all transcripts. It is possible that the bi-phasic change in the BDNF gene following chronic treatment with 5-HT re-uptake inhibitors or TCP indicates differences in the expression of individual BDNF exons. In support of this idea, we have recently shown that an acute injection of paroxetine or TCP down regulates expression of exon IV but not exon I, 4 h after the injection (Khundakar and Zetterstro¨ m, 2002b). Interestingly, exon IV belongs to a group of exons that share properties with early–immediate genes and therefore would be expected to give rise to fast and transient changes (Timmusk et al., 1993). In comparison, the upregulation of BDNF gene expression at 24 h following repeated treatment with TCP has been shown to be particularly pronounced for exon I ( Russo-Neustadt et al., 2000). Exon I has previously been shown to require ongoing protein synthesis for its transcription and therefore would be slower in response, compared to the exons sharing properties with IEGs. The concept of differential expression of BDNF mRNAs following drug treatment is not unusal. For example, kainic acid induced seizure activity has been shown to elicit differential expression of BDNF mRNAs in the rat brain (Timmusk et al., 1993).
5. Summary and functional implications In summary, the main finding of the present study is that repeated treatment with a monoamine oxidase inhibitor or 5-HT re-uptake inhibitors, evokes a time-
909
dependent, bi-phasic change in BDNF gene expression in the hippocampus (especially dentate gyrus). These antidepressants decrease BDNF mRNA 4 h after the injection but increase it 24 h after the last injection. The effect at 4 h also occurred following a single injection. None of the above effects were seen with drugs blocking noradrenaline re-uptake or with an atypical antidepressant. Although recent studies demonstrate neurotrophic effects following chronic administration of antidepressant drugs (Duman et al., 2001; Norrholm and Quimet, 2001), future studies investigating changes in the BDNF protein after antidepressant drugs are crucial for these effects to be linked to increased BDNF expression.
References Ainsworth, K., Smith, S.E., Zetterstro¨ m, T.S.C., Pei, Q., Franklin, M., Sharp, T., 1998. Effect of antidepressant drugs on dopamine D1 and D2 receptor expression and dopamine release in the nucleus accumbens of the rat. Psychopharmacology 140, 470–477. Amaral, D.G., Witter, M.P., 1995. Hippocampal formation. In: Paxinos, P. (Ed.), The Rat Nervous System, 2nd edn. Academic Press, pp. 443–486. Aghajanian, G.K., Marek, G.J., 1997. Serotonin induces excitatory postsynaptic potentials in apical dendrites of neocortical pyramidal cells. Neuropharmacology 36, 589–599. Altar, C., 1999. Neurotrophins and depression. Trends Pharmacol. Sci. 20, 59–61. Coppell, A.L., Zetterstro¨ m, T.S.C., 1999. 5-HT mediated regulation of BDNF gene expression. Fund Clin Pharmacology 38 (Suppl), 107. Coppell, A.L., Zetterstro¨ m, T.S.C., 2000. Biphasic expression of brainderived neurotrophic factor gene in rat hippocampus following fluoxetine treatment. BR J. Pharmacol. Proc. 129 (Suppl), 134. Duman, R.S., 1998. Novel therapeutic approaches beyond the serotonin receptor. Biol Psychiatry 44, 324–335. Duman, R.S., Nakagawa, S., Malberg, J., 2001. Regulation of adult neurogenesis by antidepressant treatment. Neuropsychopharmacology 25, 836–844. Ferrer, A., Artigas, F., 1994. Effects of single and chronic treatment with tranylcypromine on extracellular serotonin in rat brain. Eur J Pharmacol 263, 227–234. Freund, T.F., Gulyas, A.I., Acsady, L., Gores, T., Totu, K., 1990. Serotonergic control of the hippocampus via local inhibitory interneurons. Proc Natl Acad Sci USA 87, 8501–8505. Gartside, S.E., Umbers, V., Hajos, M., Sharp, T., 1995. Interaction between a selective 5-HT1A receptor antagonist and an SSRI in vivo: effects on 5-HT cell firing and extracellular 5-HT. Br J Pharmacol 115, 1064–1070. Hjort, S., Bengtson, H.J., Kullberg, A., Carlzon, D., Peilot, H., Auerbach, S.B., 2000. Serotonin autoreceptor function and antidepressant drug action. J Psychopharmacol 14, 177–185. Khundakar, A.A., Mazumdar, A., Zetterstro¨ m, T.S.C., 2002a. Gene expression for brain derived neurotrophic factor in rat hippocampus is reduced by baclophen, tranylcypromine and paroxetine but not flunitrazepam. Br J Pharmacol 136 (Proc Suppl), 57. Khundakar, A.A., Zetterstro¨ m, T.S.C., 2002b. Differential expression of BDNF exons in rat brain after systemic administration of paroxetine and tranylcypromine. Br J Pharmacol 137 (Proc Suppl), 122. Lesch, K.P., 2001. Serotonergic gene expression and depression: implications for developing novel antidepressants. J Affective Disorders 62, 57–76. Mackowiak, M., O’Neill, M., Hicks, C., Bleakman, D., Skolnick, P.,
910
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
2002. An AMPA receptor potentiator modulates hippocampal expression of BDNF: an in vivo study. Neuropharmacol 43, 1–10. Manier, D.H., Shelton, R.C., Sulser, F., 2002. Noradrenergic antidepressants: does chronic treatment increase or decrease nucleaar CREB-P? J Neural Trans 109, 91–99. Nibuya, M., Morinobu, S., Duman, R.S., 1995. Regulation of BDNF and trkB mRNA in rat brain by chronic electroconvulsive seizure and antidepressant drug treatment. J Neurosci 15, 7539–7547. Nibuya, M., Nestler, E.J., Duman, R.S., 1996. Chronic antidepressant administration increases the expression of camp response element binding protein (CREB) in rat hippocampus. J Neurosci 16, 2365–2372. Norrholm, S.D., Quimet, C.C., 2001. Altered dendritic spine density in animal models of depression and in response to antidepressant treatment. Synapse 42, 151–163. Pei, Q., Burnet, P.W.J., Grahame-Smith, D.G., Zetterstro¨ m, T.S.C., 1997. Differential effects of acute and chronic electroconvulsive shock on the abundance of messanger RNAs for voltage-dependent potassium channel subunits in the rat brain. Neurosci 78, 343–350. Pryor, J.C., Sulser, F., 1992. Evolution of monoamine hypothesis of depression. In: Horton, R.W., Katona, C. (Eds.), Biological Aspects of Affective Disorders. Academic Press, London, pp. 77–94. Russo-Neustadt, A.A., Beard, R.C., Cottman, C.W., 1999. Exercise, antidepressant medications, and enhanced brain derived neurotrophic factor expression. Neuropsychopharmacol 21, 679–682. Russo-Neustadt, A.A., Beard, R.C., Huang, Y.M., Cottman, 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. Shirayama, Y., Chen, A.C., Nagakawa, S., Russell, D.S., Duman, R.S., 2002. Brain-derived neurotrophic factor produces antidepressant effects in behavioural models of depression. J Neurosci 22, 3251–3261. Siuciak, J.A., Lewis, D.R., Wiegand, S.J., Lindsay, R., 1997. Antidepressant like effect of brain derived neurotrophic factor (BDNF). Pharmacol Biochem Behav 56, 131–137.
Smith, M.A., Makino, S., Kvetnansky, R., Post, R.M., 1995. Stress and glucocorticoids affect the expression of brain-derived neurotrophic factor and neurotrophin-3 mRNA in the hippocampus. J Neurosci 15, 1768–1777. 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. Vaidya, V.A., Marek, G.J., Aghajanian, G.K., Duman, R.S., 1997. 5HT2 receptor-mediated regulation of brain-derived neurotrophic factor mRNA in the hippocampus and the neocortex. J Neurosci 17, 2785–2795. Wetmore, C., Olson, L., Bean, A.J., 1994. Regulation of brain-derived neurotrophic factor (BDNF) expression and release from hippocampal neurons is mediated by non-NMDA type glutamate receptors. J Neurosci 14, 1688–1700. Zafra, F., Castren, E., Thoenen, H., Lindholm, D., 1991. Interplay between glutamate and g-aminobutyric acid transmitter systems in the physiological regulation of brain-derived neurotrophic factor and nerve growth factor synthesis in hippocampal neurons. Proc Natl Acad Soc USA 88, 10037–10041. Zetterstro¨ m, T.S.C., Pei, Q., Grahame-Smith, D.G., 1998a. Repeated electroconvulsive shock extends the duration of enhanced gene expression for BDNF compared to a single administration. Mol Brain Res 57, 106–110. Zetterstro¨ m, T.S.C., Pei, Q., Ainsworth, K., Grahame-Smith, D.G., 1998b. Effects of antidepressant treatments on BDNF gene expression in rat brain. Br J Pharmacol 123 (Proc Suppl), 211. Zetterstro¨ m, T.S.C., McQuade, R., Pei, Q., 1998c. Pretreatment with the non-NMDA antagonist DNQX attenuates ECS-induced gene expression for BDNF in rat brain. Br J Pharmacol 124 (Proc Suppl), 69. Zetterstro¨ m, T.S.C., Pei, Q., Madhav, T.R., Coppell, A.L., GrahameSmith, D.G., 1999. Manipulations of brain 5-HT levels affect gene expression for BDNF in rat brain. Neuropharmacology 38, 1063– 1073.
Bi-phasic change in BDNF gene expression following antidepressant drug treatment A.L. Coppell a, Q. Pei a, T.S.C. Zetterstro¨m b,∗ a
University Department of Clinical Pharmacology, University of Oxford, Radcliffe Infirmary, Oxford OX2 6HE, UK b School of Pharmacy, De Montfort University, The Gateway, Leicester LE1 9BH, UK Received 7 November 2002; received in revised form 5 February 2003; accepted 6 February 2003
Abstract The gene for brain derived neurotrophic factor (BDNF) has recently received attention in relation to the therapeutic action of antidepressant treatment. This study aimed to clarify the influence of post drug interval on the effect of acute and repeated treatment with antidepressant drugs on BDNF gene expression in the rat brain. It was found that repeated administration of either the monoamine oxidase inhibitor tranylcypromine (TCP) or 5-hydroxytryptamine (5-HT) re-uptake inhibitors (fluoxetine, paroxetine and sertraline), evoke a bi-phasic and time-dependent effect on BDNF gene expression in the rat hippocampus (especially dentate gyrus). A down-regulation of the BDNF gene was detected at 4 h (TCP and fluoxetine) and an up-regulation at 24 h (TCP, paroxetine, fluoxetine, sertraline) after the last of twice daily injections for 14 days. After a single injection the down-regulation was detected at 4 h (TCP, fluoxetine, paroxetine and sertraline) but BDNF mRNA levels were not altered at 24 h post drug (TCP, fluoxetine and paroxetine). Administration of inhibitors of noradrenaline re-uptake (desipramine and maprotiline) or the atypical antidepressant mianserin had no effect on BDNF mRNA levels at either single (4 h post drug, desipramine) or repeated (24 h post drug, desipramine, maprotiline, mianserin) treatment. The gene expression for NT-3, which is distributed in a high density in the dentate gyrus, was not affected by single or repeated injections of antidepressant drugs (TCP, fluoxetine, paroxetine, sertraline, desipramine, maprotiline or mianserin) at 4 or 24 h post drug. In conclusion, these data show that the effect of antidepressant drugs on BDNF gene expression may be more complex and less widespread across treatments than previously thought. Thus, in this study drugs interacting with the central 5-HT system altered BDNF expression but the effect was bi-phasic over the 24 h post drug period. 2003 Published by Elsevier Science Ltd. Keywords: Brain-derived neurotrophic factor (BDNF); Antidepressants; Hippocampus
1. Introduction Antidepressant drugs require a course of treatment of several weeks to achieve full therapeutic effect. Although, the reason for this remains unresolved, the explanation is thought to lie in the induction of neuroadaptive changes in gene expression following an increase in brain monoamine function. Thus, antidepressant treatments (drugs and electroconvulsive shocks) increase brain monoamine function and activate second messenger signal transduction mechanisms resulting in changes of gene expression (Duman, 1998; Lesch, 2001). One gene that has recently received considerable
∗
Corresponding author. Fax: +1-44-116-2577135. E-mail address: [email protected] (T.S.C. Zetterstro¨m).
0028-3908/03/$ - see front matter 2003 Published by Elsevier Science Ltd. doi:10.1016/S0028-3908(03)00077-7
attention in relation to antidepressant action is that encoding for brain derived neurotrophic factor (BDNF; see Altar, 1999). For example, stress reduces BDNF mRNA levels in rat hippocampus (Smith et al., 1995), an effect blocked by antidepressant treatment (Nibuya et al., 1995). Moreover, repeated administration of electroconvulsive shock (ECS) as well as antidepressant drugs increase BDNF gene expression (Nibuya et al, 1995, 1996; Zetterstro¨m et al, 1998a, b; Russo-Neustadt et al., 1999, 2000). Further, administration of BDNF produces antidepressant effects in behavioural models of depression (Siuciak et al., 1997; Shirayama et al., 2002). There are however, some differences in BDNF mRNA responses after antidepressant drugs among previously published studies. Thus, some studies demonstrate, after a wide range of drugs (5-HT re-uptake inhibitors, tricyclics monoamine oxidase inhibitors and atypical
904
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
antidepressants) robust increases in BDNF mRNA levels in several parts of the brain implicated in depressive disorders, e.g. cortical and hippocampal brain regions (Nibuya et al., 1995, 1996). While, others have shown this effect to be less pronounced, anatomically wide spread and not common to all antidepressant drugs (Russo-Neustadt et al., 1999, 2000; Zetterstro¨ m et al., 1998b). Interpretation of the increase in BDNF gene expression following chronic antidepressant drug treatment is complicated by a recent report that acute administration of drugs which increase extracellular 5-HT levels, reduces BDNF gene expression in hippocampus (Zetterstro¨ m et al., 1999). This effect was observed within 2–3 h of administering: 5-HT releasing agents, a 5-HT precursor load or antidepressant drugs such as a selective 5-HT re-uptake inhibitor or a monoamine oxidase inhibitor. Whether this acute inhibitory effect on BDNF mRNA level is common to most antidepressant drugs and remains after chronic treatment needs clarification. To clarify the influence of post drug interval on BDNF gene expression, this study investigated the effect of single and repeated administration of antidepressant drugs on rat brain BDNF gene expression at 4 and 24 h after the last injection. A range of antidepressant drugs were tested, specifically inhibitors of 5-HT (fluoxetine, paroxetine, sertraline) and noradrenaline (desipramine, maprotiline) re-uptake, a monoamine oxidase inhibitor tranylcypromine (TCP) and the atypical antidepressant, mianserin. Drug effects on gene expression of neurotrophin NT-3 were also tested in some experiments. A preliminary account of these experiments was presented previously (Coppell and Zetterstro¨ m, 2000).
maprotiline (Tocris, UK) and the atypical antidepressant mianserin (Tocris, UK), were dissolved in normal saline. Paroxetine (GlaxoSmithKline, UK) and sertraline (Pfizer, UK) were dissolved in sterile water and 2% Tween-80 in sterile water, respectively. Drugs were injected at either 5 mg/kg i.p. (paroxetine, sertraline and tranylcypromine) or 10 mg/kg i.p. (fluoxetine, desipramine, maprotiline and mianserin). 2.2.2. Acute treatment For acute treatment, rats were given single injections of antidepressant drugs at the same doses as in the chronic experiments (see Section 2.2.1), or the appropriate vehicles. Rats were killed 4 h (fluoxetine, paroxetine, sertraline, TCP and desipramine) or 24 h (fluoxetine, paroxetine and TCP) after single injections. 2.3. ISHH procedure
All procedures were carried out in accordance with the UK Animals Scientific Procedures Act, (1986). Male Sprague–Dawley rats (250–280 g, Harlan-Olac Ltd, Bicester, UK) were housed in cages (4–6 rats/cage) under a 12 h light/dark cycle in a temperature controlled environment with free access to food and water.
Brain sections (12 µm) were cut on a cryostat, thawmounted onto gelatine-subbed slides and pre-treated for ISHH using a standard protocol as described previously (Pei et al., 1997). Oligonucleotides complementary to mRNAs encoding genes for all forms of BDNF (5’GGT CTC GTA GAA ATA TTG CTT CAG TTG GCC TTT TGA TAC CGG GAC 3’) or NT-3 (5’GCG TTT CCT CCG TGG TGA TGT TCT ATT GGT TAC CAC CGG GTT GCC CAC 3’) were 3’-tail labelled with 35S-dATP with terminal deoxynucleotide transferase. The labelled oligonucleotide probes (specific activity ⬎ 109 cpm/µg) was added to each section (1 × 106 cpm/section) in hybridisation buffer as previously described (Pei et al., 1997). After incubation in humid chambers (containing 50% formamide in 4 × SSC) at 35 °C (BDNF probe) or 42 °C (NT-3 probe) for 14–16 h, slides were washed in 1 × SSC (BDNF probe) or 0.5 × SSC (NT-3 probe) buffer at 55 °C for 3 × 20 min followed by 2 × 60 min at room temperature. Sections were then air-dried and exposed to autoradiography film (Hyperfilm b-max, Amersham, Buckinghamshire, UK) for 3 weeks at room temperature. Controls included the use of oligonucleotides in the sense orientation and displacement with unlabelled probes. Search of Genebank data-base using the BLAST programs revealed no significant homology of the nucleotide sequences with other previously characterised genes.
2.2. Drugs and experimental protocol
2.4. Data analysis and statistics
2.2.1. Chronic antidepressant treatment Rats were injected with antidepressant drugs or the corresponding vehicles, twice daily for 14 days and killed 4 h (fluoxetine and TCP) or 24 h (fluoxetine, paroxetine, sertraline, TCP, desipramine, maprotiline and mianserin) after the last injection. Fluoxetine (Eli. Lilly, UK) desipramine HCl (Sigma), TCP (Sigma),
The relative abundance of mRNAs was determined by densitometric quantification of autoradiograms (MCID, St Catherine, Canada). Grey density values were calibrated to 14C tissue equivalents (Amersham, UK) and corrected for 35S decay. Optical densities were converted to nCi/g tissue, and mRNA abundance in drug treated groups was expressed as a percentage of respective con-
2. Materials and methods 2.1. Animals
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
905
trols and presented as mean ± SEM values. Statistical analysis of mRNA abundance was performed using Student’s t-test for comparisons between two groups (vehicle control vs. drug treatment). Multiple group comparisons were analysed using one-way ANOVA with Dunnett’s t-test.
3. Results 3.1. Effect of antidepressant drugs on BDNF gene expression at 4 h after single or repeated injection Single injection of the 5-HT re-uptake inhibitors, fluoxetine (10 mg/kg, i.p.), paroxetine (5 mg/kg, i.p.) and sertraline (5 mg/kg, i.p.), as well as the monoamine oxidase inhibitor TCP (5 mg/kg, i.p.), reduced BDNF mRNA in the hippocampus when measured 4 h post drug. The dentate gyrus showed the most marked and consistent reduction (⫺26–30%) in BDNF gene expression (Fig. 1A). Fluoxetine and sertraline, but not paroxetine or TCP also decreased BDNF mRNA in the CA1 and CA3 subfields of the hippocampus (Table 1). In contrast, a single injection of the noradrenaline reuptake inhibitor, desipramine (10 mg/kg, i.p.), had no significant effect on BDNF mRNA expression in any of the hippocampal subfields at 4 h post-drug (Fig. 1A and Table 1). The effect of fluoxetine and TCP was also tested 4 h after the last of twice daily injections for 14 days. Both drugs reduced BDNF mRNA expression the dentate gyrus at 4 h post-drug (Fig. 1B and Fig. 2A and B). Neither single nor repeated injections of any of the drugs tested influenced BDNF mRNA in cortical regions (piriform, parietal and cingulate of frontal cortex) at 4 h after the injection (data not shown). 3.2. Effect of antidepressant drugs on BDNF mRNA levels 24 h after single or repeated injections Single injection of the 5-HT uptake inhibitor, fluoxetine (10 mg/kg, i.p.) and paroxetine (5 mg/kg, i.p.) or the monoamine oxidase inhibitor TCP (5 mg/kg, i.p.) had no effect on BDNF mRNA levels when measured 24 h post drug (Fig. 3A). In contrast, repeated administration (twice daily for 14 days) of the 5-HT uptake inhibitor, fluoxetine, paroxetine or sertraline (5 mg/kg, i.p.) and the monoamine oxidase inhibitor TCP significantly increased BDNF mRNA levels in the dentate gyrus of the hippocampus when measured 24 h after the last injection. This increase ranged from +26% for fluoxetine to +44% for sertraline, compared to the corresponding vehicle injected control groups (Fig. 3B). The fluoxetine effect is illustrated as representative autoradiograms in Fig. 2C and D. After repeated TCP there was also an increase of
Fig. 1. Effects of antidepressant drugs on BDNF gene expression in the rat dentate gyrus of the hippocampus at (A) 4 h after a single injection (fluoxetine, paroxetine, sertraline, tranylcypromine and desipramine) and at (B) 4 h after the last injection following repeated administration for 2 weeks (fluoxetine and TCP). Data are expressed as % of controls and each column represents a mean ± SEM value from 5–6 rats ∗∗∗p ⬍ 0.001; ∗∗p ⬍ 0.01; ∗p ⬍ 0.05 vs. corresponding vehicle injected controls (Student’s t-test).
BDNF mRNA in the CA1 of the hippocampus (+ 33 ± 11%, n = 6, p ⬍ 0.05 vs. controls) when measured 24 h post drug. Neither the 5-HT re-uptake inhibitors nor TCP altered BDNF mRNA in any other subfields of the hippocampus or in any cortical regions (data not shown) at 24 h after the injections. The noradrenaline reuptake inhibitors, desipramine and maprotiline (10 mg/kg, i.p.) and the atypical antidepressant, mianserin (10 mg/kg, i.p.) did not alter BDNF mRNA levels in any of the brain regions measured 24 h after the last injection (Fig. 4). 3.3. Effects of antidepressant drugs on NT-3 mRNA levels The rat brain distribution of NT-3 mRNA was in agreement with previous reports (see Smith et al., 1995).
906
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
Table 1 Effects of antidepressant drugs on BDNF mRNA expression in the CA1 and CA3 of rat hippocampus, 4 h following a single injection. Data are given as mean ± SEM (n = 5–6) and expressed as % of controls CA1
CA3
Saline Fluoxetine
100 ± 1.2 80.4 ± 2.7a
100±2.2 76.9 ± 3.1a
Vehicle Paroxetine
100 ± 4.5 76.3 ± 10.3
100 ± 4.3 84.0 ± 5.8
Vehicle Sertraline
100 ± 3.4 87.4 ± 3.0c
100 ± 2.7 79.2 ± 1.8b
Saline Desipramine
100 ± 3.4 104.3 ± 5.3
100 ± 2.7 100.3 ± 2.3
Saline Tranylcypromine
100 ± 9.8 88.3 ± 6.4
100 ± 14.6 91.0 ± 4.8
a p⬍0.001 vs. corresponding vehicle injected controls (Student’s t-test). b p⬍0.01 vs. corresponding vehicle injected controls (Student’s ttest). c p⬍0.05 vs. corresponding vehicle injected controls (Student’s ttest).
Thus, high levels of NT-3 mRNA were detected in the dentate gyrus with lower levels in the CA2 and the medial CA1 (mCA1). NT-3 gene expression was unchanged in hippocampus after either acute or chronic antidepressant treatment (fluoxetine, paroxetine, sertraline, TCP, desipramine, maprotiline or mianserin) at either 4 or 24 h (data not shown). 4. Discussion Here we report the effect of acute and repeated administration of antidepressant drugs on BDNF gene expression in rat brain. The main finding is that drugs which inhibit either the 5-HT transporter (fluoxetine, paroxetine, sertraline) or monoamine oxidase (TCP), influence BDNF gene expression in the rat hippocampus in a bi-phasic manner following repeated injections. Thus, a down-regulation in the abundance of BDNF mRNA was detected at 4 h, and an up-regulation was detected at 24 h, after the last injection of twice daily injections for two weeks. After a single injection of these drugs a decrease in BDNF mRNA was also detected at 4 h but there was no change at 24 h post drug. Repeated administration of drugs which inhibit noradrenaline reuptake (desipramine and maprotiline), or an atypical
Fig. 2. Representative autoradiogram, showing the effect of repeated fluoxetine administration (10 mg/kg i.p. twice daily for 14 days), 4 h (A, B) and 24 h (C, D) after the last injection on rat BDNF mRNA in hippocampus. Dentate gyrus (DG).
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
907
Fig. 4. Effects of desipramine, maprotiline and mianserine on BDNF gene expression in the dentate gyrus of rat hippocampus at 24 h after the last injection following a 2 weeks treatment. Data are expressed as % of controls and each column represents a mean ± SEM value from 5–6 rats.
Fig. 3. Effects of antidepressant drugs on BDNF gene expression in the rat dentate gyrus of the hippocampus at 24 h (A) after a single injection (fluoxetine, paroxetine, tranylcypromine) and at 24 h (B) after the last injection following repeated administration for 2 weeks (fluoxetine, paroxetine, sertraline and tranylcypromine). Data are expressed as % of controls and each column represents a mean ± SEM value from 5–6 rats). ∗∗∗p ⬍ 0.001; ∗∗p ⬍ 0.01; ∗p ⬍ 0.05 vs. corresponding vehicle injected controls (Student’s t-test).
antidepressant drug (mianserin) did not change BDNF gene expression at neither 24 h (repeated administration) nor 4 h (single administration, desipramine only) post drug. Gene expression for NT-3 was not affected by any of these treatments. 4.1. Effect of antidepressant drugs on BDNF gene expression 24 h after administration Repeated injection of either a monoamine oxidase inhibitor (TCP) or 5-HT re-uptake inhibitors (fluoxetine, paroxetine or sertraline) increased BDNF mRNA
expression in the hippocampus 24 h after the last injection. This effect was restricted to the dentate gyrus, with the exception of TCP, which also caused an increase in the CA1. Importantly, these increases in BDNF mRNA were not detected after a single injection, which is in agreement with previous studies and indicates the occurrence of neuroadaptive effects (Nibuya et al., 1995, 1996). In contrast, both single and repeated ECS increase hippocampal BDNF mRNA at 24 h after treatment, although the effect by repeated ECS is longer lasting (Zetterstro¨ m et al., 1998a). Repeated injection with noradrenaline uptake inhibitors (desipramine and maprotiline) or an atypical antidepressant (mianserin) did not increase BDNF mRNA levels in hippocampus 24 h after the injection. This is in contrast to a previous study by Nibuya et al. (1996), which found an up-regulation in hippocampal BDNF gene expression (18 h after the last injection) common to all antidepressant drugs tested including, a monoamine oxidase inhibitor (TCP), tricyclics (desipramine, imipramine), 5-HT re-uptake inhibitors (fluoxetine, sertraline) and an atypical antidepressant (mianserin). However, a more recent study found that repeated administration with TCP but not the tricyclic imipramine significantly increased BDNF mRNA levels in dentate gyrus (Russo-Neustadt et al., 1999). The increase in BDNF mRNA levels in the present study, 24 h after the last of repeated injections was present in the dentate gyrus, but apart from TCP not in other hippocampal subfields. In comparison, previous studies by Nibuya et al. (1995, 1996) demonstrated overall increases in BDNF gene expression in dentate gyrus,
908
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
CA1, CA3 subfields of the hippocampus, as well as in the frontal cortex. The reason for this discrepancy is not clear, but some differences in drug doses and regimes might contribute. For example, twice daily injections for 2 weeks were used here, while once daily injections for 3 weeks were used by Nibuya et al. While most drugs were given in similar doses this was however, not the case for TCP: 5 mg/kg in present study and 7.5 mg/kg for 7 days then 10 mg/kg for 14 days in studies by Nibuya et al and sertraline: 5 mg/kg and 10 mg/kg respectively. In addition, the dose of the noradrenaline reuptake inhibitor desipramine was lower in the current study (10 mg/kg) compared to previous studies by Nibuya et al. (15 mg/kg). It is unlikely that the dose of desipramine used in the present study is too low, since previous studies have shown that this dose is both pharmacologically active and achieves plasma drug levels in the therapeutic range (Ainsworth et al., 1998). The explanation for the increase in BDNF gene expression following chronic, but not acute administration of 5-HT re-uptake inhibitors or TCP is not clear. However, these drugs cause greater or more prolonged increases in extracellular levels of 5-HT following repeated compared to acute administration (Ferrer and Artigas, 1994; Hjort et al., 2000). A prolonged increase of extracellular 5-HT levels would result in long-term activation of second messenger signal transduction systems such as the cAMP/protein kinase A (PKA, 5HT4,6,7), as well as diacylglycerol (DAG)/protein kinase C (PKC, 5-HT2). Gene expression for the trancription factor cAMP response element binding protein (CREB) is up-regulated by both of these pathways and has been linked to antidepressant drug induced elevations of the BDNF gene (Duman, 1998). It is possible that increased expression of the CREB gene accounts for the increased BDNF expression seen after chronic treatment with 5HT uptake inhibitors or TCP in the present study. In this context, it is however, difficult to explain the lack of change in BDNF mRNA levels in rats treated with desipramine and maprotiline in the present study. Since, adrenergic receptors of b and a1 activate cAMP and DAG transduction systems, respectively, these drugs might also be expected to induce the CREB gene. However, chronic administration of noradrenergic antidepressants is well known to cause desensitisation of b receptors (see Pryor and Sulser, 1992). Interestingly, a recent study shows that noradrenergic antidepressant drugs (desipramine and reboxetine) decreases nuclear CREBP (regulator of CREB-CRE directed gene transcription), suggesting that noradrenergic drugs do not stimulate CRE-mediated gene expression (e.g. BDNF, Manier et al., 2002). Interestingly, there are previous reports demonstrating glutamate-dependent induction of the BDNF gene (Zafra et al., 1991; Wetmore et al., 1994; Zetterstro¨ m et al., 1998c; Mackowiak et al., 2002). In this respect, it is
possible that the enhanced BDNF expression at 24 h following repeated administration by 5-HT uptake inhibitors or TCP is due to increased activity of glutamatergic input into the dentate gyrus via the perforant path of the entorhinal cortex (Amaral and Witter, 1995). In support of this idea, 5-HT2 receptor-stimulation increases glutamate activity in the neocortex (Aghajanian and Marek, 1997). 4.2. Effect of antidepressant drugs on BDNF gene expression 4 h after administration Repeated and single administration of TCP or the 5HT re-uptake inhibitors decreased BDNF gene expression in the hippocampus at 4 h after the last injection, while the noradrenaline re-uptake inhibitor desipramine had no effect. The decrease of BDNF gene expression is likely to be mediated by different mechanisms to the increase, as the latter effect was seen only 24 h after repeated administration. Previous microdialysis studies have shown that single injections of TCP or 5-HT uptake inhibitors increase extracellular levels of 5-HT in rat hippocampus (Gartside et al., 1995) and it is possible that this causes the down regulation in BDNF gene expression. Consistent with this, we found that 5-HT precursor loading and a 5-HT releasing agent drug decrease BDNF gene expression in rat hippocampus 2-3 h after the injection (Zetterstro¨ m et al., 1999). Also, a single injection of fluoxetine does not reduce BDNF mRNA levels in 5-HT lesioned rats (Coppell and Zetterstro¨ m, 1999). Inconsistent with the present study, a previous report failed to detect a decrease in BDNF gene expression at 3 h after either single (no effect) or 3 weeks administration (an increase), (Nibuya et al., 1995). It is possible that some differences in doses and regimes contribute to these differences (see previous section). It should however, be noted that the acute dose of TCP (10 mg/kg) used in the Nibuya study is expected to increase extracellular 5-HT levels sufficiently to stimulate 5-HT receptors (Ferrer and Artigas, 1994). Including 5-HT2 receptors which have previously been shown to reduce BDNF gene expression upon activation (Vaidya et al., 1997). Alternatively the differences could be accounted for by the use of a Northern blotting technique in the Nibuya study, probing the entire hippocampus for BDNF mRNA changes, which could fail in detecting regional differences within the hippocampus. In comparison, the ISHH techniques used in the present study yields better anatomical resolution. Thus, the most consistent and pronounced decrease was detected in the dentate gyrus representing only a small part of the entire hippocampus. Electrophysiological studies indicate that excitatory 5HT receptors are localized on GABA-ergic inter-neurons in the hippocampus (Freund et al., 1990). A possible explanation of our data is that excitatory 5-HT receptors
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
(5-HT2,4,6or7) on GABA-ergic interneurons in the polymorphic layer of the hippocampus, increase the release of GABA, which inhibits glutamatergic neurons and leads to a decrease of BDNF gene expression. In support of this idea, we have recently shown that the GABAB receptor agonist baclophen inhibits hippocampal BDNF gene expression with a similar magnitude to paroxetine and TCP (Khundakar et al., 2002a). 4.3. Bi-phasic BDNF gene expression following repeated antidepressant drug treatment The present data indicate that observed changes in the expression of the BDNF gene following repeated administration of 5-HT re-uptake inhibitors (fluoxetine, paroxetine and sertraline) and a monoamine oxidase inhibitor (TCP) varies according to the elapsed time between the last injection and the BDNF mRNA measurements. The rat BDNF gene gives rise to five separate exons (I–V) which have distinct promoters and a different cellular distribution (Timmusk et al., 1993). All BDNF transcripts contain the coding region for the mature BDNF protein and therefore, should be of equal functional importance. In the present study we used a probe that binds to the coding region for the BDNF protein (exon V) that is common to all transcripts. It is possible that the bi-phasic change in the BDNF gene following chronic treatment with 5-HT re-uptake inhibitors or TCP indicates differences in the expression of individual BDNF exons. In support of this idea, we have recently shown that an acute injection of paroxetine or TCP down regulates expression of exon IV but not exon I, 4 h after the injection (Khundakar and Zetterstro¨ m, 2002b). Interestingly, exon IV belongs to a group of exons that share properties with early–immediate genes and therefore would be expected to give rise to fast and transient changes (Timmusk et al., 1993). In comparison, the upregulation of BDNF gene expression at 24 h following repeated treatment with TCP has been shown to be particularly pronounced for exon I ( Russo-Neustadt et al., 2000). Exon I has previously been shown to require ongoing protein synthesis for its transcription and therefore would be slower in response, compared to the exons sharing properties with IEGs. The concept of differential expression of BDNF mRNAs following drug treatment is not unusal. For example, kainic acid induced seizure activity has been shown to elicit differential expression of BDNF mRNAs in the rat brain (Timmusk et al., 1993).
5. Summary and functional implications In summary, the main finding of the present study is that repeated treatment with a monoamine oxidase inhibitor or 5-HT re-uptake inhibitors, evokes a time-
909
dependent, bi-phasic change in BDNF gene expression in the hippocampus (especially dentate gyrus). These antidepressants decrease BDNF mRNA 4 h after the injection but increase it 24 h after the last injection. The effect at 4 h also occurred following a single injection. None of the above effects were seen with drugs blocking noradrenaline re-uptake or with an atypical antidepressant. Although recent studies demonstrate neurotrophic effects following chronic administration of antidepressant drugs (Duman et al., 2001; Norrholm and Quimet, 2001), future studies investigating changes in the BDNF protein after antidepressant drugs are crucial for these effects to be linked to increased BDNF expression.
References Ainsworth, K., Smith, S.E., Zetterstro¨ m, T.S.C., Pei, Q., Franklin, M., Sharp, T., 1998. Effect of antidepressant drugs on dopamine D1 and D2 receptor expression and dopamine release in the nucleus accumbens of the rat. Psychopharmacology 140, 470–477. Amaral, D.G., Witter, M.P., 1995. Hippocampal formation. In: Paxinos, P. (Ed.), The Rat Nervous System, 2nd edn. Academic Press, pp. 443–486. Aghajanian, G.K., Marek, G.J., 1997. Serotonin induces excitatory postsynaptic potentials in apical dendrites of neocortical pyramidal cells. Neuropharmacology 36, 589–599. Altar, C., 1999. Neurotrophins and depression. Trends Pharmacol. Sci. 20, 59–61. Coppell, A.L., Zetterstro¨ m, T.S.C., 1999. 5-HT mediated regulation of BDNF gene expression. Fund Clin Pharmacology 38 (Suppl), 107. Coppell, A.L., Zetterstro¨ m, T.S.C., 2000. Biphasic expression of brainderived neurotrophic factor gene in rat hippocampus following fluoxetine treatment. BR J. Pharmacol. Proc. 129 (Suppl), 134. Duman, R.S., 1998. Novel therapeutic approaches beyond the serotonin receptor. Biol Psychiatry 44, 324–335. Duman, R.S., Nakagawa, S., Malberg, J., 2001. Regulation of adult neurogenesis by antidepressant treatment. Neuropsychopharmacology 25, 836–844. Ferrer, A., Artigas, F., 1994. Effects of single and chronic treatment with tranylcypromine on extracellular serotonin in rat brain. Eur J Pharmacol 263, 227–234. Freund, T.F., Gulyas, A.I., Acsady, L., Gores, T., Totu, K., 1990. Serotonergic control of the hippocampus via local inhibitory interneurons. Proc Natl Acad Sci USA 87, 8501–8505. Gartside, S.E., Umbers, V., Hajos, M., Sharp, T., 1995. Interaction between a selective 5-HT1A receptor antagonist and an SSRI in vivo: effects on 5-HT cell firing and extracellular 5-HT. Br J Pharmacol 115, 1064–1070. Hjort, S., Bengtson, H.J., Kullberg, A., Carlzon, D., Peilot, H., Auerbach, S.B., 2000. Serotonin autoreceptor function and antidepressant drug action. J Psychopharmacol 14, 177–185. Khundakar, A.A., Mazumdar, A., Zetterstro¨ m, T.S.C., 2002a. Gene expression for brain derived neurotrophic factor in rat hippocampus is reduced by baclophen, tranylcypromine and paroxetine but not flunitrazepam. Br J Pharmacol 136 (Proc Suppl), 57. Khundakar, A.A., Zetterstro¨ m, T.S.C., 2002b. Differential expression of BDNF exons in rat brain after systemic administration of paroxetine and tranylcypromine. Br J Pharmacol 137 (Proc Suppl), 122. Lesch, K.P., 2001. Serotonergic gene expression and depression: implications for developing novel antidepressants. J Affective Disorders 62, 57–76. Mackowiak, M., O’Neill, M., Hicks, C., Bleakman, D., Skolnick, P.,
910
A.L. Coppell et al. / Neuropharmacology 44 (2003) 903–910
2002. An AMPA receptor potentiator modulates hippocampal expression of BDNF: an in vivo study. Neuropharmacol 43, 1–10. Manier, D.H., Shelton, R.C., Sulser, F., 2002. Noradrenergic antidepressants: does chronic treatment increase or decrease nucleaar CREB-P? J Neural Trans 109, 91–99. Nibuya, M., Morinobu, S., Duman, R.S., 1995. Regulation of BDNF and trkB mRNA in rat brain by chronic electroconvulsive seizure and antidepressant drug treatment. J Neurosci 15, 7539–7547. Nibuya, M., Nestler, E.J., Duman, R.S., 1996. Chronic antidepressant administration increases the expression of camp response element binding protein (CREB) in rat hippocampus. J Neurosci 16, 2365–2372. Norrholm, S.D., Quimet, C.C., 2001. Altered dendritic spine density in animal models of depression and in response to antidepressant treatment. Synapse 42, 151–163. Pei, Q., Burnet, P.W.J., Grahame-Smith, D.G., Zetterstro¨ m, T.S.C., 1997. Differential effects of acute and chronic electroconvulsive shock on the abundance of messanger RNAs for voltage-dependent potassium channel subunits in the rat brain. Neurosci 78, 343–350. Pryor, J.C., Sulser, F., 1992. Evolution of monoamine hypothesis of depression. In: Horton, R.W., Katona, C. (Eds.), Biological Aspects of Affective Disorders. Academic Press, London, pp. 77–94. Russo-Neustadt, A.A., Beard, R.C., Cottman, C.W., 1999. Exercise, antidepressant medications, and enhanced brain derived neurotrophic factor expression. Neuropsychopharmacol 21, 679–682. Russo-Neustadt, A.A., Beard, R.C., Huang, Y.M., Cottman, 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. Shirayama, Y., Chen, A.C., Nagakawa, S., Russell, D.S., Duman, R.S., 2002. Brain-derived neurotrophic factor produces antidepressant effects in behavioural models of depression. J Neurosci 22, 3251–3261. Siuciak, J.A., Lewis, D.R., Wiegand, S.J., Lindsay, R., 1997. Antidepressant like effect of brain derived neurotrophic factor (BDNF). Pharmacol Biochem Behav 56, 131–137.
Smith, M.A., Makino, S., Kvetnansky, R., Post, R.M., 1995. Stress and glucocorticoids affect the expression of brain-derived neurotrophic factor and neurotrophin-3 mRNA in the hippocampus. J Neurosci 15, 1768–1777. 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. Vaidya, V.A., Marek, G.J., Aghajanian, G.K., Duman, R.S., 1997. 5HT2 receptor-mediated regulation of brain-derived neurotrophic factor mRNA in the hippocampus and the neocortex. J Neurosci 17, 2785–2795. Wetmore, C., Olson, L., Bean, A.J., 1994. Regulation of brain-derived neurotrophic factor (BDNF) expression and release from hippocampal neurons is mediated by non-NMDA type glutamate receptors. J Neurosci 14, 1688–1700. Zafra, F., Castren, E., Thoenen, H., Lindholm, D., 1991. Interplay between glutamate and g-aminobutyric acid transmitter systems in the physiological regulation of brain-derived neurotrophic factor and nerve growth factor synthesis in hippocampal neurons. Proc Natl Acad Soc USA 88, 10037–10041. Zetterstro¨ m, T.S.C., Pei, Q., Grahame-Smith, D.G., 1998a. Repeated electroconvulsive shock extends the duration of enhanced gene expression for BDNF compared to a single administration. Mol Brain Res 57, 106–110. Zetterstro¨ m, T.S.C., Pei, Q., Ainsworth, K., Grahame-Smith, D.G., 1998b. Effects of antidepressant treatments on BDNF gene expression in rat brain. Br J Pharmacol 123 (Proc Suppl), 211. Zetterstro¨ m, T.S.C., McQuade, R., Pei, Q., 1998c. Pretreatment with the non-NMDA antagonist DNQX attenuates ECS-induced gene expression for BDNF in rat brain. Br J Pharmacol 124 (Proc Suppl), 69. Zetterstro¨ m, T.S.C., Pei, Q., Madhav, T.R., Coppell, A.L., GrahameSmith, D.G., 1999. Manipulations of brain 5-HT levels affect gene expression for BDNF in rat brain. Neuropharmacology 38, 1063– 1073.