SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Abstract:  A large body of evidence indicates that lithium, the prototype mood stabilizer in the treatment of bipolar disorder, has diverse neuroprotective and neurotrophic actions, and the actions are associated with its efficacy in treating bipolar disorder. It has been suggested that up-regulation of neurotrophic and neuroprotective factors including brain-derived neurotrophic factor (BDNF) and B-cell CLL/lymphoma 2 (Bcl-2) may underlie these neuroplastic actions of the drug. Olanzapine, an atypical anti-psychotic drug, has been shown to be an effective mood stabilizer. Olanzapine also has neurotrophic and neuroprotective actions, and these actions may underlie the efficacy of the drug for bipolar disorder and schizophrenia. However, the molecular mechanism by which the drug produces the neuroplastic actions is poorly understood. To understand a common molecular mechanism underlying the neuroplastic actions of lithium and olanzapine, we assessed the effect of 4-week lithium and olanzapine treatment on the levels of BDNF, Bcl-2 and cyclic adenosine monophosphate response element-binding protein (CREB), a transcription factor involved in expression of BDNF and Bcl-2, in the dentate gyrus and hippocampal area CA1. Our results show that 4-week treatment with both olanzapine and lithium increases the levels of Bcl-2 and CREB in the dentate gyrus and hippocampal area CA1. Four-week lithium treatment up-regulates BDNF in the dentate gyrus, and 4-week olanzapine treatment marginally did so. Neither drug altered BDNF levels in area CA1. These results suggest that the up-regulation of Bcl-2 and CREB may underlie the neuroplastic actions of olanzapine and lithium.

An accumulating body of evidence from brain imaging and postmortem studies indicates that detrimental neuroplastic changes such as the brain atrophy and neuronal loss are associated with the pathogenesis of schizophrenia [1] and bipolar disorder [2,3]. Lithium, the prototype mood stabilizer, has robust neuroplastic actions such as protection of neurons and glial cells from various neurotoxicities [4–6], an increase in neuronal survival, the enhancement of adult neurogenesis [7] and the facilitation of synaptic plasticity [8,9]. These diverse neuroplastic actions may underlie the therapeutic effects of lithium [3,10]. Interestingly, some atypical anti-psychotic drugs have been shown to be efficacious in treating bipolar disorder [11,12]. Among them, olanzapine has been most extensively studied as a mood stabilizer and an add-on drug for bipolar depression [12–14]. Olanzapine, in contrast to typical anti-psychotics, has neurotrophic [15] and neuroprotective [16–18] actions. This unique profile of olanzapine may counteract the loss of neurons and attenuate the progressive atrophy of the brain in patients with schizophrenia [17] as well as bipolar disorder [12]. However, the molecular mechanism underlying these neuroplastic actions of olanzapine is unclear.

It has been suggested that lithium produces these neuroplastic changes by up-regulating various neurotrophic, neuroprotective and transcription factors including Bcl-2, BDNF and CREB. It is well-known that these molecules play crucial roles in neuroprotective and neurotrophic actions and synaptic plasticity. For example, BDNF increases neuronal survival, neuronal growth actions and neurogenesis [19–21]. BDNF is required to maintain normal synaptic transmission and plasticity [20,21]. Bcl-2 regulates apoptosis by blocking pro-apoptotic process [22,23]. CREB is required for gene expression of neurotrophic and neuroprotective factors including BDNF and Bcl-2 [24,25]. CREB also plays crucial roles in maintaining normal synaptic transmission and plasticity such as hippocampal long-term potentiation and memory [26,27]. Since these molecules have diverse neuroplastic actions, the up-regulation of Bcl-2, CREB and BDNF may underlie the neuroplastic actions of lithium and olanzapine.

In the present study, we suggested that both lithium and olanzapine up-regulate BDNF, Bcl-2 and CREB in the subregions of the hippocampus, major regions for the pathology of bipolar disorder and schizophrenia, as a molecular basis for the neurotrophic and neuroprotic actions of the drugs. To test this hypothesis, we compared the effect of 4-week treatment with lithium and olanzapine on the levels of these molecules in the dentate gyrus and area CA1 of the hippocampus.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Animals and drug treatment.  Male Sprague Dawley rats weighing 175–200 g (Harlan, Indiana) were used. Animals were housed in a temperature-regulated and light-controlled colony room with food and water available ad libitum for 1 week. All animal procedures were carried out in strict accordance with the guidelines of the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the Cleveland Veterans Affairs Medical Center Institutional Animal Care and Use. Eighteen Animals were randomly divided into the lithium (n = 6), olanzapine (n = 6) and control (n = 6) treatment groups. Lithium-treated animals were fed with rat diet chow containing 0.24% lithium carbonate (BioServ®, Frenchtown, NJ) for 4 weeks. The cages of lithium-treated animals were equipped with two water bottles. One of these contained 0.9% saline since free choice saline consumption has been shown to reduce renal toxicity associated with lithium administration [28]. Olanzapine-treated animals were provided with tap water containing 0.1 mg/ml olanzapine dissolved in acetic acid. The stock solutions of 10 ml were produced by dissolving 100 mg olanzapine in 2 ml tap water containing 57 µl glacial acetic acid, then filling to 10 ml. Subsequently, each 10 ml of (100×) stock solution was diluted 100-fold with tap water and provided fresh every 2 days for 4 weeks. Thus, rats were administered with olanzapine 1.5–2 mg/kg daily in tap water or control vehicle for 4 weeks. Olanzapine-treated animals and lithium control animals were fed a control rat chow diet. The cages of these animals were equipped with only tap water bottles.

ELISA.  At the end of the feeding period, animals were decapitated, and trunk blood was collected to determine the blood levels of lithium. The brains were removed and stored at –70°. Gentle excisions of both hippocampi were carried out after near complete thawing of the whole brains, and each hippocampus was further dissected at 4° into the dentate gyrus and CA1 sub-regions by gently opening the hippocampal fissure with sweeping jets of ice cold PBS from a 30 ml syringe fitted with 30 Ga. (0.3 mm, OD) flexible blunt tubing (fig. 1). After the fissure was opened with the jets (1 or 2 min. of back-and-forth movements along the entire length of the fissure), the hippocampi were easily unrolled and the dentate gyrus was separated from the rest of the hippocampus using gently curved micro-dissecting forceps. The CA1 sub-region was then separated from the remaining regions of the hippocampus using a No. 10 or 11 scalpel blade (margins were determined by carefully consulting drawings by Cajal, Lorente de Nó and others). Right and left hemisphere tissue samples of dentate gyrus were combined for each individual animal because of the small size of the sub-region. The CA1 sub-region samples were similarly prepared.

image

Figure 1.  Location of hippocampal fissure in excised hippocampus. Jets of ice cold PBS opened the fissure allowing micro-dissection of sub-regions dentate gyrus (DG) (near surface), and CA1 (far surface). Combined weight (from both hemispheres of same animal) of DG averaged 28 mg, while combined weight of CA1 averaged 55 mg.

Download figure to PowerPoint

Nine volumes of cell protein extraction buffer were added to each sample to achieve the first 1 : 10 dilution. The extraction buffer consisted of (in mM) Tris Cl pH 7.4 10; NaCl 100; EDTA 1.0; EGTA 1.0; NaF 1.0; Na4P2O7 20; Na3VO4 2.0; PMSF (3.0 M in DMSO stock) 1.0; (in %) Triton® X-100 1.0; glycerol 10; SDS 0.1; deoxycholate 0.5; and protease inhibitor cocktail 5.0. Tissue samples were homogenized with an ultrasonic homogenizer, lysed for 20 min. and centrifuged at 6262 ×g for 10 min. at 4°. The clear lysate was transferred to another set of microtubes for acid treatment (BDNF only) and storage at –70°. The BDNF acid treatment (in four volumes of Dulbecco's PBS) was carried out with 1 N HCl at 100 µl/ml for 20 min. on ice followed by neutralization with 1 N NaOH for a final BDNF sample dilution of 1 : 50. The final dilution for pCREB samples was 1 : 100 (in standard diluent buffer from the kit) to reduce the matrix effect of the protein extraction buffer, and the final dilution for Bcl-2 was 1 : 50. The tissue levels of BDNF, Bcl-2 and pCREB were assayed using commercially available ELISA kits. The pCREB and Bcl-2 sandwich ELISA kits were obtained from Sigma (Phospho-CREB [pSer133 ELISA, Bcl-2 ELISA) and the BDNF sandwich ELISA from Promega (BDNF Emax® ImmunoAssay System). All other reagents and materials were obtained from commercially available suppliers. Both sandwich ELISA manufacturers protocols were closely adhered for the assays. Following processing, optical density or absorbance was read at 450 nm in a multiwell plate reader (FLUOstar Optima, BMG LabTech, Germany) and compared to the kit-supplied standards to assess levels of BDNF protein (pg/ml) and Bcl-2 (ng/ml) and phosphorylated CREB (pCREB).

Statistical analysis.  The resulting tissue levels of BDNF, Bcl-2 and pCREB obtained from each of the lithium-treated and olanzapine-treated animals and control animals were processed using FluoStar Optima Evaluation software®, Microsoft Excel® spreadsheets, and graphs were generated using Sigma Plot 8.0®. The resulting tissue levels of protein/peptide were expressed as mean ± S.E.M. and were analysed statistically using one-way anova with LSD post hoc pair-wise comparisons in SPSS 10.0®. The criterion of significance was set at P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

To examine the chronic effects of lithium treatment on the levels of pCREB, Bcl-2 and BDNF in the dentate gyrus and hippocampal area CA1, young adult rats were administered with lithium chow (0.24% Li2CO3, Harlan Teklad®, Madison, WI) or control chow for 4 weeks. The commercially available lithium chow was proved to maintain lithium levels stable within the therapeutic range in human. At the end of 4-week treatment with the lithium or control diet chows, the animals were killed, and trunk blood was collected to determine the blood levels of lithium. We confirmed that the blood levels of lithium at this dose were within the therapeutic range in human (0.71–0.89 mEq/L). Olanzapine was diluted in tap water daily, instead of intraperitoneal injection, to keep levels of olanzapine stable. It has been demonstrated that 1.0–2.5 mg/kg daily dose of olanzapine is relevant to the efficacy of the drug and produces neuroplastic actions without noticeable adverse effects [29,30].

To examine the chronic effects of lithium treatment on the levels of pCREB, Bcl-2 and BDNF in the dentate gyrus and hippocampal area CA1, young adult rats were administered with lithium chow (0.24% Li2CO3, Harlan Teklad®, Madison, WI) or control chow for 4 weeks. The commercially available lithium chow was proved to maintain lithium levels stable within the therapeutic range in human. At the end of 4-week treatment with the lithium or control diet chows, the animals were killed, and trunk blood was collected to determine the blood levels of lithium. We confirmed that the blood levels of lithium at this dose were within the therapeutic range in human (0.71–0.89 mEq/L). Olanzapine was diluted in tap water, instead of intraperitoneal injection, daily to keep the stable levels of olanzapine. It has been demonstrated that 1.0–2.5 mg/kg daily dose of olanzapine is relevant to the efficacy of the drug and produces neuroplastic actions without noticeable adverse effects [29,30].

The effects of 4-week treatment with lithium or olanzapine on the levels of BDNF, Bcl-2 and pCREB were examined in the dentate gyrus and area CA1 using ELISA. The baseline level of BDNF protein in the dentate gyrus was approximately three times higher than that in area CA 1. In the dentate gyrus (Effects of drug treatment compared with controls, F(2,15) = 4.010, P = 0.040), 4-week lithium treatment significantly increased the level of BDNF (P = 0.016, fig. 2A). Four-week olanzapine treatment marginally increased the level of BDNF (P = 0.054), and this level did not significantly differ from that induced by lithium (P = 0.550, fig. 2A). However, neither lithium- nor olanzapine treatment altered the levels of BDNF in hippocampal area CA1 (F(2,15) = 0.871, P = 0.438, fig. 2B).

image

Figure 2. Effects of 4-week treatment with lithium or olanzapine on the levels of BDNF in the dentage gyrus (DG) and area CA1 of the hippocampus. A. Effects of 4-week drug treatment on the levels of BDNF in DG (F(2,15) = 4.010, P = 0.040). Difference between lithium treatment and controls was significant (lithium: 4442 ± 286.2 pg/ml; control: 3352 ± 364.3 pg/ml; P = 0.016). Difference between olanzapine treatment and controls was also marginally significant (olanzapine: 4194.9 ± 172.3 pg/ml; P = 0.054), and lithium and olanzapine treatment groups did not differ (P = 0.550). B. Effects of 4-week drug treatment on the levels of BDNF in area CA1 (F(2,15) = 0.871, P = 0.438). Differences between lithium treatment and controls (lithium: 1552 ± 61.3 pg/ml; control: 1434 ± 64.9 pg/ml; P = 0.227) or olanzapine treatment (olanzapine: 1525.1±72.9 pg/ml; P = 0.347) were not significant, and lithium and olanzapine treatment groups did not differ (P = 0.777).

Download figure to PowerPoint

In the dentate gyrus (F(2,15) = 33.017, P < 0.001), 4-week lithium treatment (P < 0.001) as well as 4-week olanzapine treatment (P < 0.001) significantly increased the level of Bcl-2 in the dentate gyrus (fig. 3A). In area CA1 (F(2,15) = 8.459, P = 0.003), both 4-week lithium treatment (P = 0.025) and olanzapine treatment (P < 0.001) also significantly increased the level of Bcl-2 (fig. 3B).

image

Figure 3.  Effects of 4-week treatment with lithium or olanzapine on the levels of Bcl-2 in the dentage gyrus (DG) and area CA1 of the hippocampus. A. Effects of 4-week drug treatment on the levels of Bcl-2 in DG (F(2,15) = 33.017, P < 0.001). Difference between lithium treatment or olanzapine treatment and controls (lithium: 129.0 ± 6.0 pg/ml; olanzapine: 147.30 ± 22.9 pg/ml; control: 86.2 ± 0.6 pg/ml; P < 0.001) was significant. Lithium and olanzapine treatments did not differ (P = 0.703). B. Effects of 4-week drug treatment on the levels of Bcl-2 in area CA1 (F(2,15) = 8.459, P = 0.003). Difference between lithium treatment and controls (lithium: 83.9 ± 3.1 pg/ml; control: 70.7 ± 4.5 pg/ml; P = 0.025) or olanzapine treatment was significant (olanzapine: 92.4 ± 3.54 pg/ml; P = 0.001). Lithium and olanzapine treatments did not differ (P = 0.131).

Download figure to PowerPoint

In the dentate gyrus (F(2,15) = 10.461, P = 0.001), both 4-week lithium treatment (P = 0.001) and olanzapine treatment (P = 0.002) significantly increased the level of pCREB in the dentate gyrus (fig. 4A). The same treatment with lithium (P = 0.001) as well as 4-week olanzapine treatment (P = 0.003) significantly increased the level of pCREB in hippocampal CA1 area (F(2,15) = 9.269, P = 0.002; fig. 4B).

image

Figure 4.  Effects of 4-week treatment with lithium or olanzapine on the levels of pCREB in the dentate gyrus (DG) and area CA1 of the hippocampus. A. Effects of 4-week drug treatment on the levels of pCREB in DG (F(2,15) = 10.461, P = 0.001). Difference between controls and lithium treatment (lithium: 107.6 ± 8.1 pg/ml; control: 73.8 ± 2.8 pg/ml; P = 0.001) or olanzapine treatment (olanzapine: 92.4 ± 10.9 pg/ml; P = 0.002) was significant. Lithium and olanzapine treatments did not differ (P = 0.644). B. Effects of 4-week drug treatment on the levels of pCREB in area CA1 (F(2,15) = 9.269, P = 0.002). Difference between lithium treatment (lithium: 115.3 ± 3.2 pg/ml; control: 82.2 ± 7.1 pg/ml; P = 0.001) or olanzapine treatment and controls was significant (olanzapine: 111.5 ± 6.8 pg/ml, P = 0.003). Lithium and olanzapine treatments did not differ (P = 0.653).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Our results show that 4-week treatment with both olanzapine and lithium up-regulates Bcl-2 and CREB in the dentate gyrus and hippocampal area CA1 (figs 3 and 4). Four-week lithium treatment also up-regulates BDNF in the dentate gyrus, and 4-week olanzapine treatment marginally does so (fig. 2A). In hippocampal CA1 area, 4-week treatment with either olanzapine or lithium did not up-regulate BDNF (fig. 2B). The hippoccampal formation is made up of separate and interconnected sub-regions. Each hippocampal sub-region expresses unique molecular, cellular and neurophysiological profiles [31,32] and is differently vulnerable to the pathology of bipolar disorder [2] and schizophrenia [1], and also responds differently to psychotropic drugs [7]. Although molecular studies have reported the effects of lithium on these molecules in the hippocampus [33], few studies have examined the effects of lithium treatment on the regulation of these molecules on the sub-regions of the hippocampus. The present results are not in agreement with a previous report that lithium treatment for 4 weeks increased the level of pCREB, but did not alter the levels of BDNF and Bcl-2 in the dentate gyrus [9]. The authors injected lithium to the rats intraperitoneally and showed a higher range of blood lithium levels (0.97–1.2 mEq/l) than in the current report, when measured 24 hrs post-treatment. Lithium has a narrow therapeutic index (the therapeutic range: 0.6 –1.0 mEq/l), and small increases beyond its therapeutic range in serum concentrations can produce toxic effects [34]. Since intraperitoneal injections of lithium can produce spikes of toxic lithium levels shortly following treatment, their data could be contaminated by lithium toxicity. We fed animals with rat chow containing 0.24% lithium carbonate and confirmed that blood levels of lithium were within the therapeutic range (0.71–0.89 mEq/l).

Recently, we reported that 2-week lithium treatment did not up-regulate BDNF, Bcl-2 or CREB in the dentate gyrus and hippocampal area CA1 [35]. The present study combined with our previous report [35] suggests that chronic, but not subacute, lithium treatment is required to up-regulate these proteins. Our findings are consistent with preclinical and clinical findings that 4-week lithium treatment is required for hippocampal neurogenesis [7], behavioural changes [36] and increases in the grey matter volume in patients with bipolar disorder [37–39]. Our findings are also consistent with the clinical observation that the efficacy of lithium is delayed to 2–4 weeks of the drug administration. These temporal profiles of lithium suggest that the efficacy of lithium is probably mediated by neuroplastic changes via the up-regulation of proteins at the gene level, and these neuroplastic changes produce behavioural changes and clinical efficacy [40]. The present results are consistent with other studies, which have shown that chronic lithium treatment increased the levels of pCREB [9] and Bcl-2 [7] in the dentate gyrus. Previous studies have shown that chronic lithium treatment increased the levels or the expression of BDNF [35,41], Bcl-2 [5,42] and p-CREB [6,33,43] in the hippocampus, although two have not [44,45]. However, these studies have not examined the effects of lithium treatment on these molecules in the subregions of the hippocampus.

Our results indicated that chronic olanzapine treatment increased the levels of Bcl-2 and pCREB in the dentate gyrus and hippocampal area CA1. Previously published reports have also shown that long-term olanzapine treatment increased the expression and production of Bcl-2 in the hippocampus and frontal cortex [46], blocked stress-induced reduction in Bcl-2 in the hippocampus [47] and reversed methamphetamine-induced decrease in Bcl-2 in the striatum [48]. However, a recent study reported that single olanzapine treatment produced dose-dependent decreases of pCREB in medial prefrontal cortex [49]. These results suggest that chronic olanzapine treatment, rather than acute olanzapine treatment, may up-regulate Bcl-2 and CREB. Our findings that chronic olanzapine treatment marginally increased the levels of BDNF in the dentate gyrus are consistent with previously published reports [47,50]. However, both Luo et al. [47] and Bai et al. [50] reported that chronic olanzapine treatment increased the mRNA expression of BDNF in the dentate gyrus and area CA1, while we did not. In fact, another report [51] showed that chronic olanzapine treatment reduced BDNF levels in the hippocampus and frontal cortex. Interestingly, recent studies have shown that chronic olanzapine treatment restored reduced BDNF levels produced by haloperidol [52] or MK-801 treatment, but did not alter basal BDNF levels [53]. Perhaps, chronic olanzapine treatment is more prominent in blocking the down-regulation of BDNF in pathological or neurotoxic conditions rather than in normal states.

Olanzapine was originally developed as an atypical anti-psychotic drug. However, the drug has been shown to be effective in treating bipolar disorder as a primary or adjunct drug [12–14,54]. The multiple lines of evidence from preclinical and clinical studies suggest that olanzapine produces neuroprotective and neurotrophic effects, and these effects may be associated with its clinical efficacy in treating bipolar disorder and schizophrenia [15–18]. However, the molecular bases of the neurotrophic and neuroprotective actions of the drug are unclear. Our results indicate that olanzapine has positive influences on the regulation of Bcl-2 and CREB and possibly BDNF, suggesting that the up-regulation of these molecules may underlie the neuroprotective and neurotrophic actions of olanzapine.

Preclinical and postmortem studies have shown that a high ratio of pro-apoptotic factor (eg; Bad or Bax) over anti-apoptotic factor (eg; Bcl-2) triggers excessive apoptotic neuronal death and leads to synaptic pathology, and this pathological apoptosis may be associated with clinical deterioration in schizophrenia [17,55,56]. In this context, our results that chronic olanzapine treatment up-regulates Bcl-2 suggest that olanzapine may contribute to attenuating pathological apoptosis and atrophy of the brain.

Of particular interest, our findings that chronic lithium treatment enhances the activation of CREB in the dentate gyrus and hippocampal area CA1, the crucial regions involved in memory formation, have the clinical implication that lithium can improve learning and memory, since CREB activity is crucial in memory and learning [26,27,57]. In fact, we recently found that chronic lithium treatment magnified hippocampal long-term potentiation, the major neurophysiological basis of memory and learning [58,59] (Shim and Hammonds, submitted), and enhanced spatial memory and learning [60]. In conclusion, this study shows that chronic lithium treatment and chronic olanzapine treatment up-regulate Bcl-2 and CREB in the dentate gyrus and hippocampal area CA1. Both chronic lithium treatment and chronic olanzapine treatment up-regulate BDNF in the dentate gyrus. These molecular findings may account for neuroprotective and neurotrophic actions of lithium and olanzapine as well as their actions on synaptic plasticity.

Acknowledgement

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Funding for this study was provided by The Lilly Investigator Initiated Research Award. The Lilly pharmaceutical company had no further role in studying design; in the collection, analysis and interpretation of data; in the writing for the report; and in decision to submit the paper for publication. The authors have no conflict of interest. Thanks to Tim Teyler, Ph.D. for providing technical advice in the hippocampal sub-dissection.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References