Peripheral PDLIM5 expression in bipolar disorder and the effect of olanzapine administration
© Zain et al.; licensee BioMed Central Ltd. 2012
Received: 22 June 2012
Accepted: 28 September 2012
Published: 2 October 2012
Skip to main content
© Zain et al.; licensee BioMed Central Ltd. 2012
Received: 22 June 2012
Accepted: 28 September 2012
Published: 2 October 2012
One of the genes suggested to play an important role in the pathophysiology of bipolar disorder (BPD) is PDLIM5, which encodes LIM domain protein. Our main objective was to examine the effect of olanzapine treatment on PDLIM5 mRNA expression in the peripheral blood leukocytes of BPD patients.
We measured the expression of PDLIM5 mRNA from 16 patients with BPD Type I after 0, 4, and 8 weeks of treatment with olanzapine using quantitative real-time PCR. The Young Mania Rating Scale was used to evaluate the severity of manic symptoms in BPD patients. We also compared PDLIM5 mRNA expression in treatment-naïve BPD patients with that in healthy control subjects.
No significant difference was found in PDLIM5 mRNA expression between patients before olanzapine treatment and following 4 and 8 weeks of treatment (p>0.05). Although we observed a significant reduction in the severity of manic symptoms in all BPD patients (p<0.05), the effectiveness of the medication did not significantly correlate with the expression of PDLIM5 mRNA (p>0.05). Interestingly, PDLIM5 mRNA expression differed significantly between treatment-naïve BPD patients and healthy control subjects (p=0.002).
PDLIM5 mRNA expression did not appear to be a reflection of the efficacy of olanzapine in reducing the manic symptoms of BPD. The significant difference in expression of PDLIM5 mRNA in the peripheral blood leukocytes of treatment-naïve BPD patients versus that of healthy control subjects, however, suggests that it may be a good biological marker for BPD.
Bipolar disorder (BPD) is a mental illness that specifically affects mood and has an estimated prevalence of 0.8-2.6% . Strong evidence from twin, adoption, family, and linkage studies has shown that BPD is associated with genetic factors, with an estimated heredity of 63%. On the basis of evidence reported in genetic association and expression studies, PDLIM5 is a promising candidate gene for BPD .
The PDLIM5 gene lies on chromosome 4q22 and consists of a PDZ (post-synaptic density-95/discs large/zone occludens-1) domain and three LIM (Lin-11 Isl-1 mec03) domains. LIM domains are cysteine-rich double zinc fingers composed of 50 to 60 amino acids that are involved in protein-protein interactions such as cytoskeleton organization, cell lineage specification, organ development, and oncogenesis. LIM domains are also members of the Enigma class of proteins, including the Enigma homologue (ENH), a family of proteins that possess a 100-amino acid PDZ domain in the N terminus and one to three LIM domains in the C terminus. The PDLIM5 gene is ubiquitously expressed in the brain and its cellular localization is identical to that of synapsin I, which is known to be involved in neurotransmitter release. The LIM domain is an intermediate protein that can interact with various isoforms of protein kinase C (PKC). LIM domains have been reported to modulate intracellular calcium levels by linking N-type calcium channels with PKC epsilon, thereby promoting the phosphorylation-dependent modulation of channel properties . The formation of the PKCε-ENH-N-type Ca2+ channel complex has been hypothesized to have an important role in the molecular basis and efficiency of cellular signalling [4, 5].
In genetic association studies, single nucleotide polymorphisms (SNPs) that are located in the upstream region of PDLIM5 have been found to be associated with BPD [6, 7], schizophrenia , and major depressive disorder . In addition, previous studies showed a significant association between BPD and haplotypes of PDLIM5 polymorphisms located in the upstream region [7, 10]. However, other studies failed to replicate these findings, possibly because of genetic heterogeneity or methodological inconsistencies [11–13]. Expression studies had found that the mRNA expression of the PDLIM5 gene was increased in post-mortem brain tissue  and decreased in lymphoblastoid cell lines [14, 15] derived from peripheral blood leukocytes (PBLs) of BPD patients. The difference might be caused by a state-related rather than a trait-related change in PDLIM5 gene action .
Alterations in PKC activity have been suggested to play an important role in BPD and may be the core pathological characteristic [16, 17]. Novel findings confirmed the molecular function of the LIM domain in regulating PKC activity in a PKC isoform-specific manner . PDLIM5 was also shown to promote shrinkage of dendritic spines that may be associated with cognitive impairment . Previous pharmacotherapy studies showed an interaction between lithium and the LIM domain in improving manic symptoms in BPD through the normalization of PKC activity [16, 17]. On the other hand, one study showed no interaction between lithium prophylaxis and the LIM domain . Currently, atypical antipsychotics that have mood stabilizer properties are frequently used to treat BPD because of their improved tolerability and efficacy . Olanzapine, an atypical antipsychotic used in the treatment of BPD, has been shown in several studies to improve the symptoms of mania [22–25]. We hypothesize that this effect of olanzapine reflects in part its indirect interaction with the LIM domain in the signalling cascade, leading to an improvement in manic symptoms by normalizing PKC activity. In this study, we compared PDLIM5 mRNA expression in BPD patients before olanzapine treatment and at 4 and 8 weeks after treatment. We also compared PDLIM5 mRNA expression between treatment-naïve BPD patients and control subjects. In addition, we examined the association between PDLIM5 mRNA expression in treatment-naïve patients with BPD Type I and four SNPs of the PDLIM5 gene located in the upstream region: rs10008257, rs2433320, rs2433322, and rs2438146.
The subjects in this study consisted of 16 patients with BPD Type I (6 males, 10 females; mean age 40.25±14.6) and 16 control subjects (8 females, 8 males; mean age 44.5±10.25). Patients were recruited from the inpatient psychiatric clinic of University of Malaya Medical Centre (UMMC), and control subjects were recruited from the medical clinic of the UMMC. The protocol used in this study was approved by the Medical Ethics Committee of UMMC, and all subjects signed consent forms before they participated in the study. All subjects reported an absence of inter-ethnic marriage for at least three prior generations. Demographic data were recorded, including age, gender, marital status, education, occupation, ethnicity, duration of illness, previous non-compliance, previous substance abuse, and family history of mental illness. Only patients diagnosed with BPD Type I were included in this study. Patient diagnosis was undertaken by a qualified psychiatrist using the Mini International Neuropsychiatric Interview assessment scale . The patients were confirmed as being treatment-naïve or as having no history of psychotropic medication for at least 3 months. The progress of all patients was extensively monitored during the 8 weeks of treatment with olanzapine. Patients were given 5 to 20 mg of oral olanzapine daily, as assessed by the psychiatrist during the 8 weeks of treatment. Severity of illness was evaluated by the psychiatrist using the Young Mania Rating Scale (YMRS) at three time points, namely, before medication and after 4 and 8 weeks of olanzapine treatment. Control subjects were confirmed as having no history of mental illness and no first-degree relatives with mental illness, having no chronic disease, and not taking psychotropic medication or any other medications for chronic diseases.
Patient blood samples were collected in two tubes: a Tempus BloodTM RNA Tube and an ethylenediaminetetraaceticacid (EDTA) tube. The Tempus BloodTM RNA Tube contains 6 ml of Applied Biosystems Stabilizing Reagent, which immediately lyses blood cells after the blood is drawn into the tube, whereas the sample in the EDTA tube was used for DNA extraction. The RNA and DNA were extracted from the blood using the Tempus BloodTM RNA Extraction kit and QIAamp DNA Mini Kit (Qiagen, Hilden, Germany), respectively. All procedures of blood collection, storage, handling, and DNA and RNA extraction of the samples followed standard protocols provided by the manufacturers. The quantity and purity of the RNA and DNA were measured using NanoDrop (NanoDrop Technologies, Wilmington, DE, USA).
Expression of the PDLIM5 gene was quantified with the StepOnePlusTMReal-Time PCR System with the TaqMan® Gene Expression Assay and TaqMan® Gene Expression Master Mix according to the manufacturer’s instructions. Primers and probes were purchased from Applied Biosystems (Assay ID: PDLIM5, Hs00179051_m1). Glyceraldehyde-3-phosphate dehydrogenase was used as an internal control for normalization. Samples were run in real-time PCR in triplicate. We used the comparative Ct method to calculate relative changes in PDLIM5 gene expression following the subject’s response to olanzapine medication. Data were collected and analysed using StepOnePlusTM software version 2.1 (Applied Biosystems, Foster City,CA, USA).
In our preliminary study to investigate the association between genotype and expression level of the PDLIM5 gene, we genotyped four SNPs of PDLIM5 (rs10008257, rs2433320, rs2433322, and rs2438146) for the 16 treatment-naïve BPD patients using an inventoried TaqMan probes assay, TaqMan Genotyping Master Mix, and StepOnePlusTMReal-Time PCR, following the protocol provided by the manufacturer (Applied Biosystems).
Statistical calculations in this study were performed using SPSS statistical software version 16.0. The distributions of YMRS scores and PDLIM5 mRNA expression were checked for normality using the Kolmogorov-Smirnov test. The differences between PDLIM5 mRNA expression in controls and patients, as well as in patients before and after 4 and 8 weeks of olanzapine treatment, were calculated using the Wilcoxon signed rank test. The differences between the YMRS scores, however, were determined using the Wilcoxon signed rank test only in patients before and after 4 and 8 weeks of olanzapine treatment. A Spearman correlation coefficient test was performed to investigate the association between PDLIM5 mRNA expression and YMRS scores following 8 weeks of treatment. The SHEsis web-based platform was used to calculate linkage disequilibrium between the four SNPs of PDLIM5. A one-way analysis of variance test was used to determine the association between PDLIM5 mRNA expression and genotypes of the four SNPs of the PDLIM5 gene, as well as between PDLIM5 mRNA expression and patient demographic data, including duration of illness, previous non-compliance, previous substance abuse, and family history with mental illness, in treatment-naïve BPD patients. Hardy-Weinberg equilibrium was checked for all four SNPs using the Hardy-Weinberg calculator online tools .
Demographic data of the 16 BPD patients in this gene expression study
Duration of illness
Previous substance abuse
Family mental illness
Association between genotypes of PDLIM5 gene polymorphisms and PDLIM5 mRNA expression in treatment-naïve BPD patients*
Correlation between YMRS score and RQ level
YMRS score vs. RQ
Adjusted for age, gender,and ethnicity
At 4 weeks
At 8 weeks
The mechanism of antipsychotic drug action at the receptor level in reducing the symptoms of mania in BPD is well explained as being due to the blockade of the D2 receptor; the antidepressant effect of the antipsychotics, on the other hand, relates to the action of the 5-HT receptors . However, the effect of antipsychotics on events downstream of the receptor is unclear. Many studies relate mood disorder to abnormalities of neuroplasticity in the central nervous system. Changes in the cellular signalling cascade are an example of a neuroplastic event. The cellular signalling cascade plays an important role in the regulation of neuropeptide and neurotransmitter systems and is also a target of psychotropic drugs and hormones that are implicated in the pathophysiology of BPD. Abnormality in the signalling cascade could explain various neurovegetative symptoms and medical comorbidity associated with BPD . Olanzapine has been shown to activate certain signalling cascade such as the JAK/STAT, Akt/PKB, p38, and ERK1/2 pathways . From these findings, it was suggested that activation of the signalling cascade by olanzapine could alter the expression of genes involved in the signalling cascade, leading to a positive therapeutic effect .
We hypothesized that olanzapine administration causes changes in the expression of PDLIM5 mRNA, which encodes a protein that functions in the regulation of PKC activity, leading to improvement of manic symptoms in BPD. We observed a significant reduction in YMRS scores over 8 weeks of medication, indicating the effectiveness of olanzapine medication in this BPD group. Previous studies have also confirmed the effectiveness of olanzapine in the treatment of BPD [23, 24, 33]. However, we observed no significant changes in PDLIM5 mRNA expression over 8 weeks of treatment with olanzapine, nor did we find any correlation between improvement in manic symptoms of BPD and PDLIM5 mRNA expression. Interestingly, in a study of schizophrenia patients, Numata et al. showed that the expression of PDLIM5 was significantly higher in the leukocytes of treatment-naïve patients than it was in control subjects. However, when these investigators compared PDLIM5 expression between controls and schizophrenia patients who were stable after receiving antipsychotic medication for at least 3 months, they observed no significant difference. They reported that this finding may be a consequence of the pharmacological effects of antipsychotics or of clinical improvement and suggested that the expression of PDLIM5 mRNA may not be a trait-oriented change, but a state-related one that may respond to environmental factors such as antipsychotic medication . It is notable that the study was cross-sectional and that no specific antipsychotic medication was mentioned. In contrast, our study did not show any correlation between PDLIM5 expression and olanzapine treatment. These contrasting findings in the association between antipsychotics and PDLIM5 expression may relate to differences in the disease condition, that is, BPD and schizophrenia, or to antipsychotic medication other than olanzapine. Our results indicate a lack of interaction between PDLIM5 and olanzapine medication in improving the manic symptoms of BPD.
Gene expression in peripheral blood is assumed to reflect the expression of the gene in the prefrontal cortex, making it potentially useful as a surrogate marker for mental disorder . Moreover, components of the immune system found in peripheral blood, such as cytokines, lymphocytes, and interleukins, have also been found in the central nervous system . Thus, studying mRNA expression in peripheral blood has been suggested to reveal disease-specific immune changes and provide a marker of disease progression. In our study, we observed that PDLIM5 mRNA expression in PBLs was significantly reduced in treatment-naïve BPD patients compared with that in healthy control subjects. Our results support previous findings that showed a significant decrease in PDLIM5 mRNA expression in the lymphoblastoid cell lines of BPD and schizophrenia patients [14, 15], as well as in the PBLs of patients with major depressive disorder , compared with that in control subjects. Our results also support previous studies that suggested that the difference in expression level of the PDLIM5 gene in PBLs between patients and control subjects could be used as a potential biological marker for mental disorder [12, 13]. Further functional studies with larger sample sizes are needed to confirm the effectiveness of PDLIM5 in such a role.
One important factor that could affect gene expression is genetic polymorphism. To examine the effect of SNPs on the expression level of PDLIM5 mRNA in untreated BPD, we genotyped four SNPs of PDLIM5 that have previously been found to be significantly associated with BPD [6, 7]. We observed no significant correlation between these four SNPs, located in the upstream region of the PDLIM5 gene, with PDLIM5 mRNA expression in peripheral leukocytes. From these preliminary data, we conclude that none of these SNPs significantly affected the expression of PDLIM5 mRNA in PBLs. However, we cannot exclude the possibility of a type II error, as our sample size in this association study was relatively small.
In summary, we showed that there was no correlation between PDLIM5 mRNA expression and clinical improvement in BPD patients during 8 weeks of olanzapine treatment. We also found that PDLIM5 mRNA expression in PBLs was significantly lower in treatment-naïve patients than in control subjects, but found no association between each of the four SNPs of PDLIM5 and mRNA expression level. Our results suggest that further studies are warranted of PDLIM5 mRNA expression in PBLs as a potential biological marker for BPD.
We would like to thank all patients and staff from the University of Malaya Medical Centre and the National University of Malaysia Medical Centre who participated in this study. Special thanks to Dr Nazariah Harun for her assistance during blood collection. This study was supported by University of Malaya Research Grant RG162/09HTM and RG184/10HTM, Postgraduate Research Grant of the University Malaya PS226/09HTM, and HIR MOHE Grant E000025-20001. Language editing was provided by Barbara Every, ELS, of BioMedical Editor.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.