Psychoneuroendocrinology (2015) 51, 201—208
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/psyneuen
The interplay between BDNF and oxidative stress in chronic schizophrenia Xiang Yang Zhang a,b,∗, Da-Chun Chen a, Yun-Long Tan a, Shu-ping Tan a, Zhi-Ren Wang a, Fu-De Yang a, Olaoluwa O. Okusaga b, Giovana B. Zunta-Soares b, Jair C. Soares b,∗∗ a
Beijing HuiLongGuan Hospital, Peking University, Beijing, China Department of Psychiatry and Behavioral Sciences, Harris County Psychiatric Center, The University of Texas Health Science Center at Houston, Houston, TX, USA
b
Received 10 July 2014; received in revised form 6 September 2014; accepted 29 September 2014
KEYWORDS Schizophrenia; Brain-derived neurotrophic factor; Oxidative stress; Psychopathology; Interaction; Cognition
Summary Neurodegenerative processes may be involved in the pathogenesis of schizophrenia. Brain-derived neurotrophic factor (BDNF), the most widely distributed neurotrophin and oxidative stress (OS) may be critical for several pathological manifestations of neurodegenerative disorders. Accumulating evidence suggests that both BDNF and OS may be involved in the pathophysiology of schizophrenia. However, the possible interaction between BDNF and OS has been under-investigated. Serum BDNF, plasma malondialdehyde (MDA) levels and superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase (CAT) activities were analyzed using established procedures in 164 chronic medicated schizophrenia and 50 healthy controls. Schizophrenic symptoms were assessed by the Positive and Negative Syndrome Scale (PANSS) with cognitive and depressive factors derived from the five factor model of the PANSS. Compared to the control group, the patients exhibited a significant decrease in BDNF levels, in the activities of SOD and GSH-Px but a significant increase in MDA levels. In patients, but not in controls, we observed a significant negative correlation between BDNF and SOD. Furthermore, the interaction between BDNF and CAT was associated with the PANSS cognitive factor, and the interaction between BDNF and GSH-Px with the PANSS depressive factor. Both decreased BDNF levels and OS may be implicated in the pathophysiology of chronic schizophrenia. Their inverse association only in the schizophrenia group may reflect a pathological mechanism involving an interaction of oxidative damage and neurotrophin dysfunction. Moreover, OS may interact with
∗ Corresponding author at: University of Texas, Harris County Psychiatric Center, 2800 South MacGregor Way, Houston, TX 77021, USA. Tel.: +1 713 741 6047. ∗∗ Corresponding author at: Department of Psychiatry and Behavioral Sciences, The University of Texas Health Science Center at Houston, UT Houston Medical School, 1941 East Road, Ste. 3219, Houston, TX 77054, USA. Tel.: +1 713 486 2507; fax: +1 713 486 2552. E-mail addresses: [email protected] (X.Y. Zhang), [email protected] (J.C. Soares).
http://dx.doi.org/10.1016/j.psyneuen.2014.09.029 0306-4530/© 2014 Elsevier Ltd. All rights reserved.
202
X.Y. Zhang et al. the BDNF system to influence the clinical symptoms and cognitive impairment in schizophrenia, which is line with the neurodevelopmental hypothesis of schizophrenia. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction The neurodevelopment hypothesis of schizophrenia has postulated that interaction between genetics and environmental events during critical early periods in neuronal growth may negatively influence the way by which nerve cells are laid down, differentiated, and selectively culled by apoptosis (Nagahara and Tuszynski, 2011). Brain-derived neurotrophic factor (BDNF), a member of the neurotrophin family, plays an important role in supporting neuronal survival and function during development and in adulthood (He and Katusic, 2012). Accumulating preclinical and clinical data indicate that dysfunctions of BDNF may contribute to impaired brain development, neuroplasticity and synaptic connectivity leading to schizophrenia (Buckley et al., 2011; Pillai and Buckley, 2012; Nieto et al., 2013). Numerous recent studies have shown decreased serum or plasma BDNF levels in chronic antipsychotic-treated, neuroleptic free or neuroleptic naive, first-episode patients with schizophrenia (Chen et al., 2009; Xiu et al., 2009; Pillai et al., 2010; Nurjono et al., 2012), although some authors failed to replicate these findings (Green et al., 2011). Further, several studies have reported that BDNF was found to be associated with positive symptoms (Buckley et al., 2007; Xiu et al., 2009), negative symptoms (Rizos et al., 2008; Chen et al., 2009), and tardive dyskinesia (TD) (Zhang et al., 2012a,b). Taken together, these findings provide evidence that BDNF may be involved in psychopathology of schizophrenia. Free radicals are highly reactive chemical species generated during normal metabolic processes, and, in excess, can damage lipids, proteins, and DNA, causing cellular dysfunction and even death (Lohr and Browning, 1995). Oxidative stress (OS) involves a disequilibrium between pro-oxidant processes and the antioxidant defense system in favor of the former (Lohr et al., 2003; Yao and Keshavan, 2011). An unbalanced accumulation of oxidized proteins in the brain potentiates neurodegeneration and impairs cognitive function (Radak et al., 2007). Numerous studies have confirmed that the accumulation of oxidative damage such as oxidized proteins and lipid peroxides in aged mammalian brains underlies the molecular basis of brain aging and neurodegenerative disorders like Parkinson’s disease, Alzheimer’s disease and Huntington’s disease (Federico et al., 2012). Increasing evidence suggests that OS may be involved in the pathophysiology of patients with schizophrenia (Ng et al., 2008; Yao and Reddy, 2011). For example, patients with schizophrenia have abnormal activities of critical antioxidant enzymes (Zhang et al., 2003), reduced levels of antioxidants (Raffa et al., 2009; Chittiprol et al., 2010), and increased levels of lipid peroxidation in plasma, red blood cells, and cerebrospinal fluid (Padurariu et al., 2010). Furthermore, antioxidant enzymes or lipid peroxidation are correlated with psychopathology in schizophrenia, including negative symptoms, positive symptoms and with TD (Zhang et al., 2003). In addition, some symptoms of
schizophrenia improve with antioxidants, such as vitamins, extract of Ginkgo biloba or essential polyunsaturated fatty acids (Zhang et al., 2001; Yao and Keshavan, 2011). These findings provide further evidence that free radicals may be involved in the pathology of schizophrenia. Recently, some preclinical and clinical studies have shown the complex and reciprocal interactions between neurotrophins, antioxidant enzymes and OS. For example, a previous rat study showed that regular exercise training improves memory, decreases the level of reactive oxygen species, and increase the production of BDNF and nerve growth factor (Radak et al., 2007). Wu et al. reported that a diet high in saturated fat (HF) induced increased oxidative stress, and HF-induced oxidative damage was associated with reduced expression of BDNF in rats. Furthermore, treatment with antioxidant vitamin E completely counteracted the HF-elicited reduction in levels of BDNF mRNA through its antioxidant effect (Wu et al., 2004). Interestingly, a negative correlation between serum BDNF and thiobarbituric acid reactive substances (TBARS — a measure of lipid peroxidation) was found in a bipolar disorder cohort during manic episodes (Kapczinski et al., 2008). Moreover, a positive correlation between serum BDNF and TBARS was found in chronically mediated patients with schizophrenia (Gama et al., 2008). Recently, He and Katusic (2012) reported that BDNF protects circulating angiogenic cells by increasing expression of manganese superoxide dismutase (MnSOD) thereby enhancing their antioxidant capacity. These findings suggest that oxidative stress can interact with the BDNF system, suggesting the need of further investigation with regards interactions of BDNF and oxidative markers in the mental disorders. In view of the previously mentioned studies regarding BDNF and OS in schizophrenia and the potential interaction between OS and the BDNF system, we tested the hypothesis that decreased BDNF serum levels may be related to oxidative damage in schizophrenia. Also, we speculate that interaction of BDNF and OS might be associated with schizophrenia symptoms. Therefore, the purpose of the study was to investigate (1) whether decreased BDNF serum levels and altered antioxidant enzyme activities and lipid peroxidation occurred simultaneously in patients with schizophrenia; (2) an interaction between BDNF and OS parameters, or between symptom severity and BDNF and OS parameters; (3) a significant difference between typical and atypical antipsychotic drugs in the influence on the BDNF and OS parameters.
2. Methods 2.1. Subjects One hundred and sixty four patients (male/female = 122/42) were recruited from among the inpatients of Beijing Hui-Long-Guan Hospital, a Beijing City owned psychiatric
BDNF and oxidative stress in schizophrenia hospital. All patients met the DSM-IV diagnosis of schizophrenia, which was confirmed by two independent experienced psychiatrists based on the Structured Clinical Interview for DSM-IV. Their clinical subtypes were: paranoid, 63 (38.4%); disorganized, 19 (11.6%), undifferentiated, 18 (11.0%); residual 60 (36.6%) and others, 4 (2.4%). Patients were between 25 and 75 years old and had a mean duration of illness of 23.4 ± 8.8 years. All patients had been receiving stable doses of oral antipsychotic drugs for at least 12 months before entry into this study. Antipsychotic treatment consisted mainly of monotherapy with clozapine (n = 76), risperidone (n = 24), and other typical antipsychotics (n = 64), including perphenazine (n = 21), haloperidol (n = 18), sulpiride (n = 12), chlorpromazine (n = 10), or others (n = 3). Mean antipsychotic dose (as chlorpromazine equivalents) was 372 ± 315 mg/day. Normal controls (male/female = 36/14) were recruited from the local community through advertisement. Current mental status and personal or family history of any mental disorder was assessed by a clinical psychiatrist. None of the healthy control subjects presented a personal or family history of psychiatric disorder. We obtained a complete medical history, physical examination and laboratory tests from all subjects. Any subjects with test abnormalities or major medical illnesses were excluded. Neither patients nor control subjects suffered from drug or alcohol abuse/dependence. All subjects were Han Chinese, and they gave written informed consent, which was approved by the Institutional Review Board of Beijing Hui-Long-Guan hospital.
2.2. Blood sampling Venous blood from forearm vein was collected between 7 and 9 am following an overnight fast. The serum or plasma was separated, aliquoted, and stored at −70 ◦ C before use. The samples of the patients and the controls were assayed in the same assay batches. The identity of all subjects was indicated by a code number maintained by the investigator until all biochemical analyses were completed. Each evaluated parameter was assayed in duplicate for all samples.
2.3. Determination of serum BDNF levels We measured fasting serum BDNF levels by sandwich ELISA using a commercially available kit as described in our previous report (Chen et al., 2009; Xiu et al., 2009). All samples were assayed by a research assistant blind to the clinical situation. Inter- and intra-assay variation coefficients were 7 and 5%, respectively.
2.4. Determination of OS parameters All antioxidant enzymes and lipid peroxidation products in plasma were measured by the spectrophotometer. Among these OS parameters, superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase (CAT) are antioxidant enzymes, while malondialdehyde (MDA) is lipid peroxidation products. A full description of the assays has
203 been given in our previous report (Zhang et al., 2006). Briefly, lipid peroxidation levels were monitored by determining the end product of lipid peroxidation MDA using the thiobarbituric acid (TBA) method. MDA results were expressed as nmol/ml. The total SOD activity was determined using a standard assay involving spectrophotometric determination of the inhibition of superoxide-induced formation of nitrite from hydroxylamine. Xanthine—xanthine oxidase provided the superoxide source. One unit is defined as the amount of SOD that inhibits 50% of nitrite formation under the assay conditions. GSH-Px activity was measured by adding H2 O2 to the reaction mixture containing reduced glutathione (GSH), reduced nicotinamide adenine dinucleotide phosphate (NADPH) and glutathione reductase. One unit of GSH-Px is defined as micromoles of NADPH oxidized per minute. CAT activity was assayed based on the decomposition of hydrogen peroxide by CAT. CAT catalyzes the transformation of hydrogen peroxide to water and oxygen. CAT activity was determined by monitoring the decreased absorbance spectrophotometrically at 240 nm due to degradation of hydrogen peroxide. One unit of CAT was defined as the amount of enzyme that decomposes 1 mol H2 O2 /min. SOD, GSH-Px and CAT activities were expressed as units per milliliter plasma (U/ml). The intra- and inter-assay coefficient of variation for these antioxidant enzymes ranged from 3.2% to 5.9%.
2.5. Clinical assessment Four psychiatrists who had simultaneously attended a training session in the use of the Positive and Negative Syndrome Scale (PANSS) rated patients on this scale. After training, repeated assessment showed that the inter observer correlation coefficient was maintained at >0.8 for the PANSS total score. We obtained blood samples to assess BDNF and OS parameters at the time of PANSS ratings. In its original form, the PANSS was divided in three scales: positive (items P1 to P7), negative (items N1 to N7) and general psychopathology (items G1 to G16) (Kay et al., 1987). Later factorial analyses have presented data in favor of five-factor components, with factors commonly labeled as ‘positive’, ‘negative’, ‘cognitive’, ‘depression’ and ‘excitement’ (Wallwork et al., 2012). The cognitive factor (sometimes called ‘disorganization’) refers to the patient’s cognitive functioning, and is composed of several PANSS items that vary partially in the different factorial analyses (Rodriguez-Jimenez et al., 2013). Recently, Wallwork et al. (2012) have proposed a new consensus model of the cognitive factor that is only made up of three PANSS items: ‘Conceptual disorganization’ (P2), ‘Difficulty in abstract thinking’ (N5), and ‘Poor attention’ (G11). Subsequently, the cognitive factor was confirmed in the clinical assessment of patients with schizophrenia (Rodriguez-Jimenez et al., 2013), suggesting that the cognitive component of the PANSS is a valid measure of cognitive deficits in schizophrenia. The other factors in the five-factor model proposed by Wallwork et al. (2012) include an excited factor (items P4, P7, G8, G14) and a depressed factor (items G2, G3, G6).
204
2.6. Statistical analysis Since all the oxidative markers and BDNF were normally distributed in both the patient and control groups (Kolmogorov—Smirnov one-sample test: all p > 0.5), the main models consisted of a univariate analysis of covariance (ANCOVA) with gender, age, education, smoking and body mass index (BMI) as covariates for comparison of these biomarkers in the patient and control groups. Group differences were compared using Student’s t test or one-way analysis of variance (ANOVA) for continuous variables and Chi squared for categorical variables. Correlation between variables was studied using Pearson product moment correlations. Bonferroni corrections were applied to adjust for multiple testing. Lastly, exploratory regression analyses were used to examine the relationships between clinical phenotypes and BDNF or OS parameters in patients. Stepwise multiple regression analysis used the PANSS total or subscale scores including the PANSS cognitive or depressive factor as dependent variables with BDNF, antioxidant enzyme and MDA as the independent variables. Covariates in these stepwise forward entry models included age, gender, education, smoking, duration of illness, age of onset, and antipsychotic medication dosage, type (typical vs. atypical antipsychotics) and duration. Two-tailed significance values were used and significance levels were set at 0.05.
3. Results 3.1. Demographic data
X.Y. Zhang et al. Table 1 Demographics of patients and normal control subjects.
Sex, M/F Age (years) Education Smokers/nonsmokers Smoked cigarettes/day Body mass index (kg/m2 ) Age of onset (years) Duration of illness (years) Hospitalization numbers PANSS Positive symptom Negative symptom General psychopathology Total *
Table 1 shows the demographic data of the subjects in the present study. There was a significant difference in age between patient and control groups (p < 0.05), which was adjusted in the following analysis. No other significant difference in demographic data was noted between patient and control groups. A significant gender difference in BDNF levels was noted in patients (7.2 ± 2.3 ng/ml for males vs. 5.9 ± 2.2 ng/ml for females; F = 7.58, df = 1, 148, p = 0.007). In addition, there was a significant inverse relationship between age of onset of psychosis and SOD activities (r = −0.17, df = 162, p = 0.03), and a significant positive relationship between age and MDA (r = 0.17, df = 162, p = 0.03). No other significant association was observed between clinical parameters and BDNF or OS parameters.
3.2. OS parameters and BDNF in schizophrenia and controls Table 2 showed that both SOD and GSH-Px activities were significantly lower in patients than in controls (both p < 0.001), while MDA levels were markedly higher in patients than in controls (p < 0.001). When the effect of age, sex, education, smoking and BMI were examined by adding them to the ANOVA as covariates, a significant difference between patient and control groups was still observed in SOD (p < 0.005), GSH-Px (p < 0.001) and MDA (p < 0.001). No significant difference was noted in CAT between patients and controls.
Schizophrenia (n = 164)
Control subjects (n = 50)
122/42 48.3 ± 6.4 9.5 ± 2.4 95/58 13.6 ± 9.1
36/14 45.6 ± 5.1* 9.5 ± 1.4 31/18 12.6 ± 6.5
24.4 ± 3.5
23.5 ± 3.6
24.9 ± 6.4 23.4 ± 8.8
3.8 ± 2.4
16.0 ± 5.8 25.1 ± 6.0 34.4 ± 8.3 75.5 ± 15.7
p < 0.05.
BDNF serum levels were markedly lower in serum of patients than in controls (p < 0.001). When the effect of age, sex, education, smoking and BMI were examined by adding them to the ANOVA as covariates, a significant difference between patients and controls was still observed (p < 0.001).
3.3. Interaction of BDNF and OS parameters in schizophrenia and controls To detect the relationship between BDNF and OS parameters, partial correlation (controlled for age, gender, education, smoking and BMI) was performed. This analysis showed a significant negative association between BDNF levels and SOD activity in the patient group (r = −0.23, df = 142, p = 0.006; Fig. 1). No significant correlation was observed between BDNF and the other OS parameters (all p > 0.05). In addition, no significant correlations between BDNF levels and OS parameters were observed in healthy controls (all p > 0.05).
3.4. Association between BDNF, OS parameters and clinical symptoms in schizophrenia There was a significant positive correlation between SOD and positive subscore (r = 0.18, df = 164, p = 0.02), or PANSS total score (r = 0.16, df = 164, p = 0.04). A significant and negative correlation was noted between CAT and general
BDNF and oxidative stress in schizophrenia Table 2
Oxidative stress parameters and BDNF levels in schizophrenia and controls.
Markers
Schizophrenia (n = 164)
SOD (U/ml) GSH-Px (U/ml) CAT (U/ml) MDA (nmol/ml) BDNF (ng/ml)
85.9 108.8 2.6 11.3 6.8
a
205
± ± ± ± ±
19.5 31.5 3.2 7.6 2.4
Controls (n = 50) 97.2 147.2 2.7 2.6 9.5
± ± ± ± ±
10.9 31.0 2.4 1.9 4.4
F
df
13.3 39.5 0.05 56.4 29.2
1, 1, 1, 1, 1,
212 212 212 212 201
p
Adjusted pa
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/psyneuen
The interplay between BDNF and oxidative stress in chronic schizophrenia Xiang Yang Zhang a,b,∗, Da-Chun Chen a, Yun-Long Tan a, Shu-ping Tan a, Zhi-Ren Wang a, Fu-De Yang a, Olaoluwa O. Okusaga b, Giovana B. Zunta-Soares b, Jair C. Soares b,∗∗ a
Beijing HuiLongGuan Hospital, Peking University, Beijing, China Department of Psychiatry and Behavioral Sciences, Harris County Psychiatric Center, The University of Texas Health Science Center at Houston, Houston, TX, USA
b
Received 10 July 2014; received in revised form 6 September 2014; accepted 29 September 2014
KEYWORDS Schizophrenia; Brain-derived neurotrophic factor; Oxidative stress; Psychopathology; Interaction; Cognition
Summary Neurodegenerative processes may be involved in the pathogenesis of schizophrenia. Brain-derived neurotrophic factor (BDNF), the most widely distributed neurotrophin and oxidative stress (OS) may be critical for several pathological manifestations of neurodegenerative disorders. Accumulating evidence suggests that both BDNF and OS may be involved in the pathophysiology of schizophrenia. However, the possible interaction between BDNF and OS has been under-investigated. Serum BDNF, plasma malondialdehyde (MDA) levels and superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase (CAT) activities were analyzed using established procedures in 164 chronic medicated schizophrenia and 50 healthy controls. Schizophrenic symptoms were assessed by the Positive and Negative Syndrome Scale (PANSS) with cognitive and depressive factors derived from the five factor model of the PANSS. Compared to the control group, the patients exhibited a significant decrease in BDNF levels, in the activities of SOD and GSH-Px but a significant increase in MDA levels. In patients, but not in controls, we observed a significant negative correlation between BDNF and SOD. Furthermore, the interaction between BDNF and CAT was associated with the PANSS cognitive factor, and the interaction between BDNF and GSH-Px with the PANSS depressive factor. Both decreased BDNF levels and OS may be implicated in the pathophysiology of chronic schizophrenia. Their inverse association only in the schizophrenia group may reflect a pathological mechanism involving an interaction of oxidative damage and neurotrophin dysfunction. Moreover, OS may interact with
∗ Corresponding author at: University of Texas, Harris County Psychiatric Center, 2800 South MacGregor Way, Houston, TX 77021, USA. Tel.: +1 713 741 6047. ∗∗ Corresponding author at: Department of Psychiatry and Behavioral Sciences, The University of Texas Health Science Center at Houston, UT Houston Medical School, 1941 East Road, Ste. 3219, Houston, TX 77054, USA. Tel.: +1 713 486 2507; fax: +1 713 486 2552. E-mail addresses: [email protected] (X.Y. Zhang), [email protected] (J.C. Soares).
http://dx.doi.org/10.1016/j.psyneuen.2014.09.029 0306-4530/© 2014 Elsevier Ltd. All rights reserved.
202
X.Y. Zhang et al. the BDNF system to influence the clinical symptoms and cognitive impairment in schizophrenia, which is line with the neurodevelopmental hypothesis of schizophrenia. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction The neurodevelopment hypothesis of schizophrenia has postulated that interaction between genetics and environmental events during critical early periods in neuronal growth may negatively influence the way by which nerve cells are laid down, differentiated, and selectively culled by apoptosis (Nagahara and Tuszynski, 2011). Brain-derived neurotrophic factor (BDNF), a member of the neurotrophin family, plays an important role in supporting neuronal survival and function during development and in adulthood (He and Katusic, 2012). Accumulating preclinical and clinical data indicate that dysfunctions of BDNF may contribute to impaired brain development, neuroplasticity and synaptic connectivity leading to schizophrenia (Buckley et al., 2011; Pillai and Buckley, 2012; Nieto et al., 2013). Numerous recent studies have shown decreased serum or plasma BDNF levels in chronic antipsychotic-treated, neuroleptic free or neuroleptic naive, first-episode patients with schizophrenia (Chen et al., 2009; Xiu et al., 2009; Pillai et al., 2010; Nurjono et al., 2012), although some authors failed to replicate these findings (Green et al., 2011). Further, several studies have reported that BDNF was found to be associated with positive symptoms (Buckley et al., 2007; Xiu et al., 2009), negative symptoms (Rizos et al., 2008; Chen et al., 2009), and tardive dyskinesia (TD) (Zhang et al., 2012a,b). Taken together, these findings provide evidence that BDNF may be involved in psychopathology of schizophrenia. Free radicals are highly reactive chemical species generated during normal metabolic processes, and, in excess, can damage lipids, proteins, and DNA, causing cellular dysfunction and even death (Lohr and Browning, 1995). Oxidative stress (OS) involves a disequilibrium between pro-oxidant processes and the antioxidant defense system in favor of the former (Lohr et al., 2003; Yao and Keshavan, 2011). An unbalanced accumulation of oxidized proteins in the brain potentiates neurodegeneration and impairs cognitive function (Radak et al., 2007). Numerous studies have confirmed that the accumulation of oxidative damage such as oxidized proteins and lipid peroxides in aged mammalian brains underlies the molecular basis of brain aging and neurodegenerative disorders like Parkinson’s disease, Alzheimer’s disease and Huntington’s disease (Federico et al., 2012). Increasing evidence suggests that OS may be involved in the pathophysiology of patients with schizophrenia (Ng et al., 2008; Yao and Reddy, 2011). For example, patients with schizophrenia have abnormal activities of critical antioxidant enzymes (Zhang et al., 2003), reduced levels of antioxidants (Raffa et al., 2009; Chittiprol et al., 2010), and increased levels of lipid peroxidation in plasma, red blood cells, and cerebrospinal fluid (Padurariu et al., 2010). Furthermore, antioxidant enzymes or lipid peroxidation are correlated with psychopathology in schizophrenia, including negative symptoms, positive symptoms and with TD (Zhang et al., 2003). In addition, some symptoms of
schizophrenia improve with antioxidants, such as vitamins, extract of Ginkgo biloba or essential polyunsaturated fatty acids (Zhang et al., 2001; Yao and Keshavan, 2011). These findings provide further evidence that free radicals may be involved in the pathology of schizophrenia. Recently, some preclinical and clinical studies have shown the complex and reciprocal interactions between neurotrophins, antioxidant enzymes and OS. For example, a previous rat study showed that regular exercise training improves memory, decreases the level of reactive oxygen species, and increase the production of BDNF and nerve growth factor (Radak et al., 2007). Wu et al. reported that a diet high in saturated fat (HF) induced increased oxidative stress, and HF-induced oxidative damage was associated with reduced expression of BDNF in rats. Furthermore, treatment with antioxidant vitamin E completely counteracted the HF-elicited reduction in levels of BDNF mRNA through its antioxidant effect (Wu et al., 2004). Interestingly, a negative correlation between serum BDNF and thiobarbituric acid reactive substances (TBARS — a measure of lipid peroxidation) was found in a bipolar disorder cohort during manic episodes (Kapczinski et al., 2008). Moreover, a positive correlation between serum BDNF and TBARS was found in chronically mediated patients with schizophrenia (Gama et al., 2008). Recently, He and Katusic (2012) reported that BDNF protects circulating angiogenic cells by increasing expression of manganese superoxide dismutase (MnSOD) thereby enhancing their antioxidant capacity. These findings suggest that oxidative stress can interact with the BDNF system, suggesting the need of further investigation with regards interactions of BDNF and oxidative markers in the mental disorders. In view of the previously mentioned studies regarding BDNF and OS in schizophrenia and the potential interaction between OS and the BDNF system, we tested the hypothesis that decreased BDNF serum levels may be related to oxidative damage in schizophrenia. Also, we speculate that interaction of BDNF and OS might be associated with schizophrenia symptoms. Therefore, the purpose of the study was to investigate (1) whether decreased BDNF serum levels and altered antioxidant enzyme activities and lipid peroxidation occurred simultaneously in patients with schizophrenia; (2) an interaction between BDNF and OS parameters, or between symptom severity and BDNF and OS parameters; (3) a significant difference between typical and atypical antipsychotic drugs in the influence on the BDNF and OS parameters.
2. Methods 2.1. Subjects One hundred and sixty four patients (male/female = 122/42) were recruited from among the inpatients of Beijing Hui-Long-Guan Hospital, a Beijing City owned psychiatric
BDNF and oxidative stress in schizophrenia hospital. All patients met the DSM-IV diagnosis of schizophrenia, which was confirmed by two independent experienced psychiatrists based on the Structured Clinical Interview for DSM-IV. Their clinical subtypes were: paranoid, 63 (38.4%); disorganized, 19 (11.6%), undifferentiated, 18 (11.0%); residual 60 (36.6%) and others, 4 (2.4%). Patients were between 25 and 75 years old and had a mean duration of illness of 23.4 ± 8.8 years. All patients had been receiving stable doses of oral antipsychotic drugs for at least 12 months before entry into this study. Antipsychotic treatment consisted mainly of monotherapy with clozapine (n = 76), risperidone (n = 24), and other typical antipsychotics (n = 64), including perphenazine (n = 21), haloperidol (n = 18), sulpiride (n = 12), chlorpromazine (n = 10), or others (n = 3). Mean antipsychotic dose (as chlorpromazine equivalents) was 372 ± 315 mg/day. Normal controls (male/female = 36/14) were recruited from the local community through advertisement. Current mental status and personal or family history of any mental disorder was assessed by a clinical psychiatrist. None of the healthy control subjects presented a personal or family history of psychiatric disorder. We obtained a complete medical history, physical examination and laboratory tests from all subjects. Any subjects with test abnormalities or major medical illnesses were excluded. Neither patients nor control subjects suffered from drug or alcohol abuse/dependence. All subjects were Han Chinese, and they gave written informed consent, which was approved by the Institutional Review Board of Beijing Hui-Long-Guan hospital.
2.2. Blood sampling Venous blood from forearm vein was collected between 7 and 9 am following an overnight fast. The serum or plasma was separated, aliquoted, and stored at −70 ◦ C before use. The samples of the patients and the controls were assayed in the same assay batches. The identity of all subjects was indicated by a code number maintained by the investigator until all biochemical analyses were completed. Each evaluated parameter was assayed in duplicate for all samples.
2.3. Determination of serum BDNF levels We measured fasting serum BDNF levels by sandwich ELISA using a commercially available kit as described in our previous report (Chen et al., 2009; Xiu et al., 2009). All samples were assayed by a research assistant blind to the clinical situation. Inter- and intra-assay variation coefficients were 7 and 5%, respectively.
2.4. Determination of OS parameters All antioxidant enzymes and lipid peroxidation products in plasma were measured by the spectrophotometer. Among these OS parameters, superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase (CAT) are antioxidant enzymes, while malondialdehyde (MDA) is lipid peroxidation products. A full description of the assays has
203 been given in our previous report (Zhang et al., 2006). Briefly, lipid peroxidation levels were monitored by determining the end product of lipid peroxidation MDA using the thiobarbituric acid (TBA) method. MDA results were expressed as nmol/ml. The total SOD activity was determined using a standard assay involving spectrophotometric determination of the inhibition of superoxide-induced formation of nitrite from hydroxylamine. Xanthine—xanthine oxidase provided the superoxide source. One unit is defined as the amount of SOD that inhibits 50% of nitrite formation under the assay conditions. GSH-Px activity was measured by adding H2 O2 to the reaction mixture containing reduced glutathione (GSH), reduced nicotinamide adenine dinucleotide phosphate (NADPH) and glutathione reductase. One unit of GSH-Px is defined as micromoles of NADPH oxidized per minute. CAT activity was assayed based on the decomposition of hydrogen peroxide by CAT. CAT catalyzes the transformation of hydrogen peroxide to water and oxygen. CAT activity was determined by monitoring the decreased absorbance spectrophotometrically at 240 nm due to degradation of hydrogen peroxide. One unit of CAT was defined as the amount of enzyme that decomposes 1 mol H2 O2 /min. SOD, GSH-Px and CAT activities were expressed as units per milliliter plasma (U/ml). The intra- and inter-assay coefficient of variation for these antioxidant enzymes ranged from 3.2% to 5.9%.
2.5. Clinical assessment Four psychiatrists who had simultaneously attended a training session in the use of the Positive and Negative Syndrome Scale (PANSS) rated patients on this scale. After training, repeated assessment showed that the inter observer correlation coefficient was maintained at >0.8 for the PANSS total score. We obtained blood samples to assess BDNF and OS parameters at the time of PANSS ratings. In its original form, the PANSS was divided in three scales: positive (items P1 to P7), negative (items N1 to N7) and general psychopathology (items G1 to G16) (Kay et al., 1987). Later factorial analyses have presented data in favor of five-factor components, with factors commonly labeled as ‘positive’, ‘negative’, ‘cognitive’, ‘depression’ and ‘excitement’ (Wallwork et al., 2012). The cognitive factor (sometimes called ‘disorganization’) refers to the patient’s cognitive functioning, and is composed of several PANSS items that vary partially in the different factorial analyses (Rodriguez-Jimenez et al., 2013). Recently, Wallwork et al. (2012) have proposed a new consensus model of the cognitive factor that is only made up of three PANSS items: ‘Conceptual disorganization’ (P2), ‘Difficulty in abstract thinking’ (N5), and ‘Poor attention’ (G11). Subsequently, the cognitive factor was confirmed in the clinical assessment of patients with schizophrenia (Rodriguez-Jimenez et al., 2013), suggesting that the cognitive component of the PANSS is a valid measure of cognitive deficits in schizophrenia. The other factors in the five-factor model proposed by Wallwork et al. (2012) include an excited factor (items P4, P7, G8, G14) and a depressed factor (items G2, G3, G6).
204
2.6. Statistical analysis Since all the oxidative markers and BDNF were normally distributed in both the patient and control groups (Kolmogorov—Smirnov one-sample test: all p > 0.5), the main models consisted of a univariate analysis of covariance (ANCOVA) with gender, age, education, smoking and body mass index (BMI) as covariates for comparison of these biomarkers in the patient and control groups. Group differences were compared using Student’s t test or one-way analysis of variance (ANOVA) for continuous variables and Chi squared for categorical variables. Correlation between variables was studied using Pearson product moment correlations. Bonferroni corrections were applied to adjust for multiple testing. Lastly, exploratory regression analyses were used to examine the relationships between clinical phenotypes and BDNF or OS parameters in patients. Stepwise multiple regression analysis used the PANSS total or subscale scores including the PANSS cognitive or depressive factor as dependent variables with BDNF, antioxidant enzyme and MDA as the independent variables. Covariates in these stepwise forward entry models included age, gender, education, smoking, duration of illness, age of onset, and antipsychotic medication dosage, type (typical vs. atypical antipsychotics) and duration. Two-tailed significance values were used and significance levels were set at 0.05.
3. Results 3.1. Demographic data
X.Y. Zhang et al. Table 1 Demographics of patients and normal control subjects.
Sex, M/F Age (years) Education Smokers/nonsmokers Smoked cigarettes/day Body mass index (kg/m2 ) Age of onset (years) Duration of illness (years) Hospitalization numbers PANSS Positive symptom Negative symptom General psychopathology Total *
Table 1 shows the demographic data of the subjects in the present study. There was a significant difference in age between patient and control groups (p < 0.05), which was adjusted in the following analysis. No other significant difference in demographic data was noted between patient and control groups. A significant gender difference in BDNF levels was noted in patients (7.2 ± 2.3 ng/ml for males vs. 5.9 ± 2.2 ng/ml for females; F = 7.58, df = 1, 148, p = 0.007). In addition, there was a significant inverse relationship between age of onset of psychosis and SOD activities (r = −0.17, df = 162, p = 0.03), and a significant positive relationship between age and MDA (r = 0.17, df = 162, p = 0.03). No other significant association was observed between clinical parameters and BDNF or OS parameters.
3.2. OS parameters and BDNF in schizophrenia and controls Table 2 showed that both SOD and GSH-Px activities were significantly lower in patients than in controls (both p < 0.001), while MDA levels were markedly higher in patients than in controls (p < 0.001). When the effect of age, sex, education, smoking and BMI were examined by adding them to the ANOVA as covariates, a significant difference between patient and control groups was still observed in SOD (p < 0.005), GSH-Px (p < 0.001) and MDA (p < 0.001). No significant difference was noted in CAT between patients and controls.
Schizophrenia (n = 164)
Control subjects (n = 50)
122/42 48.3 ± 6.4 9.5 ± 2.4 95/58 13.6 ± 9.1
36/14 45.6 ± 5.1* 9.5 ± 1.4 31/18 12.6 ± 6.5
24.4 ± 3.5
23.5 ± 3.6
24.9 ± 6.4 23.4 ± 8.8
3.8 ± 2.4
16.0 ± 5.8 25.1 ± 6.0 34.4 ± 8.3 75.5 ± 15.7
p < 0.05.
BDNF serum levels were markedly lower in serum of patients than in controls (p < 0.001). When the effect of age, sex, education, smoking and BMI were examined by adding them to the ANOVA as covariates, a significant difference between patients and controls was still observed (p < 0.001).
3.3. Interaction of BDNF and OS parameters in schizophrenia and controls To detect the relationship between BDNF and OS parameters, partial correlation (controlled for age, gender, education, smoking and BMI) was performed. This analysis showed a significant negative association between BDNF levels and SOD activity in the patient group (r = −0.23, df = 142, p = 0.006; Fig. 1). No significant correlation was observed between BDNF and the other OS parameters (all p > 0.05). In addition, no significant correlations between BDNF levels and OS parameters were observed in healthy controls (all p > 0.05).
3.4. Association between BDNF, OS parameters and clinical symptoms in schizophrenia There was a significant positive correlation between SOD and positive subscore (r = 0.18, df = 164, p = 0.02), or PANSS total score (r = 0.16, df = 164, p = 0.04). A significant and negative correlation was noted between CAT and general
BDNF and oxidative stress in schizophrenia Table 2
Oxidative stress parameters and BDNF levels in schizophrenia and controls.
Markers
Schizophrenia (n = 164)
SOD (U/ml) GSH-Px (U/ml) CAT (U/ml) MDA (nmol/ml) BDNF (ng/ml)
85.9 108.8 2.6 11.3 6.8
a
205
± ± ± ± ±
19.5 31.5 3.2 7.6 2.4
Controls (n = 50) 97.2 147.2 2.7 2.6 9.5
± ± ± ± ±
10.9 31.0 2.4 1.9 4.4
F
df
13.3 39.5 0.05 56.4 29.2
1, 1, 1, 1, 1,
212 212 212 212 201
p
Adjusted pa
