Abnormal transbilayer distribution of phospholipids in red blood cell membranes in schizophrenia

Psychiatry Research 169 (2009) 91–96

Contents lists available at ScienceDirect

Psychiatry Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s

Abnormal transbilayer distribution of phospholipids in red blood cell membranes in schizophrenia Philippe Nuss a,b,c,⁎, Cedric Tessier a,b,c, Florian Ferreri a,c, Marc De Hert d, Joseph Peuskens d, Germain Trugnan a,b,c, Joelle Masliah a,b,c, Claude Wolf a,b,c a

UPMC Univ Paris 06, UMR_S 538, CHU St Antoine, 27, rue de Chaligny, 75012, Paris, France INSERM, UMR_S 538, 27, rue de Chaligny, 75012, Paris, France AP-HP, Hôpital Saint Antoine, 184, rue du faubourg St Antoine, 75012, Paris, France d University Psychiatric Center Katholieke Universiteit Leuven, Campus Kortenberg, Belgium b c

a r t i c l e

i n f o

Article history: Received 28 October 2007 Received in revised form 2 January 2009 Accepted 5 January 2009 Keywords: Schizophrenia Erythrocyte Membrane Phosphatidylethanolamine Lipids

a b s t r a c t Abnormalities in membrane lipids have been repeatedly reported in patients with schizophrenia. These abnormalities include decreased phosphatidylethanolamine (PE) and n−3 and n−6 polyunsaturated fatty acids in peripheral and brain cell membranes. The present study investigates the hypothesis of an overrepresentation of PE in the external leaflet of the red blood cell (RBC) membrane in patients with schizophrenia. The assumption was that this modification of PE asymmetrical distribution could explain the reported lipid membrane abnormalities. Phosphatidylethanolamine located in the external leaflet was specifically labeled in RBC membranes from 65 medicated patients with schizophrenia and 38 healthy controls. Labeled (external) and non-labeled (internal) PE and their respective fatty acid composition were analyzed by mass spectrometry. A significant increase in the percentage of external leaflet PE was found in RBC membranes in 63.1% of the patients. In this subgroup, a significant depletion of n−3 and n−6 polyunsaturated fatty acids from internally located PE was also observed. Age, sex and antipsychotic treatment were not associated with the transbilayer membrane distribution of PE. Potential mechanisms underlying these abnormalities may involve membrane phospholipid transporters or degradative enzymes involved in phospholipid metabolism. The anomaly described could characterize a subgroup among patients with schizophrenia. © 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Schizophrenia is a heterogeneous disorder in which the diversity of symptom presentation, clinical course and treatment outcome may arise from a combination of numerous inherited and environmental vulnerability factors (McGlashan and Hoffman, 2000; Mueser and McGurk, 2004; Maki et al., 2005; Riley and Kendler, 2006). The identification of accessible endophenotypes in schizophrenia could be of great useful to researchers and clinicians in the attempt to delineate more homogeneous subtypes of the disorder, and thus could help to improve prognosis and guide treatment choice. Several lines of evidence point towards lipid modifications (Keshavan et al., 2000) on both peripheral and central nervous cells in patients with schizophrenia (Yao et al., 2002). The relevance of these lipid modifications for schizophrenia may relate to their possible structural or functional implication in the physiopathology of this disease. For example, they may reflect modifications of lipid-derived neuromodulators such as the central endocannabinoid system which has been postulated to be ⁎ Corresponding author. UPMC Univ Paris 06, UMR_S 538, CHU St Antoine, 27, rue de Chaligny, 75012, Paris, France. Tel.: +33 149 282 655; fax: +33 140 011 347. E-mail address: [email protected] (P. Nuss). 0165-1781/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2009.01.009

perturbed in schizophrenia (Leweke et al., 1999; De Marchi et al., 2003; Giuffrida et al., 2004; Lafourcade et al., 2007). Alternatively, changes in the polyunsaturated fatty acid (PUFA) content of neuronal membranes may modify excitatory amino acid-mediated neurotransmission, which is known to be altered in schizophrenia (Chalon et al., 2001) and to interfere with 5-HT2 receptor functioning (Mato et al., 2007). In this context, several studies using n−3, n−6 diet restriction in rats or dietary supplementation with PUFAs suggest that PUFAs can modulate neurotransmission in the cortex (Chalon et al., 1998; Chalon et al., 2001), striatum (Yao et al., 2000) and hippocampus (Farkas et al., 2002). Modifications of PUFA intake have also been shown to interfere with the binding properties of 5-HT2, D2 and muscarinic receptors (Yao et al., 1996; Zimmer et al., 2000; du Bois et al., 2005). In patients with schizophrenia, magnetic resonance spectroscopy has provided evidence for a reduction in phosphomonoester and phosphodiester fatty acids in the medial and lateral prefrontal cortex, indicative of disrupted phospholipid metabolism (Smesny et al., 2007). These alterations have been confirmed by post-mortem and other in vivo studies (Horrobin, 1999; Jensen et al., 2002;Yao et al., 2002; Jayakumar et al., 2003; Giuffrida et al., 2004; McNamara et al., 2007; Smesny et al., 2007). In post-mortem brain tissues, lower amounts of phosphatidylcholine and phosphatidylethanolamine (PE),

92

P. Nuss et al. / Psychiatry Research 169 (2009) 91–96

accompanied by a marked reduction in total PUFAs and saturated fatty acids, have been found (Yao et al., 2000). In particular, a pronounced reduction of docosahexaenoic acid was seen in the orbitofrontal cortex (McNamara et al., 2007). The reduction in PUFA content could be attributed principally to a loss of n−6 PUFA (Yao et al., 2000). An increased activity of phospholipase A2 (PLA2), an enzyme involved in maintaining membrane phospholipid homeostasis as well as in cell signaling, has also been observed in serum and brain of patients with schizophrenia (Smesny et al., 2005; Law et al., 2006). Increased peroxidation may also be involved in the PUFA decrease. These results were found both in medicated and unmedicated patients. Interestingly, membrane lipid abnormalities have also been found in peripheral cells of patients with schizophrenia, with one study showing a correlation between central and peripheral membrane lipid abnormalities (Yao et al., 2002; Reddy and Keshavan, 2003). Modifications of overall membrane lipid composition in terms of phospholipids and PUFAs in red blood cells (RBCs) from patients with schizophrenia have been described in many studies (see review by Fenton et al., 2000). The most consistent findings have been a substantial decrease in PE content (Ryazantseva et al., 2002), accompanied by a parallel increase in sphingomyelin content (Sengupta et al., 1981; Lautin et al., 1982; Keshavan et al.,1993; Mahadik et al.,1994; Yao et al., 2000; Ponizovsky et al., 2001). Such changes have also been described in platelets, fibroblasts, and brain tissue membranes. With respect to fatty acids, a decrease in three n−3 and n−6 PUFAs, namely linoleic (C18:2), arachidonic (C20:4) and docosahexaenoic (C22:6) acids, has been observed in RBCs (Assies et al., 2001; Khan et al., 2002; Arvindakshan et al., 2003). Taken together, these findings suggest that PE is critical to the phospholipid abnormalities observed in schizophrenia. Altered PE handling may also account for the observed PUFA abnormalities, since this phospholipid is a major reservoir for n−3 and n−6 unsaturated fatty acids. The underlying molecular mechanism may involve alterations in phospholipid distribution between the external and internal membrane leaflets which are normally asymmetric. This asymmetry is tightly controlled by lipid transporters that specifically translocate particular lipid classes from one monolayer to the other against their concentration gradient, as well as by lipases (Daleke, 2003; Pomorski et al., 2004). The activity of these enzymes and transporters may be abnormal in schizophrenia. Although the evidence for membrane lipid abnormalities in patients with schizophrenia is strong, little is known about the phospholipid distribution abnormalities between the membrane inner and outer leaflets. In the present work, we focused our interest on the identification of the PE distribution between the inner and outer leaflets and their respective PUFA constituents on the RBC membranes of patients with schizophrenia. 2. Methods 2.1. Sampling and classification A fasting blood sample was taken from 65 medicated and stabilized patients with schizophrenia and 38 healthy controls. Blood samples were collected, on citric acid dextrose tubes, from psychiatric units in Reims and Amiens (France) and Kortenberg (Belgium) tertiary-care university hospitals. Lipid extraction, phospholipid and fatty acid analysis and quantification were performed at Saint-Antoine Hospital (tertiarycare University Hospital, Paris, France). All patients met strict DSM-IV criteria for schizophrenia.

clozapine, olanzapine, quetiapine, sertindole, risperidone). The majority of patients (64%) were receiving atypicals. No mood stabilizer was allowed. All patients gave their written informed consent in a protocol approved by the Medical Ethics Committee of Reims University Hospital (France) and the University Psychiatric Centre, Katholieke Universiteit Leuven, campus Kortenberg (Belgium). The control group composed of 38 healthy subjects matched for age was recruited among hospital staff and students. Family psychiatric history was investigated in controls; those with a family history of psychosis and/or bipolar disorder were excluded. Exclusion criteria for both groups were metabolic diseases, cholesterol-lowering treatments and dietary supplementation with PUFAs. Food intake was assessed through a questionnaire on diet (Garnier, 1991). The demographic and clinical characteristics of the study sample are presented in Table 1. 2.3. Labeling 2.3.1. External PE labeling Erythrocytes isolated from fresh citrated blood were harvested and washed three times within 3 h of blood puncture by centrifugation (10 min at 100 ×g) and resuspended in five volumes of a buffer consisting of 150 mM NaCl and 5 mM Tris HCl (pH 7.5). Phosphatidylethanolamine contained in the external membrane leaflet was labeled with trinitro-benzylsulfonic acid (TNBS, Sigma-Aldrich, France) under non-permeant conditions at 5 °C during 30 min (8 mM TNBS, 500 µl RBC) in Tris HCl buffer supplemented with NaHCO3 (5% W/V). The reaction was stopped by the addition of 0.5 M HCl. 2.3.2. Lipid extraction and phospholipid analysis Membrane phospholipids were extracted by the method of Bligh and Dyer (1959). Thin-layer chromatography on silica gel plates (Merck, France) with a solvent system consisting of chloroform/methanol/acetic acid/H2O (70:25:3.5:1.5 v/v) was used to separate external from internal PE, since the TNB-labeled PE (external) migrates faster than non-labeled PE (internal) in this system. After chromatography, the location of the free amino group phospholipid spots was revealed by spraying the plates with ninhydrin. The two spots were scraped off the plate and extracted with chloroform/ methanol 2:1 (v/v). Phosphatidylethanolamine from both fractions was saponified in 0.5 M KOH methanol at 60 °C for 15 min. Fatty acids released from their ester bonding were methylated using boron trifluoride–methanol complex reagent (Merck, 14%) for 15 min at 60 °C. 2.3.3. Fatty acid analysis Methylated fatty acids originating from the external and internal PE extracts were separated as a function of carbon number and unsaturation by gas chromatography on a highly polar capillary column coated with Supelcowax-10-bound phase (i.d. 0.32 mm, length 30 m, film thickness 0.25 μm (Supelco, Bellafonte, PA, USA)) fitted in a HewlettPackard (Palo Alto, CA, USA) 5890 Series II gas chromatograph. Methylated fatty acids were detected with picomolar sensitivity by mass spectrometry (Nermag 10-10 C, Quad-Service, Poissy, France) in the chemical ionisation mode with ammonia as the reagent gas. Positive quasimolecular ions [M + NH4] were selectively monitored and time integrated. Methylated fatty acids from each PE spot were quantified by spiking the extracts with heptadecanoic methyl ester as an internal standard. This artificial FA allowed quantification of the saturated (C16:0, C18:0) and unsaturated fatty acids (n−3: C20:5, C22:5, C22:6; n−6: C18:2, C20:4, C22:4 and n−9: C18:1). The FA response factor of the variety of FA was normalized in calibration mixtures (Supelco Mix-37). The total amount of PE was calculated from the sum of the individual methylated fatty acids. Fatty acids were measured in triplicate in each blood sample. 2.4. Statistical analysis The distribution of PE composition, defined as the % PE in the external leaflet, was quantified in the control group, assuming a Gaussian distribution. Mean values and standard deviations were calculated, as well as 95% confidence intervals (95%CI). The upper limit of the 95%CI was found to be 6.32. This cut-off was used to discriminate normal from abnormal PE distributions in the patient group. The number of subjects with abnormal PE distributions with respect to this threshold was then determined for the control and schizophrenia groups. The populations studied were compared using the chi-square test for categorical variables and Student's t-test for quantitative

Table 1 Population characteristics.

2.2. Patients and controls

Characteristic

Healthy controls

Patients with schizophrenia

The patients' sample comprised clinically stabilized patients with schizophrenia treated with antipsychotic medication. Minimum treatment duration was more than 1 year, with no change in treatment regiment within the last 3 months. They were referred by their psychiatrist (in- and out-patients) and fulfilled criteria for DSM-IV schizophrenia. Patients' functioning was assessed using the Global Assessment of Functioning (GAF) scale (adapted from DSM-IV). The overall psychiatric state based on the clinician's judgment was measured using the Clinical Global Impression (CGI) scale (Guy, 1976). Antipsychotic medication consisted of either classical neuroleptics (haloperidol, chlorpromazine, flupentixol) or atypicals (amisulpride, aripiprazole,

N Gender Male Female Age (years) Age of onset (years) Length of illness (years) Mean duration of antipsychotic therapy (years)

38

65

16 22 37 ± 10.9 – – –

37 28 37 ± 8.6 23 ± 4.4 14 ± 6.3 14 ± 6.3

P. Nuss et al. / Psychiatry Research 169 (2009) 91–96

Fig. 1. Percentage of phosphatidylethanolamine (PE) in the external leaflet of red blood cell membranes in patients with schizophrenia (SCZ) and matched controls. Data are represented as box plots with horizontal dark lines identifying mean values, white surfaces standard deviation and vertical T-lines data distribution. The total external PE percentage mean value in the schizophrenia population (7.4% ± 3.1) is significantly higher (P = 0.001) than in the healthy control group (4.0% ± 1.4).

variables. A risk level α = 0.05 for PE and of α = 0.001 for fatty acids was taken as the threshold for statistical significance. Comparisons were performed using SPSS software.

3. Results

93

Fig. 2. Existence of two clusters in the distribution of percentage of phosphatidylethanolamine (PE) among the group of patients with schizophrenia (SCZ) compared to a healthy control group. The nPE-SCZ group corresponds to patients with a similar distribution as healthy controls while the aPE-SCZ group refers to patients who differ significantly from controls (nPE-SCZ and aPE-SCZ standing for normal and abnormal external membrane PE percentage, respectively, in the schizophrenia population). Data are represented as box plots with horizontal dark lines identifying mean values, stripped surfaces standard deviation and vertical T-lines the data distribution. Mean percentage of PE in the external leaflet of the aPE-SCZ is significantly higher (P = 0.05) compared to both the healthy control and nPE-SCZ groups.

3.1. Phosphatidylethanolamine The mean percentage of PE in the external leaflet in healthy controls was 4.0% ± 1.4. The mean percentage in the total population of patients with schizophrenia (7.4% ± 3.1) was significantly higher (P = 0.001) (Fig. 1). In fact, the % PE values in the schizophrenia group were widely distributed, with 41 patients (63.1%) presenting a %PE above the upper 95% confidence interval for the control group (6.32%). This cut-off was thus used to define two populations of patients, a “normal group” of 24 patients (nPE-SCZ), with a mean percentage of PE in the external leaflet of 4.1% ± 1.1, and an “abnormal group” of 41 patients (aPE-SCZ) with a mean percentage of 9.4% ± 2.2, significantly higher (P = 0.05) compared with both the healthy control and the nPE-SCZ groups (Table 2 and Fig. 2). In subsequent analyses of the membrane asymmetry of PE and PUFA, these two groups were analyzed separately. Two patients (patients 13 and 42) in whom mean %PE values were very high (15.67 and 13.3) have been excluded for the calculation of mean values and standard deviations.

n−9. For each class, data are expressed as a percentage of the total amount of fatty acids for each leaflet. In both leaflets, n−6 PUFAs were the predominant fatty acid species, whereas n−3 PUFAs were the least abundant fatty acids measured. Saturated fatty acids (C16:0, C18:0) were equally distributed between the two leaflets in healthy controls, as well as in both the aPE-SCZ and nPE-SCZ populations. In healthy controls, n−9 fatty acids (C18:1) were significantly (P b 0.001) more abundant in the internal leaflet compared to the external leaflet. This was also the case in RBC membranes of the aPESCZ population, whereas a non-significant trend in the same direction was observed in the nPE-SCZ population. In contrast, both n−3 fatty acids (C20:5, C22:5, C22:6) and n−6 fatty acids (C18:2, C20:4, C22:4) were significantly (P b 0.001) less abundant in the internal than in the external leaflet in samples isolated from the aPE-SCZ population, whereas such a difference was not observed in the healthy control or nPE-SCZ groups. 4. Discussion

3.2. Fatty acid composition of PE from external and internal leaflets Fatty acids from PE extracts of the external and internal leaflets were classified into saturated and unsaturated species (Fig. 3). The latter group was further divided into three classes: n−3, n−6, and

Table 2 Distribution of subjects in the schizophrenia and control groups as a function of the cut-off value for the normal percentage of PE in the external leaflet. External PE%

Controls nPE-SCZ aPE-SCZ Total

b6.32 (normal) (nPE-SCZ)

N 6.32 (abnormal) (aPE-SCZ)

Male

Female

Male

Female

15 14 0

20 10 0

1 0 23

2 0 18

59

The gender ratio in each sub-group is indicated.

44

Total 38 24 41 103

The present results are consistent with previously described anomalies in membrane phospholipids and PUFA in schizophrenia. On the basis of the transbilayer distribution of PE, we were able to discriminate between two subgroups of patients; in subgroup (aPESCZ), which corresponds to two-thirds of the included patients, PE was overrepresented in the external membrane leaflet whereas, in the other (nPE-SCZ), the transbilayer distribution of PE did not differ from healthy controls and was consistent with normative values (Marinetti and Love, 1976). In addition, a significant decrease of internal n−3, n−6 PUFAs was also seen in the aPE-SCZ subgroup. The observed modifications of PLs and PUFAs in the aPE-SCZ subgroup were not related to age or gender, which were comparable between the two subgroups. This is consistent with previous findings showing lipid anomalies to be independent of age and gender (Keshavan et al., 1993; Ponizovsky et al., 2001; Arvindakshan et al., 2003). Diet and liver function are modulating the availability of essential PUFAs. Lipid intake was thus evaluated in the diet of patients

94

P. Nuss et al. / Psychiatry Research 169 (2009) 91–96

Fig. 3. Comparison of PE fatty acids between the inner and outer leaflets (in % of total fatty acids). Dark grey bars: inner leaflet; light grey bars: outer leaflet. Controls, nPE-SCZ (nPE standing for normal external PE percentage) and aPE-SCZ (aPE standing for abnormal external PE percentage) are represented for each of the fatty acid families. a) Saturated fatty acids (C16:0, C18:0), b) n−9 fatty acid (C18:1), c) n−3 fatty acids (C20:5, C22:5, C22:6), d) n−6 fatty acids (C18:2, C20:4, C22:4). A significant decrease (P b 0.001) of n−3 and n−6 fatty acids is observed in the internally versus externally located PE of the aPE-SCZ population compared to control and nPE-SCZ populations.

and controls; no dietary differences were observed between the groups. Red blood cell membrane PUFA content was compared to serum PUFA value for each schizophrenia subgroup. The value of the overall membrane PE PUFA decrease in the aPE-SCZ population was not correlated with serum PUFA value, suggesting that membrane PE PUFA content is not directly correlated with serum PUFA value in the aPE-SCZ population. Of interest, no correlation was identified between antipsychotic drug treatment regimen and membrane PE distribution, as all patients had been treated with antipsychotic medications for more than 1 year. Previous work indicates that the abnormalities in phospholipids and PUFAs detected in RBC membranes from patients with schizophrenia are similar for medicated and unmedicated patients (Schmitt et al., 2001; Levant et al., 2006). However, since certain antipsychotic medications may interfere with membrane lipid handling (Fukuzako et al., 1999; Schmitt et al., 2001; Tessier et al., 2008), we studied whether any individual antipsychotic drug could be associated with the abnormal PE distribution observed. Clozapine-treated patients were more numerous in the aPE-SCZ group but were also found in the nPE-SCZ group (respectively 25% and 4.2%) (Table 3). However, in experimental animals, clozapine did not modify the fatty acid composition of brain membranes (Thomas and Yao, 2007). This over-representation of clozapine-treated patients in the aPE-SCZ population may be related to the fact that clozapine is prescribed for treatment-resistant patients. Clinical parameters of the schizophrenia population were examined in order to identify clinical differences between aPE-SCZ and nPE-SCZ patients. We determined that patients from the aPE-SCZ population more frequently were characterized by clinical markers of disease severity. While 47% of the aPE-SCZ

population had a Global Assessment of Functioning (GAF) score less than 60, only 20% of the nPE-SCZ group met this criterion. Further, a CGI score of 4 or more (significantly or severely ill) was found only for patients from the aPE-SCZ population. Age of onset was earlier in the aPE-SCZ population: while a first episode before the age of 20 was found in 20% of the patients from this subgroup, no patient in the nPESCZ population had a first episode before that age. The existence of a large proportion of schizophrenia patients with an abnormal transbilayer distribution of PE is puzzling. Indeed, in all described diseases in which an abnormal phospholipid distribution has been described, only phosphatidylserine (PS) has been involved (Zwaal et al., 2005). Membrane asymmetry is a dynamic state that is maintained by specific membrane proteins which continuously transfer phospholipids from one leaflet to another (Seigneuret and Devaux, 1984). The abnormality in lipid distribution found in RBC membranes

Table 3 Antipsychotic medication prescription in each subgroup of patients with schizophrenia.

Amisulpride Olanzapine Risperidone Haloperidol Chlorpromazine Clozapine Quetiapine Flupentixol Sertindole Aripiprazole

nPE-SCZ (%)

aPE-SCZ (%)

25 25 20.8 16.7 4.2 4.2 0 0 4.2 0

7.5 17.5 30 5 2.5 25 2.5 2.5 5 2.5

P. Nuss et al. / Psychiatry Research 169 (2009) 91–96

from the aPE-SCZ group is likely to result from a permanent modification of the dynamic equilibrium in the distribution of PE and n−3 and n−6 PUFAs between the two leaflets. Over-representation of PE in the external leaflet in patients with schizophrenia may result either from decreased PE internalization or increased PE externalization or both mechanisms. These processes involve two distinct families of PL transporter, namely aminophospholipid translocases (AMPT), responsible for inward movement of PE and PS, and scramblases, responsible for outward movement of these same PLs. Aminophospholipid translocase is more efficient at translocating PS compared to PE, resulting in a higher internalization rate for PS, and the absence of measurable PS in the outer leaflet (Morrot et al., 1989). If a deficit in AMPT activity was responsible for the excess PE in the external leaflet, it would thus be expected that PS would also be present in the external leaflet, which was not the case. The alternative hypothesis would be increased PE externalization due to overactivity of scramblases. Among scramblases, calcium-dependent scramblase is the only transporter located in the cytoplasm. This enzyme is activated in stimulated cells and induces a rapid exposure of PS in the external leaflet (Daleke, 2003). Since PS is not found in the external leaflet in the aPE-SCZ population, stimulation of the activity of this scramblase cannot explain the observed increase in PE externalization. A more complex mechanism involving interaction between several transporters is thus presumed. Furthermore, the observed decrease of n−3 and n−6 PUFAs from the internally located PE pool may result from mechanisms involving PLA2 isozymes. Previous data have shown an increased activity of PLA2 in brain and peripheral tissues of schizophrenia patients (Smesny et al., 2004; Barbosa et al., 2007). Calcium-independent phospholipase A2 (iPLA2) is an intracellular enzyme specific for PUFA-rich PE. An increased activity of this enzyme could explain the altered distribution of n−3 and n−6 PUFAs from PE. This hypothesis is supported by results showing a significant allelic (P = 0.0085) and genotypic (P = 0.02) association with iPLA2 gene polymorphism in a case control study comparing 240 schizophrenia patients and 312 healthy controls (Junqueira et al., 2004). An alternative mechanism to account for depletion of PUFA from the internal leaflet would be excessive extracellular PLA2 activity. This would lead to depletion of PE from the extracellular leaflet, which would however be immediately replaced by translocation of PE from the internal leaflet pool, resulting in overall loss of n−3 and n−6 enriched PE from the internal leaflet (Bitbol and Devaux, 1988; Bevers et al., 1999; Florin-Christensen et al., 2001). Our study has certain limitations. The sample size was relatively small. This study applied in a non-medicated patients population is needed to confirm the fact that medication does not interfere with PE distribution. Analysis of RBC membranes from patients with other mental diseases is required to get a better understanding of the usefulness of this potential peripheral biomarker for schizophrenia. Other PLs such as PC, SM and PI have not been evaluated as potential candidates to account for the membrane lipid anomalies. Future experimentation will compare the transbilayer distribution of PE between peripheral and central PL turnover within the same patient using the 31P-MRS technique. Association between the existence of this biomarker and transition to psychosis in high-risk individuals could help identify if this potential schizophrenia endophenotypic marker could also be used as a vulnerability marker. In conclusion, our results show that a subgroup of patients with schizophrenia is characterized by abnormal transbilayer distribution of PE in RBC membranes. Identification of such a phenotypic marker could help individualize patient subgroups exhibiting specific clinical outcome, cognitive dysfunction, brain abnormalities and treatment response.

Acknowledgments The authors thank Mr. Hugo Trespalacios who provided patients, Mr. David Sanger for his encouragements, Mr. Alexandre Salvador and Mr. Jean Cougnard for their critical

95

advice in the statistical analysis and Mrs. Françoise Chevy for her interest on lipidomic for human diseases.

References Arvindakshan, M., Sitasawad, S., Debsikdar, V., Ghate, M., Evans, D., Horrobin, D.F., Bennett, C., Ranjekar, P.K., Mahadik, S.P., 2003. Essential polyunsaturated fatty acid and lipid peroxide levels in never-medicated and medicated schizophrenia patients. Biological Psychiatry 53, 56–64. Assies, J., Lieverse, R., Vreken, P., Wanders, R.J., Dingemans, P.M., Linszen, D.H., 2001. Significantly reduced docosahexaenoic and docosapentaenoic acid concentrations in erythrocyte membranes from schizophrenic patients compared with a carefully matched control group. Biological Psychiatry 49, 510–522. Barbosa, N.R., Junqueira, R.M., Vallada, H.P., Gattaz, W.F., 2007. Association between BanI genotype and increased phospholipase A2 activity in schizophrenia. European Archives of Psychiatry and Clinical Neuroscience 257, 340–343. Bevers, E.M., Comfurius, P., Dekkers, D.W., Zwaal, R.F., 1999. Lipid translocation across the plasma membrane of mammalian cells. Biochimica et Biophysica Acta 1439, 317–330. Bitbol, M., Devaux, P.F., 1988. Measurement of outward translocation of phospholipids across human erythrocyte membrane. Proceedings of the National Academy of Sciences of the United States of America 85, 6783–6787. Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology 37, 911–917. Chalon, S., Delion-Vancassel, S., Belzung, C., Guilloteau, D., Leguisquet, A.M., Besnard, J.C., Durand, G., 1998. Dietary fish oil affects monoaminergic neurotransmission and behavior in rats. Journal of Nutrition 128, 2512–2519. Chalon, S., Vancassel, S., Zimmer, L., Guilloteau, D., Durand, G., 2001. Polyunsaturated fatty acids and cerebral function: focus on monoaminergic neurotransmission. Lipids 36, 937–944. Daleke, D.L., 2003. Regulation of transbilayer plasma membrane phospholipid asymmetry. Journal of Lipid Research 44, 233–242. De Marchi, N., De Petrocellis, L., Orlando, P., Daniele, F., Fezza, F., Di Marzo, V., 2003. Endocannabinoid signalling in the blood of patients with schizophrenia. Lipids in Health and Disease 2, 5. du Bois, T.M., Deng, C., Huang, X.F., 2005. Membrane phospholipid composition, alterations in neurotransmitter systems and schizophrenia. Progress in NeuroPsychopharmacology & Biological Psychiatry 29, 878–888. Farkas, E., de Wilde, M.C., Kiliaan, A.J., Meijer, J., Keijser, J.N., Luiten, P.G., 2002. Dietary long chain PUFAs differentially affect hippocampal muscarinic 1 and serotonergic 1A receptors in experimental cerebral hypoperfusion. Brain Research 954, 32–41. Fenton, W.S., Hibbeln, J., Knable, M., 2000. Essential fatty acids, lipid membrane abnormalities, and the diagnosis and treatment of schizophrenia. Biological Psychiatry 47, 8–21. Florin-Christensen, J., Suarez, C.E., Florin-Christensen, M., Wainszelbaum, M., Brown, W.C., McElwain, T.F., Palmer, G.H., 2001. A unique phospholipid organization in bovine erythrocyte membranes. Proceedings of the National Academy of Sciences of the United States of America 98, 7736–7741. Fukuzako, H., Fukuzako, T., Kodama, S., Hashiguchi, T., Takigawa, M., Fujimoto, T., 1999. Haloperidol improves membrane phospholipid abnormalities in temporal lobes of schizophrenic patients. Neuropsychopharmacology 21, 542–549. Garnier, D., 1991. EDI2, Eating Disorder Inventory—2. Psychological Assessment Resources, Florida. Giuffrida, A., Leweke, F.M., Gerth, C.W., Schreiber, D., Koethe, D., Faulhaber, J., Klosterkotter, J., Piomelli, D., 2004. Cerebrospinal anandamide levels are elevated in acute schizophrenia and are inversely correlated with psychotic symptoms. Neuropsychopharmacology 29, 2108–2114. Guy, W., 1976. Clinical global impression (CGI) in assessment manual for psychopharmacology. National Institute of Mental Health 218, 22. Horrobin, D.F., 1999. Lipids and schizophrenia. British Journal of Psychiatry 175, 88. Jayakumar, P.N., Gangadhar, B.N., Subbakrishna, D.K., Janakiramaiah, N., Srinivas, J.S., Keshavan, M.S., 2003. Membrane phospholipid abnormalities of basal ganglia in never-treated schizophrenia: a 31P magnetic resonance spectroscopy study. Biological Psychiatry 54, 491–494. Jensen, J.E., Al-Semaan, Y.M., Williamson, P.C., Neufeld, R.W., Menon, R.S., Schaeffer, B., Densmore, M., Drost, D.J., 2002. Region-specific changes in phospholipid metabolism in chronic, medicated schizophrenia: (31)P-MRS study at 4.0 Tesla. British Journal of Psychiatry 180, 39–44. Junqueira, R., Cordeiro, Q., Meira-Lima, I., Gattaz, W.F., Vallada, H., 2004. Allelic association analysis of phospholipase A2 genes with schizophrenia. Psychiatric Genetics 14, 157–160. Keshavan, M.S., Mallinger, A.G., Pettegrew, J.W., Dippold, C., 1993. Erythrocyte membrane phospholipids in psychotic patients. Psychiatry Research 49, 89–95. Keshavan, M.S., Stanley, J.A., Pettegrew, J.W., 2000. Magnetic resonance spectroscopy in schizophrenia: methodological issues and findings—part II. Biological Psychiatry 48, 369–380. Khan, M.M., Evans, D.R., Gunna, V., Scheffer, R.E., Parikh, V.V., Mahadik, S.P., 2002. Reduced erythrocyte membrane essential fatty acids and increased lipid peroxides in schizophrenia at the never-medicated first-episode of psychosis and after years of treatment with antipsychotics. Schizophrenia Research 58, 1–10. Lafourcade, M., Elezgarai, I., Mato, S., Bakiri, Y., Grandes, P., Manzoni, O.J., 2007. Molecular components and functions of the endocannabinoid system in mouse prefrontal cortex. PLoS ONE 2, e709. Lautin, A., Cordasco, D.M., Segarnick, D.J., Wood, L., Mason, M.F., Wolkin, A., Rotrosen, J., 1982. Red cell phospholipids in schizophrenia. Life Sciences 31, 3051–3056. Law, M.H., Cotton, R.G., Berger, G.E., 2006. The role of phospholipases A2 in schizophrenia. Molecular Psychiatry 11, 547–556.

96

P. Nuss et al. / Psychiatry Research 169 (2009) 91–96

Levant, B., Crane, J.F., Carlson, S.E., 2006. Sub-chronic antipsychotic drug treatment does not alter brain phospholipid fatty acid composition in rats. Progress in NeuroPsychopharmacology & Biological Psychiatry 30, 728–732. Leweke, F.M., Giuffrida, A., Wurster, U., Emrich, H.M., Piomelli, D., 1999. Elevated endogenous cannabinoids in schizophrenia. Neuroreport 10, 1665–1669. Mahadik, S.P., Mukherjee, S., Correnti, E.E., Kelkar, H.S., Wakade, C.G., Costa, R.M., Scheffer, R., 1994. Plasma membrane phospholipid and cholesterol distribution of skin fibroblasts from drug-naive patients at the onset of psychosis. Schizophrenia Research 13, 239–247. Maki, P., Veijola, J., Jones, P.B., Murray, G.K., Koponen, H., Tienari, P., Miettunen, J., Tanskanen, P., Wahlberg, K.E., Koskinen, J., Lauronen, E., Isohanni, M., 2005. Predictors of schizophrenia—a review. British Medical Bulletin 73–74, 1–15. Marinetti, G.V., Love, R., 1976. Differential reaction of cell membrane phospholipids and proteins with chemical probes. Chemistry and Physics of Lipids 16, 239–254. Mato, S., Aso, E., Castro, E., Martin, M., Valverde, O., Maldonado, R., Pazos, A., 2007. CB(1) knockout mice display impaired functionality of 5-HT(1A) and 5-HT(2A/C) receptors. Journal of Neurochemistry 103 (5), 2111–2120. McGlashan, T.H., Hoffman, R.E., 2000. Schizophrenia as a disorder of developmentally reduced synaptic connectivity. Archives of General Psychiatry 57, 637–648. McNamara, R.K., Jandacek, R., Rider, T., Tso, P., Hahn, C.G., Richtand, N.M., Stanford, K.E., 2007. Abnormalities in the fatty acid composition of the postmortem orbitofrontal cortex of schizophrenic patients: gender differences and partial normalization with antipsychotic medications. Schizophrenia Research 91, 37–50. Morrot, G., Herve, P., Zachowski, A., Fellmann, P., Devaux, P.F., 1989. Aminophospholipid translocase of human erythrocytes: phospholipid substrate specificity and effect of cholesterol. Biochemistry 28, 3456–3462. Mueser, K.T., McGurk, S.R., 2004. Schizophrenia. Lancet 363, 2063–2072. Pomorski, T., Holthuis, J.C., Herrmann, A., van Meer, G., 2004. Tracking down lipid flippases and their biological functions. Journal of Cell Science 117, 805–813. Ponizovsky, A.M., Modai, I., Nechamkin, Y., Barshtein, G., Ritsner, M.S., Yedgar, S., Lecht, S., Bergelson, L.D., 2001. Phospholipid patterns of erythrocytes in schizophrenia: relationships to symptomatology. Schizophrenia Research 52, 121–126. Reddy, R., Keshavan, M.S., 2003. Phosphorus magnetic resonance spectroscopy: its utility in examining the membrane hypothesis of schizophrenia. Prostaglandins, Leukotrienes and Essential Fatty Acids 69, 401–405. Riley, B., Kendler, K.S., 2006. Molecular genetic studies of schizophrenia. European Journal of Human Genetics 14, 669–680. Ryazantseva, N.V., Novitskii, V.V., Kublinskaya, M.M., 2002. Changes in the lipid phase of erythrocyte membranes in patients with paranoid schizophrenia. Bulletin of Experimental Biology and Medicine 133, 84–86.

Schmitt, A., Maras, A., Petroianu, G., Braus, D.F., Scheuer, L., Gattaz, W.F., 2001. Effects of antipsychotic treatment on membrane phospholipid metabolism in schizophrenia. Journal of Neural Transmission 108, 1081–1091. Seigneuret, M., Devaux, P.F., 1984. ATP-dependent asymmetric distribution of spinlabeled phospholipids in the erythrocyte membrane: relation to shape changes. Proceedings of the National Academy of Sciences of the United States of America 81, 3751–3755. Sengupta, N., Datta, S.C., Sengupta, D., 1981. Platelet and erythrocyte membrane lipid and phospholipid patterns in different types of mental patients. Biochemical Medicine 25, 267–275. Smesny, S., Rosburg, T., Klemm, S., Riemann, S., Baur, K., Rudolph, N., Grunwald, S., Sauer, H., 2004. The influence of age and gender on niacin skin test results — implications for the use as a biochemical marker in schizophrenia. Journal of Psychiatric Research 38, 537–543. Smesny, S., Kinder, D., Willhardt, I., Rosburg, T., Lasch, J., Berger, G., Sauer, H., 2005. Increased calcium-independent phospholipase A2 activity in first but not in multiepisode chronic schizophrenia. Biological Psychiatry 57, 399–405. Smesny, S., Rosburg, T., Nenadic, I., Fenk, K.P., Kunstmann, S., Rzanny, R., Volz, H.P., Sauer, H., 2007. Metabolic mapping using 2D (31)P-MR spectroscopy reveals frontal and thalamic metabolic abnormalities in schizophrenia. Neuroimage 35, 729–737. Tessier, C., Nuss, P., Staneva, G., Wolf, C., 2008. Modification of membrane heterogeneity by antipsychotic drugs: an X-ray diffraction comparative study. Journal of Colloid Interface Science 320, 469–475. Thomas, E.A., Yao, J.K., 2007. Clozapine specifically alters the arachidonic acid pathway in mice lacking apolipoprotein D. Schizophrenia Research 89, 147–153. Yao, J.K., van Kammen, D.P., Moss, H.B., Sokulski, D.E., 1996. Decreased serotonergic responsivity in platelets of drug-free patients with schizophrenia. Psychiatry Research 63, 123–132. Yao, J.K., Leonard, S., Reddy, R.D., 2000. Membrane phospholipid abnormalities in postmortem brains from schizophrenic patients. Schizophrenia Research 42, 7–17. Yao, J., Stanley, J.A., Reddy, R.D., Keshavan, M.S., Pettegrew, J.W., 2002. Correlations between peripheral polyunsaturated fatty acid content and in vivo membrane phospholipid metabolites. Biological Psychiatry 52, 823–830. Zimmer, L., Delion-Vancassel, S., Durand, G., Guilloteau, D., Bodard, S., Besnard, J.C., Chalon, S., 2000. Modification of dopamine neurotransmission in the nucleus accumbens of rats deficient in n−3 polyunsaturated fatty acids. Journal of Lipid Research 41, 32–40. Zwaal, R.F., Comfurius, P., Bevers, E.M., 2005. Surface exposure of phosphatidylserine in pathological cells. Cellular aand Molecular Life Sciences 62, 971–988.