Erythrocyte membrane nervonic acid in drug-naive first-episode psychosis and chronic medicated schizophrenia: Implication for impaired myelination and prognosis

Abstract

BACKGROUND

Nervonic acid (NA, C24: 1, w9) is a monounsaturated fatty acid that plays a crucial role in myelination and motor function. It also regulates cognitive and metabolic functions, suggesting that impaired NA metabolism may contribute to the pathophysiology of schizophrenia. Although several studies have measured erythrocyte membrane NA in first-episode psychosis (FEP), findings are conflicting, and the fate of NA in patients with chronic schizophrenia (CSZ) or under long-term antipsychotic treatment schedules remains unknown.

AIM

To measure erythrocyte membrane NA and determine its association with psychopathology and metabolic parameters in drug-naive patients with FEP and antipsychotic-treated patients with CSZ.

METHODS

In this study, twenty-one drug-naive patients with FEP, twenty patients with CSZ treated with atypical antipsychotics, and fourteen healthy male subjects were analyzed. Erythrocyte membrane NA was measured using ultrathin capillary gas chromatography, plasma leptin was measured using enzyme-linked immunosorbent assay, and body mass index (BMI) was calculated by using the formula: Weight (kg)/height (m²). Psychiatric symptoms were evaluated using the brief psychiatry rating scale and the positive and negative syndrome scale (PANSS). Pearson correlation coefficient (r) was computed to find the association between erythrocyte membrane NA, PANSS scores, plasma leptin, and BMI.

RESULTS

In patients with FEP, erythrocyte NA was non-significantly increased (about 12%) and negatively correlated with negative symptoms (PANSS-negative symptom scores, r = -0.4323, P = 0.023) but not with positive symptoms (PANSS-positive symptom scores, r = -0.2915, P = 0.09). In patients with CSZ, erythrocyte NA was reduced considerably (about 40%, P < 0.001 vs FEP and about 30% vs control (CNT) subjects, P = 0.037) and negatively correlated with both PANSS-negative symptom scores (r = -0.4562, P = 0.013) and PANSS-positive symptom scores (r = -0.3911, P = 0.041). Additionally, in patients with FEP, erythrocyte NA was not significantly correlated either with BMI (r = -0.2532, P = 0.231) or plasma leptin (r = -0.3001, P = 0.102). However, in patients with CSZ, it did negatively correlate with both BMI (r = -0.4721, P = 0.029) and plasma leptin (r = -0.4701, P = 0.031).

CONCLUSION

Erythrocyte membrane NA level could be used for predicting the development of metabolic abnormalities, treatment resistance, and prognosis in schizophrenia.

Key Words: Nervonic acid; First-episode psychosis; Chronic schizophrenia; Demyelination; Metabolic abnormalities

Core Tip: Nervonic acid (NA, C24: 1, w9) is a monounsaturated fatty acid that plays a crucial role in myelination and motor function. It also regulates cognitive and metabolic functions, suggesting that impaired NA metabolism may contribute to the pathophysiology of schizophrenia. Although several studies have measured erythrocyte membrane NA in first-episode psychosis, findings are conflicting, and the fate of NA in patients with chronic schizophrenia or under long-term antipsychotic treatment schedule remains unknown. This study was designed to analyze erythrocyte NA composition and its association with psychopathology and metabolic comorbidities in drug-naive patients with first-episode psychosis and antipsychotic-treated patients with chronic schizophrenia.

INTRODUCTION

Lipid abnormalities are prominent features of psychiatric disorders, including schizophrenia[1-6]. In this regard, increased levels of saturated fatty acids (SFAs), monounsaturated fatty acids (MUFAs), their ceramide derivatives, and reduced levels of polyunsaturated fatty acids (PUFAs) have been observed in the membrane phospholipids from patients with first-episode psychosis (FEP)[1,3,5-7]. Among various MUFAs, nervonic acid (NA, C24: 1; w-9) plays a crucial role in myelination, cognitive function, and brain repair[8-11]. It also regulates metabolic functions, including insulin sensitivity and body weight[12], suggesting that impaired NA metabolism could be a potential pathological factor in schizophrenia.

In recent years, several studies have analyzed the therapeutic potential of NA in laboratory animals. In this context, NA supplementation has been found to enhance neurogenesis, insulin sensitivity, and improve motor function[11-13]. Additionally, NA supplementation in mice has also been found to reduce weight gain/adiposity and glucose intolerance induced by an NA-deficient high-fat diet[12]. These evidence suggest that NA supplementation could be effective in improving therapeutic outcomes in schizophrenia. Although several clinical trials have been conducted with PUFA supplementation in schizophrenia, the outcome remains controversial and unconvincing[14,15]. And while no attention is given to the NA supplementation in schizophrenia, it could play a crucial role in developing personalized treatment for patients with NA deficiency, pre-existing or induced during the development of metabolic abnormalities, including diabetes and obesity, as a result of long-term treatment with antipsychotic drugs (APDs), which remains to be analyzed.

Over the years, several studies have measured erythrocyte membrane NA in patients with FEP; however, the findings are conflicting, and the fate of membrane NA in patients with chronic schizophrenia (CSZ) (or treatment-resistant schizophrenia) remains unclear[1,3-6,16,17]. Moreover, the association of erythrocyte NA with metabolic parameters, including plasma leptin and body mass index (BMI), has not been studied in patients with FEP or CSZ. Additionally, no study has ever analyzed the association of NA with plasma leptin and BMI in the same group of patients with schizophrenia. These findings will play a crucial role in assessing the nature of association between membrane NA, plasma leptin, BMI, and psychopathology in schizophrenia and revealing the role of NA in metabolic comorbidities, especially weight gain induced by long-term treatment with APDs.

In this study, erythrocyte membrane NA was measured, and its association with psychopathology, plasma leptin, and BMI was studied in patients with FEP and CSZ. Also, the role of NA in various pathological landmarks, including obesity, impaired neurogenesis, and demyelination in schizophrenia, is discussed. Since NA could be obtained through the diet as well as via synthesis through de novo lipogenesis (DNL), the role of DNL in NA abnormalities in patients with FEP and CSZ is also discussed.

MATERIALS AND METHODS

Patients and control subjects

In this study, a total of twenty-one (n = 21) drug-naive patients with FEP, twenty (n = 20) medicated patients with CSZ, and fourteen (n = 14) male control (CNT) subjects were analyzed. These patients were also used earlier by our group for other studies[1,2]. Patients with FEP were enrolled from consecutive admissions at the Department of Psychiatry, Dwight David Eisenhower Army Medical Centre (DDEAMC), Fort Gordon, GA. The patients were mostly active-duty army personnel diagnosed with schizophrenia or schizophreniform disorder using Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria and after a six-month follow-up period during subsequent hospitalization. The BMI was calculated according to the formula BMI = kg/m², where kg is body weight in kilogrammes and m is the height in metres[1,2].

Symptom evaluation

We evaluated the symptoms of the patients with FEP at the first visit using the Brief Psychiatric Rating Scale (BPRS) and the positive and negative syndrome scale (PANSS)[18,19]. We examined total BPRS scores, PANSS-positive symptom scores (PANSS-PSS comprised of sum of scores on unusual thoughts, contents and suspiciousness, delusion, hallucination, and conceptual disorganization), and PANSS-negative symptom scores (PANSS-NSS comprised of sum of scores on motor retardation, emotional withdrawal, and blunted effect). The mean age of the patients with FEP at the time of diagnosis was 22.40 ± 4.08 years. Patients with CSZ were recruited and analyzed at the Mental Health Service, Veteran Affair Medical Centre (VAMC), Augusta, GA, United States. The same methodology of BPRS and PNASS scoring system, as applied in case of FEP, was used for the evaluation of clinical symptoms of patients with CSZ. At the time of sample collection, patients with CSZ were on scheduled treatment with different atypical APDs, including clozapine (n = 14, duration 3.5 ± 1.5 years), olanzapine (n = 4, duration 6.4 ± 2.1 years), or risperidone (n = 2, duration 5 months). It is to mention that after discharge from Army Medical Centre, FEP patients are admitted to the Psychiatry Clinic at the VAMC. Therefore, both FEP and CSZ patients included in this study are a unique population having all the similarities other than antipsychotic treatment and duration of the illness. The healthy volunteers representing CNT subjects (n = 14) were recruited at the Medical College of Georgia, VAMC, and DDEAMC via advertisements. The age and gender of the CNT subjects were matched with the patients with FEP. The clinical parameters along with demographic properties of these patients have been published earlier and are also included in this paper in Table 1 after some modifications[1,2]. The research protocol was approved by the Institutional Review Board of DDEAMC and Medical College of Georgia, Augusta University, Augusta, GA. All patients and CNT subjects were asked to sign a consent form before taking blood samples.

Table 1 Demographic and clinical characteristics of first-episode psychosis and chronic schizophrenia patients, mean ± SD.

Characteristics	CNT	FEP	CSZ
Age (year)	25 ± 7.6	23.54 ± 4.65	42.23 ± 5.12
Gender (male:female)	14:0	21:0	20:0
Age at onset of psychosis	-	22.30 ± 4.48	23.15 ± 6.55
Years of illness	-	≤ 5.0 days	22.77 ± 7.21
Total BPRS total	-	45.18 ± 12.53	38.17 ± 6.96
Total PANSS-PSS	-	21.03 ± 4.81	12.88 ± 4.10
Total PANSS-NSS	-	20.91 ± 5.10	07.82 ± 2.31
Plasma leptin (ng/mL)	5.56 ± 0.78	4.28 ± 1.42	08.33 ± 1.42
BMI (kg/m2)	25.1 ± 2.10	23.3 ± 2.22	29.66 ± 3.73
Smoking	-	2/21	3/20
Antipsychotic use	-	-	+++
Tobacco	-	-	-
Cannabis	-	-	-

CNT: Control subjects; FEP: First-episode psychosis; CSZ: Chronic schizophrenia; BPRS: Brief Psychiatric Rating Scale; PANSS-PSS: Positive and negative syndrome scale-positive symptom scores; PANSS-NSS: Positive and negative syndrome scale-negative symptom scores; BMI: Body mass index; +++: Treated with various antipsychiotic drugs.

Patient selection and exclusion criteria

All FEP and CSZ patients were analyzed in the present study based on the following criteria: All were physically and medically fit in all aspects except psychosis, and no patient with FEP or CSZ had previous history of seizures or severe head or traumatic brain injury and experienced consciousness loss, or had drug abuse history in the previous year. The patients having any of these conditions were not included in the study protocol. Moreover, as all patients with FEP were followed up for six months after the diagnosis, and the patients who did not meet Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition diagnosis criteria after six months or who showed characteristics of primary bipolar disorder or major depression were also not included in this study. A total of 29 patients (23 male, 6 female) with FEP were found to be suitable for selection after six months of followed up period. To remove the influence of circulating female hormones, only male patients were analyzed; therefore, 6 female patients with FEP were excluded from the study. Also, 1 male patient whose plasma sample developed turbidity and 1 patient whose NA data was lost were excluded. So, in total only 21 male patients with FEP were analyzed. The procedure for measuring plasma leptin and BMI has been published earlier[1,2] it is not discussed here. The procedure for measuring erythrocyte membrane NA is briefly discussed below.

Analysis of erythrocyte membrane NA

The fasting blood samples were collected in Vacutainers containing EDTA and immediately processed. The samples were centrifuged at 2500 rpm and 5 °C for 10 minutes and the plasma was separated gently. The erythrocyte pellets were washed three times with an equal amount of normal saline at 5 °C for 10 minutes at 2500 rpm. The samples were stored at -70 °C for long term storage or until analyzed. The procedure of membrane fatty acid (NA) analysis was used after modification of the previously described procedure[1,20]. In brief, an amount of 0.2 mL of packed erythrocytes were taken in 5 mL clean glass vials. The samples were treated afterward with 0.8 mL of methanolic-HCl (0.6 N) mixed with 1 mL of 0.018% Butylated hydroxtoluene in methanol for reducing oxidation. Then after sealing the vials, samples were incubated at 80 °C for 2 hours. Afterward, samples were mixed with 1 mL hexane followed by centrifugation at 3000 g in a swinging rotor for 10 minutes at room temperature. After centrifugation, the top clear layer was gently separated in glass vials, and the above step was repeated two-three times and all fractions were pooled, which contained mainly methylated fatty acid-hexane mixture. This mixture was completely dried at 55 °C under an atmosphere of nitrogen, and the dried material was kept at -20 °C or analyzed soon after.

The methylated fatty acid samples were then solubilized in 20 mL chloroform, and 1 mL of this sample was analyzed by gas chromatography on a Hewlett-Packard gas chromatograph, model 5890 (SERIES-II), using a capillary column of 30 m × 0.32 mm × 0.20 mm dimensions (Supelco). A flame ionization detector was used with an oven temperature set at 175 ^oC for 15 minutes with a program set at 10 °C/minute up to 220 °C with a final hold at 220 °C for 10 minutes. The injector temperature was 240 °C whereas the detector A was set at 275 °C. For these parameters, the column was first calibrated by allowing a mixture of standard fatty acids in roughly equal proportions to pass through the column. The helium gas was used as carrier passing through the column at a rate of 30 mL/minute and the data was recorded on a Hewlett-Packard Reporting Integrator, model 3390A. The peak patterns of isolated fatty acids in samples were identified individually from the retention time of standard fatty acids run under identical experimental settings. The levels of fatty acids are presented as percent of the total fatty acids as 100%. This procedure has been used by various authors and found to be reproducible and very reliable for erythrocytes. On the other hand, expressing fatty acid concentration in terms of per unit of packed erythrocytes has been found to be variable as a result of uneven packing of cells and problem in making aliquots for measurement.

Statistical analysis

A Prism6 software was used for performing statistical analyses and the values are expressed as mean ± SD. The values were further treated for obtaining significance between groups by applying Student’s t-test (two-tailed variance) with Bonferroni post-hoc test or by applying one-way analysis of variance with Tukey’s post-hoc test. A P value < 0.05 was accepted as significant. The values of Pearson correlation coefficient (r) were determined to find the types and significance of association between different variables such as membrane NA, plasma leptin, BMI, and clinical symptoms comprising of PANSS-PSS and PANSS-NSS. To verify sample size (number of patients in FEP and CSZ groups), statistical power analysis was performed by measuring effect size and applying Cohen’s d test.

RESULTS

In this study, a total of twenty-one patients with FEP, twenty patients with CSZ, and fourteen CNT subjects, all male, were analyzed. Although a table with demographic and clinical characteristics of these patients has been published earlier, it is also presented here as Table 1 for reference purposes only[1,2]. These patients and controls represent a unique population in terms of education (average 12^th grade) and smoking pattern. Only 2/21 FEP and 4/20 CSZ patients, and no controls has smoking habit. Both the patient groups and CNT subjects come from surrounding community, so, they had very similar socioeconomic backgrounds. The CNT subjects were matched to FEP patients but not to CSZ patients. However, under the similar dietary patterns, membrane fatty acids have been found in the normal range between the age 20 to 55.

Figure 1A shows statistical analyses of erythrocyte membrane NA in CNT subjects and in patients with FEP and CSZ. The level of NA was non-significantly elevated in patients with FEP (about 12.5% increase, P = 0.08) compared to the CNT subjects (Figure 1A, empty bar). However, the erythrocyte membrane NA level was significantly reduced in medicated CSZ patients (about 42% decrease vs FEP, P < 0.001, and about 30% decrease vs CNT, P = 0.037, Figure 1A, black bar). Intriguingly, oleic acid (C18:1), another fatty acid of the w-9 series, was significantly increased in patients with CSZ compared to both FEP and CNT subjects (data not shown). This suggests that synthesis and/or incorporation of NA into membrane phospholipids could be impaired in chronic treatment-resistant schizophrenia because CSZ patients were treated with various atypical APDs for several years, as mentioned in the method section.

Figure 1 Statistical analyses of erythrocyte membrane nervonic acid level and its association with psychopathology in schizophrenia. ^aP < 0.05 was considered significant. A: The level of erythrocyte membrane nervonic acid (NA) was non-significantly elevated in patients with first-episode psychosis (FEP) (grey bar) compared to the control subjects (empty bar). However, membrane NA level was significantly reduced in medicated patients with chronic schizophrenia compared to both the control subjects and FEP patients (black bar); B: In patients with FEP, erythrocyte membrane NA did not show significant association with positive and negative syndrome scale (PANSS)-positive symptom scores; C: It showed significantly negative association with PANSS-negative symptom scores; D: In chronic schizophrenia patients, erythrocyte membrane NA showed significant negative association with both PANSS-positive symptom scores; E: PANSS-negative symptom scores. CNT: Control; FEP: First-episode psychosis; CSZ: Chronic schizophrenia; PANSS-PSS: Positive and negative syndrome scale-positive symptom score; PANSS-NSS: Positive and negative syndrome scale-negative symptom score.

Figure 1B-E show the association of NA with PANSS scores. In patients with FEP, erythrocyte membrane NA did not show a significant association with PANSS-PSS (Figure 1B, r = -0.2915, P = 0.09), but it did show a significant negative association with PANSS-NSS (Figure 1C, r = -0.4323, P = 0.023). In CSZ patients, erythrocyte membrane NA showed a negative association with both PANSS-PSS (Figure 1D, r = -0.3911, P = 0.041) and PANSS-NSS (Figure 1E, r = -0.4562, P = 0.013). Since membrane NA level has also been found to be reduced and negatively correlated with clinical symptoms in patients with depression, the negative association between NA and PANSS-NSS in schizophrenia could be an indication of irreparable structural damage, most likely impaired myelination.

Figure 2 shows the association of NA with plasma leptin and BMI. The data on plasma leptin and BMI was published recently and is used here only for analyzing association with NA[2]. In patients with FEP, NA showed a non-significant negative association with both plasma leptin (Figure 2A, r = -0.3001, P = 0.102) and BMI (Figure 2B, r = -0.2532, P = 0.231). However, in medicated patients with CSZ, NA showed a significant negative association with both plasma leptin (Figure 2C, r = -0.4701, P = 0.031) and BMI (Figure 2D, r = -0.4721, P = 0.029). This could be due to treatment-induced opposite changes in the erythrocyte membrane NA to that of plasma leptin and BMI as compared to the patients with FEP.

Figure 2 Association of erythrocyte membrane nervonic acid with plasma leptin and body mass index in schizophrenia. ^aP < 0.05 was considered significant. A: In patients with first-episode psychosis, erythrocyte membrane nervonic acid (NA) did not show significant association with plasma leptin; B: In patients with first-episode psychosis, erythrocyte membrane NA did not show significant association with body mass index; C: In medicated patients with chronic schizophrenia, erythrocyte membrane NA showed significant negative association with plasma leptin; D: In medicated patients with chronic schizophrenia, erythrocyte membrane NA showed significant negative association with body mass index. FEP: First-episode psychosis; CSZ: Chronic schizophrenia.

DISCUSSION

Association of membrane NA with schizophrenia psychopathology and metabolic parameters

In this study, significant changes in the erythrocyte membrane NA were found in drug-naive patients with FEP and CSZ compared to the CNT subjects. These changes also showed significant association with psychopathology in patients with both FEP and CSZ. The main findings include: In patients with FEP, erythrocyte membrane NA was non-significantly elevated and negatively associated with the negative symptoms (PANSS-NSS) but not with positive symptoms (PANSS-PSS). In patients with CSZ, erythrocyte membrane NA was significantly reduced and showed a negative association with both PANSS-PSS and PANSS-NSS scores. These results support previous findings, which reported a negative association between erythrocyte NA and PANSS scores[1-7,16,17].

Further, in patients with FEP, erythrocyte membrane NA did not show significant association with plasma leptin or BMI, whereas it showed significant negative association with both plasma leptin and BMI in the medicated patients with CSZ, in whom both plasma leptin and BMI were significantly increased compared to the patients with FEP. This suggests that NA deficiency could be a major risk factor associated with the development of metabolic comorbidities following antipsychotic treatment in patients with schizophrenia.

This is the first report that shows erythrocyte membrane NA was increased in FEP, reduced significantly in medicated patients with CSZ, and showed a strong association with clinical symptoms and metabolic markers. Thus, NA deficiency could be a potential contributing factor in the development of metabolic comorbidities, specifically, treatment-induced obesity in schizophrenia. In support of this, it was observed that mice kept on a high-fat diet with NA deficiency showed increased weight gain and adiposity compared to the mice given the same diet supplemented with NA[12]. These findings suggest that there is a strong link between NA deficiency and obesity; therefore, further large-scale observational studies are urgently required to study the long-term antipsychotic treatment effect on membrane NA levels in patients with schizophrenia. If membrane NA level is reduced following antipsychotic treatment, NA supplementation could be a potential adjunctive treatment option, which remains to be tested in schizophrenia.

Role of NA deficiency in impaired neurogenesis and demyelination in schizophrenias

The adult human brain generates an enormous number of new neurons daily via a process called neurogenesis, which plays a crucial role in memory and cognition, and adaptive behaviour[21-23]. During neurogenesis, a sizable fraction of neural stem cells also transformed into astrocytes and oligodendrocytes, a process called gliogenesis[24]. While astrocytes play a crucial role in the development and function of both excitatory and inhibitory synapses, oligodendrocytes play a main role in the formation of the myelin sheath around the axons of certain types of pyramidal neurons in the hippocampus and cortex[25,26]. NA (C24:1) and lignoceric acid (C24:0) are major fatty acids required for myelin synthesis[8,27]. As discussed below, several lines of evidence suggest that both neurogenesis and gliogenesis could be impaired in schizophrenia, and NA deficiency could be one of the potential risk factors.

Reduced neurogenesis has been reported in the anterior dentate gyrus and hilus regions of the hippocampus[28-30]. In addition, reduced neurogenesis has been reported also in the cingulate cortices and in the prefrontal cortex area 24 of the brain in patients with schizophrenia[31-33]. Furthermore, reduced neurogenesis has been reported in the subependymal zone, which provides new neurons to adjacent cortical and subcortical areas[34]. Likewise, several studies have reported reduced oligodendrocyte density in the brains of patients with both FEP and CSZ[35,36]. Oligodendrocytes are the major component of the white matter and form myelin sheaths around the axons of pyramidal neurons in the central nervous system[37,38]. Myelination increases axon diameter and is periodically interrupted. It has been shown that myelinated axons show substantially increased pulse propagation velocity, which plays a crucial role in the information processing and retrieval speed within the cellular and synaptic circuits involved in facilitating higher-order cognitive functions[37,38]. Evidence suggests that reduced myelination is considered a potential pathological factor, as patients with schizophrenia display significantly reduced information processing and retrieval speed[39].

Regarding the consequences of NA deficiency, it has been shown that NA supplementation in mice alleviates the adverse effects of senile dementia on motor ability and improves the capacity of autonomous exploration in mice[9-11,13]. These findings suggest that NA may increase neurogenesis, as both dementia and exploratory behaviours are associated with reduced neurogenesis. In an in vitro experimental study, NA treatment significantly protected neural cells (increased cell viability/density) from oxidative stress-induced injury by increasing the expression of NGF, NF-200, and S100 protein[10,11]. Further, it has been reported that treating oligodendrocyte culture with NA supplementation improves oligodendrocyte function by accelerating maturation of oligodendrocyte precursors and improving both myelin synthesis and the remyelination process[8]. Since NA deficiency has been linked to various psychiatric and neurodegenerative diseases, it will be reasonable to perform clinical trials with NA supplementation to improve therapeutic outcome and to use membrane NA as a prognostic marker.

Mechanism underlying NA abnormalities in schizophrenia: Role of DNL

Although membrane NA could be synthesized from oleic acid (a C18:1 MUFA) obtained through the diet, it could also be synthesized in the body via DNL. It has been shown that DNL is indispensable because, genetic knockdown of fatty acid synthase, the central enzyme of DNL, induces embryonic and stem cell lethality[40-42]. Thus, while DNL could be a significant source of endogenous NA for membrane phospholipid synthesis, dysregulated DNL could be the cause of membrane NA abnormalities in schizophrenia. In this regard, we have recently published evidence suggesting that DNL could be enhanced in the early stage of psychosis and, therefore, could be the leading cause of increased fatty acid synthesis, including NA, in patients with FEP[2,5,6,43]. Evidence suggests that this could be a response for maintaining myelin integrity in the early stage of psychosis that may be lost over the years as the illness advances to the chronic stage[44].

Antipsychotic treatment has been shown to further stimulate DNL, leading to an increase in membrane SFA, MUFA, and PUFA contents[1,2,4,7,17,41]. However, the reduced membrane NA level that we observed in long-term antipsychotic-treated patients with CSZ is intriguing, which suggests that either the pathway that leads to NA synthesis in the endoplasmic reticulum (ER) or its incorporation into membrane phospholipids could be impaired in patients with CSZ, which most likely represents a treatment-resistant state that is not reparable by antipsychotic treatment[45]. In support of this, evidence suggests that patients with schizophrenia may develop ER stress coinciding with the onset of psychosis[46]. Thus, further large-scale observational studies are required to find out if people with psychosis carry a genetic predisposition to develop ER stress and NA abnormalities, which may predict treatment resistance and poor recovery in schizophrenia. Since people with psychosis may develop insulin resistance from the childhood or adolescence stage, ER stress can directly or indirectly disrupt insulin signaling and cause lipid abnormalities by altering DNL[43,47].

Although the trend of NA elevation that we observed in patients with FEP could be a compensatory lipid remodeling response in the early psychosis, it appears to be exhausted in patients with CSZ. Evidence suggests that it could be due to further deterioration in ER function, as chronic antipsychotic treatment has been shown to increase ER stress[48,49]. As a result of ER stress, conversion of oleic acid (C18:1) into NA, which involves a cycle of many elongation steps, could be impaired. Thus, the clinical relevance of NA as a biomarker for treatment resistance should be explicitly tied to therapeutic strategies, such as dietary NA supplementation or adjunctive therapies, which target lipid metabolism leading to an increase in NA synthesis.

Limitations and strengths of the study

This study has some limitations and strengths. Regarding the limitations: (1) The sample size of patients was modest; however, statistical power calculation suggests the sample size was appropriate to get the desired significance; (2) Only male patients were included; however, it was to exclude the effect of circulating female hormones, which are known to affect fatty acid metabolism; (3) The first visit membrane NA levels in patients with CSZ were not available; however, it should not be a confounder, as oleic acid, another MUFA, was increased in CSZ patients; and (4) For CSZ patients, the BMI data at fist diagnosis was not available. However, these patients were analyzed mainly for comparison purpose and similarity in demographic and socioeconomic properties to those of FEP patients and CNT subjects. As for the strength is concerned: (1) The socioeconomic and demographic properties of the patients and CNT subjects were similar/comparable; (2) The patients with FEP were analyzed within 5 days after diagnosis, so, they had the shortest duration of illness ever reported (≤ 5 days); (3) Patients with FEP has no reported drug abuse; (4) No antipsychotic exposure prior to diagnosis; (5) Minimum number of smoking (2/21); (6) No sedentary lifestyle as all patients with FEP were active-duty army personnel; and (7) They had no binge eating, means they had restricted food diet.

CONCLUSION

The findings discussed above revealed that erythrocyte membrane NA in patients with FEP was slightly increased and showed a negative association with negative symptoms only, which is in agreement with previous reports. On the other hand, in patients with CSZ, membrane NA was significantly reduced and showed a negative association with both positive and negative symptoms. Since these patients were treated for several years with different atypical APDs and were still hospitalized at the time of sample collection, therefore, this group of patients could be considered a treatment-resistant group. And reduced membrane NA level in these patients could be an indication of structural brain abnormalities, including impaired neurogenesis and reduced myelination, which seem to be resistant to antipsychotic treatment.

Reduced erythrocyte membrane NA and its negative association with plasma leptin and BMI in medicated patients with CSZ suggest that either NA synthesis or its incorporation into membrane phospholipids could be impaired. Since, our preliminary investigation and several later studies have reported that treatment with APDs increased membrane level of SFAs, MUFAs, and PUFAs, the level of NA did not increase in patients with CSZ; rather, it was decreased significantly with a concomitant increase in BMI and plasma leptin. Therefore, erythrocyte membrane NA could be used as a biomarker for predicting treatment resistance and for prognosis in patients with schizophrenia and other psychiatric diseases where APDs are frequently used. Also, therapeutic strategies, such as dietary NA supplementation or adjunctive therapies, which target lipid metabolism leading to increased NA synthesis, could be effective in improving global outcome in schizophrenia and other psychiatric disorders.

ACKNOWLEDGEMENTS

I sincerely thank Dr. Sahebarao P Mahadik (Emeritus Professor), Department of Psychiatry and Health Behavior, Medical College of Georgia, Augusta University, Augusta, GA, United States for giving valuable suggestions, providing resources and consent for publishing this data. I also pay my thanks to Dr. Denise R Evans and previous colleagues for helping in the clinical symptom evaluation of the patients.