Letter to the Editor

Molecular Psychiatry (2012) 17, 1150–1152; doi:10.1038/mp.2011.167; published online 20 December 2011

Decreased nervonic acid levels in erythrocyte membranes predict psychosis in help-seeking ultra-high-risk individuals

G P Amminger1,2, M R Schäfer1,2, C M Klier2, J-M Slavik3, I Holzer3, M Holub3, S Goldstone1, T J Whitford4, P D McGorry1 and M Berk1,5

  1. 1Orygen Youth Health Research Centre, The University of Melbourne, Melbourne, VIC, Australia
  2. 2Department of Child and Adolescent Psychiatry, Medical University of Vienna, Vienna, Austria
  3. 3Department of Nutritional Sciences, University of Vienna, Vienna, Austria
  4. 4Melbourne Neuropsychiatry Centre, The University of Melbourne, Melbourne, VIC, Australia
  5. 5School of Medicine, Deakin University, Geelong, VIC, Australia

Correspondence: GP Amminger, E-mail: amminger@unimelb.edu.au

The use of biomarkers to predict who will develop schizophrenia (or other psychoses) is a key theme in current psychiatric research that has important implications for the development of early, or even prophylactic, treatment strategies. Here, we report that decreased levels of nervonic acid (NA), a monounsaturated omega-9 fatty acid important in the biosynthesis of myelin, correlate with prodromal symptoms, and predict conversion to psychosis in young people at high clinical risk for psychosis.

Neuronal membranes are largely made up of phospholipids. Arachidonic acid (AA, C20:4n-6), docosahexaenoic acid (DHA, C22:6n-3) and NA (C24:1n-9) are the most important fatty acids in the nervous system. In addition to its structural role in the central nervous system, AA is a precursor for the eicosanoids: prostaglandins, prostacyclins, thromboxanes and leukotrienes. These compounds have complex roles in the control of the immune and inflammatory responses, and also act as messengers in the central nervous system.1 DHA is the most abundant polyunsaturated fatty acid in the brain, and the major component of excitable membranes,2 whereas NA is the major constituent of the sphingolipids in myelin membranes.3

There is evidence of abnormal levels of polyunsaturated fatty acids in individuals with schizophrenia.4 Concordant with these findings, fatty acids, particularly the omega-3 fatty acids, may have a beneficial role in the treatment of schizophrenia, and have been shown to be effective in the prevention of psychotic disorders.5 However, it is as yet unclear if specific fatty acids are associated with prodromal symptoms or if they predict conversion to psychosis in subjects with subthreshold psychotic manifestations.

We have recently shown that supplementation with long-chain omega-3 fatty acids reduces the risk of progression to psychotic disorder, and offers a safe and efficacious strategy for indicated prevention in individuals at ultra-high risk of developing a psychotic illness.5 Extending this evidence, and following studies reporting decreased levels of DHA, NA and AA in people with schizophrenia as compared with healthy controls,6 we first examined associations between these key fatty acids with psychopathology and functioning in putatively prodromal subjects. The present study sample comprised 81 individuals at ultra-high risk of psychosis (according to criteria of Yung et al.7) (mean age=16.4, s.d.=2.1 years) who participated in a randomized controlled trial of omega-3 fatty acids vs placebo.5 The primary outcome measure in this trial was transition to psychotic disorder, which was operationally defined using the Positive and Negative Syndrome Scale (PANSS) (that is, score of 4 on hallucinations, 4 on delusions or 5 on conceptual disorganization). These levels had to be sustained for at least 1 week. The exit criteria marked the threshold (linked to positive symptoms) at which treatment with antipsychotic medication is usually initiated. Secondary outcome measures included the PANSS (positive, negative and general subscales) and the Global Assessment of Functioning. Capillary gas chromatography was used to determine the fasting erythrocyte membrane fatty acid levels. Erythrocyte membrane phospholipid composition closely reflects that of neuronal membranes6 and provides an easily accessible indicator of brain phospholipids. NA correlated negatively with negative symptoms (Pearson's correlations, r=−0.474, P<0.0001) and general symptoms (r=−0.408, P<0.0001), and positively with global functioning (r=0.331, P=0.003) (Figure 1). No significant associations with symptom or functioning measures were observed for DHA and AA. Age, sex, cigarette smoking, cannabis use, the use of antidepressants and characteristics of the metabolic syndrome common in people with schizophrenia or other psychiatric conditions (that is, alterations in delta-5 and delta-6 desaturase activity)8 can all influence fatty acid metabolism; adjustment for these factors did not alter the significance of the correlations between NA and any clinical measure.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Correlations between erythrocyte nervonic acid (NA) levels with negative symptoms, general symptoms and functioning at baseline. (a) The Positive and Negative Syndrome Scale (PANSS) negative and general subscale. (b) The Global Assessment of Functioning (GAF) measures social, occupational and psychological functioning on a single scale, with higher scores indicating better functioning.

Full figure and legend (85K)Download PowerPoint slide (870 KB)

Next, we tested if baseline levels of fatty acids predicted conversion to psychotic disorder over a 12-month period. Because the conversion rates differed significantly between the treatment groups in our treatment trial,5 and to eliminate treatment effects, we only investigated those participants who received placebo. In accordance with the literature,6 we expected to find significantly lower levels of DHA, AA and NA at baseline in individuals who subsequently converted to psychotic disorder. Of the 41 people who received placebo, 11 (27%) made a conversion to psychotic disorder (8 schizophrenia, paranoid type, 1 schizophreniform disorder, 1 schizoaffective disorder and 1 bipolar I disorder). Supplementary Table 1 shows the fatty acid composition of erythrocyte membranes in ultra-high risk patients with and without transition to psychosis (Supplementary Results). Independent sample's t-tests indicated that NA was significantly lower in the subjects who developed a psychotic disorder (psychotic 0.09, s.d.=0.04; nonpsychotic 0.14, s.d.=0.08, t=2.69, df=35.5, P=0.011) (Figure 2). The magnitude of this group difference was large (Cohen's d=0.79). A multivariate survival analysis using Cox's regression to adjust for the potentially confounding factors described above confirmed NA as significant predictor of transition to psychosis. No significant group differences were observed for DHA (psychotic 1.44, s.d.=0.42; nonpsychotic 1.70, s.d.=0.36, t=1.94, df=38, P=0.06), AA (psychotic 8.78, s.d.=1.63; nonpsychotic 9.66, s.d.=1.48, t=0.60, df=38, P=0.11) or any other fatty acid index except for the sum of the fatty acids of the omega-3 series (psychotic 2.88, s.d.=0.62; nonpsychotic 3.28, s.d.=0.47, t=2.16, df=38, P=0.037) (Supplementary Table 1; Supplementary Results). In the adjusted Cox regression model, however, this group difference was not significant.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Baseline erythrocyte fatty acids in ultra-high-risk patients who did or did not transition to psychosis by 12-month follow-upa. (a) Fatty-acid values represent percent of the total fatty acids. Error bars are means, 95% confidence interval (CI). (b) Significant difference between groups (P<0.05); the magnitude of this group difference was large (Cohen's d=0.79). (c) One subject from the nonpsychotic group had the follow-up assessment 310 days after baseline.

Full figure and legend (99K)Download PowerPoint slide (1,306 KB)

There is strong evidence of white matter abnormality in schizophrenia. Ultra-high risk subjects who later develop psychosis have differences in white matter integrity in brain areas associated with schizophrenia,9 compared with ultra-high risk subjects who do not develop psychosis and with healthy controls. The pathophysiology of schizophrenia may therefore involve a disturbance in myelin development and/or myelin maintenance, resulting in compromised connectivity between neurons and brain regions.10 Fatty acids are essential for the myelinating activity of oligodendrocytes and for myelin maintenance.11 As NA is a major constituent of the myelin sheath, decreased levels of NA could reflect suboptimal myelin status in those ultra-high risk individuals who develop a psychotic disorder. This should be further investigated in imaging studies. The present data also suggest that these affected myelin pathways may in particular contribute to the negative symptoms of schizophrenia. Finally, in light of previously observed negative associations between NA and obesity-related risk factors,12 it is unlikely that the present alterations in NA were simply an epiphenomenon of the metabolic syndrome. The results of our study would have been more convincing if fatty acid data of healthy non-help-seeking matched controls had been included in our design. Comparison with healthy well-matched controls has the potential to demonstrate the extent of the reductions in key membrane fatty acid levels and to further elucidate abnormalities of lipid metabolism in ultra-high-risk patients.

Our previous observation that supplementation with omega-3 fatty acids may prevent transition to psychosis suggests that the effects of omega-3 fatty acids may offset the risk conferred by low levels of NA. The therapeutic effects of omega-3 fatty acids are complex and not fully understood. Their benefit may result from altered membrane fluidity and receptor responses following their incorporation into cell membranes as well as from interactions with neurotransmitter systems, the antioxidant defense system or the inflammatory system.13 The efficacy of omega-3 supplementation in patients at ultra-high-risk of psychosis could also imply that the present study was underpowered to detect a significant group difference for DHA. The significant result for the sum of omega-3 fatty acids in the unadjusted analysis supports this view and underscores the importance of omega-3 fatty acids in the prodromal phase of psychosis.


Conflict of interest

The authors declare no conflict of interest.



  1. Bourre JM, Bonneil M, Chaudière J, Clément M, Dumont O, Durand G et al. Adv Exp Med Biol 1992; 318: 211–229. | PubMed |
  2. Innis SM. Brain Res 2008; 1237: 35–43. | Article | PubMed | ISI |
  3. Babin F, Sarda P, Limasset B, Descomps B, Rieu D, Mendy F et al. Lipids 1993; 28: 627–630. | Article | PubMed | ISI | CAS |
  4. Berger GE, Smesny S, Amminger GP. Int Rev Psychiatry 2006; 18: 85–98. | Article | PubMed |
  5. Amminger GP, Schäfer MR, Papageorgiou K, Klier CM, Cotton SM, Harrigan SM et al. Arch Gen Psychiatry 2010; 67: 146–154. | Article | PubMed | ISI |
  6. Assies J, Lieverse R, Vreken P, Wanders RJ, Dingemans PM, Linszen DH. Biol Psychiatry 2001; 49: 510–522. | Article | PubMed | ISI |
  7. Yung AR, Phillips LJ, McGorry PD, McFarlane CA, Francey S, Harrigan S et al. Br J Psychiatry Suppl 1998; 172: 14–20. | PubMed |
  8. Assies J, Pouwer F, Lok A, Mocking RJ, Bockting CL, Visser I et al PLoS ONE 2010; 5: e10635. | Article | PubMed |
  9. Bloemen OJN, de Koning MB, Schmitz N, Nieman DH, Becker HE, de Haan L et al. Psychol Med 2010; 40: 1297–1304. | Article | PubMed | ISI |
  10. Bartzokis G. Neuropsychopharmacology 2002; 27: 672–683. | Article | PubMed | ISI |
  11. Peters BD, Duran M, Vlieger EJ, Majoie CB, den Heeten GJ, Linszen DH et al. Prostaglandins Leukot Essent Fatty Acids 2009; 81: 61–63. | Article | PubMed | ISI |
  12. Oda E, Hatada K, Kimura J, Aizawa Y, Thanikachalam PV, Watanabe K. Int Heart J 2005; 46: 975–985. | Article | PubMed | ISI |
  13. Berk M, Kapczinski F, Andreazza AC, Dean OM, Giorlando F, Maes M et al. Neurosci Biobehav Rev 2011; 35: 804–817. | Article | PubMed | ISI |


This research was supported by grant 03T-315 from the Stanley Medical Research Institute. Dr G Paul Amminger was supported by grant 566529 from the National Health and Medical Research Council, Australia. We would like to thank Konstantinos Papageorgiou for his assistance with data collection and data entry.

Supplementary Information accompanies the paper on the Molecular Psychiatry website