Disease Biomarkers in CerebrospinalFluid of Patients with First-Onset PsychosisJeffrey T.-J. Huang1[, F. Markus Leweke2[, David Oxley3, Lan Wang1, Nathan Harris4, Dagmar Koethe2,Christoph W. Gerth2, Brit M. Nolden2, Sonja Gross2, Daniela Schreiber2, Benjamin Reed4, Sabine Bahn1*1 Institute of Biotechnology, University of Cambridge, Cambridge, United Kingdom, 2 Department of Psychiatry and Psychotherapy, University of Cologne, Cologne,Germany, 3 Proteomics Research Group, Babraham Institute, Cambridge, United Kingdom, 4 Ciphergen Biosystems Ltd., Surrey Research Park, Guildford, Surrey, UnitedKingdomFunding: This research wassupported by the Stanley MedicalResearch Institute. Further thanks tothe Henry Smith Charity and theBiotechnology and BiologicalSciences Research Council forfinancial support. The funders hadno role in study design, datacollection and analysis, decision topublish, or preparation of themanuscript.Competing Interests: The authorshave declared that no competinginterests exist.Academic Editor: John McGrath,University of Queensland, AustraliaCitation: Huang JTJ, Leweke M,Oxley D, Wang L, Harris N, et al.(2006) Disease biomarkers incerebrospinal fluid of patients withfirst-onset psychosis. PLoS Med3(11): e428. doi:10.1371/journal.pmed.0030428Received: March 6, 2006Accepted: August 14, 2006Published: November 7, 2006Copyright: Ó 2006 Huang et al. Thisis an open-access article distributedunder the terms of the CreativeCommons Attribution License, whichpermits unrestricted use,distribution, and reproduction in anymedium, provided the originalauthor and source are credited.Abbreviations: ANOVA, analysis ofvariance; CSF, cerebrospinal fluid;ELISA, enzyme-linkedimmunosorbent assay; m/z, mass-to-charge ratio; OCD, obsessive-compulsive disorder; PCA, principalcomponent analysis; PLS, partialleast-squares; PLS-DA, partial least-squares discriminate analysis; SELDI,surface-enhanced laser desorptionionization* To whom correspondence shouldbe addressed. E-mail: [email protected][ These authors contributedequally to this work.ABSTRACTBackgroundPsychosis is a severe mental condition that is characterized by a loss of contact with realityand is typically associated with hallucinations and delusional beliefs. There are numerouspsychiatric conditions that present with psychotic symptoms, most importantly schizophrenia,bipolar affective disorder, and some forms of severe depression referred to as psychoticdepression. The pathological mechanisms resulting in psychotic symptoms are not understood,nor is it understood whether the various psychotic illnesses are the result of similar biochemicaldisturbances. The identification of biological markers (so-called biomarkers) of psychosis is afundamental step towards a better understanding of the pathogenesis of psychosis and holdsthe potential for more objective testing methods.Methods and FindingsSurface-enhanced laser desorption ionization mass spectrometry was employed to profileproteins and peptides in a total of 179 cerebrospinal fluid samples (58 schizophrenia patients,16 patients with depression, five patients with obsessive-compulsive disorder, ten patients withAlzheimer disease, and 90 controls). Our results show a highly significant differentialdistribution of samples from healthy volunteers away from drug-naı¨ve patients with first-onset paranoid schizophrenia. The key alterations were the up-regulation of a 40-amino acidVGF-derived peptide, the down-regulation of transthyretin at ;4 kDa, and a peptide cluster at;6,800–7,300 Da (which is likely to be influenced by the doubly charged ions of thetransthyretin protein cluster). These schizophrenia-specific protein/peptide changes werereplicated in an independent sample set. Both experiments achieved a specificity of 95% and asensitivity of 80% or 88% in the initial study and in a subsequent validation study, respectively.ConclusionsOur results suggest that the application of modern proteomics techniques, particularly massspectrometric approaches, holds the potential to advance the understanding of thebiochemical basis of psychiatric disorders and may in turn allow for the development ofdiagnostics and improved therapeutics. Further studies are required to validate the clinicaleffectiveness and disease specificity of the identified biomarkers.The Editors’ Summary of this article follows the references.PLoS Medicine | www.plosmedicine.org November 2006 | Volume 3 | Issue 11 | e4280001PLoSMEDICINEIntroductionSchizophrenia is the most devastating and endur ingpsychotic disorder affecting as many as 1% of the populationworldwide, with a similar prevalence between the sexes andthroughout diverse cultures and geographic areas [1,2]. Todate, the etiology of schizophrenia remains elusive. Thedisorder is almost certainly the result of a complexinteraction between numerous predisposing genes andenvironmental factors. Many studies have focused on theidentification of schizophrenia ‘‘ risk genes’’ /genetic poly-morphisms, and although such ‘‘ risk genes’’ may play a role ina small number of schizophrenia cases, the great majority ofpatients do not exhibit any apparent polymorphisms [3]. Analternative approach to genetic studies is to screen for diseasemarkers (biomarkers). Biomarkers or biomarker signaturesare indicators of a disease state that are usually linked to anongoing pathophysiology and thus may also provide infor-mation and insights into the underlying molecular mecha-nisms of a given disease. Typical examples include diabetesand cardiovascular disorders, where increased glucose andlow-density lipoprotein levels, respectively, are hallmarkbiomarkers for the diseases. Both cardiovascular disordersand diabetes (particularly Type II diabetes) are very similar toschizophrenia in that th ey are a result of genetic andenvironmental factors . Thus, eve n if the etiologies f orschizophrenia and other psychotic disorders remain un-known, a biomarker or biomarker signature that accuratelyidentifies the clinical syndrome would allow for improveddiagnosis, prognosis, and disease monitoring as well as thedevelopment of novel therapeutic approaches.Surface-enhanced laser desorption ionization (SELDI) massspectrometry is a powerful tool for identifying a character-istic ‘‘ fingerprint’’ of proteins and peptides in body fluids andtissues for a given condition, e.g., drug treatments anddiseases (for review, see [4]). This technology utilizes modifiedsurfaces to