Stewart F Graham1,2, Onur Turkoglu3, Ali Yilmaz3,4, Ilyas Ustun5, Zafer Ugur3,4, Trent Bjorndhal6, BeomSoo Han6, Rupa Mandal6, David Wishart6, Ray O Bahado-Singh3. 1. Oakland University-William Beaumont School of Medicine, Rochester, MI, 48309, USA. stewart.graham@beaumont.edu. 2. Research Institute, Metabolomics Division, Beaumont Health, Royal Oak, MI, 48073, USA. stewart.graham@beaumont.edu. 3. Oakland University-William Beaumont School of Medicine, Rochester, MI, 48309, USA. 4. Research Institute, Metabolomics Division, Beaumont Health, Royal Oak, MI, 48073, USA. 5. Wayne State University, Civil and Environmental Engineering, Detroit, MI, USA. 6. Department of Biological and Computing Sciences, University of Alberta, Edmonton, AB, Canada.
Abstract
INTRODUCTION: Autism spectrum disorder (ASD) is a group of neurodevelopmental disorders characterized by deficiencies in social interactions and communication, combined with restricted and repetitive behavioral issues. OBJECTIVES: As little is known about the etiopathophysiology of ASD and early diagnosis is relatively subjective, we aim to employ a targeted, fully quantitative metabolomics approach to biochemically profile post-mortem human brain with the overall goal of identifying metabolic pathways that may have been perturbed as a result of the disease while uncovering potential central diagnostic biomarkers. METHODS: Using a combination of 1H NMR and DI/LC-MS/MS we quantitatively profiled the metabolome of the posterolateral cerebellum from post-mortem human brain harvested from people who suffered with ASD (n = 11) and compared them with age-matched controls (n = 10). RESULTS: We accurately identified and quantified 203 metabolites in post-mortem brain extracts and performed a metabolite set enrichment analyses identifying 3 metabolic pathways as significantly perturbed (p < 0.05). These include Pyrimidine, Ubiquinone and Vitamin K metabolism. Further, using a variety of machine-based learning algorithms, we identified a panel of central biomarkers (9-hexadecenoylcarnitine (C16:1) and the phosphatidylcholine PC ae C36:1) capable of discriminating between ASD and controls with an AUC = 0.855 with a sensitivity and specificity equal to 0.80 and 0.818, respectively. CONCLUSION: For the first time, we report the use of a multi-platform metabolomics approach to biochemically profile brain from people with ASD and report several metabolic pathways which are perturbed in the diseased brain of ASD sufferers. Further, we identified a panel of biomarkers capable of distinguishing ASD from control brains. We believe that these central biomarkers may be useful for diagnosing ASD in more accessible biomatrices.
INTRODUCTION:Autism spectrum disorder (ASD) is a group of neurodevelopmental disorders characterized by deficiencies in social interactions and communication, combined with restricted and repetitive behavioral issues. OBJECTIVES: As little is known about the etiopathophysiology of ASD and early diagnosis is relatively subjective, we aim to employ a targeted, fully quantitative metabolomics approach to biochemically profile post-mortem human brain with the overall goal of identifying metabolic pathways that may have been perturbed as a result of the disease while uncovering potential central diagnostic biomarkers. METHODS: Using a combination of 1H NMR and DI/LC-MS/MS we quantitatively profiled the metabolome of the posterolateral cerebellum from post-mortem human brain harvested from people who suffered with ASD (n = 11) and compared them with age-matched controls (n = 10). RESULTS: We accurately identified and quantified 203 metabolites in post-mortem brain extracts and performed a metabolite set enrichment analyses identifying 3 metabolic pathways as significantly perturbed (p < 0.05). These include Pyrimidine, Ubiquinone and Vitamin K metabolism. Further, using a variety of machine-based learning algorithms, we identified a panel of central biomarkers (9-hexadecenoylcarnitine (C16:1) and the phosphatidylcholine PC ae C36:1) capable of discriminating between ASD and controls with an AUC = 0.855 with a sensitivity and specificity equal to 0.80 and 0.818, respectively. CONCLUSION: For the first time, we report the use of a multi-platform metabolomics approach to biochemically profile brain from people with ASD and report several metabolic pathways which are perturbed in the diseased brain of ASD sufferers. Further, we identified a panel of biomarkers capable of distinguishing ASD from control brains. We believe that these central biomarkers may be useful for diagnosing ASD in more accessible biomatrices.
Authors: Merina Varghese; Neha Keshav; Sarah Jacot-Descombes; Tahia Warda; Bridget Wicinski; Dara L Dickstein; Hala Harony-Nicolas; Silvia De Rubeis; Elodie Drapeau; Joseph D Buxbaum; Patrick R Hof Journal: Acta Neuropathol Date: 2017-06-05 Impact factor: 17.088
Authors: Paul R West; David G Amaral; Preeti Bais; Alan M Smith; Laura A Egnash; Mark E Ross; Jessica A Palmer; Burr R Fontaine; Kevin R Conard; Blythe A Corbett; Gabriela G Cezar; Elizabeth L R Donley; Robert E Burrier Journal: PLoS One Date: 2014-11-07 Impact factor: 3.240
Authors: Vijay R Varma; Anup M Oommen; Sudhir Varma; Ramon Casanova; Yang An; Ryan M Andrews; Richard O'Brien; Olga Pletnikova; Juan C Troncoso; Jon Toledo; Rebecca Baillie; Matthias Arnold; Gabi Kastenmueller; Kwangsik Nho; P Murali Doraiswamy; Andrew J Saykin; Rima Kaddurah-Daouk; Cristina Legido-Quigley; Madhav Thambisetty Journal: PLoS Med Date: 2018-01-25 Impact factor: 11.069
Authors: Danielle Brister; Brianna A Werner; Geoffrey Gideon; Patrick J McCarty; Alison Lane; Brian T Burrows; Sallie McLees; P David Adelson; Jorge I Arango; William Marsh; Angelea Flores; Matthew T Pankratz; Ngoc Han Ly; Madison Flood; Danni Brown; David Carpentieri; Yan Jin; Haiwei Gu; Richard E Frye Journal: Metabolites Date: 2022-04-20
Authors: Sumeyya Akyol; Zafer Ugur; Ali Yilmaz; Ilyas Ustun; Santosh Kapil Kumar Gorti; Kyungjoon Oh; Bernadette McGuinness; Peter Passmore; Patrick G Kehoe; Michael E Maddens; Brian D Green; Stewart F Graham Journal: Cells Date: 2021-09-29 Impact factor: 6.600