To identify NMR sensitive biomarkers from body fluid (Saliva) in Parkinson's disease.Parkinson's disease (PD) is a neurodegenerative disorder, caused by progressive death of dopaminergic neurons in the substantia nigra parscompacta¹. Mitochondrial dysfunction and oxidative stress are linked to PD pathogenesis¹. Metabolomics investigates metabolite changes in bio-fluids. NMR spectroscopy plays a major role in understanding the metabolic changes (morphological and cognitive correlates) associated with neurological disorders as well as other disorders. In this study we used NMR spectroscopy to identify metabolic biomarkers in saliva sample from patients with Parkinson’s disease in comparison with that of healthy controls.We recruited PD Patients (n=8, 5M/3F, mean age: 57 ± 6.78 years) from our neurology movement disorder clinics, conforming to the UKPDS (United Kingdom Parkinson Disease Society) brain bank criteria². We recruited healthy controls (n=5, 2M/3F, mean age: 49 ± 2.4 years) from among the institute employees. Saliva were collected from patients and healthy controls after 12 hr fasting and centrifuged at 2000g for 10 min at 4^oC, samples were separated and stored at -80˚C until NMR spectroscopic experiments. A total volume of 600µl sample was prepared for NMR experiments by adding 400µl saliva, 30µl TSP (0.5mM) and 170µl phosphate buffer, containing 1mM sodium azide. TSP was added to the sample to serve both as a chemical shift reference and concentration standard for the proton NMR studies. Proton NMR spectra of saliva samples were acquired using 700 MHz (M/s Agilent Technologies, USA) at 25˚C using one dimensional (1D) NMR spectroscopy with PRESAT using 90˚ pulse sequence. Parameters for 1D were: no. of scan=128; relaxation delay=14s; spectral width= 10504.2 Hz. Assignment of resonances were carried out using 1D and 2D NMR Spectroscopy. The data was processed using the Vnmrj (version:2.3A, M/s Agilent Technologies, USA) and spectral binning data were estimated using MestReNova software (version10.0, Mestrelab Research, Spain). PLS-DA multivariate analysis was carried out using MetaboAnalyst, a web-based metabolomics data processing tool.From the obtained spectrum from saliva (representative spectrum in figure 1), 9 metabolites were assigned using 1D and 2D NMR. Comparing the spectrum of PD and healthy controls, we observed the metabolic differences among them by multivariate partial least square discriminate analysis (PLS-DA) and a clear separation was observed between PD and HC (Figures 2).Parkinson’s disease (PD) affects the nerves of the entire gastrointestinal (GI) tract and this may cause gastrointestinal malfunctions such as gastroparesis (GP), constipation and small intestinal bacterial overgrowth syndrome³. In the present study, we found significantly increased level of butyrate, glycine, phenyl alanine, tyrosine and decreased level of lactate in saliva sample in PD with respect to HC. Most of the compounds present in blood are also found in saliva, but at reduced concentration. In PD, gastrointestinal malfunction might lead to elevated levels of Butyrate (formed by bacteria residing in the colon by fermentation of carbohydrate)⁴. Higher glycine may be conferring to the phenotypes associated with PD, as it is an inhibitory neurotransmitter⁵. Phenylalanine gets converted to tyrosine under biological conditions and a decreased rate of tyrosine utilization in Parkinson's disease may result in elevated level as compared to controls⁶. This indicates the poor intestinal absorption in PD patients due to villous abnormality. The work provides an insight in determining biomarkers for assessment of PD patients.