Research On Discrimination Of Organic And Conventional Rice With Biomarkers

Organic rice is produced by sustainable cultivation methods such as use of green fertilizers and biological pest control measures without using genetically modified seeds,synthetic chemical fertilizers,pesticides and herbicides.The identification of organic rice,which remains a common requirement for the organic planting process and the basic provisions of certification process,lacks qualitative and quantitative discriminant methods and specific standards for product quality.In this research,organic rice and conventional rice from Heilongjiang Province were used as raw materials,and Label-free proteomics and untargeted metabolomics combined with chemometrics and bioinformatics were applied for determination of protein biomarkers and metabolite biomarkers.The capability of organic and conventional rice authentication was presented with the established discriminant model and method.Firstly,an extraction method for metabolites in rice and a UPLC-Q-TOF MS/MS method for metabolite determination were established.The ultrasound-assisted extraction(24 k Hz,55 W,30 min)was carried out twice with methanol-water(7:3)and acetonitrile-water(3:7);UPLC column was ACQUITY BEH AMIDE column(1.7 μm,2.1 mm × 100 mm),column temperature was set at 30 °C,flow rate was 0.3 m L/min,a 23 min gradient elution was applied with 25 mmol/L ammonia-ammonium acetate aqueous solution as phase A and acetonitrile as phase B;Q-TOF MS/MS was utilized with jet electrospray ionization source under positive and negative ion mode.Total 9443 characteristic peaks under positive ion mode and 8975 characteristic peaks under negative ion mode were obtained from rice QC samples;the precision,repeatability and linearity of the established method were found to meet the requirements of untargeted metabolite analysis.Based on the Label-free proteomic analysis,the differences in protein levels between organic rice and conventional rice were studied.A total of 1,660 proteins were identified,and 418 core proteins co-expressed in two groups were screened.According to t-test and fold change analysis,there were 114 differential proteins,of which 56 up-regulated proteins of organic rice and 58 down-regulated proteins;114 proteins were subjected to principal component analysis,hierarchical clustering analysis and similarity analysis for discrimination between organic rice and conventional rice.The results showed that rice samples could be classified into two groups by differential proteins.The orthogonal partial least squares discriminant analysis model was used to screen out 30 potential protein biomarkers.According to the KEGG analysis of 30 potential protein biomarkers,21 proteins were involved in the main metabolic processes of Oryza sativa,including starch synthesis,glycolysis,tricarboxylic acid cycle,plant stress response,and ribosome synthesis.The untargeted metabolite analysis of 20 organic rice and 20 conventional rice was carried out by the established UPLC-Q-TOF MS/MS method.A total of 16814 metabolites were identified under both positive and negative ion modes.There were 8614(ESI+)and 7526(ESI-)core metabolites;3665(ESI+)and 3376(ESI-)differential metabolites were screened by t-test and fold change analysis.Principal component analysis,partial least squares discriminant analysis and orthogonal partial least square discriminant analysis were used to establish classification and discrimination models of organic rice and conventional rice based on differential metabolites.According to the models there were 100 metabolites selected.37 potential metabolite biomarkers with significant contribution to discrimination were determined by KEGG pathway enrichment analysis,including glucose metabolism,amino acid metabolism,fatty acid metabolism,and nucleosides acid metabolism.Based on the correlation of metabolic pathways between potential protein biomarkers and potential metabolite biomarkers,as well as the contribution to sample classification and the abundance differences,6 proteins were screened from 21 potential protein biomarkers and 9 metabolites were screened from 37 potential metabolite biomarkers.Six proteins were enolase,phosphoglucose isomerase,catalase A,alcohol dehydrogenase,HSP 70,and HSP16.9;9 metabolites were thiamine,folinic acid,succinic acid,betaine,raffinose,prostaglandin D2,serotonin,sebacic acid and aminocyclopropanecarboxylic acid.According to the abundance range of 15 biomarkers and the probabilistic statistical analysis,a discrimination standard of organic rice and conventional rice was established.A sample with 11～15 identifiable organic rice biomarkers could be regarded as organic rice,and a sample with 10～15 identifiable conventional rice biomarkers could be regarded as conventional rice.Eighteen known rice samples(9 organic rice and 9 conventional rice)were used for method validation: the precision and accuracy of rice biomarkers met the requirements of the methodology,the recognition accuracy of rice samples was 88.89%.Blind sample analysis proved that the discriminant model could be used to identify organic rice and conventional rice,but failed to effectively recognize adulteration of organic rice.