Extra virgin olive oil: microbial ecology of the extractive processes and its effect on the aromatic composition of the final products Serena Trapani1, Simona Guerrini1, Eleonora Mari1, Marzia Migliorini2, Chiara Cherubini2, Giacomo Gianni2, Bruno Zanoni1, Massimo Vincenzini1 1 - Dept. Agricultural Biotechnology, University of Florence; Florence-Italy 2 - Metropoli - Special Agency of the Chamber of Commerce of Florence - Chemical Laboratory Division; Florence -Italy *serena.trapani@unifi.it Introduction Results Newly produced olive oils might harbor viable microbial cells which could affect, according to their metabolic capability, the oil quality [1]. b-glucosidase and esterase capabilities can improve the taste and the antioxidant capability of the oil, while the lipase capability can worsen the oil quality by hydrolyzing triglycerides [2]. Usually, microbial contamination of oils originates from raw material (olives) and/or from the oil-mill. However, information on the microbial occurrence in the different steps of the extra virgin olive oil production and their influence on the aromatic composition of the final product are lacking. The microbial populations were manly constituted by yeasts and moulds, while in most cases bacteria occurred at very low concentrations. The yeast and mould densities in the pastes and in the oil after the centrifugation ranged between values below 1 and of about 104 CFU/g, while in the filtrated oil below 102 CFU/100 mL (Fig. 2). Yeast and mould occurred at very low concentrations in the oils after three months of conservation (data not shown). Aim of the work The aim of this work was to investigate on the presence of microorganisms in the pastes after crushing, in the oil after centrifugation in a two phase decanter, in the oil after filtration, and in the oil after three months of conservation in dark-green bottles. A total of 16 extraction processes, carried out in the same manufacture located in Tuscany, were considered. Figure 2: Process and analysis The 16 oils after filtration were also analyzed for their aromatic composition: Correlation studies showed some positive or negative correlations between microbial densities in the different productive steps and some aromatic compounds in the oil (Table 1). Principal component analysis (Fig. 3), carried out on the microbiological and chemical data, demonstrated that all the productive processes showing the highest microbial contaminations were positively associated with the same aromatic compounds (circulated with ellipses). Materials and Methods Olive cultivars: Frantoio and Moraiolo 16 extractive processes, conduced in two different days, were considered; the sampling points are showed in Fig. 1. Sample code  F: Frantoio; M: Moraiolo; 1 and 2: first and second sampling day; a,b,c,d: different lots. Viable counts of the yeasts and moulds were carried out on MYPG agar, total microbial counts (TMC) on PCA. The aromatic composition of the oils was detected by GC analysis according to … [3]. Statistical analysis were performed using Statistica 7 software package. Figure 1: Process and analysis Microbial data YP = yeast cell densities in the pastes MP = mould in the pastes TMCP = TMC cell densities in the pastes YD = yeast cell densities in the oil after centrifugation MD = mould in the oil after centrifugation TMCD = TMC cell densities in the oil after centrifugation YO = yeast cell densities in the oil after filtration MO = mould in the oil after filtration TMCO = TMC cell densities in the oil after filtration Chemical data Figure 3: Principal component analysis Conclusion This work may be considered a preliminary study to understand the microbial ecology of the extra virgin olive oil and its effective impact on the aromatic composition of this product. Tab 1: Correlation between volatile compounds and Yeast-Moulds Esters Alchools Aldehydes Keton Acids Ethyl acetate Methyl acetate 2 and 3 methyl-1-butanol isobutanol 2-pentanol 2-heptanol Hexanol 1-octen-3-ol Nonanol Phenol Ethylguaiacol Heptanal Valeraldehyde E-2-hexenal Octanal Nonanal 2,4-hexadienal Decanal E-2-nonenal E-2-decanal 6 Methyl 5 hepten 2 one Butirric acid Octanoic acid Yeasts + no - -* +* Moulds Reference “+” and “-” positively or negatively correlated respectively  Pearson and Sperman r > 0.7 (p<0.05); “+*” or “-*” positively or negatively related respectively  Pearson and Sperman 0.6< r < 0.7 (p<0.05); no: no correlated. [1] Ciafardini G., Zullo B. A. (2002). Microbiological activity in stored olive oil. International Journal of Food Microbiology 75: 111-118. [2] Zullo B. A., Cioccia G., Ciafardini G. (2010). Distribution of dimorphic yest species in commercial extra virgin olive oil. Food Microbiology 27: 1035-1042. [3] Vichi S, Piazzale L, Conte L. S, Buxaredas S, L’opez-Tamames E.Simultaneus determination of volatile and semi-volatile aromatic hydrocarbons in virgin oil by headspace solid-phase microextraction coupled to gas chromatography/mass spectrometry. Journal Chromatography 1090, 146-154 (2005). 10th Euro Fed Lipid Congress, 23-26 September 2012, Cracow Poland The research leading to these results has received funding from “Bando Misura 124” of PSR 2007-2013 of Regione Toscana, promoted and coordinated by Gal - Start srl, published on the BURT of Regione Toscana n. 7 of 16 J 2011 for the project Oleosalusistem: “Validazione di protocolli per la produzione di oli ad elevato valore nutrizionale ed a ridotto impatto ambientale”