original scientific paper UD C 665.327.3-543.06! CHARACTERIZATIO N OF OLIVE OILS FROM SLOVENIA AN D CROATI A BY COMPOUN D SPECIFIC ISOTOPE ANALYSIS Jorge L SPANCENBERG institut de Minéralogie et Pétrographie, Université de Lausanne BFSH-2,1015 Lausanne, C H E-mail: jorge.Spangenberg@imp.unil.ch Nives OGRINC Department of Environ. Sciences, "j. Stefan" Institute, S1-1Q00 Ljubljana, Jamova 39 E-mail: nives.ogrinc@ijs.si ABSTRACT The fatty adds of olive oils from Slovenia and Croatia were chemically and isotopically characterized. The analytical approach utilized combined gas chromatography - mass spectrometry (GC-MS) and the novel technique of compound specific isotope analysis (CSIA) through gas chromatography coupled to a stable isotope ratio mass spectrometer (IRMS) via a combustion (C) interface (GC-C-IRMS). This approach provides further insights into the control of the purity and the geographical origin of oils. The differences in the d!3C values of palmitic and oleic acids are discussed as the differences in biosynthesis of these acids in the plant tissue, admixing of distinct vegetable oils, and degradation of the lipids during oil extraction and refinement. Key words: olive oil, vegetable lipids, carbon isotope, CSIA, adulteration, geographical origin INTRODUCTIO N Stable carbon isotope analyses have proven to be a powerful tool for assessing the authenticity of vegetable food products from plants of different photosynthetic pathways (Doner, 1991). During photosynthetic fixation of carbon dioxide into plant biomass, plant cells dis­criminate against the heavier stable carbon isotope The most important atmospheric C02-fixing reactions are the C.3 and C4 pathways (Farquhar ef al., 1989; O'leary, 1988, 1993). C 3 plants use the Calvin cycle for CO2 fixation, while the C4 plants use the Hatch-Stack cycle. Ail trees operate with the C3 pathway, and their carbon isotope compositions fall into the range of -34 to -22%o. C4 plants comprise most plants in the tropics, in­cluding tropical grasses, sedge, maize, sugar cane and salt marsh plants, and are isotopically heavier (-23 to -6%u). Factors other than the C02-fixation, however, may also have a less important impact on the isotopic composition of plants and their products. These include plant growth rate, local atmospheric CO2 concentration, nutritional status of the cells, water availability, and cul­tivation practices (O'Leary, 1993). Therefore, the carbon isotopic composition of bulk oil and individual lipids may record the source and geographical origin of a plant product. Food chemists are increasingly using on­line gas chromatography - combustion - stable isotope ratio mass spectrometry (GC-C-IRMS) for carbon isotope analysis of the individual lipids as a tool for assessing adulteration of vegetable oils (Woodbury et al, 1995; Kelly, 1997; Remaud, 1997). Previous work focusing on the chemical and isotopic composition of the major fatty acids (palmitic, stearic, oleic, and linoleic acids) of cold pressed (CP) olive oils from the main producing coun­tries of the Mediterranean region were reported (Spangenberg et al., 1998). We present herewith the isotopic compositions of fatty acids of the CP olive oils from Slovenia and Croatia. TORBE F SPANG ENS ESC, Nives OCRINC : CHARACTERIZATION O!' OLIVE OILS EROM SLOVENIA AND CROATIA ...,1-4 MATERIALS AN D METHODS Thirteen samples of extra virgin olive oil were ob­tained from Slovenia (Koper region, n=6) and Croatia (Istra, Daimatia, n=7). All the samples were from the 1997-1998 olive season. The analytical approach com­bined chemical characterization of the fatty acid methyl esters (FAMES) by gas chromatography - mass spec­trometry (GC-MS), and carbon isotope analyses of indi­vidual fatty acids by GC-C-1RMS. The bulk oils were analyzed for carbon isotope composition by combustion isotope ratio mass spectrometry using an on-line Carlo Erba 1108 elemental analyzer (EA) connected to a fin­nigan MAT Delta S 1RMS via a Conflo Et split interface (EA-1RMS). Ail the analyses were performed at: the De­partment of Earth Sciences of the University of Lausart­ne. The stable carbon isotope ratios are reported in the delta (6) notation as the per mil (%o) deviations relative to the Pee Dee Belemnite limestone (PDB). The repro­ducibility of the EA-IRMS analyses was better than 0.1%o (1 SD). Three to five replicate GC-C-IRMS runs were performed for each sample. The reproducibility ranged between ±0.1 and ±0.4%o (1 SD). The accuracy of the GC-C-IRMS analyses was monitored by co-injection of a FAME laboratory standard of the known isotopic com­position. The isotopic shift due to the carbon introduced in the fatty acid methylation was corrected by a mass balance equation (Spangenberg et al., 1998). RESULTS AN D DISCUSSIO N Fatty acids composition The scatter of the compositions of fatty acids for the virgin olive oils probably reflects the variation in variety, climatic conditions of the area, water-use efficiency in cultivars, salinity, temperature and pH of the irrigation water, olive-ripening stage and other factors (results not shown for brevity). Bulk isotopic composition The S13 C of the bulk olive oils (-27.7 to -30.6%o show isotopic compositions typical of C3 plants (Tab. 1). The scatter of the 5 13 C values of the oils (2.9%o may be attributed to factors affecting the chemical distribution of the fatty acids, and particularly by the physiological processes and enzymatic reactions occurring in the plant cells. Additionally, the chemical changes (transmerization of oleic acid and oxidation) during thermal degradation (natural or induced during steam washing or other refining procedures) of the olive oil, or blending of C P oil with refined olive oil or other vege­table oil may cause a further isotopic discrimination. Tab. 1: Carbon isotope composition of bulk oil and main individual fatty acids in olive oil samples from Slovenia and Croatia. Tab. 1: Izotopska sestava celokupnega ogljika in ogljika posameznih maščobnih kislin v vzorcih oljčnih olj iz Slovenije in Hrvaške. (%« PDB) Sample Country bulk oil palmitic (16:0) stearic (18:0) oleic (18:1) COIL-58 Slovenia -30.1 -34.3 -33.3 -34.1 COIL-59 Slovenia -28.4 -31.4 -31.3 -30.1 COIL-6O Slovenia -29.4 -32.1 -32.2 -30.1 COIL-75 Slovenia -29.1 -32.8 -32.2 -31.6 COIL-76 Slovenia -30.0 -33.5 - -33.6 COIL-79 Slovenia -29.1 -32.4 -32.8 -31.9 COIL-65 Croatia -29.6 -32.9 -32.4 -31.0 CUÎL-66 Croatia -29.4 -31.8 -32.0 -30.3 COIL-67 Croatia -29.1 -30.6 -31.5 -30.7 COIL-68 Croatia -29-8 -33.3 -32.7 -31.3 COIL-77 Croatia -30.6 -35.0 - -33.1 COlL-78 Croatia -27 7 -31.8 - -32.6 C01L-80 Croatia -28.0 -35.4 -31.2 -32.3 - » not analysed lotge {.. SPAN CENSE KG, Nives OGRIN C CHARACTERIZATION OF OLIVE OILS FROM 5LOVFNIA AN D CROATIA ...,1-4 Isotopic composition of individual fatty acids -29 The S13 C values of the virgin olive oil fatty acids -30 vary between -34.1 to -28.5%e> (Table 1). A substantial separation of the oils from the 1:1 line in the S 13 Qg: o -31 vs. S13C)8:1 diagram (16:0 - palmitic acid, 18:1 - oleic acid) suggests admixing of cold pressed virgin olive oil with refined olive oils or other vegetable oils of different O 18:1/16:0 concentration-ratios than the genuine olive oil. The distribution of the samples in the vs. 513 C i 8:l diagram strongly suggests the adulteration or « «> -34 inappropriate processing of some CP olive oils (Fig. 1). Virgin CP olive oils are separated from the lower grade olive oils by dedicated principal component analysis -35 performed combining the fatty acid composition and the bulk and molecular carbon isotope ratios (results not -36 shown for brevity). -37 CONCLUSION S -3? -36 -35 -34 -33 -32 -31 -30 -29 WIM W The 5 13 C values of the bulk oil and individual fatty Fig, 1: Carbon isotope composition of oleic acid acids can be used for identification of the sources of ol­ (S'-'Cyg.jj vs. palmitic acid (8i3C16:0) of olive oils from ive oil and control of their authenticity. The use of Slovenia and Croatia. S13CIL:Q vs. covariations serves to assess cases Si. 1: Izotopska sestava ogljika oleinske kisline where impurity or adulteration is suspected. Blending of (&13Cif]:i) v odvisnosti od izotopske sestave ogljika olive oil with edible oils with slightly different fatty acid palmitinske kisline (S13C}(,:o) v vzorcih oljčnih olj iz compositions (e.g., olive pomace, sunflower, hazelnut) Slovenije in Hrvaške. may be detected using this approach combined with molecular information (GC-MS) and the carbon isotope ACKNOWLEDGEMENTS composition of the bulk oil. We thank ail people who provided us oil samples. The organic geochemical laboratory including the GC ­MS facilities at the University of Lausanne was funded under a joint venture of the Swiss National Science Foundation and University of Lausanne. This is a contribution to the 1GCP Project 429. UPORAB A STABILNI H IZOTOPO V OGLJIK A PR I KARAKTERiZACI.i l OLJČNEG A OLJ A I Z SLOVENIJE IN HRVAŠKE Jorge E. SPANGENBERG institut de Minéralogie et Pétrographie, Université de Lausanne BFSH-2,1015 Lausanne, C H E-mail: jorge.Spangenberg@imp.unil.ch Nives OGRINC Odsek z a kemijo okolja, Institut "j, Stefan", SI-1000 Ljubljana, Jamova 39 E-mail: Nives.Ogrinc@ijs.si POVZETEK Meritve izotopske sestave ogljika so se izkazale kot izvrstno naravno sledilo za spremljanje različnih procesov, zato jih s pridom izkoriščamo tudi v živilski industriji pri določanju kakovosti in pristnosti (avtentičnosti) različnih živil - vin, sadnih sokov, medu, olj. f-'oleg tega se stabilni izotopi uporabljajo tudi pri določitvi geografskega porekla. Jorge E. SPANGENliERG, Nives OCRINC : CHARACTERIZATtON O F OLIV E OILS EROM SI.OVEM1A AN D CROATIA ..., V prispevku, ki je nastal v sodelovanju z Laboratorijem za stabilne izotope, Inštututa za mineralogijo in petrologijo Univerze v Luzani v Švici, smo predstavili uporabo stabilnih izotopov ogljika pri določitvi avtentičnosti in geograf­skega porekla oljčnega olja. Določili smo kemijsko in izotopsko sestavo maščobnih kislin v oljčnih oljih iz Slovenije in HrvaŠke. Koncentracije maščobnih kislin smo določili s plinskim kromatografom s kapilarno kolono (CC-MS), nji­hovo izotopsko sestavo (o13C) pa z masnim spektrometrom za stabilne izotope CC-C-IRMS. Iz korelacije med izo­topsko sestavo ogljika palmitinske kisline, 573C^:q, od izotopske sestave ogljika oleinske kisline, Sl3C^g:h lahko ugotovimo možne potvorbe in nepravilno predelavo oljčnega olja. Drugi možni vzroki, ki privedejo do razlik med vrednostmi in S,3Cig:j, so še: različna biosinteza teh kislin v rastlinskih tkivih ter razgraditev maščob pri ekstrakciji olj in nadaljnjem čiščenju ekstrakta. Ključne besede: oljčno olje, rastlinske maščobe, izotopi ogljika, CSIA, potvorjenost, geografsko poreklo REFERENCES Doner, L. W. (1991): Verifying the authenticity of plant derived materials by stable isotope ratio and chroma­tographic methodologies, journal of the Association of official analytical chemist. 74, 14-19. Farquhar, G . D., J. R. Ehleringer & K. T. Hubick, (1989): Carbon isotope discrimination and photosynthesis. An­nua! Review of Plant Physiology and Plant Molecular Biology 40, 503-537. Kelly, S.; L Parker, M. Sharman, J. Dennis & I. Goodall, I. (1997): Assessing the authenticity of single seed vegetable oils using fatty acid stable carbon isotope ra­tios (13C/12Q. Food Chemistry 59, 181-186. O'Leary, M. H. (1988): Carbon isotope in photosynthe­sis. Bioscience, 38, 328-336. O'Leary, M. H. (1993): Biochemical basis of carbon isotope fractionation. In Stable isotopes and plant water- relations; Ehleringer, j. R., Half, A. E., Farquhar, G. D., Eds., Academic Press, San Diego. Remaud, G. S., Y. L. Martin, G. G. Martin, N. Naulet, & G . J. Martin, (1997): Authentication of mustard oils by combined stable isotope analysis (SNIF-NMR and I RMS). journal of Agricultural & Food Chemistry 45, 1844­1848. Spangenberg, J. E., S. A. Macko & J. j., Honziker, (1998): Characterization of olive oil by carbon isotope analysis of individual fatty acids: implications for authentication, journal of Agricultural & Food Chemistry 46, 4179-4184. Woodbury, S. E v R, P. Evershed, J. B. Rosseli, R. E. Griffith, P. FarneU, (1995): Detection of vegetable oil adulteration using gas-chromatography combustion isotope ratio mass-spectrometry. Analytical Chemistry 67, 2685-2590.