Pesticide residues and microbiological quality of bottled wines

Abstract

In the study we determined the pesticide residues and microbiological quality of bottled wines. Pesticide residues in wine were analysed using the multiresidual method with GC–MS, the multiresidual method with LC–MS–MS and the method for determination of dithiocarbamate residues with GC–MS. Furthermore, the effect of bentonite and the combined fining agent on the concentration of boscalid in wine was tested. The microbiological analysis was performed using membrane filtration. Nine pesticide residues were determined in the samples. More than 50% of wines belonged to the category of wines with only up to two pesticide residues. The most frequently determined pesticide in wines was boscalid (76% of samples) followed by fenhexamid (44%). The highest concentrations of pesticide residues in wines were determined for cyprodinil (0.44 mg/L) and fludioxonil (0.21 mg/L). The combined fining agent was more efficient in lowering the concentration of boscalid in wine if compared with bentonite. The results of microbiological quality indicate that no less than one quarter of bottled wines on our store shelves is microbiologically unstable.

Introduction

The surface of vineyards in Slovenia extends over approximately 17.000 ha with an average annual production of 100 million litres of wine. About 6 million litres are exported. The quantity of imported wine is approximately the same. The consumption of wine amounts to 39 L per capita, which is a parameter that classifies us among five most important European wine consumers (Čuš, Vanzo, Lisjak, Bavčar, & Kocjančič, 2007). At this time, quantitative and qualitative data about pesticide residues and microbiological risks in traded wines are rare. Soleas and Goldberg (2000) assayed 26 pesticides in 1827 raw grape juices prior to fermentation, in which grapes were the source of contamination. Pesticides most frequently found in their samples were folpet, captan, guthion, carbaryl and dimethoate. A study of pesticide residues in Italian wines from vineyards with a known history showed that 64 wine samples from six Italian regions contained the residues of five active ingredients at very low concentrations: benomyl (0.05 mg/L), dimethoate (0.02–0.06 mg/L), iprodione (0.02–0.07 mg/L), metalaxyl (0.04–0.14 mg/L) and vinclozolin (0.02 mg/L) (Cabras, Garau, Melis, Pirisi, & Spanedda, 1995). Vitali, Guidotti, Giovinazzo, and Cedrone (1998) analysed 21 wine samples available on the Italian market and about 57% were positive for one or more pesticides. However, Cabras and Conte (2001) analysed 190 wine samples in 1998 and 1999 and not even one of the samples was found to contain residues. Calhelha, Andrade, Ferreira, and Estevinho (2006) reported the detection of several fungicide residues (dichlofluanid, benomyl, iprodione, procymidone and vinclozolin) in red and white bottled wines from two Portuguese wine-producing zones. Otero, Ruiz, Grande, and Gandara (2002) analysed 15 commercial white wines produced in Rias Baixas area (Spain) upon the presence of cyprodinil and fludioxonil. Most of wine samples analysed (75%) presented the two monitored fungicides. Čuš, Baša Česnik, Velikonja Bolta, & Gregorčič (unpublished results) found boscalid, cyprodinil, dimethomorph, fenhexamid, metalaxyl, procymidone and tebuconazole in finished wines after the vinification of six grapevine varieties in two years, but the concentrations were lower than 0.14 mg/L, which was similar to the results of Cabras et al. (1995). Cabras and Angioni (2000) showed that after fermentation, the quantities of pesticide residues in wine were always smaller than those on the grapes and in the must, except for those pesticides that did not have a preferential partition between liquid and solid phase (i.e. azoxystrobin, dimethoate, and pyrimethanil) and were present in wine in the same concentration as on the grapes. Among the clarifying substances commonly used in wine (bentonite, activated carbon, diatomaceous earth, gelatin, polyvinylpolypyrrolidone, potassium caseinate, and colloidal silicon dioxide), activated carbon allows the complete elimination of most pesticides, especially at low levels, whereas, the other clarifying substances are rather ineffective (Cabras and Angioni, 2000, Ruediger et al., 2004). Ruediger et al. (2004) also showed that bentonite had a major effect on the removal of carbendazim. However, pesticide residues belong to the chemical parameters of overall wine quality, therefore, data on them are definitely welcome both for the producers, who hope to reduce the presence and concentration of pesticides in wine, as well as for the consumers, who are concerned about food safety.

Numerous analytical methods for determining pesticide residues in grapes have been published until now. Gas chromatography (Angioni et al., 2005, Gonzalez et al., 2003, Navarro et al., 2000a) and liquid chromatography (Juan-García et al., 2004, Rial Otero et al., 2003, Teixeira et al., 2004) are the two most powerful techniques (Gandara et al., 1993, Lehotay et al., 2005, Pang et al., 2006) that were used in the study.

On the other hand, proper microbiological wine quality means the presence of yeast and bacteria cells below the concentration prescribed/desired for a certain wine. Prior to bottling, the producers must therefore ensure a microbiological stability that prevents the growth of microorganisms in bottles. The limits of acceptable concentrations of individual groups of microorganisms in wine depend on the chemical and microbiological parameters of the wine (i.e. pH value, concentrations of ethanol, free and total SO₂, reducing sugars and malic acid), internal control, and the control of trade chains as well as on the type of transport, storage and selling of wine (Loureiro & Malfeito-Ferreira, 2003). A microbiological survey of 68 samples of filtered and non-filtered wines from different wineries in the La Mancha region showed that 25% of the wines contained lactic acid bacteria, 13% acetic acid bacteria and 51% contained yeasts (Ubeda & Briones, 1999). Renouf, Claisse, & Lonvaud-Funel (2007) followed the wine microbial consortia at each stage of the process from grapes to bottles. They confirmed that the microbial diversity decreased during the process. Among the most resistant species in wine were Pediococcus parvulus for the bacteria and Brettanomyces bruxellensis for the yeast. Nisiotou and Gibson (2005) discussed that yeast population in market wines depended on postfermentation procedures or storage conditions. Their results show that commercial wines may possess diverse and potentially harmful yeast populations of B. bruxellensis, Saccharomyces cerevisiae and Rhodotorula pinicola yeasts. Bartowsky and Henschke (2008) claim that non-filtered red wine, especially, may often have a small resident bacterial population (<1.000 cfu/mL), which, under conducive conditions, might proliferate. Consequently, bottled red wines sealed with natural cork closures and stored in a vertical upright position may develop spoilage by acetic acid bacteria. Unfortunately, we do not have an analysis of the microbial quality of marketed wines in Slovenia. However, the wine industry is interested in easy, rapid and accurate methods to monitor microbiological contaminants during wine processing. Microbial population in wine can be assessed on the basis of membrane filtration and cultivation (Ubeda & Briones, 1999) or with direct analyses without cultivation step, for example with the use of molecular tools like PCR–DGGE (Cocolin et al., 2000, Renouf et al., 2007b).

In the current study, we sampled bottled wines from store shelves in order to determine the type and quantity of pesticide residues and the presence of viable and cultivable microorganisms. At the same time, we tested the effect of two fining agents on the decrease of boscalid concentration in wine.