Low Occurrence of Patulin-Producing Strains of Penicillium in Grapes and Patulin Degradation during Winemaking in Chile

Vineyard and winery sampling.

Specimens were isolated from apparently healthy clusters of Vitis vinifera, 280 clusters from red grape cultivars and 110 clusters from white grape cultivars. The samples were obtained from 13 different vineyards located within 255 km along a north to south axis in central Chile (lat: 33°27′ to 35°34′ S; long: 70°42′ to 71°43′ W) between 2006 and 2008. To study the frequency of occurrence of Penicillium spp. in the vineyards, samples were taken at the pea-sized berry stage, at the beginning of berry ripening (veraison), and at harvest stage (Eichhorn and Lorenz 1977). Each sample consisted of 10 apparently healthy grape clusters that were transported in cold chests to the laboratory. Fifty berries were randomly selected from each cluster and washed with 50 to 75 mL sterile 0.05% Tween 80 for 5 min. Aliquots (100 μL) of the resulting suspension were plated on modified potato dextrose agar plus 0.005% tetracycline, 0.01% streptomycin, and 0.1% Igepal CO-630 (Sigma-Aldrich, St. Louis, MO) (MPDA) (Díaz et al. 2009). Data were expressed as colony forming units (cfu) per grape surface. It was assumed that Cabernet Sauvignon berries were spherical.

Air samples (n = 20) were obtained 0.5 m from the grape clusters and 1.80 m aboveground in the vineyard at full flowering, berries small, berries pea-sized, beginning of bunch closure, and veraison in the Alto Jahuel and Alhué vineyards (Eichhorn and Lorenz 1977). Samples were obtained with the aid of a portable spore trap (model PASA/B; Burkard, Rickmansworth, UK) equipped with 90-mm-diam petri dishes containing MPDA. On each sampling date, the spore trap sampled the air for 10, 30, and 60 s at a rate of 10 L min⁻¹. The presence of Penicillium colonies was determined by examining the samples under a stereoscopic microscope after 5 days incubation at 20°C. Isolates were maintained on MPDA at 5°C. The air samples in the wineries (n = 30) were obtained by sampling the air at five wineries at different times during the winemaking process (first working day and 30 and 60 days later). Air samples were taken for 60 s with the aid of a portable spore trap that was placed 1.5 m aboveground and ~1.0 m from destemming and crushing machinery.

Identification of isolates.

Isolates of Penicillium were identified based both on the morphological characteristics of the conidia and conidiophores and on colony morphology on Czapek agar (Cz), Cz yeast extract (CYA), CYA-20% sucrose (CYAS), malt extract agar (MEA), creatine sucrose agar (CREA), and yeast extract sucrose agar (YES) (Samsom and Frisvad 2004, Pitt and Hocking 2009). Patulin-producing isolates of Penicillium morphologically identified as P. brevicompactum Dierckx (isolate PenUC-14) and P. expansum (isolates PenUC-21 and PenUC-35) and one patulin nonproducer Penicillium isolate, tentatively identified as P. glabrum (Wehmer) Westling (isolate PenUC-08), were selected for molecular identification using the internal transcribed spacer region (ITS1-5.8S-ITS2) of rDNA. These isolates were deposited in the Commonwealth Agricultural Bureau International (CABI, Egham, Surrey, UK).

Patulin-producing strains.

All Penicillium isolates (n = 132) were tested for patulin production in broth medium. Each isolate was cultivated in 30 mL Czapek yeast broth (CYB) with constant shaking in an orbital shaker for 7 days at 20°C. The cultures were filtered (glass fiber filter circles G6; Fisher Scientific, Pittsburgh, PA), centrifuged at 8944 × g for 25 min, and stored in 1.5 mL and 5 mL sterile tubes at −20°C until the patulin determinations were performed.

Patulin production was determined by high-performance liquid chromatography using a detector with UV and diode array detector capabilities (HPLC-UV/DAD) with a solid-phase extraction (SPE) method (Li et al. 2007). Samples were prepared by washing 1 mL culture extract from the CYB cultures with 0.5 mL acetic acid and agitating for 10 min. Then, 1.5 mL of treated sample was filtered through a solid-phase column (2–3 mL min⁻¹), eluted with 5 mL hexane (2 to 3 mL min⁻¹), and air dried for 15 min. Patulin was eluted three times with 5 mL of hexane/ethyl acetate/acetone (1:5:4; 1:4:5; 1:3:6, v/v/v). The 15 mL of eluted material was acidified with 15 μL glacial acetic acid and then evaporated to complete dryness at 40°C under a flow of nitrogen. The patulin concentration was calculated using the peak area at λ = 276 nm of a patulin standard solution. A stock standard solution of patulin was prepared by dissolving 1 mg pure crystalline patulin (Sigma-Aldrich) in 10 mL double-distilled water (pH 4.0) acidified with acetic acid. Working standard solutions were prepared by appropriate dilution of this solution with water (pH 4.0).

Patulin in wine made with grapes contaminated with a patulin-producing strain of P. expansum.

Cabernet Sauvignon grapes harvested at maturity (24% total soluble solids) in April were mixed with artificially infected grapes to obtain 20 kg grape lots that included 0.0, 0.5, 1.0, and 3% (w/w) infected grapes. Infected grapes were obtained after inoculating 100, 200, or 600 g of grape clusters that had been superficially disinfected (75% ethanol for 30 s) before spraying with 3 mL of a conidial suspension of P. expansum (PenUC-35) adjusted to 10⁶ conidia mL⁻¹. Berries were aseptically punctured with a sterile hypodermic needle before inoculation. Inoculated clusters were incubated in humid chambers at 20°C for 5 days. Grape lots were subjected to microvinification at 28 to 30°C in 40 L jars under semi-aerobic conditions using destemmed grapes that were manually crushed and fermented in the presence of skins and seeds. Total soluble solids content in the musts was 24%, pH 3.45. The nitrogen content of the musts was augmented with 300 mgL⁻¹ ammonium phosphate. Musts were then treated with 20 mgL⁻¹ pectolytic enzymes (Rapidase color; Gist-Brocades, Beverages Ingredients Group, Seclin, France) and 30 mgL⁻¹ sulfur dioxide (SO₂). The musts were homogenized, and the temperature and density were determined before the addition of 200 mgL⁻¹ Saccharomyces cerevisiae (Anchor WE 372; Anchor Bio-Technologies, Eppindust, South Africa) that had been prehydrated at 37°C for 30 min. At the end of fermentation, the wine was transferred into 5 L glass bottles that were kept at 20 to 24°C until the end of malolactic fermentation. The wine obtained was decanted into clean containers and was treated with 35 mgL⁻¹ SO₂. The wines were stored at 0°C for 30 days, and then the free SO₂ was adjusted to 35 mgL⁻¹. The final ethanol concentration of the wines was 14%. The patulin concentration in must and wine was determined with HPLC-UV/DAD, as described previously (Li et al. 2007). Data obtained from the musts and wines made from grapes infected with a patulin-producing strain of P. expansum were subjected to analysis of variance (p ≤ 0.05). The relationship between the proportion of infected grapes and the amount of patulin detected in musts and wines was analyzed by regression analysis. The statistical software SigmaStat (Systat Software, San Jose, CA) was used.