Journal of Food Engineering 99 (2010) 485490

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier.com/locate/jfoodeng

Application of biosensors in early detection of contamination with lactic acid bacteria during apple juice and concentrate production
Magorzata Przybyt a,*, Jan Iciek b, Agnieszka Papiewska b, Joanna Biernasiak b
a b

Institute of General Food Chemistry, Faculty of Biotechnology and Food Sciences, Technical University of Lodz, Stefanowskiego 4/10, 90-924 Lodz, Poland Institute of Chemical Technology of Food, Faculty of Biotechnology and Food Sciences, Technical University of Lodz, Stefanowskiego 4/10, 90-924 Lodz, Poland

a r t i c l e

i n f o

a b s t r a c t
The aim of Collective Research Project QUALI-JUICE (COLL-CT-2005, Co Nr 012461) was to introduce the biosensors in control of microbiologically pure apple juice production. Three commercial lactate biosensors were used and compared with enzyme kits assays. Parallel the microbiological analysis of the production process at one juice producing enterprise was done. The results of lactic acid assay with biosensors were in good correlation with those obtained by enzyme kits. The main benet of biosensor use was shortening of measurement time as compared with assay by enzyme kit and possibility to measure at line. The concentration of L-lactic acid in apple pulp could be correlated with the number of lactic acid bacteria. Pasteurization process was eliminating lactic acid bacteria and the concentration of lactic acid was at this stage not exceeding 0.1 g L1. The nal product (apple concentrate) contained in some cases very high amounts of lactic acid indicating secondary microbiological contamination after pasteurization step. Parallel microbiological analysis of production process and lactate assay indicated that the critical point during production was prolonged vacuum ltration after pasteurization. 2010 Elsevier Ltd. All rights reserved.

Article history: Received 20 February 2009 Received in revised form 17 February 2010 Accepted 18 February 2010 Available online 24 February 2010 Keywords: Lactate biosensor Lactic acid bacteria Juice production control

1. Introduction Europe is one of the worlds largest markets for fruit products and fruit juice and the demand is still on rise. The fruit juice industry is very important for the economy of the European Union counting more than 20,000 companies, of which about 90% are small- or medium-sized enterprises. The undesired microbial contamination is a problem in fruit juice industry. The origin and source of infection is often unknown and the beginning of juice spoilage is difcult to be detected. The infections of juices in storage tanks by alcohol producing microorganisms can be detected by increase of pressure due to production of CO2, while the lactic acid fermentation caused by lactic acid bacteria (LABs) is often unnoticed. When such fermentation is detected by sampling and tasting of the stored juice, organoleptic properties of the juice do not permit a further use for human consumption. The spoiled juice has to be poured away provoking problems for the waste water treatment systems and nancial losses for the producer. There are two stereoisomers of lactic acid: D- and L-lactic acid. Most of LABs are producing only L-lactic acid but some strains can produce D isomer or racemate. The A.I.J.N. (European Fruit Juice Association) Code of Practice laid down the upper limit of lactic acid in juice for human consumption to 0.5 g L1 (sum of both

* Corresponding author. Tel.: +48 42 6313426. E-mail address: mprzybyt@snack.p.lodz.pl (M. Przybyt). 0260-8774/$ - see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jfoodeng.2010.02.022

forms). The detection methods for LABs (microbial cultivation or DNA ngerprinting) are time and work consuming and expensive. The increase of L-lactic acid concentration in juice can be a marker for fermentation by LABs (Trir et al., 1997) and in some cases it could be a measure of index of Escherichia coli number (Casimiri and Burnstein, 1998). Lactic acid can be assayed using HPLC, ion exclusion chromatography, colorimetric test strips or enzymatic test kits. All these methods are time consuming, expensive and demanding trained laboratory stuff. Only enzymatic kit methods can distinguish the stereoisomers of lactic acid, the colorimetric test strips measure only L-lactate and chemical methods the sum of both forms. The alternative method for monitoring lactic acid is the use of biosensors, mainly amperometric enzyme electrodes. The state of art of amperometric lactate biosensors was reviewed recently (Nikolaus and Strehlitz, 2008). The use of biosensors can shorten the assay time and lower the cost of analysis. The objective of Collective Research Project QUALI-JUICE (COLLCT-2005 Co Nr 012461) is to develop a biosensor based detection and early warning system for juice spoilage by LABs for the European fruit juice industry. For purpose of the project four commercial lactate biosensors were chosen and tested for measurements of lactate concentration in apple juices during their production and in nal products. This work presents testing of three of them BIOSEN_C line sport (EKF, Germany), LactatPro 3000 (ABT, Germany) and Sensytec 1 (Tectronik, Italy) and application of them in control

486

M. Przybyt et al. / Journal of Food Engineering 99 (2010) 485490

of apple juice concentrate production. The second goal of the research was to show how the biosensors with parallel microbiological analysis can be used to nd the origin of secondary contamination by LABs on the production line. 2. Experimental 2.1. Reagents Polyamide (nylon) 6 (for column chromatography) was purchased from Fluka (Germany), polyvinylpolypyrrolidone powder (PVPP) was purchased from Sigma (Germany). As the standards L-lactic acid, approx. 98% (Sigma, Germany) or lithium L-lactate, approx. 97% (Sigma, Germany) were used dissolved in distilled water or 0.1 M phosphate buffer, pH 7. For enzymatic assay of L- and Dlactate, ethanol and acetic acid in juice samples the enzyme kits produced by Megazyme (Ireland) were used. All other chemicals were of analytical grade. The samples of apple pulp (crushed apples), must (juice pressed from pulp with tissue particles, not claried), juice (claried, after ltration) and concentrate were a kind gift of VIN-KON company (Konin, Poland). The samples were deep frozen after taking them from the production line (except concentrates). The samples of apple pulp, must and juice were signed by the producer with date of the production. The samples of concentrates were signed by the producer symbols depending on the production stock or storage tank from which they were taken. These symbols are used throughout the text. 2.2. Measurements The apple juice concentrates were dissolved with distilled water to 11.18Brix. The samples of apple must or pulp were centrifuged at 14 000 rpm for 15 min. For enzymatic assay (spectrophotometric) of lactate 1 mL of sample was added by 100 mg of PVPP, shacked vigorously, left for 5 min and then centrifuged at 14 000 rpm for 15 min. The assay was done according to the procedure given by the producer of kits. Each assay was triplicated. The absorbance was measured at 340 nm in disposable PMMA Plastibrand cuvettes (Sigma, Germany) with UV spectrophotometer Nicolet Evolution 300 from Thermo Electron Corporation (MA, USA). 2.3. Biosensors The assay of lactate concentration with the use of biosensors was done according to the producer manual with standards and buffers purchased from them. Each measurement was triplicated (except Sensytec 1). BIOSEN_C line sport (EKF, Germany) and LactatPro 3000 (ABT, Germany) devices use the biosensors in form of screen printed electrodes (biochips) with immobilized L-lactate oxidase (LOD). L-lactate is oxidized to pyruvate according to the reaction:

for 50 days or 6000 measurements. In case of LactatPro 3000 5 lL of sample is sucked by a capillary which is then placed in the apparatus. The sample is sucked from capillary to chamber where the biochip is placed and the sample is analyzed. Time of measurement is about 2 min. One biochip can be used for 15 days or 400 measurements. Sensytec 1 (Tectronik, Italy) use the screen printed electrodes with L- or D-lactate dehydrogenase. Lactates are oxidized by them according to the reactions:

l-lactate NAD d-lactate NAD

l-lactate dehydrogenase

pyruvate NADH pyruvate NADH

2 3

d-lactate dehydrogenase

The details of electrochemical reaction cannot be given while they are secret due to patent. The device is dedicated to assay lactates in wines. The biochip is dipped in 10 mL of buffer solution. After equilibration of the biochip (240 s) 200 lL of standard solution is added, after waiting 60 s sample (200 lL) is added, and after next 60 s standard solution once again. After next 60 s the result of measurement is given. The biosensors used with Sensytec 1 are disposable and can be used only for three measurements. The samples were used as received in case of juices or centrifuged in case of pulp and must. For comparison the measurements were done for samples puried by adsorption with PVPP or polyamide 6. Amount of 100 mg of the was added to 1 mL of sample, shacked vigorously, left for 5 min and then centrifuged at 14 000 rpm for 15 min. 2.4. Microbiological analysis Enumeration of microorganisms was done according to Polish standards: aerobic mesophilic and psychrophilic microorganisms (PN-90/A-75052/05); yeasts and moulds (PN-90/A-75052/08); lactic acid bacteria (PN-90/A-75052/07) and aerobic thermophilic bacteria (PN-90/A-75052/06, 2006). 2.5. Fermentations Raw apple must was obtained from apples bought on local market by crushing them and pressing. Then 200 mL must was left for fermentation for 168 h at different temperatures (20 and 30 C). For model LABs fermentations two types of bacteria were chosen: Lactobacillus plantarum and Lactobacillus brevis. Reconstructed apple juice was obtained by dissolving apple concentrate to 11.18Brix and sterilized. 699 mL of the juice was inoculated with 1 mL of bacteria suspension. Fermentations were done in closed bottles at 20 and 30 C for 168 h. For comparison the apple juice was inoculated with wild strains using fermented apple must as inoculum. 3. Results and discussion Performance characteristic of BIOSEN_C line sport and LactatPro 3000 is reported elsewhere (Przybyt and Biernasiak, 2008). The results presented there indicated that the two biosensors can be used to assay L-lactate in apple juice after some modications. The results are more accurate when samples are puried with polyamide 6 as compared with typical adsorbent is PVPP which is used in wine and juice industry (Spagna et al., 1996; Borneman et al., 2001) and in enzymatic assay of lactate to decolorize samples. Also the disturbing inuence of vitamin C which is oxidized at the electrode at the same potential as H2O2 (Garca Armada et al., 2003; Ledru and Boujtita, 2004; Wang et al., 2006) causing the apparent increase of current leading to overestimated results of measurements is partially eliminated by this adsorbent. The

l-lactate O2

l-lactate oxidase

pyruvate H2 O2

H2O2 produced in reaction (1) is oxidized on the electrode surface producing current which is proportional to L-lactate concentration. The concentration of L-lactate in the sample is calculated by comparison of current produced with standard solution of known concentration. BIOSEN_C line sport and LactatPro 3000 are dedicated for assay of L-lactate in blood and serum and their analytical performance is optimized for these matrices. In case of BIOSEN_C line sport 20 lL of sample is added to 1 mL of buffer in test tube purchased by the producer. Fifteen test tubes are placed in the apparatus and the measurement is done automatically. Time of one measurement is about 15 s. One biochip can be used at least

M. Przybyt et al. / Journal of Food Engineering 99 (2010) 485490

487
o o

application of PVPP is leading to signicant increase of measured L-lactate results for both devices. PVPP dissolves to some extend in juice and most probably is electroactive. Additionally PVPP is strongly swelling in water and sedimenting slowly. Polyamide 6 has advantage as the purifying agent in practical use while it is not swelling and is sedimenting quickly so there is no need of centrifugation of samples. Important modication in measurement procedure for LactatPro 3000 was the calibration of the device before each measurement instead every 8 h as it is indicated by the manual. The results of performance characteristics showed that BIOSEN_C line sport is more accurate, less work and time consuming and more user friendly. The optimization of Sensytec 1 for assay of lactates in apple juice was done by the producer of the device (Tectronik) and is not presented in this paper. 3.1. L- and D-lactate assay in apple juice concentrates Table 1 summarizes the results of lactic acid assay by enzyme kits and biosensors at different analytical conditions. The results of L-lactate assay with enzyme kit and biosensors are in good agreement. Purication of samples with PVPP is leading to great errors in the results (Przybyt and Biernasiak, 2008). Application of polyamide 6 as purifying agent is leading to more accurate results. The greatest differences between results of enzyme kit assay and measurements with biosensors are observed for concentrates signed by the producer as 65/VII, 66/VII and A1/FB/07. These concentrates had very dark brown colour as compared with others. Probably the products of enzymatic and non-enzymatic browning could interfere with the biosensors. For D-lactate values of concentrations assayed by enzyme kits and measured with Sensytec 1 are also in good agreement. Characteristic is that concentrates with high concentration of L-lactate have also elevated level of D-lactate. 3.2. Laboratory fermentations To check if the increase of L-lactate concentration can be connected with the growth of lactic acid bacteria the fermentation of apple must and apple juice reconstructed from concentrate were carried on. During spontaneous fermentation of apple must the growth of LABs is causing the increase of L-lactate especially at temperature 20 C (Fig. 1) in the 4th day. The concentration of D-lactate was unchanged indicating that in this case the wild LABs
Table 1 D- and L-lactate concentration in apple concentrates measured by different methods. Concentrate (symbols of the producer)
L-Lactate

20 C 30 C
109
o

20 C 30 C
5

lactic acid bacteria [CFU mL ]

108

-1

107

L-lactate [g L ]

10

105
1

-1

104

103 0 20 40 60 80 100 120 140 160 180

time [hours]
Fig. 1. Dynamics of lactic acid bacteria growth and L-lactate evolution during spontaneous fermentation of apple must (L-lactate assayed by enzyme kit).

are not producing this isomer of lactic acid. This could not be truth always while populations of LABs from different apple stocks could be different. Some preliminary experiments with spontaneous fermentations of samples of apple pulp and unpasteurized must indicated that sometimes also the D-lactic acid could be produced. Also the growth of yeast during wild fermentation was observed leading to alcoholic fermentation which was predominant at temperature 30 C. In this case after 7 days the concentration of ethanol was 17.6 g L1. Also the production of acetic acid was observed, once again more signicant at 30 C. The inoculation of apple juice with L. plantarum and L. brevis caused no increase of lactic acid when the initial inoculum was about 103 CFU mL1. The microbiological analysis conrmed these results showing no increase of bacteria even decrease of their numbers. Only in case of L. brevis when the initial inoculum was 2.6 106 CFU mL1 some production of L-lactic acid was observed. L. brevis was growing during rst 2 days and then the number of them was decreasing. These results indicate that laboratory strains of LABs cannot adapt to apple juice environment. The results of apple juice inoculation with wild bacteria strains were very similar to those of spontaneous fermentation of unpasteurized apple must. The increase of L-lactate concentration and

(g L1) BIOSEN_C line sport PVPP Nylon 6 0.037 0.007 0.061 0.005 0.072 0.003 0.230 0.005 0.598 0.006 0.543 0.002 0.021 0.002 0.011 0.001 0.038 0.002 0.014 0.001 0.010 0.001 0.830 0.008 0.08 0.01 0.11 0.14 0.02 039 0.03 0.83 0.01 0.73 0.01 0.17 0.01 0.14 0.02 0.10 0.02 0.17 0.01 0.13 0.01 0.96 0.04 LactatPro 3000 PVPP 0.13 0.18 0.18 0.01 0.37 0.01 0.85 0.05 0.76 0.03 0.12 0.01 0.11 0.01 0.16 0.02 0.13 0.11 1.06 0.04 Nylon 6 0.08 0.01 0.12 0.13 0.01 0.36 0.01 0.84 0.04 0.67 0.03 0.14 0.01 0.14 0.01 0.11 0.01 0.16 0.01 0.17 0.01 1.11 0.12 Senzytec 1 PVPP 0.048a 0.076a 0.101a 0.242 0.686 0.575 0.057a 0.086a 0.064a 0.003a 0.051a 0.880a Nylon 6 0.084a 0.090a 0.133a 0.259 0.810 0.819 0.086a 0.127a 0.023a 0.015a 1.058

D-Lactate

(g L1) Senzytec 1 PVPP Nylon 6 0.023 0.081 0.044 0.038 0.217 0.091 0.049 0.052 0.041 0.027 0.266

Enzyme kit

Enzyme kit

46/VII 47/VII 48/VII 64/VII 65/VII 66/VII 71/VII T 154 T 156 T 262 T 348 A1/FB/07

0.052 0.019 0.099 0.019 0.119 0.035 0.248 0.013 0.884 0.093 0.915 0.106 0.044 0.020 0.039 0.015 0.041 0.005 0.033 0.005 0.048 0.008 1.068 0.018

0.046 0.002 0.080 0.002 0.081 0.004 0.254 0.006 0.718 0.021 0.632 0.016 0.030 0.003 0.021 0.004 0.054 0.005 0.013 0.001 0.012 0.001 0.986 0.025

0.113 0.009 0.139 0.003 0.152 0.006 0.297 0.003 0.714 0.001 0.648 0.009 0.099 0.004 0.102 0.003 0.121 0.004 0.096 0.003 0.077 0.004 0.949 0.004

0.029 0.004 0.051 0.001 0.048 0.002 0.083 0.005 0.269 0.009 0.130 0.009 0.031 0.002 0.025 0.001 0.027 0.001 0.027 0.001 0.025 0.001 0.249 0.003

0.055 0.057 0.059 0.022 0.113 0.125 0.029 0.069 0.021 0.006 0.254

Not detectable. All concentrates were dissolved to 11.18Brix. In case of D-lactate assay with Sensytec 1 all measurements were done by internal standard method. a Measured by internal standard method.

488

M. Przybyt et al. / Journal of Food Engineering 99 (2010) 485490

0.07

0.22 0.20 0.18 0.16

L-lactate [g L ]

0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00 0 1 2 3 4 5 6 7

L-lactate [g L ]

enzyme kit LactatProfi 3000 BIOSEN_C line sport

0.06 0.05 0.04 0.03 0.02 0.01 0 1.00E+01

y = 0.0058Ln(x) - 0.0156 R 2 = 0.7412

-1

-1

1.00E+02

1.00E+03

1.00E+04

1.00E+05
-1

1.00E+06

time [hours]
Fig. 2. L-Lactate concentration during fermentation of apple juice inoculated by wild strains at temperature 20 C.

lactic acid bacteria number [CFU g ]

Fig. 3. Dependence of L-lactate concentration in apple pulp on number of lactic acid bacteria (L-lactate assayed by enzyme kit).

LABs number was not so signicant because lower initial concentration of them (one order of magnitude). Also the production of ethanol and acetic acid was observed. Fig. 2 presents the evolution of L-lactate during fermentation with wild strains measured with different methods. The results indicate that biosensors can be used to follow the lactic acid fermentation in apple juice.

3.3. Application of biosensors in juice production control Lactic acid bacteria that can cause increase of lactate concentration during apple juice production can originate from the raw material apples. Microbiological analysis of crushed apples (apple pulp) showed presence of bacteria (mainly mesophilic aerobes), lactic acid bacteria, yeasts and moulds. The values for different microorganisms were varying in limits: Mesophilic aerobes from 1.5 102 to 4.5 104 CFU g1 Thermophilic aerobes from less then 1 to less than 10 101 CFU g1 Lactic acid bacteria from 4 101 to 5 105 CFU g1 Yeasts from 6.5 101 to 5.8 105 CFU g1 Moulds from 5 101 to 2 104 CFU g1. Highest number of microorganisms was found in samples from September 2007 (freshly harvested apples), lowest from January 2007 (stored apples). Bacteriological analysis of residual water after its use in apple washing showed very high level of microbiological contamination even with thermophilic aerobes (which cannot be eliminated in pasteurization process). This indicates that

washing could be critical step to remove soil bacteria contamination from fresh apples, and should be improved. L-Lactate level in apple pulp (assayed in supernatant after centrifugation of the pulp as it is described in experimental), was always below 0.1 g L1 and showed some correlation with LABs number (Fig. 3). Concentration of L-lactate in raw must after pressing was always similar as in apple pulp and sometimes increased slightly after pasteurization (Table 2) due to evaporation of some water at this production step. Concentration of D-lactate in all studied samples was very similar (between 0.03 and 0.05 g L1) and was independent on bacterial contamination of raw material and stage of production process. The values of lactate concentrations measured by biosensors were always similar to that assayed by enzyme kit method. The bacteriological analysis of the production process showed that pasteurization was carried out properly. Yeasts, moulds and mesophilic aerobes among them LABs were practically completely eliminated but in some cases the thermophilic aerobes were still present after pasteurization (Table 2). This phenomenon is probably caused by thermal activation of dormant spores of thermophilic aerobes originating from soil (Iciek et al., 2006). The pasteurized must is then ltrated and directed to evaporator to obtain concentrate. Most of the concentrate samples had also low concentration of L- and D-lactate corresponding to values in the must after pasteurization. Surprisingly some concentrates had very high content of L-lactate accompanied with increased amount of Dlactate (Table 1). These results indicate indirectly that there must be a source of secondary microbiological contamination after pasteurization step.

Table 2 Examples of the results of bacteriological analysis illustrating the effectiveness of pasteurization process. Raw must 9.01.2007 Number of mesophilic aerobes Number of thermophilic aerobes Number of yeast Number of moulds Number of lactic acid bacteria
D-Lactate L-Lactate

Must after pasteurization 9.01.2007 Less than 1 CFU mL1

Raw must 15.11.2006 1.8 103 CFU mL1

Must after pasteurization 15.11.2006 Less than 1 CFU mL1

Raw must 15.11.2006 afternoon 7 103 CFU mL1

Must after pasteurization 15.11.2006 afternoon Less than 1 CFU mL1

6.5 102 CFU mL1

Less than 10 CFU mL1 4 101 CFU mL1 4 102 CFU mL1 4.5 102 CFU mL1 0.030 0.001 0.031 0.013

2 101 CFU mL1

Less than 1 CFU mL1 3.5 102 CFU mL1 2.8 102 CFU mL1 1.3 103 CFU mL1 0.036 0.002 0.060 0.010

5.62 102 CFU mL1

Less than 1 CFU mL1 6.5 102 CFU mL1 1.9 102 CFU mL1 2.2 103 CFU mL1 0.029 0.001 0.054 0.012

5.22 102 CFU mL1

Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 0.040 0.001 0.023 0.004

Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 0.056 0.003 0.054 0.015

Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 0.056 0.001 0.068 0.028

(g L1) (g L1)

M. Przybyt et al. / Journal of Food Engineering 99 (2010) 485490 Table 3 Results of microbiological analysis and lactate concentration assays in juice during ltration. Must after pasteurization 27.03.2007 Number of Number of aerobes Number of Number of Number of
D-lactate D-lactate L-lactate L-lactate L-lactate

489

Juice after vacuum lter 28.03.2007 5 101 CFU mL1 Less than 1 CFU mL1 7 101 CFU mL1 Less than 10 CFU mL1 6 101 CFU mL1 0.067 0.005 0.080 0.006 0.029 0.110 0.002 0.15 *

Juice after ultralter 28.03.2007 Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 0.047 0.001 0.053 0.061 0.011 0.064 0.123 0.002 0.16 Typical apple smell and taste

Juice after ultralter afternoon 28.03.2007 Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 10 CFU mL1 Less than 1 CFU mL1 0.138 0.004 0.077 0.309 0.027 0.403 0.402 0.007 0.52 0.01 Taste and smell changed

mesophilic aerobes thermophilic yeasts moulds lactic acid bacteria

Less than 10 CFU mL1 Less than 10 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 Less than 1 CFU mL1 0.041 0.001 0.020 0.002 0.073 0.091 0.002 0.14 0.01 *

(g L1) (enzyme kit) (g L1) (Sensytec 1) (g L1) (enzyme kit) (g L1) (Sensytec 1) (g L1) (BIOSEN_C line

sport)
L-lactate

(g L1) (LactatPro 3000) Organoleptic properties

, Unmeasurable; *, not investigated. The measurements with biosensors were done for samples treated with PVPP. D-Lactate was measured by Sensytec 1 using internal standard method.

Further examination of the production process showed that the critical point is ltration of the juice after pasteurization. Two types of ltration are applied: ultraltration and vacuum ltration. Ultralter is used always during production process. Vacuum lter is used only occasionally when in the buffer tank must rich in sediment (rest of apple tissues) is present. The juice after vacuum ltration is directed back to the buffer tank where the must is waiting for ultraltration. While the vacuum lter is sucking the non sterile air from the production room the juice can be contaminated by microorganisms. Such contamination is dangerous for production process especially when the vacuum lter is used for a long time. Even if the juice is contaminated by low number of microorganisms the bacteria can grow in buffer tank leading to fermentation of juice and production of lactic acid. The microbiological analysis results indicating secondary contamination after vacuum ltration leading to increased level of lactic acid are shown in Table 3. The results of Llactate measurements by biosensors are overestimated but all measurements were done for samples treated with PVPP. The fermentation and spoilage of the juice was conrmed by organoleptic analysis. When the vacuum lter was not used no increase of L-lactate after ultraltration was observed (analysis of the production during one week) and the concentrates had typical low concentration of lactic acid. These results indicate that the critical point in the production process is vacuum ltration. 4. Conclusions

 The critical points in apple concentrate production are: washing of apples (important from soil bacteria contamination point of view) and ltering of juice after pasteurization. At this point on vacuum lter there is possibility of secondary bacterial contamination.  The increase of L-lactate in apple juice during production process and storage could be used as indicator of juice spoilage by LABs and could be monitored with biosensors.  The nal choice of the device (biosensor) by the future user (juice producing company) would be dened by its particular demands (simplicity of the measurements, possibility of usage at line) and economical impact (price of the device and consumables).

Acknowledgements This work was supported by the EU under 6th Framework Programme Collective project QUALI-JUICE No. COLL-CT-2005-012461 and supporting Grant of Ministry of Science and Higher Education in Poland SPUB-246. References
Borneman, Z., Gkmen, V., Nijhuis, H.H., 2001. Selective removal of polyphenols and brown colour in apple juices using PES/PVP membranes in a single ultraltration process. Separation and Purication Technology 2223, 5361. Casimiri, V., Burnstein, C., 1998. Biosensor for L-lactate determination as an index of E. coli number in crude culture medium. Analytica Chimica Acta 361 (12), 45 53. Garca Armada, M.P., Losada, J., Cuadrado, I., Alonso, B., Gonzlez, B., Casado, C.M., 2003. A siloxane homopolymer with interacting ferrocenes as a new material for the preparation of sensors based on the detection of hydrogen peroxide. Electroanalysis 15 (13), 11091114. Iciek, J., Papiewska, A., Molska, M., 2006. Inactivation of Bacillus stearothermophilus spores during thermal processing. Journal of Food Engineering 77, 406410. Ledru, S., Boujtita, M., 2004. Electrocatalytic oxidation of ascorbate by heme-FeIII/ heme-FeII redox couple of the HRP and its effect on the electrochemical behaviour of an L-lactate biosensor. Bioelectrochemistry 64 (1), 7178. Nikolaus, N., Strehlitz, B., 2008. Amperometric lactate biosensors and their application in (sports) medicine, for life quality and wellbeing. Microchimica Acta 160 (12), 1555. Polish standard PN-90 A-75052/05. Przetwory owocowe, warzywne i warzywno mikrobiologicznych Oznaczanie obecnos ci i liczby mie sne Metody badan drobnoustrojw tlenowych mezolnych i psychrolnych. Fruits, vegetables and vegetable-meat products. Methods of microbiological analyses. Detection and enumeration of aerobic, mesophilic and psychrophilic microorganisms (in Polish).

 The commercial lactate biosensors could be used to measure lactate concentration in juices during production of apple concentrate and in apple juices shortening the time of analysis and lowering costs.  The apparent value of L-lactate measured by biosensors is inuenced by interferants present in juices (vitamin C and polyphenols).  The results of lactate measurements with commercial biosensors could be improved by purication of samples with polyamide 6.  The L-lactate concentration in raw material (apple pulp) could be correlated with contamination with lactic acid bacteria.  The process of pasteurization eliminates bacterial contamination with exception of thermophilic aerobes.

490

M. Przybyt et al. / Journal of Food Engineering 99 (2010) 485490 Przybyt, M., Biernasiak, J., 2008. Zastosowanie biosensorw do oznaczania mleczanw w owocowych sokach komercyjnych i koncentratach. Application _ c . of biosensors to L-lactate assai in commercial juices and concentrates. Zywnos c . 5 (60), 168177 (in Polish). Nauka. Technologia. Jakos Spagna, G., Pifferi, P.G., Rangoni, C., Mattivi, F., Nicolini, G., Palmonari, R., 1996. The stabilization of white wines by adsorption of phenolic compounds on chitin and chitosan. Food Research International 29 (34), 241248. Trir, A., Saccani, G., Gherardi, S., Vicini, E., Spotti, E., Previdi, M.P., Ndagijimana, M., Cavalli, S., Reschiotto, C., 1997. Use of ion chromatography for monitoring microbial spoilage in the fruit juice industry. Journal of Chromatography A 770 (12), 243252. Wang, K., Xu, J.J., Chen, H.Y., 2006. Biocomposite of cobalt phthalocyanine and lactate oxidase for lactate biosensing with MnO2 nanoparticles as an eliminator of ascorbic acid interference. Sensors and Actuators B 114 (2), 10521058.

Polish standard PN-90 A-75052/07. Przetwory owocowe, warzywne i warzywno mikrobiologicznych Oznaczanie liczby bakterii mie sne. Metody badan kwasza cych typu mlekowego. Fruits, vegetables and vegetablemeat products. Methods of microbiological analyses. Enumeration of lactic acid bacteria (in Polish). Polish standard PN-90 A-75052/06. Przetwory owocowe, warzywne i warzywno mikrobiologicznych Oznaczanie liczby bakterii mie sne. Metody badan tlenowych termolnych. Fruits, vegetables and vegetablemeat products. Methods of microbiological analyses. Enumeration of aerobic thermophilic bacteria (in Polish). Polish standard PN-90 A-75052/08. Przetwory owocowe, warzywne i warzywno_ zy _ i mikrobiologicznych Oznaczanie liczby drozd mie sne. Metody badan ni. Fruits, vegetables and vegetablemeat products. Methods of ples microbiological analyses. Enumeration of yeasts and moulds (in Polish).