Introduction

Polygalacturonase enzyme (PG) is a pectic enzyme produced by fungi and high plants that catalyses the hydrolytic split between galacturonosyl units in pectins. According to its mode of action, two forms of PG have been described: exo-polygalacturonase (exo-PG) and endo-polygalacturonase (endo-PG).The exo form, exo-poly(1,4-α-D-galacturonide)galacturonohydrolase (EC 3.2.1.67), acts on the non-reducing end unit of polygalacturonic chains, releasing mainly galacturonic acid. Endo-PG, endo-poly(1,4-α-D-galacturonide)glycanohydrolase (EC 3.2.15), randomly breaks the polygalacturonic chain of pectins, generating small oligomers. Consequently, viscosity in pectin solutions decreases considerably [1, 2].

PG plays a major role during ripening, post harvest storage, industrial processing, and retailing of fruits and vegetables [3, 4, 5, 6]. The main interest and applications of this enzyme are in the fruit and vegetable juice industry, where the use of commercial enzyme preparations (CEP) with PG, as processing aids, has become customary. Depending on the goals pursued concerning the final product, induced changes by PG must be either prevented, as they cause undesirable effects, or enabled and enhanced, as they are essential in many current processes [7, 8, 9, 10]. Thus, rapid inactivation of native or added PG is necessary to ensure a cloudy product with high viscosity. Conversely, to obtain total liquefaction of fruit flesh, reduced viscosity of juice or better clarification of juices, higher PG activities will be needed [11, 12]. Research studies on the biological importance of, and new technological applications for PG, within and beyond food technology, are ongoing [13, 14].

Heat treatment is normally used to avoid adverse effects (for example, cloud loss, viscosity reduction, fruit softening) in processed foods caused by indigenous or fungal PG in raw materials [15]. It is also often used to destroy residual PG after PG is added during a processing step in order to cause certain changes when the correct extent of the changes has been achieved [11, 12]. Unfortunately, thermal methods lead to important undesirable effects such as colour changes, cooked flavours, and loss of vitamins and nutrients. Therefore, inactivation of PG by non-thermal treatments becomes a matter of interest.

Among non-thermal treatments, high-intensity pulsed electric fields (HIPEF) has tentatively emerged as a worthwhile technique to potentially inactivate PG, especially where the effects of HIPEF on PG have not been yet described in scientific literature. In fact, HIPEF has shown an uneven degree of efficacy in inactivating enzymes, such as plasmin [16], protease [17], alkaline phosphatase [18, 19], α-amylase, glucose-oxidase, peroxidase [19], polyphenoloxidase [19, 20, 21], papaine [22], lipase [19, 23], and pectin methylesterase [24, 25]. Levels of inactivation reached for these enzymes, and particularly for enzymes belonging to the same group as PG, encourage research on the effects of HIPEF on PG.

The main purpose of this work was to assess the capability of HIPEF treatments to inactivate PG, as well as to state empirical kinetic models that could describe the evolution of PG activity during HIPEF treatments at several electrical conditions.

Materials and methods

PG source and medium

The CEP Pectinex 100 L (Novo Nordisk Ferment, Neumatt, Switzerland) was used as an enzyme source. Enzyme solution (ES) that was to be HIPEF-treated was prepared by diluting the CEP in distilled water at 5% mass fraction. The conductivity of the ES at 16±1 °C was 13.17±0.01 mS/cm and pH 4.62±0.05. These measurements were carried out with a Testo-240 conductivimeter (Testo, Lenzkirch, Germany) and a Crison micropH 2000 pH-meter (Crison, Alella, Barcelona, Spain), respectively.

Assays for PG activity

PG activity in HIPEF-treated ES was determined by the reduction in viscosity that occurs when a substrate of pectin solution reacts with the enzyme. Therefore, viscosity at different stages, after ES was added to the substrate, was measured with an Ostwald-Cannon-Fenske capillary viscometer No. 100 (Afora, L´Ametlla del Vallés, Barcelona, Spain) until viscosity became steady (Fig. 1). Measurements of viscosity were done at 47.5 °C by plunging and maintaining the viscometer in a thermostatic water-bath during all viscosity assays; this was the optimum temperature for PG activity. The mixture of reactants in the viscometer contained 10 ml of warmed substrate at assay temperature and 25 μl of ES. The substrate consisted of a solution of 1% fractional mass low-metoxyle apple pectin (Pomosin pectin Type LM 22 CG; Pomosin, Grossenbrode, Germany) in distilled water. The ratio and concentration of reactants and optimal temperature resulted from preliminary experimental assays.

Fig. 1.
figure 1

Time course of viscosity in a pectin solution mixed with untreated control and high-intensity pulsed electric field (HIPEF)-treated (400 pulses of 40 μs-1 at 10.09 kV/cm) polygalacturonase in an aqueous solution of a commercial enzyme preparation. Plotted lines correspond to slopes at zero time whereas η0 and η∞ are the values for viscosity at initial and steady time, respectively

Full size image

Viscosity data for each enzyme assay were fitted to Eq. 1 [26]:

$$\eta _t = \eta _\infty + \left( {\eta _\infty - {\rm \eta }_0 } \right) \cdot e^{ - \beta \cdot t} $$
(1)

where ηt is the viscosity (millipascal seconds) at time t (minutes), η∞ is the asymptotical value of viscosity when the hydrolysis extent of pectin by PG makes the contribution of pectin to viscosity almost negligible, and η0 is the initial value of viscosity (mPa.s) and β is a constant (minutes-1). Both η∞ and β were estimated as parameters of the non-linear regression used to fit the experimental data to Eq. 1. PG activity was calculated from the absolute value of the slope of the viscosity course curves at zero time \( \left( {{\rm \dot \eta }} \right) \), which is given by Eq. 2.

$$\dot \eta \left. { = {{ - {\rm d}\eta _t } \over {{\rm dt}}}} \right|_{{\rm t} = {\rm 0}} = \beta \cdot \left( {\eta _\infty - {\rm \eta }_{\rm 0} } \right)$$
(2)

One unit of PG activity was defined as a decrease of 1 mPa.s on viscosity per minute and microliter ES at the conditions of the assay. Measurements of PG activity were carried out immediately following each HIPEF treatment.

HIPEF treatments

Exponential decay pulses generated by a Bio-Rad Gene Pulser II (Bio-Rad, Hercules, Calif., USA) were discharged to aliquots of ES in a batch processing. Gene Pulser cuvettes (Bio-Rad) were used as treatment chambers. The electrical field distribution obtained across these chambers is parallel plate geometry. The conditions under which HIPEF treatments on ES were carried out are shown in Table 1. Pulses of 40 and 160 μs pulse width were obtained working at 3 and 10 μF capacitance of capacitor, respectively. The electric field intensity (E, kilovolts/centimetre) applied to ES at 4 °C initially was procured by setting the voltage selector of the pulse generator (1.8, 2.0, or 2.4 kV) in combination with the electrode gap of the cuvettes (0.1, 0.2, or 0.4 cm). Set voltages differed slightly from actual peak voltages of pulses, which were monitored and reported by the pulse equipment on its display after each individual pulse during HIPEF treatments. To take into account these deviations, possibly due to unavoidable induced changes on the electrical properties of ES by small temperature shifts, the displayed U 0-values of individual pulses were annotated and then averaged separately for each HIPEF treatment. Thus, actual E for a given HIPEF treatment was calculated as the quotient between its mean U 0 and the electrode gap of the cuvette used during the treatment. Cuvettes were always filled with the necessary volume of ES to wholly cover their aluminium electrodes (80, 375, or 850 μl, depending on the cuvette gap). The number of pulses discharged to samples (N) ranged up to 400. Polarity of pulses was reversed manually by plugging the cuvette wires into the inverse poles of the pulse generator after each pulse to operate in bipolar mode. Each HIPEF treatment, which meant a combination of capacitance of capacitor, electric field intensity and, N pulse was replicated five times.

Table 1. Experimental electric conditions of pulse treatments applied to polygalacturonase in a commercial enzyme preparation. (Treatment temperature always below 25 °C)
Full size table

Preliminary studies were conducted to determine the temperature history of ES at 4 °C initial temperature within the cuvettes and, as a result, to fix the number of pulses so that temperature never exceeded 25 °C maximum during HIPEF treatments. Even for the electrical conditions that caused the highest heating of ES, a sequence of five pulses followed by submerging cuvettes in a cold-bath (melted ice) for 1 min was proved to guarantee not surpassing the maximum temperature. No pH changes were observed in ES after HIPEF treatments.

For exponential decay pulses, the energy density of each pulse (q, joules/metre3) was approximated by Eq. 3 [27]:

$$ q = {{U_0^2 \cdot C} \over {2 \cdot v}}$$
(3)

where U 0 is the peak voltage (volts), C the capacitance of the capacitor (farads), and v the volume of ES in the HIPEF treatment chamber (metres3). Thus, total density energy input (Q) was calculated, as indicated in Eq. 4.

$$Q = q \cdot N$$
(4)

Untreated ES control

Untreated ES remained at 4 °C during HIPEF treatments. Afterwards, its PG activity was determined. The initial PG activity in ES was the mean of five non-pulsed samples of ES. The percentage of residual PG activity (RA) in HIPEF-treated ES was calculated using Eq. 5.

$$RA = 100{{A_t } \over {A_0 }}$$
(5)

where A t is the PG activity in samples after t milliseconds HIPEF treatment time (t=N∙τ) and A 0 is the PG activity in untreated samples.

Modelling of enzyme inhibition

To describe kinetically the inactivation of PG in ES by HIPEF treatments, experimental data of RA were fitted to the exponential decay model (Eq. 6). This model has already been assayed by Giner and others [20, 21] for the same purpose, when studying other enzymes.

$$RA = RA_0 \cdot exp\left( { - k_1 \cdot t} \right)$$
(6)

where RA is the residual PG activity (percent) after a HIPEF treatment time t microseconds long, RA 0 is the RA of untreated samples, and k 1 (microseconds-1) is the first-order kinetic constant. k 1 can be stated [27] as a function of applied electric field intensity (E) by Eq. 7:

$$k_1 = k_{01} \cdot exp\left( {\omega \cdot E} \right)$$
(7)

where k 01 and ω are constants that require calculation and are expressed in the same units as k 1 and in centimetres/kilovolt, respectively.

A similar relationship (Eq. 8) has been used elsewhere [20, 21, 23, 24] to correlate RA as a function of the energy density, which is delivered by the pulse generator during HIPEF treatment:

$$RA = RA_0 \cdot \exp \left( { - k \cdot Q} \right)$$
(8)

RA and RA 0 have the same meaning as that used in Eq. 4 and factor k is a constant whereby units are expressed in metres3/gigajoule.

Statistical analysis

Data were fitted to Eq. 1 using the non-linear procedure of the Statgraphics 7.0 package (Statistical Graphics, Rockville, Md., USA, 1993). Its simple regression procedure was used for Eq. 6, Eq. 7, and Eq. 8. The same statistical package was also used to construct confidence intervals around estimated parameters and to analyse variance at P= 0.05. Confidence intervals for estimated parameters were obtained by multiplying their respective standard errors by Student-t adjusted at degree of freedom.

Results and discussion

Inactivation of PG activity by HIPEF treatments

The effects of HIPEF treatments using pulses of 40 and 160 μs on PG activity in ES at the assayed electric field intensities are shown in Fig. 2 and Fig. 3, respectively. HIPEF treatments by themselves were effective in reducing PG activity in ES because cooling steps were included between the successive steps of discharge of pulses through the ES to ensure that heating caused by the pulses was never responsible for the observed enzyme depletion.

Fig. 2.
figure 2

Residual activity (RA) of polygalacturonase in aqueous solution of a commercial enzyme preparation after exposure to electric pulses of 40 μs-1. Plotted lines correspond to adjustments of experimental data to a first order kinetic model. (RA mean±standard deviation of five replicates)

Full size image
Fig. 3.
figure 3

Residual activity (RA) of polygalacturonase in aqueous solution of a commercial enzyme preparation after exposure to electric pulses of 160 μs-1. Plotted lines correspond to adjustments of experimental data to a first order kinetic model. (RA mean±standard deviation of five replicates)

Full size image

The most significant trends to note for both pulse widths were that HIPEF treatments caused significant inactivation (P=0.05) of PG activity in ES, and that higher inactivation resulted when electric field intensity, number of pulses, or treatment time increased. These effects have been also found for plasmin [16], apple and pear polyphenoloxidase [20], peach polyphenoloxidase [21], pectin methylesterase [24], and microbial lipase [23].

The minimum RA achieved in ES after HIPEF treatments was 2%, which means a 98% inactivation of initial PG activity occurred in ES. This degree of inactivation of the enzyme was attained when the ES endured a 32 ms long HIPEF treatment constituted by 200 pulses of 160 μs at 10.28 kV/cm. Such a high degree of inactivation of an enzyme subjected to pulsed electric fields had never been recorded previously. The nearest result had been reported by Giner and others [20] who inactivated polyphenoloxidase extracted from apple up to 3.15% RA at 24.6 kV/cm, after 300 pulses of 20 μs generated by the same pulse system used in the present study.

Effect of pulse width

When comparing the RA after an identical number of pulses at matching electric field intensities, the inactivation of the PG enzyme was found to be higher for pulses of 160 μs than for pulses of 40 μs. Thus, for instance, 72% RA was achieved after 200 pulses of 40 μs at 5.18 kV/cm electric field intensity, whereas the same number of pulses but with 160 μs pulse width at 5.18 kV/cm led to 41% RA. Greater inactivation due to longer pulse width was also found for apple and pear polyphenoloxidase [20], peach polyphenoloxidase [21], and tomato pectin methylesterase [24], also subjected to exponential decay pulses.

Effect of electric field intensity

For each electric field intensity and pulse width, the experimental values of RA as a function of HIPEF treatment time were adjusted to the exponentially decay model (Eq. 6). The computed rate constants (k 1) and regression parameters of the fitted model at the tested conditions are given in Table 2. The adjusted R-squared statistic indicated that the model as fitted explained between 88.1% and 98.6% of the variability in RA. In addition, analysis of variance (P=0.05) indicates that there was a statistically significant relationship between RA and HIPEF treatment time. Therefore, the exponential decay model accurately described the relationship between the observed depletion of RA and the total HIPEF treatment time for pulses of exponentially decaying shapes at fixed electric field intensity.

Table 2. Kinetic rate constants (k 1) for inactivation of polygalacturonase activity in a commercial enzyme preparation by pulsed electric fields at different electric conditions. (Treatment temperature always below 25 °C)
Full size table

Within the assayed conditions, the values of the kinetic rate constant (k 1) for PG in ES ranged from 32 to 590 μs-1. Giner and others found that the kinetics of inactivation by HIPEF for polyphenoloxidase in apple, pear [20], and peach [21] could also be successfully described by means of the first order kinetic model and reported k 1 values for the enzymes they studied. They found that k 1 ranged from 24 to 173 μs-1 for pear polyphenoloxidase when enzyme extracts were exposed to pulses within a working electric field of 5.51–22.3 kV/cm; for polyphenoloxidase extracted from apple submitted at pulses of 5.52–24.6 kV/cm, k 1 ranged from 132 to 440 μs-1; finally, for polyphenoloxidase extracted from peach, k 1 ranged from 8.7 to 234 μs-1 at electric field intensities of 2.18–24.30 kV/cm. Thus, values for the kinetic rate constant of PG in ES under assayed HIPEF treatments indicated reliable effectiveness on inactivating the enzyme.

The effect of electric field intensity on k 1 is illustrated by Fig. 4. It was observed that electric field intensity exerted a positive and strong effect on k 1. In general, the influence of E on k 1 for both pulse widths was statistically significant (P=0.05), although rate constants did not differ significantly for close values of the tested electric field intensities. To describe the relationship between k 1 and electric field intensity, the obtained values of k 1 were fitted to Eq. 7 for pulses of 40 μs. The fitted model accurately explained the variability in k 1 (R 2= 0.943) and the estimates of the parameters in the model (k 01=14±4 μs-1; ω=0.21±0.09 cm/kV) were significant at P= 0.05. This dependence of k 1 on electric field intensity was in agreement with the relationship found in previously cited studies on fruit polyphenoloxidases [20, 21]. The magnitude of the exponential factor in Eq. 7 indicates how important the effect of electric field intensity is on the kinetic rate constant, and that a lineal increase in the electric field will lead to an exponential increase in the kinetic constant. The ω-exponential factor of PG was comparable to those of polyphenoloxidase from fruits. It was close to the reported values for pear (0.153±0.024 cm/kV) and peach (0.10±0.09 cm/kV) polyphenoloxidase but significantly greater than the ω-value for apple polyphenoloxidase (0.067±0.010 cm/kV).

Fig. 4.
figure 4

Effect of electric field intensity (E) on the rate constant (k 1) in a first kinetic model used to describe the inactivation of polygalacturonase in aqueous solution of a commercial enzyme preparation. Plotted line results from fitting the points to an exponential model. (k 1 value±confidence interval, P= 0.05)

Full size image

In other works [23, 24] relating the effect of HIPEF treatments on enzyme inactivation using first order kinetics, but in which the number of pulses was considered the independent variable instead of treatment time, it was observed that electric field intensity exerted an analogous influence on first order kinetic rate constants.

Relationship between RA and Q

Input of total energy density supplied to ES during each HIPEF treatment was computed from Eq. 4. An input of 22.56 GJ/m3 led to the maximum enzyme activity reduction (98%), whereas 0.37 GJ/m3 to the minimal inactivation of PG (2%). A plot of the experimental RA of PG activity in HIPEF-treated ES against its respective Q is shown in Fig. 5, revealing an exponential depletion of RA when Q increased. The results of fitting Eq. 8 to describe the effect of Q on RA indicated a significant relationship between them both (R 2=0.936) and that the estimated parameters of the model were statistically significant at P=0.05 as well. Thus, RA could be given as a function of Q, as shown in Eq. 9.

$$RA = \left( {100 \pm 3} \right) \cdot \exp \left\langle {\left( { - 0.196 \pm 0.015} \right) \cdot Q)} \right\rangle $$
(9)

This relationship between RA of PG activity and Q agreed with one reported in earlier studies on other enzymes in a variety of media under HIPEF treatments carried out with the same equipment at similar conditions. Giner and co-workers had already described the dependency of RA on Q by using Eq. 8 for tomato pectin methylesterase [24], apple and pear polyphenoloxidase [20], and peach polyphenoloxidase [21]. On the other hand, Bendicho and others [23], who performed experiments with different HIPEF devices working either in batch or continuous mode, reported the same kind of relationship for microbial lipase in simulated milk ultrafiltrate. The value of the exponential factor (k) in Eq. 8 for the mentioned enzymes ranged from 18.8E−3 to 1900E-3 m3/GJ, obtained for peach polyphenoloxidase and microbial lipase, respectively. The value of k for PG enzyme in ES (196E−3 m3/GJ) was comparable in magnitude to k for microbial lipase treated in the continuous flow HIPEF system used by Bendicho and others [23]. This result indicated that PG and lipase showed almost equal sensitivity towards HIPEF treatments. Therefore, although using different HIPEF systems, a similar degree of enzyme inactivation will be achieved for both PG and lipase when supplying the same quantity of energy per volume unit to its respective medium throughout HIPEF treatments. In addition, PG and lipase were the enzymes that theoretically would require lower energy to reduce its activity up to a fixed degree in assayed medium under HIPEF treatments because of higher sensitivity to this type of processing.

Fig. 5.
figure 5

Residual polygalacturonase activity (RA) in an aqueous solution of a commercial enzyme preparation subjected at different inputs of electric energy densities supplied by HIPEF treatments. Electric field intensities ranged from 5.18 to 19.4 kV/cm. Number of pulses (40 and 160 μs-1 pulse width) ranged up to 400. Plotted line corresponds to adjustment of experimental data to an exponentially decay model. (RA mean±standard deviation of five replicates)

Full size image

To conclude, HIPEF treatments can be used to lessen PG activity considerably in commercial enzyme preparations.The degree of pursued inactivation of the enzyme can be regulated by selecting and controlling adequate electrical conditions. The inactivation of the enzyme as a function of HIPEF treatment time is well described by an exponential decay model in which kinetic rate constants were between 32 and 590 μs-1. The same kind of mathematical model matches depletion of the enzyme at different input electric energy densities.