Original Paper
Changes in electrical properties and void distribution of mung bean sprouts during hot water heating
2021 Volume 27 Issue 2 Pages 311-318
Details
2021 Volume 27 Issue 2 Pages 311-318
We investigated the relationships between electrical properties and tissue structures, including the cell membrane, and the voids of fresh and heated bean sprouts. The electrical properties, which were obtained by electrical impedance spectroscopy and equivalent circuit analysis, changed drastically within 60 s by heating at 70 °C. Confocal laser-scanning microscopy suggested that these changes are due to cell membrane damage. We also visualized void distribution using micro X-ray computed tomography imaging and the calculated porosity. The porosity decreased from 6.71% to less than 1% after 30 s of heating. The fresh samples showed several axially elongated voids with low sphericity (< 0.3) and high volume (> 105 µm3), which then disappeared preferentially with heating. Moreover, extracellular resistance was highly correlated with porosity (R = 0.900). We suggest that the contribution of cell membrane and void conditions to the texture of bean sprouts for short heating times needs to be considered.
Bean sprouts have been consumed for hundreds of years and they are best known for their unique crispiness (Kimura et al., 1980). In recent decades, cultivation of bean sprouts has been conducted primarily in factories with an increase in production from 229 000 tons in 1974 to 512 000 tons in 2018i). Since bean sprouts can be grown in closed rooms with minimal cultivation, they have relatively low susceptibility to the effects of climate change and offer an inexpensive and stable supply for the market. In Japan, mung bean sprouts are immensely popular as a topping for ramen noodles and are also widely used in salads and other ready-to-eat foods. Different types of vegetables have various tissue structures, which confer specific textures. Therefore, to establish suitable processing conditions that do not disrupt the texture of bean sprouts, it is essential to understand structural properties such as the distribution of cells and voids, as well as processing tolerance.
The main plant tissues that affect processing or cooking in vegetables are the cell membrane, voids, and cell wall/middle lamella. Pectin, which is a component of the cell wall and middle lamella (Imaizumi et al., 2017a), has been widely investigated (Imaizumi et al., 2019; Tijskens et al., 1997; Van Buggenhout et al., 2009). Disruption of the cell membrane and loss of voids are drastic at the very early stages of heating and are considered to be the main factors that contribute significantly to the texture of several types of vegetables, including bean sprouts, that are cooked for short periods of time.
Electrical impedance spectroscopy (EIS) has generally been used for assessing the retention of the lipid bilayer of cell membranes in fruits and vegetables (Imaizumi et al., 2018; Watanabe et al., 2017; Wu et al., 2008). Recently, Ando et al. (2014) proposed a method to derive parameters for the cell membrane by equivalent circuit analysis based on a model using a constant phase element (CPE), which is now widely employed. Using a similar method, Imaizumi et al. (2015) showed that changes in cell membrane capacitance, intracellular resistance, and extracellular resistance of potato tissue were related to heating, demonstrating a useful method for determining the occurrence of specific changes in the plasma membrane during processing. In addition, visualization of the void distribution by X-ray micro-computed tomography (CT) imaging has become widespread and has been reported for various foods, including granola bars (Kelkar et al., 2015) and apples (Janssen et al., 2020).
In this study, we used EIS and X-ray micro-CT to quantify the tissue structure parameters for bean sprouts. We also assessed the correlation between electrical characteristics and porosity.
Sample preparation Mung bean sprouts [Vigna radiata (L.) R. Wilczek] purchased from a local market in Gifu City, Japan, were used for the experiment. The top 10 mm and the bottom 5 mm of the mung bean sprouts were removed, and 15 mm samples were submerged in 200 mL of distilled water in a water bath (TB-2NC, As-One) at 70 °C for periods of 15, 30, or 60 s. Following heat treatment, the samples were cooled in water (approximately 20 °C) for 3 min to prevent overheating.
Electrical impedance spectroscopy (EIS) EIS was conducted using an LCR meter (IM3536, HIOKI). A pair of needle electrodes connected to the device was inserted into the sample at intervals of 10 mm. The impedance (|Z|), resistance (R), and reactance (X) of the sample were measured at 200 points over a frequency range of 50 Hz to 5 MHz at a measurement voltage of 1 V and recorded by a computer. The measurements were replicated seven times for individual samples, which were selected randomly.
Equivalent circuit analysis was performed to quantitatively evaluate the Cole-Cole plots obtained by the impedance measurements. The simplified Hayden model, which assumes that plant tissue is a homogeneous collection of cells, has been widely used for equivalent circuit analysis of the cell membrane (Wu et al., 2008; Juansah et al., 2012). However, this model has limitations of low fitting accuracy because actual cell structures are not homogeneous and the dielectric relaxation phenomenon varies among individual cells. Therefore, a model using a CPE (see Fig. 1) was developed to overcome this problem, and the impedance of the CPE was calculated using the following equation:
 |
Equivalent circuit model of biological tissues: constant phase element (CPE) model.
where j is the imaginary unit, ω is the angular frequency, and T and p are constants (0 ≤ p ≤ 1). The impedance of the CPE model can be derived from the following equation:
 |
where Re is the extracellular (intercellular) resistance and Ri is the intracellular resistance. The parameters Re, Ri, T, and p were calculated by complex non-linear least-squares curve fitting (Macdonald, 1992). In addition, the cell membrane capacitance (Cm) was defined using Eq. (3) (Ando et al., 2014):
 |
In this study, we conducted an equivalent circuit analysis using the CPE model in accordance with previous studies (Ando et al., 2014; Imaizumi et al., 2015) because we obtained better fitting results (R2 > 0.99) with the CPE model in a preliminary experiment.
Confocal laser-scanning microscopy (CLSM) To visualize the cell membrane condition of the samples, a confocal laser-scanning microscope (LSM710, Carl Zeiss) was used in accordance with the methods described by Imaizumi et al. (2015). A 500-µm-thick section was obtained from the sample and then stained with 1,1′-dioctadecyl-3,3,3′,3′-tetramethyl-indocarbocyanine perchlorate (DiI). DiI has been widely used to stain cell membranes as it penetrates membrane lipids and emits strong fluorescence (Oka and Tanishita, 1995; Haidekker et al., 2002). The stained membranes were excited by an HeNe laser (543 nm).
X-ray micro-CT X-ray micro-CT observations were conducted in accordance with the methods of Kuroki et al. (2004) with a slight modification. The sprout sample was inserted into a sample holder and wrapped in polyvinylidene chloride film. The sample was then placed in a Skyscan 1172 CT scanner (Bruker). Scanning was conducted with a tube voltage of 50 kV and a tube current of 150 µA. The rotation step was 0.675°. These operating conditions resulted in a cross-sectional pixel size (spatial resolution) of 2.43 µm. NRecon 1.6.10.4 software (Bruker) was used to reconstruct two-dimensional (2D) images of the samples from the X-ray-transmitted images. The resulting 2D images were encoded with 8-bit precision corresponding to a grayscale level of 0–255.
Ten images at 243-µm intervals were selected and used to determine the porosity with ImageJ image analysis software. Porosity was defined as the percentage of the area of the sprout tissue in the image that could be considered to contain voids. The resulting images indicated two peaks in the histogram of the grayscale values, and the lower peak (less than GS136) was considered to correspond to voids. The porosity was calculated for each of the 10 images, and the average value was taken as the porosity of the sample. Kuroki et al. (2004) set the region at a threshold of 4 × 4 × 4 pixels or more to provide sufficient information in three dimensions to identify voids, and the region below this limit was considered to be noise. Therefore, in this study, a region of 16 pixels or more was set as the threshold for identification of voids. The size distribution and circularity of the voids were analyzed by three-dimensional reconstruction of CT images using VGstudio MAX software version 1.1 (Volume Graphics GmbH, Germany).
Statistical analysis Individual results were compared using the Tukey-Kramer method with R i386 3.5.1 (R Development Core Team, New Zealand).
Changes in electrical properties of mung bean sprout during heating Fig. 2A shows the changes in the Cole-Cole plot for each heating period. A typical arc shape was observed in the unheated sample, and the arc became smaller as the heating period increased. This arc reduction was previously suggested to be caused by a decrease in membrane integrity (Watanabe et al., 2017). Fig. 2B–F show the change in the relevant circuit parameters Ri, Re, Cm, p, and T during heating. Re was significantly reduced by heating. Ri tended to increase and Cm tended to decrease as the heating period increased.
Changes in the Cole–Cole plot (A) and equivalent circuit parameters (B–F) of bean sprout tissue during heating at 70 °C. Bars: SE (n = 7); different lowercase letters indicate a significant difference (p < 0.05).
The impedance (|Z|) depends on the concentration of ions in the solution (Lew, 1996). In addition, the equivalent circuit parameters, Ri and Re, are affected by the ion content of the intracellular and extracellular fluids (Imaizumi et al., 2015). Disruption of the cell membrane structure due to a phase transition of the lipid bilayer at high temperature is considered to cause leakage of electrolytes to the extracellular fluid in the cytoplasm, leading to a decrease in Re and an increase in Ri. By contrast, the cell membrane is mainly composed of a double layer of phospholipids and is therefore considered to be an insulator. The cell membrane, which separates the extracellular and intracellular fluids, acts as a capacitor (Cm) (Imaizumi et al., 2015). Thus, the decrease in Cm was attributed to cell membrane damage caused by heating at 70 °C. Fig. 3 shows CLSM images of the heated samples. Imaizumi et al. (2015) observed cell membrane fragments in CLSM images of DiI-stained potato tissue and assumed that membrane damage had occurred. In addition, heating at temperatures higher than 60 °C leads to loss of turgor pressure, resulting in tissue shrinkage (Kamat et al., 2018; Imaizumi et al., 2017). In this study, there were no obvious fragments detected by CLSM; however, for heating periods of 30 and 60 s, whole tissue distortions were apparent. This was considered to be due to the loss of turgor pressure resulting from membrane damage, which reduced the rigidity of the tissue.
Confocal laser-scanning microscopy images of the cell membrane of bean sprouts stained by 0.05% DiI at different time points: A, 0 s; B, 15 s; C, 30 s; and D, 60 s. Arrows: Damaged cell membranes.
Changes in void states of mung bean sprout during heating Figs. 4A–D show the reconstructed cross-sectional images binarized at GL136 obtained by X-ray micro-CT. The black area in each image represents air space, and the white area represents the solid or liquid cellular region. Small voids were dispersed in the unheated tissues, but there were few obvious voids visible in the heated tissues. Fig. 5 shows the porosity at each heating period. For the unheated samples, the porosity was 6.71 ± 0.93%, which decreased to less than 1% by heating for more than 30 s. Compared to the unheated samples, the heated samples showed a significant reduction in void area. Based on the reduced membrane capacity and tissue distortions described above, we assumed that membrane damage occurred due to a lipid bilayer phase transition at high temperature. The reduction in tissue rigidity would have reduced air retention.
X-ray micro-computed tomography images (A–D) of bean sprouts and threshold images (a–d) heated at 70 °C for (A, a) 0 s, (B, b) 15 s, (C, c) 30 s, and (D, d) 60 s.
F: wrapping film, SH: sample holder.
Changes in the closed porosity of bean sprout tissue during heating. Bars: SE (n = 7); different lowercase letters indicate a significant difference (p < 0.05).
Fig. 6 shows plots of sphericity and volume obtained by three-dimensional analysis using VGstudio. Sphericity was calculated by measuring the ratio between the surface area of a sphere with the same volume as the void and the actual surface area of the void. Unheated samples showed high porosity, and there were many long vertical voids, which are considered to be the intercellular voids of plants that are used for gas exchange during growth (Yamauchi et al., 2018; Kuroki et al., 2004). Additionally, the relationship between the sphericity and volume of unheated samples was highly correlated (R = −0.897). This suggests that the formation of voids in the sprouts proceeded mainly in one direction with minimal spread in the direction perpendicular to the axis. After heat treatment, the intracellular fluid flowed out of the cells due to membrane damage. This intracellular fluid flowed into the voids because the cell walls, which are the main part of the cell matrix in plants, can be permeated by fluids from both inside and outside the cell (Andou et al., 2006). Upon heat treatment, the elongated voids were immediately fragmented or had disappeared, leaving 103 to 105 voids. These relatively small voids are highly isolated in nature and are likely to have little access to the outside of the tissue. These figures show an increase in sphericity and a decrease in porosity with larger volumes in response to the decrease in porosity with heating.
Void-shape plots of sphericity vs. volume in bean sprout samples heated for (A) 0 s, (B) 15 s, (C) 30 s, and (D) 60 s.
Relationships between electrical properties and porosity The correlation coefficient between the porosity and Ri, Re, and Cm calculated by CT analysis was 0.238, 0.900, and 0.605, respectively, with a particularly strong correlation for extracellular resistance, Re. Fig. 7 shows the relationship between porosity and extracellular resistance for the fresh and heated samples. Extracellular resistance tended to decrease with decreasing porosity not only due to electrolyte leakage from membrane damage but also because of reduction of intercellular voids caused by leakage and shrinkage of the entire tissue, which reduces the volume of insulating air in the intercellular environment. Conditions such as hydrothermal treatment at 70 °C for several tens of seconds have previously been used for microbial removal treatment (Phua et al., 2014), and our CLSM observations suggest that such conditions can cause significant changes in tissue structures. Similarly, significant changes in porosity were observed in the region for heating for periods up to 30 s. This was consistent with previous experimental results obtained for sweet potatoes (Imaizumi et al., 2017b), suggesting that it is possible to estimate changes in porosity based on electrical measurements.
Relationship between closed porosity and extracellular resistance (Re).
The electrical properties of bean sprouts changed significantly and cell membranes were readily damaged after a short period of heating for less than 60 s. The shape of the voids in the tissue could be quantitatively characterized by X-ray CT, which provides important information for the evaluation of voids. These changes are thought to be responsible for tissue weakening in bean sprouts, although changes in the cell wall and other factors should also be considered. Electrical properties and CT imaging were shown to be useful for representing some aspects of tissue weakening during the cooking and processing of bean sprouts. Future studies are needed to evaluate changes in mechanical properties, pectin, cellulose, and water content to obtain a more comprehensive assessment. Our study also proposes a link to voids as a factor that alters extracellular resistance. These findings can contribute to the development of quality assessment techniques using electrical properties.
