Introduction

There has been increasing interest in the consumption of fresh fruits and vegetables in the USA during the past decade (Cisneros-Zevallos 2003; Cook 2014). Regular intake of fruits and vegetables, which provide vitamins, minerals, and phytochemicals, was linked to a reduced incidence of cancer and other degenerative diseases (Heber 2004). As the largest category of phytochemicals in plant foods, phenolic compounds are highly associated with the antioxidant capacity of fruits and vegetables (Liu 2003). The health benefits of phenolic compounds are mainly attributed to their ability to relieve oxidative stress in human cells. Phenolic compounds are classified as plant secondary metabolites. Plants will produce and accumulate phenolic compounds as defense responses when they are exposed to biotic and abiotic elicitors, which may result in improved nutritional quality of the plants (Bartwal et al. 2012). Meanwhile, fresh produce items are perishable during distribution and storage due to growth of quality degradation organisms and losses in sensory and nutritional quality (Beuchat 1998). Processing technologies, such as treatment with electrolyzed water (Gil et al. 2015), UV light (Yaun et al. 2004), ozone (Chitravathi et al. 2015), and cold plasma (Misra et al. 2014) have been proposed and tested for use in the produce industry to enhance the safety and quality of fresh produce.

High-intensity ultrasound or power ultrasound with frequency ranging from 20 to 100 kHz is among the emerging technologies applicable in food processing and preservation (Kentish and Feng 2014). Ultrasonication has been tested to enhance such unit operations as extraction, emulsification, cutting, crystallization, heat and mass transfer, surface decontamination, and microbial and enzyme inactivation (Raviyan et al. 2005; Baumann et al. 2009; Patist and Bates 2011; Zhou et al. 2012). Recovery of bioactive compounds with ultrasound is another promising application (Roselló-Soto et al. 2015). Generally, ultrasound treatments at high acoustic power density (APD) find applications where, for instance, cell destruction (Lee et al. 2013; Bevilacqua et al. 2014; Zinoviadou et al. 2015) is required, while treatments at low APD have been demonstrated to stimulate cell growth and increase enzyme activities (Barton et al. 1996; Pitt and Ross 2014).

The mode of action of a power ultrasound treatment is often attributed to acoustic cavitation, i.e., the generation, growth, and implosion of tiny bubbles when ultrasound travels through a liquid (Kentish and Ashokkumar 2011). The implosion of cavitating bubbles results in generation of a spectrum of chemical and physical events in the vicinity of the bubbles, including localized high temperature and high pressure, shear forces, shock waves, water jets, and free radicals (Leighton 1994). These physical and chemical effects are utilized to perform different food processing and preservation processes. The application of ultrasound for preservation of fresh produce is a relatively new endeavor (Xu et al. 2013). Among the work reported in the literature, most of them focused on the use of ultrasound to ensure microbial safety of produce. When used alone or combined with sanitizers, ultrasonication enhanced the destruction and/or removal of bacteria, molds, and yeasts on the surfaces of produce such as apples (Wang et al. 2009), cantaloupes (Cao et al. 2010a), strawberries (Wang et al. 2007; Cao et al. 2010b), plum fruits (Chen and Zhu 2011), lettuce (Seymour et al. 2002; Salgado et al. 2014), and spinach (Zhou et al. 2009, 2012). Some studies also reported that ultrasound treatment was effective maintaining the sensory and nutritional quality (Chen and Zhu 2011; Alexandre et al. 2012). Meanwhile, other reports documented an increase of total phenolics and antioxidant capacity in sonicated produce samples (Rudolf and Resurreccion 2005; Potrebko and Resurreccion 2009; Sale and Resurreccion 2010). However, no effort so far has been made to study the physiological mechanism that leads to an increase in total phenolics and antioxidant capacity.

On the other hand, the elicitation effect of ultrasond treatment has been well investigated in biotechnology research. Ultrasound as a form of physical energy has various physical, chemical, and biological effects (Kentish and Feng 2014). Several studies using algal, plant cell, and animal cell cultures confirmed the defense response triggered by low APD ultrasonication. Wu and Lin (2002a, b) reported that ultrasound treatment triggered cross-membrane ion fluxes, production of reactive oxygen species (ROS), and a rapid increase of phenylalanine ammonia-lyase (PAL) activity, followed by increased production of polyphenols (PP) and phenolic compounds in a Panax ginseng cell suspension. Wang et al. (2006) discovered another signaling molecule of the defense response pathway, nitric oxide (NO), in a Taxus cell culture after ultrasonication. Induced production of phenolic secondary metabolites was also reported in Morinda citrigolia cell cultures (Komaraiah et al. 2005), hazelnut cell cultures (Rezaei et al. 2011), and Vitis vinifera cell cultures (Santamaria et al. 2012).

We hypothesized that low APD ultrasonication could act as an abiotic elicitor for whole plants, triggering the defense response system and stimulating production of secondary metabolites, which would enhance antioxidant capacity and maintain postharvest quality of fresh produce. In the present study, we examined the effect of ultrasonication on overall quality, phenolic production, and antioxidant capacity of fresh Romaine lettuce right after treatment and during storage. The mechanism of the production and accumulation of secondary metabolites was investigated by examining the responses of a defense-related enzyme in ultrasound-treated samples.

Materials and Methods

Plant Material

Romaine lettuce (Lactuca sativa, var. longifolia) was purchased from a local supermarket (Champaign, IL) and stored in cold room at 1 °C (34 °F). Samples were pre-selected based on size, color, and visual quality to ensure the initial consistency. Before processing, the outer three leaves, as well as any other leaves with visible damage were discarded. The next four or eight undamaged leaves were carefully removed, and the bottom part of the stem was cut with a sterile sharp stainless steel knife to make each leaf a 10-g sample. They were then randomly divided into different treatment groups for at least three replications.

Ultrasound Treatment and Sample Preparation

A specially designed ultrasound treatment unit was utilized in this study. A water tank with a dimension of 400 × 660 × 460 mm was used as a treatment chamber. Two transducer boxes (400 × 400 × 50 mm each) were vertically placed in the tank face to face, with a space of 280 mm between the boxes. The pair of transducer boxes was at the frequency of 25 kHz and 2 kW nominal power. Unlike traditional ultrasound washing baths or probe units that are with non-uniform sound intensity distribution in the treatment chamber (Ando and Kagawa 1989; Klíma et al. 2006), the acoustic field distribution of the ultrasound treatment chamber utilized here was nearly uniform, as determined by an aluminum foil test (Zhou et al. 2012). Uniform exposure of samples to ultrasonication was important to assure that samples placed at different locations of the tank received the same ultrasonic energy. A single layer of Romaine lettuce leaves was placed with no overlapping in a sample holder (400 × 400 × 30 mm), made by a wooden frame and two pieces of stretchable polyethylene mesh to fix samples without blocking sound waves. The ultrasound treatment set up is shown in Fig. 1.

Fig. 1
figure 1

Ultrasound treatment set up. a Generator, b water tank, c transducer box, and d sample holder

Full size image

Prior to an ultrasound treatment, the water tank was filled with 44.5 L room temperature tap water. In each test run, Romaine lettuce samples (~60 g, 6 leaves) were properly fixed in the holder that was submerged in the water tank parallel to the transducer boxes (Fig. 1). Ultrasound at 100 % power output from a generator with an APD of 26 W/L was applied to the samples in the water for 1, 2, or 3 min in each treatment. The acoustic energy delivered into the water and samples was 69.4, 138.8, and 208.3 kJ for treatments at 1, 2, and 3 min, respectively. Samples washed with tap water were used as the control. The APD was measured with the calorimetric method described by Baumann et al. (2005).

After sonication, fresh-cut Romaine lettuce samples were used for sensory evaluation, and whole leaf Romaine lettuce was used for all other analyses to eliminate the influence of wounding by cutting (Reyes et al. 2007). Treated whole leaf lettuce samples were air-dried (30 min) under room temperature and stored in plastic gallon zipper storage bags (GLAD, Oakland, CA). Fresh-cut lettuce samples were prepared by carefully cutting into square pieces (25.4 × 25.4 mm) and dried with a salad spinner for 1 min (OXO, New York, NY) before being transferred into plastic zipper storage bags. All samples were sealed in bags, covered with thick brown paper to prevent light exposure, and stored at room temperature for up to 150 h. Room temperature was chosen in this study to speed up the metabolite production in lettuce. Each group had at least three replicates.

Total Phenolics

Whole leaf Romaine lettuce samples (10 g) taken 0, 30, 60, and 90 h after storage were homogenized in 50 mL methanol at high speed for 1 min in a kitchen blender. The homogenate was transferred into a 250-mL covered plastic bottle and incubated at 40 °C for 1 h. After incubation, the homogenate was filtered under a vacuum (Whatman no. 1 filter paper) and the filtrate was collected as the sample extract and stored at −18 °C for later analysis. Three sample extractions were prepared for each treatment condition. Total phenolic content was determined followed the methods described by Kang and Saltveit (2002) with slight modifications.

Absorption at 320 nm

The absorbance (320 nm) of each sample extract prepared as described above was directly read by a UV/Vis/NIR spectrophotometer (Perkin Elmer Lamda 950). Pure methanol was set as the blank.

Folin–Ciocalteu Reaction

Total phenolics were also determined by the Folin–Ciocalteu method and expressed as gallic acid equivalents (GAE). For each test, 0.2 mL sample extract was mixed with 1.0 mL of Folin–Ciocalteu reagent (0.2 M), kept at room temperature for 3 min, and followed by addition of 0.8 mL sodium carbonate solution (7.5 %, w/v). After 2 h incubation (~23 °C), the absorbance (765 nm) was recorded. Samples were compared with a standard curve prepared with garlic acid (7.81, 15.63, 31.25, 62.5, 125, 250, 500, and 1000 mg/L). Blank contained methanol in place of sample extract.

Antioxidant Capacity

The antioxidant capacity of sample extracts was measured by the free radical scavenging activity using the 1,1-diphenyl-2-picryl-hydrazyl (DPPH) method (Blois 1958) with slight modifications. Sample extract (0.4 mL) was added to 3.6 mL freshly prepared DPPH solution (0.024 g DPPH in 100 mL methanol) and shaken well. Initial absorbance (515 nm) was immediately recorded prior to room temperature incubation for 30 min, and then the final absorbance at 515 nm reading was taken. Pure methanol was set as the blank. The antioxidant capacity was calculated by the following formula:

$$ \%\mathrm{DPPH}\ \mathrm{inhibition}=\left(1-\frac{A_{\mathrm{f}}}{A_{\mathrm{i}}}\right)\times 100 $$

where A f is the final absorbance after 30 min incubation and A i is the initial absorbance.

Phenylalanine Ammonia-Lyase Activity

Sample extraction was conducted according to the procedure of Yu et al. (2012). One leaf of Romaine lettuce (10 g) was homogenized in 50 mL sodium borate buffer (20 mM, pH 8.7) containing 20 mM ß-mercaptoethanol at high speed for 1 min in a kitchen blender. The homogenate was filtered through four layers of cheesecloth and centrifuged (10,000×g, 30 min, 4 °C). The supernatant was used to determine protein content and PAL enzymatic activity. Extractions were prepared at least in triplicate for each group.

The Bio-Rad Protein Assay was used for protein content determination. The dye reagent (1:4, v/v) was freshly prepared from Dye Reagent Concentrate. Samples were compared with standard curves generated using bovine serum albumin (BSA; 0.2, 0.4, 0.6, and 0.8 mg/mL). Based on preliminary experiments, enzyme extracts were diluted (1:1, v/v) with sodium borate buffer (20 mM, pH 8.7), containing 20 mM ß-mercaptoethanol to fit into the linear range of the BSA standard curve. Diluted enzyme extract 0.1 mL or protein standard was combined with 5.0 mL of dye reagent, vortexed (5 s), and incubated at room temperature (15 min) prior to measuring absorbance (595 nm). The dye reagent (1:4, v/v) was set as the blank.

PAL activity was determined as described by Qin et al. (2003). Enzyme extract (1 mL), 2.0 mL sodium borate buffer (20 mM, pH 8.7) containing 20 mM ß-mercaptoethanol, and 0.5 mL of l-phenylalanine (20 mM) were mixed together in a test tube. After incubation at 40 °C (1 h), the enzymatic reaction was stopped by adding 0.1 mL HCl (6 M) and vortexed (5 s). To measure the difference between absorbance before and after reaction, absorbance (290 nm) was measured using 1.0 mL enzyme extract, 2.0 mL sodium borate buffer (20 mM, pH 8.7) containing 20 mM β-mercaptoethanol, 0.5 mL l-phenylalanine (20 mM), and 0.1 mL HCl (6 M), all combined to serve as blank.

Color Measurement

To represent color quality of whole leaf Romaine lettuce, leaf surface color and rib cut edge color of samples were measured at the same time, with a Minolta CR-300 Chroma Meter connected to a DP-301 Data Processor (Minolta Corporation, Ramsey, NJ). Color analysis was based on the CIELAB color space: briefly, it is a color-opponent three-dimension space with L * (lightness), a* (positive values indicate red and negative values indicate green), and b * (positive values indicate yellow and negative values indicate blue). Three readings were taken on the adaxial (upper) surface from the left, middle, and right part of each leaf, and the average was used to represent sample surface color. One color reading for cut edge was taken directly from the rib edge damaged by cutting before treatment. To better express sample color, hue angle and chroma (color saturation) were calculated (McGuire 1992):

$$ \mathrm{hue}\;\mathrm{angle}={ \tan}^{-1}\left(\raisebox{1ex}{${b}^{*}$}\!\left/ \!\raisebox{-1ex}{${a}^{*}$}\right.\right);\mathrm{chroma}=\sqrt{a^{*2}+{b}^{*2}}. $$

Texture

Texture of Romaine lettuce samples was determined by firmness measurements utilizing a TA-XT2 Texture Analyzer with a five-blade Kramer Shear Press (Texture Technologies Corporation, Scarsdale, NY). Before each test, samples were cut into 25.4 × 25.4 mm square pieces, and 25 g sample was weighed and evenly spread in the holder (internal dimension, 82 × 63 × 89 mm). The height of lettuce sample in the holder was about 55 mm, so the five blades (blade thickness, 1.5 mm each) plunger was initially set at a height of 65 mm from the bottom of sample holder, and then forced down through all samples at a test speed of 1.0 mm/s until the bottom layer of lettuce was cut broken. Data was collected by Texture Expert software, v.2.55 (Texture Technologies Corporation, Scarsdale, NY). The maximum force (N) was recorded to represent firmness (Bourne 1997). Three replicates were measured for each treatment group at each storage time.

Sensory Evaluation of Fresh-Cut Romaine Lettuce

Descriptive sensory evaluation was performed to assay the quality change of fresh-cut Romaine lettuce due to different treatments and storage times. Sensory quality parameters including overall visual quality, surface browning/discoloration, cut edge browning, sogginess/watery, and off-odor were evaluated by a trained panel of 14 individuals (six males and eight females, all students or scholars in the Department Food Science and Human Nutrition at the University of Illinois) after 0, 30, 60, 90, and 150 h storage at room temperature. Different samples were placed on separate white plates labeled with a random three-digit code, and randomly set on the booth. For overall visual quality, a scale from 1 to 9 (1 = “poor/inedible,” 9 = “excellent and no difference from the fresh reference”) was applied; a score of 5 was set as the limit of acceptance. For surface browning/discoloration and cut edge browning, a scale of 1–5 was applied (1 = “no browning”, 5 = “significant browning”, 3 was the limit of acceptance). Similar scales were used for sogginess/watery and off-odor evaluation (1 = “crispy” or “fresh odor”, 5 = “severe watery” or “severe off-odor”, 3 was the limit of acceptance).

Statistical Analysis

All treatments were replicated at least three times. Data were compiled by Microsoft Excel (Microsoft Corporation, Redmond, WA) and statistical analysis was performed using SAS software (SAS Institute, Cary, NC). At each storage time, treatments were compared by analysis of variance (ANOVA) to determine if there was at least one treatment significantly different from the others (p < 0.05). Least significant difference (Fisher’s LSD) test was conducted to further analyze the difference between each pair of treatments (p < 0.05) when ANOVA gave a significant result. ANOVA and LSD tests were also applied to the analysis of the difference of the same treatment during different storage times (p < 0.05).

Results and Discussion

Effect of Ultrasound on Total Phenolic Content

The phenolic content of the treated samples expressed in Abs320 values is shown in Fig. 2a. After room temperature storage for 60 h, Romaine lettuce treated by ultrasound for 1 min and 2 min showed significantly higher phenolic contents (35.28 and 26.74 %) than the control (3 min water wash). The samples treated with ultrasound for 3 min also exhibited a higher phenolic content than the control, though the increase was not significant. Results from the Folin–Ciocalteu method (Fig. 2b) showed the same changes in phenolic content. After 60 h, storage samples treated using ultrasound for 1, 2, and 3 min had phenolic contents of 22.50, 16.25, and 17.92 % higher than the control, respectively; however, only the 1-min exposure was statistically significant (Fig. 2b). There was no significant difference among ultrasound-treated group and the control immediately after the treatment, indicating that the generation and accumulation of phenolic compounds in response to a treatment takes time.

Fig. 2
figure 2

Total phenolic content of Romaine lettuce treated with and without ultrasound (25 kHz) determined by a absorbance at 320 nm and b Folin–Ciocalteu reaction. Means within time results with different letters (a and b) are different at α = 0.05. Means within treatments results with different letters (x–z) are different at α = 0.05

Full size image

After room temperature storage for 30 h, the samples treated with ultrasound for 1 min had significantly lower phenolic content than the control (Fig. 2a). This decrease might be attributed to consumption of phenolics to overcome the oxidative stress from ROS caused by ultrasound treatment (Adyanthaya et al. 2009). The total phenolic content in plants is determined by the interplay between factors that promote the production of phenolic compounds and those that consume them. The reduction in phenolic content thus indicated that after 30 h storage the consumption of phenolics surpassed the generation and accumulation. After 90 h storage there was no difference in phenolic content between the ultrasound-treated samples and the control. This indicates the need to study when to apply stimulation, when the phenolics production and accumulation will surpass the consumption, and if multiple stimulations at a specified time frame will produce a significant and long-lasting increase in phenolics. The Folin–Ciocalteu reaction (FCR) method is a non-specific assay, which measures not only phenolic compounds but also other oxidation substrates such as ascorbic acid and aromatic amines in the extract (Ainsworth and Gillespie 2007). The changes of the other chemicals in the samples over time may influence the FCR values, causing the difference between the FCR measurements and that from the Abs320.

Effect of Ultrasound on Antioxidant Capacity

Antioxidant capacity of whole leaf Romaine lettuce was expressed by percentage DPPH inhibition (Fig. 3), with higher percentage inhibition indicating higher antioxidant capacity. After 60 h storage, the DPPH inhibition of samples sonicated for 1, 2, and 3 min was 97.84, 75.22, and 75.87 %, significantly higher than the control, respectively. After 90 h storage, only the antioxidant capacity of samples sonicated for 2 min remained significantly higher than the control. Similar to total phenolic content, no significant difference was found in antioxidant capacity immediately after sonication.

Fig. 3
figure 3

Antioxidant capacity of Romaine lettuce treated with and without ultrasound (25 kHz) determined by the DPPH method. Means within time results with different letters (a and b) are different at α = 0.05. Means within treatments results with different letters (x–z) are different at α = 0.05

Full size image

The antioxidant capacity of lettuce sonicated for 1 and 2 min decreased by 50.87 and 64.24 % compared with the control, respectively, during the first 30 h storage, followed by a significant increase in antioxidant capacity by 97.07 and 83.67 %, respectively, during the next 30 h. No such pattern of change was observed in the 3-min ultrasound-treated group or the control group. As an abiotic elicitor, ultrasound can trigger oxidative bursts in vivo (Wu and Lin 2002a), leading to defense responses of plant cells. The oxidative stress caused by ultrasonication was harmful to the cells, so the natural antioxidants already present in the plants were directly used to combat reactive oxygen species at the beginning. During storage, additional antioxidants were continuously synthesized by lettuce cells joining force with the natural antioxidants to further eliminate the adverse effects of the oxidative burst (Reyes et al. 2007). The ultrasound-induced production of antioxidants as shown in Fig. 2a is likely responsible for higher antioxidant capacity in 30 and 60 h samples than the control (Fig. 3).

Samples without ultrasound treatment also showed a significant decrease in antioxidant capacity after storage for 30 h and continued declining afterwards. Since a small amount of ROS can be produced even with normal cellular metabolism of postharvest leafy vegetables (Toivonen 2004), antioxidants previously stored in cells were likely used to regulate production of ROS leading to the slow decrease of antioxidant capacity during postharvest storage.

Effect of Ultrasound on Phenylalanine Ammonia-Lyase Activity

PAL is one of the most important antioxidant enzymes in plants and is directly related to secondary metabolite production (Dixon and Paiva 1995). PAL catalyzes the first step in phenylpropanoid metabolism, which is the synthetic pathway of phenolic compounds in plant cells. As a key regulatory enzyme in plant defense and in secondary metabolite synthesis, PAL serves as an indicator of the defense response triggered by environmental stresses (Dixon and Paiva 1995). No significant difference in PAL activity was found among different groups in the first 30 h (Fig. 4). Afterwards, the PAL activity increased at 60 h in all groups. Noticeably, the 2 and 3 min ultrasound-treated groups expressed significantly higher PAL activity than the control. With reference to posttreatment time, a bell-shaped curve of PAL activity was observed in all three ultrasound-treated groups, similar to the wound-induced pattern reported by Choi et al. (2005). Compared with the total phenolic content (Fig. 2) and the antioxidant capacity (Fig. 3) of ultrasound-treated samples, an increase in PAL correlated well with an increase in Abs320 and DPPH values. The changes in PAL activity after sonication might be explained by the similar physiological defense responses triggered by wounding. However, the peaks of PAL activity in ultrasound-treated lettuce occurred at later time during storage than that in wound plants. For instance, iceberg lettuce treated by wounding exhibited a peak PAL activity within 6–16 h during storage at 15 °C, depending on the wounding level (Lopez-Galvez et al. 1996; Ritenour et al. 1995). The difference is probably due to the fact that the stimulation of ultrasound was not as strong as wounding by cutting or processing.

Fig. 4
figure 4

Phenylalanine ammonia-lyase (PAL) activity of Romaine lettuce treated by ultrasound at 25 kHz. Means within time results with different letters (a and b) are different at α = 0.05. Means within treatment results with different letters (x–z) are different at α = 0.05

Full size image

Effect of Ultrasound on Color and Texture of Romaine Lettuce

Texture and color are two critical physical parameters that determine postharvest quality of fresh produce. Lower maximum shear force is associated with a decrease in firmness and crispness of leaf tissue, which is a good indicator for loss of textural quality (Bourne 1997). Immediately after treatment, the lower maximum shear force of all treated groups decreased compared with the untreated raw samples (Table 1). Lettuce samples treated by ultrasound exhibited higher firmness (maximum force, N) than the control (water washed) right after treatment and during storage. Xu et al. (2013) reported a delay in softening of fresh produce after ultrasound treatment, but the physiological mechanism is still unknown. Interestingly, there was an increase of maximum shear force at the end of 90 h storage. The plant’s self-recovery system and production of phenolic compounds might help the plant to regain tissue firmness, which took as long as 90 h (Wang 2004). Similar to the present study, Martin-Diana et al. (2005) recorded an increase of maximum puncture load of fresh-cut iceberg lettuce after 10 days storage at 4 °C. They observed a correlation between maximum water loss and maximum crispiness readings (maximum puncture load), so they suggested the loss of moisture in cell could dehydrate tissues and increase tissue elasticity resulting in an elevated maximum puncture force. Further studies are needed to monitor the moisture loss of produce and its relation to texture characteristics during storage.

Table 1 Maximum force (N) of Romaine lettuce treated with and without ultrasound (25 kHz) during storage
Full size table

Another major problem in quality loss during processing and storage of lettuce is enzymatic browning. Color of both surface and cut edges of Romaine lettuce was measured during storage to evaluate color changes. The “L,” “hue” and “chroma” values of the surface color in all groups did not change significantly during 90 h storage, and no significant difference was observed among groups (Table 2). Changes in color of the cut edge, the place most susceptible to enzymatic browning, revealed a decrease in “hue” value and an increased “chroma” value, while no significant differences in “L” values (Table 3). The “hue” value can be used to evaluate browning by estimating the change of color from green to red in lettuce, and a lower value of “hue” indicated more severe browning (Castaner et al. 1999). A decrease in “hue” value of all samples indicated that enzymatic browning happened during storage regardless of different treatments (Table 3). According to Cantos et al. (2001), wounding at the cut edge accelerated enzymatic browning. Some early studies suggested that phenolic compound accumulation via an elevated level of PAL activity was correlated with an increased susceptibility of plant tissue enzymatic browning (Cantos et al. 2001; Peiser et al. 1998; Pereyra et al. 2005). In this study, however, no difference in browning was observed among sonicated treatments after 60 h storage, when both increased PAL activity and higher phenolic content were observed.

Table 2 Effect of ultrasound (25 kHz) on the leaf surface color (Hunter color) of Romaine lettuce
Full size table
Table 3 Effect of ultrasound (25 kHz) on the cut edge color (Hunter color) of Romaine lettuce
Full size table

Effect of Ultrasound on Sensory Quality of Fresh-Cut Romaine Lettuce

Fresh-cut Romaine lettuce samples were used in a sensory test to evaluate the effect of ultrasound on product quality and the average scores of the 14 panel members are tabulated in Table 4. Generally, different times of exposure to sonication did not change sample quality at 0 h, but the overall quality of all groups significantly declined over time. Samples treated with 1 min ultrasonication were consistently rated higher than the control during storage, and the quality of this group maintained an acceptable score (5.80) at the end of 150 h. The samples exposed to ultrasound for 2 and 3 min received quality scores similar to the control. Similar to overall quality, the browning on the surface or cut edge was not detected right after treatment. Samples sonicated for 1 min exhibited a significantly lower surface browning in 30 h and a lower cut edge browning in 60 h than the control. No difference in the trend of browning was found between the control and either the 2- or 3-min ultrasound-treated groups. All samples lost their crispness and thus received a higher score of sogginess over time, but no difference appeared among treatments. Neither ultrasound treatment nor water washing caused changes of off-odor to lettuce at 0 h, but slight off-odor was detected at similar level in all groups at the end of 150 h storage.

Table 4 Effect of ultrasound (25 kHz) on overall quality, surface browning, cut edge browning, sogginess, and off-odor of Romaine lettuce determined by sensory evaluation
Full size table

Hypothetical Model on the Effect of Ultrasound on Fresh Produce

The data reported in this study demonstrated that ultrasound treatment at low APD acted as an abiotic elicitor and triggered the defense systems of Romaine lettuce in a way similar to common abiotic stresses such as wounding. Although more complicated, whole plants share similar physiological foundations with plant cell cultures. Based on this study on ultrasound treatment of whole lettuce and previous work about plant cell culture, a hypothesized mechanism for the effect of low APD ultrasound on whole fresh produce was proposed, as shown in Fig. 5. Ultrasound may first trigger the oxidative burst, including the increase of cross-membrane ion fluxes (Ca2+ and K+/H+) and the production of ROS, which are two early and important events in plant response to an elicitor (Wu and Lin 2002a). The induced production of NO may also be involved in oxidative burst by the activation of nitric oxide synthase (NOS) (Wang et al. 2006). Acting as signaling molecules, hydrogen peroxide (H2O2), and NO may activate other signal pathways such as increasing the biosynthesis of jasmonic acid (JA) and its derivatives (Wu and Ge 2004). These signal pathways work together as a network for activating the subsequent defense responses including antioxidant enzyme activation, defense-related gene expression, and secondary metabolite production (Chen et al. 2008; Safari et al. 2013). Production of phenolic compounds may be increased with an elevated level of PAL activity via phenylpropanoid pathway induced by ultrasound elicitor (Wu and Lin 2002b), which ultimately achieves an elevated antioxidant capacity and nutrition quality improvement of fresh produce.

Fig. 5
figure 5

Hypothetical mechanism of an ultrasound-stimulated secondary metabolism response in Romaine lettuce

Full size image

Conclusions

Fresh Romaine lettuce exhibited different responses to low APD ultrasound treatments at different storage times. Generally, ultrasonic treatment did not affect the sensory or nutritional quality right after a treatment. Activation of PAL and production of phenolic compounds became apparent during storage, and thus the antioxidant capacity of lettuce at 60 h storage at room temperature was higher than that of the control. The sensory quality analyzed by both instrumental methods and human panelists indicated that ultrasound treatment did not adversely affect lettuce quality during storage. Ultrasonic exposure for 1 min delayed lettuce tissue browning and in turn maintained better quality and a longer shelf-life. Further studies are needed to elucidate the physiological responses of fresh produce to ultrasonic stimuli, especially the accumulation and consumption of secondary metabolites as affected by sonication duration and timing of the treatment. This knowledge will be useful to regulate the defense system of produce to obtain improved nutritional quality.