Stimulation of probiotics by ultrasound pre-treatment and their subsequent incorporation in bread system: A new strategy for bread containing live probiotic

Abstract

This study investigates the potential of ultrasound treatment to enhance the viability and efficacy of Lacticaseibacillus paracasei (L. paracasei) in bread production. Given the increasing consumer demand for probiotic-enriched foods, bread represents a promising vehicle for these beneficial microorganisms. Nevertheless, the high temperatures involved in traditional baking processes significantly reduce probiotic cell viability, limiting the functional effectiveness of probiotic-enriched breads. High-intensity ultrasound (HIU) has emerged as a potential strategy to enhance probiotic viability and stimulate microbial metabolic pathways, thus potentially improving probiotic performance throughout bread fermentation and baking processes. Results indicated that certain controlled ultrasound treatments promoted L. paracasei growth and metabolic activity. Specifically, treatments at lower power did not affect the immediate viability of the cells, while higher power caused a significant reduction. However, after recovery incubation, certain treated groups exhibited improved growth, indicating that specific ultrasound parameters could foster cell growth and propagation. During bread proofing, ultrasound-treated probiotics showed a slight but significant increase in cell growth. In the baking phase, while the overall viability of L. paracasei decreased, ultrasound-treated cells demonstrated enhanced heat resistance. Furthermore, the addition of L. paracasei in the dough positively impacted dough height, stability, and gas retention. Bread with added L. paracasei, both treated and untreated, revealed improvements in texture, specific volume, moisture content, and pH. In conclusion, carefully optimised ultrasound pre-treatment demonstrates substantial potential for improving probiotic-related viability and processing performance within bread systems. Further research is needed to optimise ultrasound parameters and to elucidate the underlying mechanisms responsible for the protective effects observed during bread processing.

1. Introduction

The definition of a probiotic is ‘live microorganisms that, when administered in adequate amounts, confer a health benefit on the host’ [1]. Given the recent trend of consumers moving towards healthier lifestyles and dietary patterns, various probiotic products, including supplements and functional foods with probiotics, have been developed to cater to this demand [2]. Although both probiotic supplements and foods have been clinically proven to be effective carriers of probiotics for the prevention and treatment of diseases, food-based carriers are generally more acceptable to consumers due to their convenience and taste [3], [4]. To date, most probiotic foods have been developed using dairy products and beverages as bases. Bread, a widely consumed baked product, has recently garnered escalating attention as a potential novel probiotic carrier [5], [6], [7].

The viability of probiotics in food depends on the composition of the food matrix, processing conditions, and storage conditions [8]. Traditional bread is made by mixing ingredients (primarily wheat flour, water, sugar, salt, and yeast) followed by resting, proofing, and baking. However, the viability of probiotics in bread is often compromised due to heat damage caused by high temperatures during baking [6], [7]. Over the past decade, scientists have attempted to improve the viability of probiotics during baking and storage by using various processing methods and techniques. Micro-encapsulation and edible coatings are common strategies employed to protect probiotic cells during thermal processing. For example, Soukoulis et al. [9] demonstrated the feasibility of incorporating Lactobacillus rhamnosus GG into pan bread using an edible film composed of alginate and whey protein concentrate, achieving delivery of up to 8 log CFU/g. Ghasem et al. [10] reported that the survival of Lactobacillus acidophilus and Lactiplantibacillus plantarum (L. plantarum) increased by 2 logs during bread baking and storage when probiotic cells were micro-encapsulated with gum and sago starch. However, the complexities of extra processing steps and the potential negative impact on the physicochemical and organoleptic properties of bread from these current methods are driving the search for novel and simpler processing techniques to ensure probiotic activity and promote health benefits [11], [12], [13].

Ultrasound is defined as sound waves with frequencies above the upper limit of human hearing (typically > 20 kHz). From an application perspective, high-intensity ultrasound (HIU), typically ranging from 10 to 1000 W/cm², has emerged as a green, rapid, and non-thermal processing technology in the food industry. It has been widely applied for purposes such as homogenization, defoaming, microbial inactivation, fermentation acceleration, and structural modification. These effects are primarily attributed to acoustic cavitation generated by the energy from the ultrasonic system [14], [15], [16].

Beyond processing efficiency, HIU has also attracted growing interest for its potential to modulate probiotic physiology and metabolism in dairy products [14], [17]. By exposing probiotics to adverse environmental stresses during their growth, the specific stress-sensing system of probiotics can improve their tolerance to extreme environmental stress by regulating physiological and metabolic behaviours. Cross-protection effects have been observed in response to various stressors such as heat shock, cold stress, acid stress, osmotic pressure, and nutrient starvation [18], [19].

To our knowledge, limited studies have focused on the changes in heat resistance or activity of probiotic-related microorganisms following ultrasonic stressors. Therefore, this study aimed to stimulate Lacticaseibacillus paracasei (L. paracasei) using ultrasound as a pre-treatment and investigate their subsequent incorporation into a bread system. Specifically, we investigated the activity of L. paracasei under different ultrasound conditions. In addition, the metabolic activity and viability of L. paracasei that participated in the entire bread manufacturing process (from an experimental perspective) were also studied, as they may have a positive effect on the bread system. What is more, the dough fermentation properties and the quality of bread made with the addition of ultrasonically-treated cells were evaluated as well. In this study, probiotic-related functionality is defined in terms of processing performance, including viability retention and stress tolerance during proofing and baking.

2. Material and methods

2.1. Strain preparation

L. paracasei is a common type of bacteria with probiotic-related properties, generally used in food fermentation and dairy products [20]. Both L. paracasei and its metabolites have been reported to have potential health benefits for consumers, making it a suitable candidate for functional bread production [21]. The L. paracasei JLPF-176 used in this study was obtained as a commercially supplied, freeze-dried single-strain culture (Xi’an Ebos Biotech Co., Ltd.). Approximately 1 g of freeze-dried culture was dissolved in 10 ml of sterile de Man Rogosa Sharpe (MRS) broth (Merck Pte Ltd, SG) supplemented with 0.05% L-cysteine (Thermo Fisher Scientific Inc, SG) and incubated overnight at 37 °C. Subsequently, 100 μl of the culture was transferred to 10 ml of sterile MRS broth and incubated at 37 °C, and the subculturing procedure was repeated twice. The subculture bacteria were streaked onto MRS agar (Merck Pte Ltd, SG) supplemented with 0.05% L-cysteine and incubated at 37 °C for 36 h to obtain isolated colonies. Morphologically uniform single colonies were selected and used as the working stock. The plates were stored at 4 °C for up to 2 weeks. For each experiment, cultures were freshly prepared from single colonies derived from the same purification batch.

2.2. Ultrasound system set up

A colony from the MRS plate was transferred to 10 ml of sterile MRS broth and incubated at 37°C for 6 h. Then, 100 µl of the culture was transferred to 10 ml of sterile MRS broth and incubated at 37°C for an additional 6 h. After incubation, 1 ml of the L. paracasei subculture was transferred into a sterile 100 ml-capacity blue-cap bottle containing 99 ml of sterile MRS broth. The ultrasound system (Fig. 1), consisting of a 20 kHz probe, a generator, an ultrasonic transducer, and a booster (UIP 1000, Hielscher, Germany), was controlled by a computer. The sonication was conducted under controlled temperature through a water-cooling system to prevent thermal damage.

Fig. 1.

Ultrasound system set up.

For the HIU treatment groups, cell suspensions were subjected to ultrasound in pulse mode (50% duty cycle) using a probe operated at amplitudes of 25%, 50%, and 100% for 5, 10, or 15 min, respectively. The treatments are denoted as UT25-5, UT25-10, UT25-15, UT50-5, UT50-10, UT50-15, UT100-5, UT100-10, and UT100-15. The actual total energy input (Ws) for each treatment was obtained from the ultrasound system, which automatically calculates the cumulative energy delivered to the sample. The ultrasonic power (W) dissipated was measured and calculated for each group using the calorimetric method [22]. A non-sonicated control group (Control) was included for comparison. Following treatment, all samples were incubated at 37 °C for up to 24 h. Each treatment, including the control, was prepared in triplicate from the initial culture preparation step.

2.3. Measurement of growth kinetics

The growth kinetics of L. paracasei was monitored by measuring the optical density value (OD value) at a wavelength of 600 nm by using a 96-well-plate in a microplate reader (Spectrafluor Plus, Tecan, Durham, NC, USA) at 37 °C with continuous circinate shaking for over 18 h (10 min intervals). The OD value of each sampling time was recorded by the program (Spectrafluor Plus), and the lag time for each well was calculated automatically.

2.4. Mathematical modelling

The Gompertz models: N = N₁ exp {−exp [(k₁ e) (t₂ − t)/N₁ + 1]} was used to estimate kinetic parameters of microorganism growth curves [23]. A modified Gompertz model mentioned in the study of [24] was used to assess kinetic parameters from the growth curve of L. paracasei based on the DMFit excel based program. The primary kinetic parameters including the specific growth rate (μmax, OD unit/h) and lag phase (λ, h), were calculated for all groups.

2.5. Viable cell counts after ultrasound treatment

To monitor the changes in the cell population, 100 μl of the bacterial suspension was spread onto MRS agar supplemented with 0.05% L-cysteine. Prior to spreading, the suspension was subjected to serial dilution using a sterile 0.1% peptone water solution (Merck Pte Ltd, SG). This dilution step was performed immediately after ultrasound treatment and again after 12 h of incubation. The plates were then incubated at 37°C for 36 h, and the enumeration results were recorded as colony-forming units (CFU) per ml.

2.6. Biomass measurement

The biomass of the control and all treatment groups was measured after 24 h of incubation. The cells were collected by centrifugation (10,000×g, 10 min, 4°C) and washed twice with sterile 0.1% peptone water (Merck Pte Ltd, SG). After freeze-drying, the biomass was precisely quantified gravimetrically using an analytical balance.

2.7. Bread sample preparation

Bread samples were prepared following a modified recipe from Zhang et al. [25]: bread flour (1 kg), sugar (40 g), fine salt (9.6 g), shortening (30 g), instant yeast (10 g), and 0.9% sterile saline (600 g) with the addition of L. paracasei. The L. paracasei added to the bread was treated with ultrasound using the method described in Section 2.2, and five treated conditions were selected: non-treated (Control), UT25-10, UT50-10, UT75-10, and UT100-10. Following treatment, the cells were harvested by centrifugation (10,000 × g, 10 min, 4 °C), washed twice with 0.9% sterile saline, and re-suspended in 0.9% sterile saline for subsequent incorporation into the bread formulation. The cells were then harvested by centrifugation (10,000 × g, 10 min, 4 °C) and washed twice with 0.9% sterile saline. Subsequently, the cells were re-suspended in 0.9% sterile saline for bread dough preparation.

A total of six bread groups were prepared: (i) Blank, without probiotic addition; (ii) Control, with non-sonicated L. paracasei; and (iii-vi) HIU-treated probiotic groups designated as UT25, UT50, UT75, and UT100, respectively. Dough preparation was performed using a planetary mixer (Varimixer Globe, USA): all dry ingredients were mixed at low speed for 1 min, followed by high-speed mixing for 6 min after water addition. The dough was rested for 10 min, then divided and shaped into 55 ± 1 g portions using a dough moulder (DR ROBOT2, Holland) and proofed at 40 °C and 85% relative humidity for 75 min in a prover (DR ROBOT2, Holland). Finally, baking was conducted at 200 °C for 10 min in a convection oven (Eurofours®, France).

2.8. Microbiological analysis of proofing and baking process

The methods for microbiological analysis of the proofing and baking process were adapted from Zhang et al. [25]. The viable cell count of L. paracasei was determined at various time points: before proofing (0 min) and during proofing (40 °C, 85% RH) at 15, 30, 45, and 75 min, as well as after baking (200 °C) for 0, 5, and 10 min. At each sampling time, samples of dough or bread weighing 5.0 g were collected from the core and crust. These samples were then suspended and aseptically homogenised with 95 ml of sterile 0.1% peptone water in a sterile stomacher bag (Practical Mediscience Pte Ltd, SG) for 1 min. Following serial dilution, 1 mL of the suspension was spread onto MRS agar supplemented with 0.05% L-cysteine and 100 µg/mL cycloheximide (Merck Pte Ltd, SG). After incubation, the enumeration results were recorded as CFU per gram.

2.9. Rheofermentometer analysis

The effect of probiotic addition on dough development and gas production was monitored using a Rheofermentometer F3 (Chopin Technologies, France) [26]. A 315 g piece of dough from each sample group was weighed and placed into the basket of the rheofermentometer. Subsequently, a 2 kg cylindrical piston was placed on the dough, and the basket was transferred to the fermentation chamber set at 40 °C for 180 min. Throughout the fermentation process, the dough development curve was recorded, including measurements such as the total volume of CO₂ produced (V_T), the volume of CO₂ lost (V_L), gas retention coefficient, maximum dough height (Hm), final dough height (h), and dough weakening coefficient.

2.10. Texture profile analysis

Texture profile analysis (TPA) of bread samples was measured using a texture analyser (TA-XT2i, Stable Micro System, UK) to obtain hardness, chewiness, cohesiveness and springiness. A 20 mm vertical slice of bread crumb was sliced mechanically by a bread splicer, with measurements taken using the middle of the bread crumb. The tests were conducted with a P20 aluminium plunger, under the following conditions: compression force: 5 g; compression speed: 10 mm/s; and distance: 8 mm (i.e., 40% compression). The texture analyser generated a two-cycle compression TPA force–time graph, providing information on the texture profile characteristics.

2.11. Colour measurement

The colour characteristics of bread crust and crumb were measured using a spectrophotometer (CM-5, Konica Minolta Sensing Inc., Tokyo, Japan) across a wavelength range of 360–740 nm with a resolution of 10 nm. Bread samples were placed in transparent Petri dishes (CM-A128) for measurement. As shown in Fig. 2, colour measurements were taken at five predefined positions on each bread sample (both crust and crumb), and the mean of these five readings was calculated to represent the overall colour value of the individual bread crust or crumb. A total of nine replicate breads were analysed per group.

Fig. 2.

Sampling positions for colour measurement on bread.

Colour values were recorded in the CIE L*a* b* colour space, where L* indicates lightness (0 = black, 100 = white), a* denotes the red-green axis (−a* = greenness, +a* = redness), and b ∗ represents the blue-yellow axis (−b* = blueness, +b* = yellowness). To quantify the overall colour difference between groups, the total colour difference (ΔE) was calculated using the following equation (1):

$$\mathrm{\Delta }E=\sqrt{\left\{{\left(L_t-L_c\right)}^2+{\left(a_t-a_c\right)}^2+{\left(b_t-b_c\right)}^2\right\}}$$ (1)

where L_t, a_t, and b_t refer to the colour values of the treatment group, and L_c, a_c, and b_c correspond to the values of the control group.

2.12. Physicochemical analysis

pH: The dough samples (5.0 g) were suspended in 95 ml of distilled deionized water at each sampling time and then homogenized. The pH value of the suspension was measured using a pH meter (Metrohm 744 pH Meter, Switzerland). The pH value of bread samples after baking was also measured using the same methods [25].

Specific volume: After baking, the bread samples were cooled to room temperature for 30 min, and the specific volume of the bread was measured using a Vol-scan Profiler 600 (Stable Micro Systems, UK) [26].

Moisture content: The moisture content of the bread was measured using a moisture analyser (ShimadzuMOC63u, SG) [25].

Water activity: The water activity of the bread was measured using a water activity meter (AQUA LAB® Series 3 TE, USA) [25].

2.13. Statistical analysis

All experiments were conducted at least in triplicate, and data are presented as mean values ± standard deviation. Statistical comparisons among treatment groups were performed using one-way analysis of variance (ANOVA), followed by Tukey’s honestly significant difference (HSD) post hoc test. All analyses were conducted using IBM SPSS Statistics version 26 (IBM Corp., Armonk, NY, USA), and a significance level of p ≤ 0.05 was considered statistically significant.

3. Results and discussion

3.1. Effect of ultrasound conditions on L. paracasei growth curve and kinetic parameters

The growth kinetics of L. paracasei were characterised by plotting by plotting the average OD600 value against the incubation time. Fig. 3 (A) shows the growth curves of the control group and all groups treated under different HIU conditions with various amplitudes and times at a constant frequency of 20 kHz. Fig. 3 (B) and Fig. 3 (C) present the maximal specific growth rate (µmax, OD unit/h) and lag phase (λ, h) of each group, respectively, calculated based on the growth curve.

Fig. 3.

(A) The growth curve of L. paracasei as measured by OD600 value of control (untreated) and after treatments under different HIU conditions, incubated at 37 °C over 18 h; (B) Lag time of L. paracasei of control and after different HIU treatments; (C) The highest µmax of L. paracasei of control and after different HIU treatments. ^{a, b, c} Different superscript letters show the significant differences between groups.

As observed, the lag phase of L. paracasei was significantly influenced by the time duration and amplitude of the ultrasound treatment. When comparing groups with the same treatment amplitude or time duration, the lag time gradually increased with higher amplitudes (from 25% to 100%) or longer treatment durations (from 5 min to 15 min). All treated groups showed significantly longer lag phases than the untreated control group (p ≤ 0.05), with the 100% 15 min treated group exhibiting the longest lag phase. Similar findings have been reported in previous studies, suggesting that the bacteria require more time to adapt to the new environment and for certain probiotic cells to recover from the higher level of injury caused by ultrasound treatment [23].

Regarding the maximal specific growth rate (µmax, OD unit/h), the groups treated with 25% 15 min, and 50% 5 min showed significantly higher rates compared to the control group (p ≤ 0.05), as shown in Fig. 3 (C). The µmax values of the other treated groups under different ultrasound conditions were not significantly different from the control group, indicating a similar growth behaviour of L. paracasei. When applied with an appropriate frequency and time duration, the controlled ultrasound treatment initially extended the lag time required for recovery from the detrimental effects of ultrasound but then stimulated faster growth. A similar effect was reported by Brito et al. [27], where probe ultrasound at 640 J/mL increased the growth rate of yogurt starter cultures, highlighting the growth-promoting potential of moderate ultrasound intensity. The stimulation effect of ultrasound treatment on probiotic cells can be attributed to enhanced nutrient utilization from the surroundings and increased metabolic activity induced by the cavitation effect of ultrasound waves [27], [28].

Based on this mechanism, various studies have used ultrasound treatment to improve the yield and production efficiency of active substances and beneficial metabolites during microbial cell growth [23]. In the food industry, ultrasound has been employed to shorten fermentation times for microorganisms [29], [30]. However, it should be noted that microorganisms are susceptible to ultrasound conditions such as intensity, duration, and amplitude, as higher treatment levels can lead to intracellular content leakage and irreversible damage [28].

3.2. Biomass and viable cell count

The biomass of all groups ranged from 0.132 g/100 ml to 0.327 g/100 ml, with the highest biomass observed in the 25% 15 min treated group as shown in Fig. 4. This concurs with the results obtained by Dahroud et al. [31], who studied the ultrasound treatment effect on the fermentation efficiency of Lacticaseibacillus casei subsp. casei ATTC 39392. Appropriate ultrasonic pretreatment resulted in the highest biomass production, an extended logarithmic growth phase, and an increased specific growth rate.

Fig. 4.

The biomass of L. paracasei after 12 h of incubation in both the control (untreated) group and the groups treated under different HIU conditions. ^{a, b, c} Different superscript letters show the significant differences between groups.

As shown in Fig. 5, there was no significant difference in the viable cell count between the groups treated at amplitudes of 25% and 50% immediately after ultrasound treatment and the control group. However, there was a significant decrease in groups treated at an amplitude of 100%. The number of live bacteria in the 100% treated 5 min group was significantly (p ≤ 0.05) lower than the control, followed by the 100% treated for 10 min and then 100% treated for 15 min groups, suggesting that the relatively strong ultrasound treatment parameters caused cell inactivation immediately after treatment. The number of cells inactivated highly depended on ultrasound amplitude without changing the ultrasonic processing time. These differences were not evident from the initial OD600 value due to the method’s inability to differentiate between viable and non-viable cells, underscoring the need for plate counting in viability assessment.

Fig. 5.

Populations of L. paracasei in control and treatment groups under different HIU conditions. ^{A, B, C} Different superscript letters show significant differences immediately after ultrasound treatments. ^{a, b, c} Different superscript letters show significant differences after 12 h incubation.

It’s worth noting that after 12 h incubation at 37 °C, the 25% 15 min treated group, and 50% 5 min treated group had significantly higher viable cell counts than the control. Similar results can be found in literature, in Behzadnia et al. [23]’s study, the L. plantarum treated by ultrasound at 25 kHz for 30 min in MRS broth had a higher population than the control group after 24h incubation. Moreover, the 100% 10 min treated group had the lowest live bacteria and biomass after 12 h incubation, indicating that strong ultrasound parameters caused irreversible damage to the cells as described above. In earlier studies, the acceleration of cell growth and propagation has been attributed to improvement in the permeability of cell membranes, which enhances ion uptake and nutrient transport [32]. However, in Dahroud et al. [31]’s study, they found the biomass of Lacticaseibacillus casei subsp. casei improved with ultrasound treatment without changes in cell morphology. In another study, the enhanced growth was attributed to the loosening of cell aggregates, which facilitated better nutrient uptake [28]. The different types of probe sonicators should also be considered when setting up an ultrasound system to treat microorganisms.

3.2.1. Effect of proofing process on L. paracasei

To investigate the effect of the proofing process on L. paracasei and ultrasound-treated L. paracasei growth in the dough, the viable cell counts of all five groups were measured every 15 min during the proofing process, and the pH values were also measured at each sampling time. Fig. 6 shows that L. paracasei cells were not detected in the Blank group dough during the entire proofing process, while the initial quantity of probiotic cells was detected at approximately 6.8–9.8 log CFU/g in other dough samples with probiotic addition. During the proofing process, there was no significant difference in viable cell counts of L. paracasei cells in Control group with untreated probiotic addition. The probiotic population of UT25 and UT50 exhibited a slight but significant increase (p ≤ 0.05) at the end of proofing, and the probiotic cells in UT25, which were treated at 25% amplitude, had the most significant growth with a 1.2 logs increase. These results suggest that the controlled ultrasound treatment applied could stimulate the growth of L. paracasei during dough proofing. Similar results can be found in some studies. For example, Wang et al. [33] found that the growth of L. plantarum during apple juice fermentation was stimulated by applied ultrasound at the lag phase of microbial growth stages. In Nguyen at al. [34]’s study, the fermentation activities of bifidobacteria were stimulated in milk by treating the sample at 50% amplitude for 15 min with a 20 kHz ultrasound probe application. Unfortunately, there are few studies that have researched microbial growth in dough proofing with ultrasound application.

Fig. 6.

Viable cell counts of L. paracasei in dough during dough proofing.

As shown in Fig. 7, the pH of the Blank dough sample remained at around 6.0, while the pH of the dough samples with untreated and treated L. paracasei was significantly lower (p ≤ 0.05) than the control during the proofing process. The decrease in pH supported the proliferation and increased metabolic activity of L. paracasei in the dough. The acidification was due to the lactic acid, acetic acid, and CO2 dissolution in the water phase (to a lesser extent) produced by the metabolism of L. paracasei during dough fermentation [35], [36].

Fig. 7.

The pH of dough during dough proofing.

3.2.2. Viability of L. paracasei in baking process

To investigate the viability of L. paracasei during the baking process, the viable cell count in bread samples (taken from both core and crust sections) was measured at 0 min, 5 min, and 10 min during baking, with no probiotic cells observed in the blank samples. As can be seen in Fig. 8, the initial number of viable cells in dough samples with probiotic addition was measured at around 10⁸ CFU/g, which substantially reduced to 10⁵ CFU/g from 0 min to 5 min of baking, then continued to decrease from 5 min to the end of baking. Thermal inactivation was expected to be the dominant mechanism of cell inactivation because the temperature reached the lethal level for L. paracasei during baking [37]. Interestingly, the HIU pre-treatment exhibited a protective effect on L. paracasei in bread during the baking process. The population of L. paracasei in UT75 and UT100 was significantly higher (p ≤ 0.05) than in the Control sample (untreated) at 5 min of baking. At the end of baking, the residual viability of L. paracasei in UT75 and UT100 was significantly higher than the Control by about 1–2 logs (p ≤ 0.05), indicating that L. paracasei treated under some ultrasound conditions possessed a higher heat-resistant ability in bread during the baking process.

Fig. 8.

Viable cell counts of L. paracasei during bread baking.

The enhanced survival of L. paracasei observed during baking is likely due to stress-adaptation mechanisms triggered by ultrasound pre-treatment. Sub-lethal stresses such as heat, cold, acid, or oxidative conditions can activate protective responses in L. paracasei, including the expression of heat shock proteins (HSPs), which improve survival under subsequent stress. This phenomenon, termed cross-protection, describes how prior exposure to a mild stressor increases resistance to a later, more severe one [18], [19]. In this context, ultrasound may act as a mild stress that elicits protective cellular responses, thereby enhancing the thermal tolerance of L. paracasei during baking. Giordano et al. [38] provided transcriptomic evidence that ultrasound exposure triggers broad transcriptional reprogramming in L. casei ATCC 393, including the differential expression of genes involved in membrane transport, energy metabolism (glycolysis and purine biosynthesis), DNA repair, and ribosome biogenesis. These shifts suggest a coordinated stress response consistent with cross-protection.

Additionally, similar ultrasound-induced physiological adjustments have been used in milk fermentation systems, where LAB modulate gene expression and metabolic activity to better withstand environmental stress [14], [17]. Zhang et al. [25] also reported partial survival of Bifidobacterium animalis subsp. lactis BB12 during baking, which was possibly attributed to thermal stress-induced HSPs expression. Collectively, these findings support the role of HIU as a non-thermal stimulus capable of inducing cross-protective mechanisms in L. paracasei within thermally processed food matrices like bread.

3.2.3. Properties of dough and bread

The Rheofermentometer analysis reflects the correlation between both the fermentation capacity of the starter culture and the development of the gluten network structure. Fig. 9 (A) and Fig. 9 (B) display the maximum dough height and final dough height of all groups, respectively. It is evident that Sample UT25, with HIU-treated L. paracasei at 25% amplitude for 10 min added, rose to a significantly greater maximum height and final height (51.40 ± 0.40 mm and 35.70 ± 0.90 mm) than the blank samples without any probiotic added (44.00 ± 1.60 mm and 29.80 ± 0.66 mm).

Fig. 9.

Characteristics of dough development and gas production. (A) Maximum dough height (Hm); (B) Final dough height (h); (C) Dough weakening coefficient; (D) Total CO₂ production volume (V_T); (E) Volume of CO₂ lost (V_L); (F) CO₂ retention coefficient. ^{a, b, c} Different superscript letters indicate significant differences between groups.

From the results of the dough weakening coefficient shown as Fig. 9 (C), it is clear that all samples with untreated or HIU-treated L. paracasei added had a significantly lower (p ≤ 0.05) dough weakening coefficient than the blank samples, indicating that the addition of L. paracasei in dough increased the dough stability. These results differ from a similar study described by Pei et al. [39], in which they found that the addition of L. plantarum LB-1 led to a higher dough weakening coefficient than when fermented with yeast. However, the result of the CO₂ retention coefficient aligns with their study, which will be discussed later. The difference may be caused by the different Lactobacillus types in the starter culture. An interesting finding is that Sample UT25 and Sample UT50, with L. paracasei treated at 25% and 50% amplitude for 10 min respectively, showed a significantly lower (p ≤ 0.05) dough weakening coefficient than the control sample with untreated L. paracasei added, and were more resistant to collapse. This indicates that the HIU treatment conditions affect the performance of L. paracasei in preventing dough collapse and promoting the formation of a gluten network structure.

The activity of yeast can be indicated by the total CO₂ production volume (V_T) [26]. During the initial stage of dough fermentation, yeast performs aerobic respiration, which lasts a few minutes until oxygen is depleted. Subsequently, yeast transitions to anaerobic respiration, producing CO₂ and ethanol [26]. L. paracasei is a type of facultative heterofermentative bacteria that produce lactic acid, CO₂, alcohol, and other organic acids as by-products during dough fermentation [40]. In this study, V_T is an indicator of the activity of the starter culture (yeast or yeast with L. paracasei). As shown in Fig. 9 (D), the Control (with untreated L. paracasei added) and UT25 (with treated L. paracasei at 25% amplitude for 10 min added) had a significantly larger V_T (p ≤ 0.05) than the blank group. There was no significant difference between the Blank and samples with stronger HIU pre-treated L. paracasei added. These results indicate that the addition of L. paracasei could increase the total CO₂ production volume, and some controlled ultrasound parameters could enhance this role, while some stronger ultrasound parameters did not affect the gas producing ability of L. paracasei in dough fermentation. This result was consistent with the pH and viable cell count results during dough proofing discussed in Section 3.2.1. Fig. 9 (F) shows the results of the CO₂ retention coefficient of all groups, in which UT25 shows a significantly higher gas retention ability than the control group (p ≤ 0.05). The higher gas retention coefficient may be associated with the increased contents of water-soluble protein [39]. A better gluten structure and improved gas retention coefficient often contribute to a better bread volume [26], [41].

The hardness and chewiness are the two most important properties that provide information on bread texture. As shown in Fig. 10 (A) and Fig. 10 (B), all bread samples enriched with L. paracasei possessed significantly greater hardness and chewiness than those in the blank group. Notably, UT25 displayed the highest level of hardness, also suggesting that the HIU pre-treatment influences the metabolic activities of L. paracasei. This effect could be attributed to the enhanced protease activity in lactic acid bacteria (LAB) or to the presence of lactic and organic acids produced by LAB, which may lead to gluten structure damage and a reduction in disulphide bonds [42], [43].

Fig. 10.

(A) Hardness and (B) Chewiness of different bread samples. Blank: bread without probiotic addition. ^{a, b, c} Different superscript letters indicate significant differences between groups.

The colour characteristics of bread crust and crumb are presented in Table 1. The incorporation of L. paracasei significantly reduced the crust brightness (L*) compared with the Blank, accompanied by a notable increase in redness (a*) (p ≤ 0.05). These changes indicate a darker and more reddish crust surface in probiotic-containing breads, whereas yellowness (b*) remained statistically unchanged. The total colour difference (ΔE) exceeded the critical threshold of 3 in all L. paracasei-containing groups, suggesting that the visual alteration was perceptible to the human eye, which is consistent with previous reports on probiotic-enriched breads [9], [44]. Notably, HIU pre-treatment of the L. paracasei did not cause significant changes in crust L*, a* or b* values compared with the untreated group (Control). This suggests that the colour change was mainly related to the presence of viable cells, rather than to the ultrasound processing itself, which is consistent with observations in probiotic breads [8].

Table 1.

Colour characteristics of bread crust and crumb.

	Samples
	Blank	Control	UT25	UT50	UT75	UT100
Bread Crust
L*	58.85 ± 0.59a	53.91 ± 2.01b	53.48 ± 3.59b	53.42 ± 1.42b	53.70 ± 1.80b	54.20 ± 2.37b
a*	13.82 ± 0.32a	15.64 ± 0.35b	16.15 ± 0.83b	16.75 ± 0.49b	17.00 ± 0.60b	17.31 ± 0.94b
b *	31.49 ± 0.50a	31.27 ± 0.87a	30.43 ± 2.00a	32.30 ± 0.78a	32.40 ± 0.90a	32.48 ± 1.04a
ΔE	−	5.42 ± 1.72a	6.97 ± 1.78a	6.30 ± 1.80a	6.10 ± 1.60a	6.06 ± 1.78a
Bread Crumb
L*	80.05 ± 1.95a	80.09 ± 2.90a	82.53 ± 1.49a	81.42 ± 2.99a	81.30 ± 2.10a	81.20 ± 1.40a
a*	0.60 ± 0.43a	0.56 ± 0.10a	0.37 ± 0.18b	0.59 ± 0.06a	0.58 ± 0.07a	0.59 ± 0.06a
b *	16.84 ± 0.52a	16.37 ± 0.60a	18.91 ± 0.54b	18.71 ± 0.71b	17.50 ± 0.75a	17.37 ± 0.80a
ΔE	−	3.74 ± 1.96a	3.38 ± 1.02a	2.89 ± 0.92a	2.50 ± 1.00a	2.19 ± 1.21a

Results are expressed as mean ± SE (n ≥ 3). ^{a, b, c} Different superscript letters indicate significant differences between groups.

For the bread crumb, neither the addition of L. paracasei nor their treatment affected brightness (L*) or redness (a*) relative to the Blank, indicating that the interior colour remained largely unaffected by L. paracasei incorporation. Interestingly, the crumb yellowness (b*) was significantly higher in UT25 and UT50 compared with the blank and other groups (p ≤ 0.05). This trend may be linked to differences in microbial cell populations and metabolic activity during the dough proofing stage, which could influence pigment precursor formation or Maillard reaction intermediates, as suggested by similar observations in L. plantarum fortified breads [45]. However, the overall ΔE values in the crumb were below 4 and did not present a consistent increasing trend across ultrasound treatments, suggesting that the visual impact inside the bread was moderate [44].

The results of dough and bread’s physical and chemical properties are summarised in Table 2. The bread samples with untreated or ultrasound-treated L. paracasei added had a larger specific volume and moisture content than the control samples (p ≤ 0.05), and there was no significant difference in this improvement between HIU-treated and untreated cells. Moreover, the addition of L. paracasei led to a significantly lower pH (p ≤ 0.05) of bread, and no significant difference was found in a_w of all bread samples. Based on the current results, it was confirmed that the addition of L. paracasei helped improve the quality of bread.

Table 2.

Physicochemical properties of dough/bread samples.

	Samples
	Blank	Control	UT25	UT50	UT75	UT100
Bread properties
Specific volume (cm3/g)	3.05 ± 0.07a	3.49 ± 0.28b	3.68 ± 0.10b	3.66 ± 0.08b	3.52 ± 0.09b	3.36 ± 0.13b
pH	6.01 ± 0.02a	5.44 ± 0.18b	5.29 ± 0.04b	5.26 ± 0.06b	5.25 ± 0.05b	5.25 ± 0.04b
Moisture (%)	41.30 ± 0.53a	43.40 ± 0.56b	43.10 ± 0.10b	42.85 ± 0.57b	43.00 ± 0.15b	43.10 ± 0.11b
aw	0.97 ± 0.00a	0.97 ± 0.00a	0.97 ± 0.01a	0.97 ± 0.01a	0.97 ± 0.01a	0.97 ± 0.00a

Results are expressed as mean ± SE (n ≥ 3). ^{a, b, c} Different superscript letters indicate significant differences between groups.

3.3. Conclusion

In conclusion, this study demonstrates the application of HIU pre-treatment on L. paracasei significantly influences its growth and metabolic activity. Whilst lower amplitudes (25% and 50%) did not result in a significant difference in viable cell count immediately after treatment compared to the control group, higher amplitudes (100%) led to a significant decrease in viable cells. Notably, after 12 h of incubation, certain treated groups exhibited significantly higher viable cell counts and biomass, suggesting that suitable ultrasound treatment parameters could promote cell growth and propagation. This growth enhancement is potentially due to the loosening of cell bundles and increased nutrient utilization. This technique has been employed in various applications, such as enhancing the yield and production efficiency of active substances and metabolites and shortening fermentation time in the food industry. However, strong ultrasound parameters were found to cause irreversible damage to the cells, emphasizing the need for careful selection of treatment conditions.

The effects of the proofing process on the growth of L. paracasei and HIU-treated L. paracasei in the dough were also investigated. The viable cell counts of L. paracasei in HIU pre-treated samples had a slight but significant increase at the end of the proofing process, with the highest growth observed in the sample treated at 25% amplitude for 10 min. This suggests that controlled HIU pre-treatment can stimulate the growth of L. paracasei during dough proofing. The pH of dough samples with untreated and pre-treated L. paracasei was significantly lower than the control during the proofing process, supporting the proliferation and increased metabolic activity of L. paracasei in the dough. The acidification was attributed to the production of lactic acid, acetic acid, and CO₂ dissolution in the water phase by L. paracasei during fermentation.

During the baking process, the viability of L. paracasei decreased substantially due to thermal inactivation. However, controlled ultrasound treatment showed a protective effect on L. paracasei during baking, with higher population numbers in treated samples compared to untreated samples. This indicates that L. paracasei treated under certain ultrasound conditions have a higher heat-resistant ability in bread during the baking process. The protective effect may be due to the formation of HSPs and activation of heat shock genes when cells are exposed to extreme conditions, providing a stress-adaptation process for L. paracasei to exhibit higher heat resistance during baking.

The addition of L. paracasei significantly improved dough height, stability, and gas retention. HIU pre-treatment influenced the ability of L. paracasei to prevent dough collapse and promote gluten network formation. The total CO₂ production volume indicated that adding L. paracasei enhanced gas generation, with certain ultrasound parameters further amplifying this effect. The incorporation of L. paracasei impacted both the appearance and the physicochemical properties of bread. It resulted in a darker and more brownish crust without markedly altering crumb colour. Furthermore, bread containing L. paracasei, regardless of ultrasound treatment, showed higher hardness and chewiness, larger specific volume, greater moisture content, and lower pH compared with the control.

However, there are limited studies on the effect of ultrasound treatment on probiotics in bread systems. Further research is needed to better understand the underlying mechanisms and to optimise ultrasound treatment parameters for improved probiotic viability and processing performance in various food applications.

CRediT authorship contribution statement

Jianing Zhai: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Lu Lu: Investigation, Formal analysis, Data curation. Jiaqi Mi: Methodology, Investigation, Formal analysis. Weibiao Zhou: Writing – review & editing, Supervision, Methodology, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.