Multi-frequency power ultrasound (MFPU) pretreatment of crayfish (Procambarus clarkii): Effect on the enzymatic hydrolysis process and subsequent Maillard reaction

Abstract

Crayfish, as a new consumer trend, has become one of the most popular freshwater aquatic foods in China, and its processed product output has been increasing year by year. In this study, the effect of multi-frequency power ultrasound (MFPU) pretreatment in various working modes including mono-, dual- and tri-frequency on the enzymatic hydrolysis and Maillard reaction of crayfish was investigated. Additionally, the underlying mechanism was also explored. Results showed that the degree of hydrolysis (DH) of crayfish was 23.89 %, while it increased to 40.78 % after MFPU pretreatment at 20/40 kHz dual-frequency, indicating that MFPU pretreatment significantly (p < 0.05) enhanced the enzymatic hydrolysis of crayfish. The crayfish protein after MFPU pretreatment exhibited a decrease in α-helix and increase in β-fold content, and the occurrence in the unfolding of the protein structure and alteration of tertiary structure. According to scanning electron microscopy (SEM) and atomic force microscopy (AFM) analyses, the MFPU pretreatment resulted in the fragmentation of crayfish protein. During the analyses of subsequent Maillard reaction, the composition and quantity of volatiles detected by the electronic nose and GC–MS indicated that MFPU pretreatment enhanced flavor of the Maillard reaction products. In conclusion, MFPU pretreatment is a promising technology for improving the degree of hydrolysis of crayfish protein and enhancing the flavor of Maillard reaction products.

1. Introduction

Crayfish, scientifically known as Procambarus clarkii, is native to the central and eastern regions of the United States. It was introduced to Japan from the United States in 1918 and spread from Japan to Nanjing, China in the 1830s [1]. In recent years, with the proliferation of human-made culture and the expansion of natural reproduction, crayfish have been distributed throughout all parts of China. Although the crayfish is an exotic species in China, it has become one of the most popular freshwater aquatic foods in the country due to its delicious meat and rich nutritional value. In 2021, the production reached 2.63 million tons in China [2].

Recently, eating crayfish has become a new consumer trend, leading to a significant growth in the crayfish industry. Seasoning, as one of crayfish processing products, is favored by consumers, which has a good market development. However, in the process of seasoning production, the enzymatic hydrolysis degree of crayfish protein is not high, and the flavor characteristic of subsequent Maillard reaction products are not good, which hinders the development of crayfish seasoning market. Physical methods including high voltage electrostatic field (HVEF), ultrasound (US) and microwave (MW), are often preferred over chemical or enzymatic treatments in food production due to their cost-effectiveness and ease of adaptation and implementation in the industry [3]. Ultrasound, as a non-thermal physical technology, involves the transmission of mechanical waves through a medium (e.g., liquid medium). It has been widely applied in the food industry, such as enzymolysis of protein [4], [5], extraction [6], drying [7], [8], modification of starch [3], inactivation of enzymes [9] as well as cleaning [10], [11].

Our previous publication reported that ultrasound can promote the enzymatic hydrolysis of protein in three ways including ultrasonic modification on enzyme, ultrasonic modification on substrate materials and ultrasonic pretreatment on enzymatic hydrolysis process [5]. In the case of ultrasonic modification on substrate materials, many of the physicochemical changes induced by ultrasound are related to alterations in protein structure. The utilization of ultrasonic pretreatment on protein-containing substrates before enzymatic hydrolysis can significantly enhance protein utilization and product quality [12]. Additionally, it can effectively reduce reaction time, thereby lowering production costs and enhancing economic efficiency. However, it is essential to ensure that the ultrasonic medium temperature, power intensity, emission frequency, reaction time, and other conditions are appropriately set to effectively promote the enzymatic reaction.

Currently, a common issue with crayfish is their poor flavor, which seriously affects their future development and marketing. The Maillard reaction refers to the condensation reaction between carbonyls and amino groups, primarily involving the carbonyls of aldehydes, ketones, and reducing sugars, as well as the free amino groups of amino acids, peptides, and proteins [13]. The Maillard reaction is the most significant type of reaction in the heating of food products to produce flavor substances. Due to its distinctive burnt flavor and color, it can mask the off odors of aquatic products, eliminate the fishy smell, and enhance both fresh and burnt flavors [14]. To the best of our knowledge, there is limited literature on the effects of multi-frequency power ultrasonic pretreatment on the enzymatic hydrolysis and subsequent Maillard reaction in crayfish.

Therefore, the current study aims to investigate the effects of multi-frequency (mono-, dual-, and tri-frequencies) power ultrasound on the enzymolysis and structural characteristics of crayfish. The underlying mechanism was also explored. Additionally, the flavor characteristics of the enzymolysis solution obtained by ultrasonic pretreatment were improved through the Maillard reaction, which provided a reference for the industrialized production of crayfish.

2. Materials and methods

2.1. Raw materials

Crayfish were purchased from the local RT-Mart supermarkets. Crayfish with uniform size, freshness, and live state were selected. The crayfish were shelled to obtain the flesh, then cleaned and stored in the refrigerator at 4 °C for standby. All reagents such as sodium hydroxide, sulfuric acid, hydrochloric acid were obtained from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China).

2.2. Experimental procedure and design

2.2.1. MFPU pretreatment of crayfish

According to our previous study, the clean crayfish flesh were mixed with distilled water at a ratio of 1:2 (m:V) and broken down using a multi-functional homogenizer [15]. The crayfish slurry was placed in a vacuum packaging bag and subjected to multi-frequency power ultrasonic pretreatment using a lab-scale MFPU machine which was self-designed by our research team (MFPU-1, Jiangsu University) [16]. The ultrasonic pretreatments were conducted in three different frequency modes including mono-frequency (20 kHz), dual-frequency (20/40 kHz) and tri-frequency (20/40/60 kHz) in the sequential working modes with the power of 250 W and duration of 40 min at 40 °C.

2.2.2. Enzymatic hydrolysis of crayfish

After ultrasonic pretreatment, alkaline protease was added to the MFPU treated crayfish slurry for enzymatic hydrolysis. The enzymatic hydrolysis were conducted for 3 h at 60 °C. The dosage of alkaline protease was 2000 U/g and pH was maintained at 8.0 for each run by 1 M NaOH. The conditions of the enzymatic hydrolysis of crayfish were optimized in our pre-experiments. After the enzymatic hydrolysis reaction was completed, the sample was promptly placed in a water bath at 95 °C for 10 min to inactivate the enzyme. Subsequently, the sample was centrifuged at speed of 5000 rpm for 10 min. The enzymatic hydrolysate was obtained for further analysis.

2.2.3. Maillard reaction

The obtained enzymatic hydrolysate of crayfish was placed in a tapered bottle with plug, and the required reducing sugar (glucose: xylose = 2:1) was added at the ratio of 5 %. The pH of the hydrolysate was adjusted to 9.0. The Maillard reaction conducted at 95℃ for 90 min in a water bath.

2.3. Analytical methods

2.3.1. Determination of the degree of hydrolysis

The enzymatic hydrolysis supernatant of crayfish protein (5 mL) was filled to 100 mL in a volumetric bottle; then 20 mL solution was poured into a 200 mL beaker and 60 mL distilled water was added. The mixture was titrated with 0.05 mol/L NaOH solution until the pH reached 8.20. Subsequently, 10 mL formaldehyde solution without precipitation were added into the mixture and stirred evenly. Then it was titrated with 0.05 mol/L NaOH solution until the pH reached 9.20. Meanwhile, 80 mL of distilled water was used for the blank. The volumes of NaOH used for the solutions were recorded.

$$\mathit{AN}(\%)=\frac{\left(V_1-V_2\right)\times c\times 0.014\times V_4}{V\times V_3}\times 100$$ (1)

where AN − the amount of amino acid nitrogen contained in the substrate; V₁ − the volume of NaOH used for the second titration, mL; V₂ − the volume of the test solution used for the blank control, mL; V₃ − the volume consumed by the dilution of the raw material, mL; V₄ − the total volume of the raw material after dilution, mL; V − volume of the test sample used, mL; c − concentration of the NaOH, mol/L; 100 − unit conversion factor.

$$\mathit{DH}(\%)=\frac{\mathit{AN}-A{N}_0}{N}\times 100$$ (2)

where DH − degree of hydrolysis; AN₀ − content of free amino acids before hydrolysis of raw crayfish protein solution; N − content of total nitrogen in the solution of crayfish protein.

2.3.2. Circular dichroism (CD)

The circular dichroism of crayfish protein with / without MFPU treatment was measured according to the methods reported by Zhang, Guan, Tang, Jiang, Sun, Manirafasha and Zhang [17] with slight modifications. 3 mL of diluted sample (0.025 mg/mL) was placed in a quartz cuvette. The scanning range was set to 190–240 nm, with a step size and bandwidth of 0.5 nm and 1 nm, respectively. The scanning speed was 100 nm/min. The measurement was repeated three times with pure water as the blank. The percentage of secondary structure was calculated using CDNN software.

2.3.3. Fluorescence spectroscopy

The fluorescence spectroscopy of MFPU treated crayfish was measured referring to the method of Xu, Zhao, Yang, Mei and Xie [18] with slight modifications. The measurements were carried out at room temperature using a fluorescence spectrophotometer (Model Cary Eclipse, Varian Inc., Palo Alto, USA) with a 10 mm optical path length and a scanning speed of 300 nm/min. The excitation wavelength was 290 nm (slit = 5 nm) and the emission wavelength range was 300–450 nm.

2.3.4. Determination of free sulfhydryl content

The free sulfhydryl content was determined according to the method reported by Cao, Sun, Huang, Zhu and Huang [19] with slight modifications. The sample solution was diluted to 4 mg/mL, and then 1 mL of the sample was mixed with 4 mL of phosphate buffer (pH 7.0). After vortexing and mixing, the reaction was conducted at 4 °C for 1 h under light protection. Subsequently, the absorbance value was measured at 412 nm. The free sulfhydryl (−SH) content was calculated as follows:

$$-SH=75.53\times A_{412}\times \frac{D}{C}$$ (2)

where 73.53 is the molar absorbance coefficient, A₄₁₂ is the absorbance value at 412 nm, D is the sample dilution, and C is the sample concentration (mg/mL).

2.3.5. Scanning electron Microscope analysis (SEM)

The sample solution was lyophilized into powder. The powder was immobilized on conductive adhesive and coated with a thin gold layer. The microstructure of the proteins was observed using a scanning electron microscopy (Hitachi S-3400 N, Hitachi High Technologies, Tokyo, Japan) at 100 × and 500 × magnifications. For the SEM observation, the acceleration voltage was 15 kV.

2.3.6. Atomic force microscopy analysis (AFM)

Images of the microstructure and morphology of the protein molecules were captured at the molecular level using atomic force microscopy. Sequentially, 5 µL of a sample (0.005 mg/mL) was dispensed as a droplet onto a clean mica sheet surface and then air-dried in a fume hood. AFM images measuring 10 μm × 10 μm were acquired using a Bruker ScanAsyst probe in peak force QNM mode. The surface roughness and height distribution of the samples were further analyzed using NanoScope Analysis V1.8 offline software. The mean roughness (R_a) and root mean square roughness (R_q) were determined according to the method reported by Xu et al. [20].

$$R_a=\frac{1}{N}\sum _{i=1}^N\left|Z_i-\overline{Z}\right|$$ (3)

$$R_q=\sqrt{\frac{\sum _{i=1}^N{\left(Z_i-\overline{Z}\right)}^2}{N}}$$ (4)

where, Z_i is the height corresponding to the ith value, Z is the average height of all points within the measurement range and N is the number of samples.

2.3.7. Determination of low molecular weight flavor intermediates formation

The Maillard reaction product was diluted 50-fold, and the absorbance value was measured at 294 nm using a UV–visible spectrophotometer (UV-1801, Beijing PUXI General Instrument Co., Ltd., China). The solution without the Maillard reaction was diluted to the same magnitude as a reference.

2.3.8. Determination of browning degree

The Maillard reaction product was diluted 20-fold, and the absorbance value was measured at 420 nm using a UV–visible spectrophotometer (UV-1801, Beijing PUXI General Instrument Co., Ltd., China). The solution without the Maillard reaction was diluted to the same magnitude as a reference.

2.3.9. TBARS and TVB-N analysis

The thiobarbituric acid reactive substances (TBARS) was determined according to the method of Zhu, Cheng, Huang, Yao, Lei, Khan, Huang and Zhou [21]. 5 g of each sample were mixed in a beaker containing 25 mL of purified water. Then, 25 mL trichloroacetic acid solution (5 %) was added and left to stand for 30 min at room temperature. The mixture was filtered and the filtrate was collected. It was then diluted with a 5 % trichloroacetic acid solution to a total volume of 50 mL. Subsequently, 5 mL of the diluted solution and 5 mL of a 0.02 mol/L thiobarbituric acid solution were thoroughly mixed and heated for 40 min in a constant-temperature water bath at 80 °C. Afterward, the solution was cooled to room temperature and the absorbance was measured at 532 nm. The TBARS value was expressed as grams of malondialdehyde per kilogram of sample. The TVB-N was determined according to the “Determination of TVB-N in food − trace diffusion method” in (GB/T 5009.228–2016) published by the Standardization Administration of China in August 31, 2016.

2.3.10. Electronic nose analysis

The electronic nose of the Maillard reaction products were measured according to method of Ning, Shasha, Pei, Weisi, Xiaoli, Jingjing and Dafeng [22]. The sample was accurately weighed into a 10 mL injection vial, and the headspace equilibrium temperature was set at 35 °C for 30 min. The conditions of the E-Nose method were as follows: injection temperature of 50 °C, air intake of 150 mL/min, determination time of 120 s, cleaning time of 100 s, and extraction of eigenvalues of 118 ∼ 120 s.

2.3.11. Headspace-solid phase micro-extraction and gas chromatography-mass spectrometry analysis

Headspace solid-phase micro-extraction coupled with gas chromatography-mass spectrometry (GC–MS) was used to analyze the volatile constituents of the Maillard reaction solution treated with different ultrasonic frequencies (20 kHz, 20/40 kHz, and 20/40/60 kHz). The solution (5 mL) was placed in a 20 mL headspace vial with 6 μL of 2-octanol (0.18 g/L; internal standard substance) and 1.8 g of sodium chloride and a magnetic rotor was added. It was sealed immediately with a polytetrafluoroethylene spacer and allowed to stand for 20 min in a constant temperature water bath at 60 °C. Once headspace equilibrium was reached, the SPME extraction needle (50/30 μm) was inserted into the headspace vial for automated extraction. The headspace adsorption was carried out for 30 min. Then the extraction needle was inserted into the gas chromatography inlet and the resolution time was 5 min.

The GC–MS analysis of the Maillard reaction products was conducted according to the method of Zhang, Ji, Peng, Ji and Gao [23] with slight modifications. It was determined by a GC–MS system (TQ8040, Shimadzu, Japan) with a Rtx-WAS column (30 m × 0.25 mm, 0.25 μm). The identifcation of volatile organic compounds was determined by comparing mass spectra with those of standard compounds using the NIST mass spectral library (NIST 17) and the NIST mass spectral search program (Version 2.2). The integration reports were accepted if matching degree was above 80 %. Confrmation of identifcation was conducted by comparing the retention indices (RI) with those of an alkane standard solution (C₇–C₃₀). Measurements were repeated three times for all samples. Quantitative analysis was performed by calculating the ratio of peak area to the peak area of the internal standard (2-octanol).

2.4. Statistical analysis

All the experiments were carried in triplicate and the results were presented as the mean and standard deviation. The data were processed using SPSS 26.0 analysis software. IBM SPSS statistics was used to perform an analysis of one-way variance (ANOVA) and the signifcance of the difference was established at 95 % confidence level (p < 0.05).

3. Results and discussion

3.1. Effects of MFPU pretreatments on the degree of hydrolysis of crayfish

The degree of hydrolysis (DH) reflects the percentage of the peptide bonds’ number cut in the hydrolysates [24]. Fig. 1A displays the DH of crayfish pretreated by MFPU with different frequency modes. The results showed that the DH of crayfish without ultrasonic treatment was 23.89 %; after MFPU pretreatments, the highest DH was MFPU-20/40 kHz (40.78 %), followed by MFPU-20/40/60 kHz (36.92 %) and MFPU-20 kHz (32.56 %). It indicates that all ultrasonic frequency modes significantly increased the DH of crayfish (p < 0.05) compared to the raw material (without ultrasonic treatment). The dual-frequency (MFPU-20/40 kHz) had the highest DH, followed by tri-frequency ultrasound (MFPU-20/40/60 kHz) and mono-frequency ultrasound (MFPU-20 kHz). These findings may be attributed to the cavitation effect intensities, which produced by varying frequency combinations and different ultrasonic frequency modes have different cavitation effect. Compared to the mono-frequency ultrasound, the dual- and tri-frequency ultrasound can generate significantly higher enhancement in cavitation yield. The multi-frequency ultrasound exhibited more resonances named “combination resonances”, which can produce a much broader range of bubble sizes resulting from the generation of more resonances in comparison with the mono-frequency ultrasound. It can be concluded from the results that ultrasonic frequency and frequency modes are important parameter of ultrasound that affect the DH values of crayfish.

Fig. 1.

Degree of hydrolysis (A), circular dichroism (B) and secondary structure content (B), fluorescence spectra (C), and free sulfhydryl content (D) of the crayfish protein with or without MFPU pretreatment.

3.2. Underlying mechanism of MFPU pretreatments on enzymolysis of crayfish

3.2.1. Circular dichroism

Circular dichroism is a type of electronic absorption spectroscopy that offers a fast and simple method to study the structure of proteins in solution [25]. The internal secondary structure can be determined from the absorption values at 190–260 nm. The circular dichroism results of various samples are depicted in Fig. 1B. All the samples exhibited a negative peak at 199 nm, indicating a clear circular dichroic feature with α-helical conformation. There is a notable variance in the absorption intensity among the samples, implying a transition between different forms of secondary structure within the protein molecule.

In order to characterize this conversion in detail, the percentage of each secondary structure was quantified using CDNN software. As shown in Fig. 1C, compared to the raw material, the α-helix content of crayfish protein decreased, while the β-folding content increased after ultrasonic pretreatment. The ultrasonic pretreatment with different frequency modes exhibited differences in the crayfish protein structure. Compared to the raw material, the MFPU-20 kHz had the lowest α-helix content, which decreased by 35 %, while the MFPU-20/40 kHz exhibited the highest β-folding content, which increased by 13 %. These changes suggested an increase in protein molecule disorder and a looser secondary structure, leading to the unfolding of the molecular structure. The alterations in β-sheet and β-turn suggest a more extended protein conformation, while the increased content of random coil indicates a further increase in disorder and looseness in the protein structure [26]. All these changes in the secondary structure suggest that enzymatic hydrolysis unfolds the protein, leading to a structure more conducive to enzymatic attack. This, in turn, further enhances protein hydrolysis, releasing more short-chain peptides [27]. Additionally, the changes in secondary structure among the different ultrasonic frequency modes were also significantly different, illustrating that ultrasound have significant effect on the protein structure by generating resonance with own frequencies of proteins [4].

3.2.2. Fluorescence spectroscopy

Since the endogenous fluorescence emitted by aromatic amino acids (i.e., tryptophan, tyrosine, phenylalanine) is present in the hydrophobic core of the protein, the intrinsic fluorescence spectrum is a useful tool for characterizing the tertiary structure of the protein [28]. The fluorescent groups within the protein molecule emit fluorescence when exposed to external ultraviolet rays. The fluorescence intensity and emission wavelength primarily rely on the polarity of tyrosine and tryptophan residues in the protein molecule, or the interactions between them. Consequently, it is possible to predict whether the tertiary structure of proteins has been altered based on changes in fluorescence intensity. As can be seen from Fig. 1C, the peak emission wavelength of the raw material was 347 nm. After sonication, the maximum absorption peak was red-shifted to around 358 nm. The red shift implies that more fluorescent groups were exposed to the surface of the protein, inducing structural unfolding of the crayfish protein molecule. This resulted in the transfer of fluorescent groups to a more polar environment, leading to alterations in fluorescence intensity and disruption of the tertiary structure of the protein. Moreover, it is shown in the Fig. 1C, the lowest value of maximum of emission peak (EX_max) was observed at the dual-frequency mode of 20/40 kHz (361 nm), followed by mono-frequency mode of 20 kHz (359 nm) and tri-frequency mode of 20/40/60 kHz (356 nm). It indicated that different ultrasonic frequencies had remarkably effects on the change in spatial conformations of the crayfish protein molecule.

3.2.3. Free sulfhydryl content

Free sulfhydryl group (−SH) is an important reactive group that affects the changes of protein properties, and it can be oxidized to form disulfide bonds, which maintains the three-dimensional spatial structure of proteins. Therefore, changes in sulfhydryl content imply changes in disulfide bond content, unfolding or clustering of enzyme structure and changes in enzyme active center. As can be seen from Fig. 1D, the free sulfhydryl content increased significantly (p < 0.05) after sonication, the higher free sulfhydryl content indicated that the sonication-induced unfolding or disruption of disulfide bonds in the natural proteins of crayfish reduced the stability of the protein molecules, which led to an increase in the free sulfhydryl content of crayfish proteins.

3.2.4. Scanning electron Microscope (SEM)

The microscopic surface morphology of crayfish proteins shown in Fig. 2 was observed by SEM. The protein structure of the native crayfish protein was large and structurally intact, with a smooth surface and an overall lumpy stacking morphology (Fig. 2A1/B1). Compared with the raw material (non-sonicated crayfish protein), the surface morphology of the sonicated proteins was significantly changed, and the protein was broken into smaller fragments with deepened folds as can be seen in Fig. 2 A2-A4 and B2-B4. These microscopic morphological changes may be due to the micro jets, shock waves and shear forces during the collapse of cavitation bubbles in the ultrasonication process, resulting in the breakage of hydrogen bonds and van der Waals forces between protein molecules. It destroyed the cross-links between protein molecules, thereby significantly altering the microscopic surface topography of the proteins. The formation of smaller fragments after MFPU pretreatment resulted in an increase in the molecular surface area of crayfish protein, thereby leading to the increase of reaction between substrate and enzyme during the subsequent enzymatic hydrolysis. Therefore, MFPU pretreatment with different frequency modes significantly increased the DH of crayfish (shown in Fig. 1A).

Fig. 2.

SEM images (100×, A and 500 × magnification, B), 3D AFM images (C) and height distribution (D) of the crayfish protein with or without MFPU pretreatment.

3.2.5. Atomic force Microscope (AFM)

In order to investigate the effect of MFPU pretreatment on crayfish proteins more deeply at the molecular level, the particle morphology, size and height of crayfish protein molecules were observed by AFM. As shown in Fig. 2C1, the crayfish proteins in the raw material had disordered aggregation and formed irregular cluster-like aggregates. After ultrasonic pretreatment, all the three groups of proteins were more evenly distributed with smaller aggregates compared to the raw material (Fig. 2C2-C4). This may be due to the cavitation effect generated by ultrasound that disrupted the aggregation between the native protein molecules, leading to consistent and orderly distribution of protein molecules. It is particularly worth mentioning that the surface microstructure of the protein under the condition of dual-frequency ultrasonic pretreatment (MFPU-20/40 kHz, Fig. 2C4) showed many slender needle structures, which increased the binding probability of substrate and enzyme in the process of enzymatic hydrolysis. This is consistent with the results of the degree of hydrolysis shown in Fig. 1A.

The roughness of the protein surface plays an important role in the enzymatic hydolysis process due to the more reaction chances between the enzyme molecules and substrates [4]. Mean roughness (R_a) and root-mean-square roughness (R_q), as indicators of surface roughness, can reflect the degree of aggregation [20], [29]. As shown in Fig. 2D, untreated crayfish proteins had the minimum surface roughness with the values of R_a and R_q of 1.77 and 5.87 nm, respectively. After ultrasonic pretreatments, the values of R_a and R_a increased significantly. This indicates that the fragmentation of protein macromolecular groups by ultrasonic pretreatment increased the surface roughness. The increase in protein surface roughness means that the increase in surface area, which is more conducive to the reaction between protein and enzyme, thereby improving the efficiency of enzymatic hydrolysis.

3.3. The absorbance values and flavor attributes of Maillard reaction crayfish products

3.3.1. Degree of formation of low molecular weight flavor intermediates and degree of browning

Numerous reactive intermediates such as glucose ketal, 3-deoxy, 3,4-dideoxy, HMF, di-reduced ketones, and unsaturated aldimines formed at the stage of the Maillard reaction can produce flavor substances such as pyrazine and imidazole rings to improve the flavor of the products. Nitrosamines are a class of brownish, structurally complex macromolecular compounds formed in the late stage of the Maillard reaction. Nigrosine-like polymers not only have antioxidant function, the most important thing is that it can bind flavor substances, thus changing the flavor of the sample, and its content can also reflect the improvement of the flavor of the sample to a certain extent [30].

The effects of the different treatments on the extent of formation of low molecular weight flavor intermediates (294 nm) and the degree of browning (420 nm) are shown in Fig. 3A. It can be seen from the figure that the formation of low molecular weight flavor intermediates and browning degree of Maillard reaction products increased significantly after enzymatic hydrolysis compared to the raw material (without enzymatic hydrolysis process). MFPU pretreatments further increased this result, especially for MFPU-20/40 kHz, which displayed the highest result. The maillard reaction promoted the formation of low molecular weight flavor intermediates and MFPU pretreatment exacerbated the formation of the products so that the maillard reaction proceeded adequately. The reason for this may be that ultrasonic pretreatment can affect the structure of food-borne peptides and confer a more desirable flavor to the products in further maillard reaction. Xu, Pan, Ma, Dabbour, Mintah, Huang, Dai, Ma and He [31] investigated the inhibitory effects of the maillard reaction (using different xylose concentrations) and sonication on the formation of cross-linked lysine alanine in silkworm pupa protein isolates, sonication promotes the maillard reaction between protein and sugar molecules. Zheng, Zhang, Liu and Liu [32] investigated the effect of ultrasonic pretreatment on the process of the Maillard reaction and found that the Maillard reaction of the samples after ultrasonic pretreatment was more intense, producing more flavor substances.

Fig. 3.

The absorbance values at 294 nm and 420 nm (A), TBARS and TVB-N content (B) of Maillard reaction products of crayfish protein with or without MFPU pretreatment.

3.3.2. TBARS and TVB-N

The degree of fat oxidation and rancidity in aquatic products is mostly reflected by the size of thiobarbituric acid value; the larger the value, the higher the degree of its own oxidation. TBARS is an important indicator to assess the quality of aquatic products, and TVB-N is an important index to determine the freshness of aquatic products. Fig. 3B shows the effect of different treatments on TBARS and TVB-N. Compared to control, TBARS values of Maillard reaction crayfish products treated by MFPU significantly (p < 0.05) increased. The MFPU-20/40 kHz exhibited the highest TBARS value, followed by MFPU-20 kHz, whereas MFPU-20/40/60 kHz was the lowest. At present, there is no clear standard for the limit of TBARS in aquatic products in China, but the relevant studies recommend that the threshold of malondialdehyde in aquatic products should not be higher than 2 mg/kg. The aquatic products may produce fishy odor when it is higher than 2.2 mg/kg, so the TBARS of the different treatments in this experiment were all in the reasonable range.

As can be also seen from Fig. 3B, the TVB-N value of Maillard reaction crayfish product without enzymatic hydrolysis (Raw material) was the highest, while it decreased significantly (p < 0.05) after enzymatic hydrolysis process (Control). This was due to the fact that the heat treatment of enzyme hydrolysis and the maillard reaction would kill part of the spoilage bacteria and inhibit the generation of volatile compounds. Additionally, MFPU treated crayfish protein further decreased the TVB-N value. The relevant aquatic product hygiene standards of China stipulate that the TVB-N of freshwater fish and shrimp ≤ 20 mg/100 g is acceptable, here the TVB-N content of the product was 16.09 mg/100 g, which is in line with the standard. In summary, the product developed in this experiment meets the relevant standards and can be used for production.

3.3.3. Electronic nose

The electronic nose can convert flavor information into digital signals. It has a small sensory threshold, which can exclude the subjectivity of sensory evaluation. The overall flavors of the Maillard reaction products with different pretreatments are shown in Fig. 4A, where the difference between the response values of W3S, W1C, W5S, W3C, W6S and W5C was very small, indicating that very few aliphatic and aromatic compounds, nitrogen oxides, ammonia, and olefinic substances were produced during the reaction. The response values of W2W, W2S, and W1W with W1S were larger, and thus had the greatest influence on the odor characteristics of the products, indicating that the samples mainly contained alcohols, sulfur compounds, and alkanes. The results of the PCA (Fig. 4B) showed that the contributions of PC1 and PC2 were 53.7 % and 37.5 %, respectively. The cumulative contribution of the two main components was 91.2 %, and the high contribution rate indicated that the electronic nose can reflect the differences of the products well. The products with different treatment conditions had no overlapping parts in the PCA plots, illustrating that the flavor substances of these samples were significantly different. The control samples were distributed farther from the other treatment groups, indicating that the flavor substances changed significantly after the heat treatment of enzymatic digestion. The samples with different ultrasonic pretreatments were distributed farther from each other group, suggesting that ultrasonic pretreatments changes the flavor substances.

Fig. 4.

Electronic nose in radar plot (A) and principal component analysis (B) of Maillard reaction products of crayfish protein with or without MFPU pretreatment.

3.3.4. GC–MS

From the results of electronic nose, it can be found that different treatments have significant effects on the flavor substances of Maillard reaction products. In order to further investigate the specific changes in volatile compounds, headspace-solid phase microextraction (HS-SPME) and gas chromatography-mass spectrometry (GC–MS) was used to qualitatively and quantitatively analyze the changes in volatile compounds of the samples with different treatments. The volatile compounds ranked in order of number (Fig. 5A) were alcohols, alkane, aldehyde, esters, ketones, pyrazine, other, furan, ethers, and acids, while ranked in order of concentration content were alcohols, aldehyde, pyridines, alkanes, lipids, other, furans, acids, ketones, and ethers (Fig. 5B). The alcohols were the most abundant, which was consistent with the electronic nose results.

Fig. 5.

GC–MS analyses in content plot (A), concentration plot (B), principal component analysis (C), partial least squares discriminant analysis (D), significance of each variable's projection scores and cross-validated results (E), hierarchical cluster analysis heatmap (F) of Maillard reaction products of crayfish protein with or without MFPU pretreatment.

In order to clarify the reasons for the differences between samples with different treatments, unsupervised PCA and supervised PLS-DA methods were used for multivariate statistical analysis of the GC–MS data. A plot of PCA scores for GC–MS volatile flavor profiles is shown in Fig. 5C. 80.4 % and 13.8 % of the total variance was explained by PC1 and PC2 for GC–MS, respectively. Fig. 5D shows the PLS-DA score plot for the volatile flavor profile of GC–MS. PC1 and PC2 of GC–MS explained 70.5 % and 22.6 % of the total variance, respectively. In the score plots, the points of the six samples were scattered and did not overlap, which was consistent with the results of the PCA analysis. The variable projection importance analysis (VIP) method of PLS-DA was used to screen different samples for important volatile flavor substances (Fig. 5E). A substance's VIP score of more than 1 was considered to be a potential marker for distinguishing differences between samples, and the higher the VIP value, the greater the difference. In Fig. 5E, 15 volatile organic compounds based on GC–MS with VIP scores > 1 were observed, which were n-octanal, 1-(1,3-dimethyl-3-cyclohexen-1-yl)ethanone, trimethylpyrazine, n-pentanol-M, isophorone, n-pentanal-D, linalool, acetic acid, propyl decalactone, iso-eudesmuscarinol, ethyl cetyl cetoate ethyl cetyl cetate, 2-heptanone, dodecanolactone, 2-hexenal, and 1,2,3,4-tetrahydro-1,1,6-trimethylnaphthalene.

Comparing with the control group, ultrasonic pretreatment significantly increased the contents of n-pentanol-M, isophorone, propyl caprolactone, and ethyl hexadecanoate in the samples. Among them, n-pentanol-M and isoeucalyptol play a key role in the formation of flavor, which can make the product have the aroma of fat; esters such as propyl caprylyl lactone and ethyl hexadecanoate are mainly produced due to the interaction between alcohols generated by the lipidation of crayfish meat and free fatty acids, which can make the product have a fruity and other sweet flavors; isophorone, which may be derived from the meladic reaction or fat oxidation, plays a non-negligible role in the formation of the meat flavor. The role of isophorone is not negligible in the formation of meat flavor. The substances with decreasing content were octanal, 1-(1,3-dimethyl-3-cyclohexen-1-yl) ethanone, trimethylpyrazine, n-pentanal-D, 2-heptanone, dodecalactone, 2-hexenal, 1,2,3,4-tetrahydro-1,1,6-trimethylnaphthalene. Among them, n-octanal, n-pentanal-D, 2-hexenal, 1-(1,3-dimethyl-3-cyclohexen-1-yl)ethanone, 2-heptanone and other aldehydes and ketones were mainly from lipid oxidation, with a low threshold and high volatility, which had an important effect on the flavor of the product, and to a certain extent can cause aquatic products to have a fishy smell; trimethylpyrazine made the product have a barbecue aroma, which had certain effect on the formation of flavor. The content of linalool and acetic acid did not change much, in which linalool would make the products have fat and rose aroma and enrich the flavor of the products; while acetic acid had a stimulating odor and brought bad flavor to the products, which might be caused by the poor processing environment. In summary, ultrasonic pretreatment can, to a certain extent, make the fishy aquatic product flavor inherent in crayfish seasoning decrease, while the fruity, floral and fatty flavors increase, thus improving the flavor of the product and increasing the richness of the flavor.

Hierarchical cluster analysis (HCA) was used to reveal the relationship between flavor compounds and Maillard reaction products (Fig. 5F). HCA color intensities were normalized to range from a maximum value of 2 (red) to a minimum value of −2 (blue), indicating high to low volatile compound content. As shown in Fig. 5F, MFPU-20 kHz, MFPU-20/40 kHz and MFPU-20/40/60 kHz can be categorized into one large group, and it can be further divided into three independent groups. This indicates that the Maillard reaction greatly changed the flavor of the samples and the different ultrasonic frequency modes (mono-, dual- and tri-frequency) also had a significant effect on the flavor of the samples, which is consistent with the electronic nose results (Fig. 4). MFPU-20 kHz, MFPU-20/40 kHz, MFPU-20/40/60 kHz, and Control can be combined into one large group, while Raw material as another group. This is probably because the small molecular peptides produced by enzymatic hydrolysis were more likely to produce flavor substances in the subsequent Maillard reaction, which made it significantly different from the other crayfish products. Moreover, compared to control MFPU pretreatment can well prevent the reduction or destruction of aromatic substances and maintain the natural flavor of the Maillard reaction products.

4. Conclusions

In this study, the effect of multi-frequency power ultrasound (MFPU) pretreatment on the enzymatic hydrolysis and Maillard reaction of crayfish (Procambarus clarkii) were investigated. It was observed that MFPU-20/40 kHz had the highest degree of hydrolysis (DH), followed by MFPU-20/40/60 kHz and MFPU-20 kHz, indicating that MFPU pretreatment significantly (p < 0.05) increased the DH of crayfish protein. This is probably because the α-helix content decreased while β-sheet content decreased after MFPU pretreatment, and ultrasonication led to the occurrence in dispersion and fragmentation of crayfish protein, enhancing the contact chance for enzyme and substrate during enzymatic hydrolysis. Moreover, MFPU and Maillard reaction could greatly promote the formation of low molecular weight flavor intermediates and browning degree. The overall flavor of the Maillard reaction products of enzymatic hydrolysis with or without MFPU was significantly improved. Meanwhile, TBARS increased significantly from 0.54 to 1.53 mg/kg, while TVB-N decreased from 18.97 to 16.09 mg/100 g, which were in accordance with the relevant standards, and therefore this experimental protocol can be safely applied in production practice.

CRediT authorship contribution statement

Weiqiang Yan: Writing – review & editing, Funding acquisition. Zhijun Chen: Writing – review & editing. Chao Zhang: Writing – original draft, Investigation. Yao Xu: Writing – review & editing. Chang Han: Writing – review & editing. Ling Yue: Writing – original draft, Investigation. Qiulian Kong: Writing – review & editing. Qi Zheng: Writing – review & editing. Wenhui Tian: Writing – review & editing. Baoguo Xu: Writing – review & editing, Supervision, Resources, 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.

Data Availability

Data will be made available on reasonable request.