Application of high-intensity ultrasound to modify the rheological properties of a guamuchil Pithecellobium dulce (Roxb.) seed protein isolate

ABSTRACT In the present study, the application of high-intensity ultrasound (USon) on the rheological properties of guamuchil seed protein isolate (GPI) was evaluated. GPI dispersions (30% w/v) were subjected to USon treatments (20 kHz; 200, 400, and 600 W) for 15 and 30 min. Strain, frequency, and temperature sweep tests showed that the storage (G) and loss (G”) moduli increased, while loss tangent (Tan δ) decreased by USon. The results showed a dominance of G’ over G”, indicating the formation of a stronger and more elastic dispersion due to USon, especially at high power (600 W for 15 and 30 min). Compared to native GPI, the USon treatments that caused the largest decreases in apparent viscosity (ηapp ) and the most notable shear thinning behavior were at low (200 W) and high (600 W) power. The modification of the rheological characteristics of GPI by USon could be associated with structural changes, which modifies the inter and intramolecular interactions. The rheologically improved GPI by USon could represent an alternative to diversify its possible use as a new food ingredient.


INTRODUCTION
The global demand for high-quality protein is increasing due to population growth and people's interest in protein-rich foods. [1]At present, food proteins in the human diet are mainly derived from plants and animals.In the case of plant proteins, many sources as by-products of the processing of oilseeds, legumes, cereals, and fruits have been recovered as protein isolates to be used in food products depending of its functional quality. [2]Nevertheless, the production processes of protein isolates can affect the functional properties and thus limit their use in the food industry. [3]he extraction process to produce protein isolates can have negative effects on the functional and nutritional properties.Exposure to solvents, heating, dehydration, homogenization, drying, and other treatments can change the structure, composition, conformation, and/or surface charge of the proteins, which together modify the functionality of these biopolymers. [4]Among the most widely used extraction methods are alkaline extraction-isoelectric precipitation, salt extraction-dialysis, ultrafiltration, and micellar precipitation. [5]Alkaline extraction followed by isoelectric precipitation is the most common method used in the food industry to obtain proteins due to its simplicity, rapidity, and low cost. [6]However, to improve yield, harsh conditions applied in this process can cause its characterization and find their potential use as food ingredient, as is the case of rheological behavior.Therefore, the objective of this work was to evaluate the effect of USon on the flow behavior and viscoelastic properties of GPI.

Materials
GPI dispersions (10% w/v) were prepared at pH 7, adding GPI in distilled water and 0.1 N NaOH, by magnetic stirring for 30 min.An ultrasound system model CPX750 (Vernon Hills, Illinois, U.S.A.) provided with a titanium probe (2.54 diameter) was employed to sonicate the GPI dispersions in a 1 L glass beaker, which was placed in an ice-water bath to maintain the temperature below 15°C.The USon treatments were generated from at frequency of 20 kHz and power levels of 200, 400 and 600 W for 15 and 30 min (pulse time: on time 5 s, off time 1 s), in addition to a control treatment without USon.The exposed treatments at 200 W, 400 W and 600 W power received an ultrasonic intensity of 36-38 W/cm −2 , 54-57 W/cm −2 and 107-109 W/cm −2 , respectively, which was measured by calorimetry. [2]Finally, all above protein dispersions were freeze-dried and stored at 4°C in airtight containers.The GPI was characterized according to, [26] methods, which presented the next proximal composition: 75.17-76.40%protein, 0.97-1.13%fat, 1.19-3.84%moisture, 2.28-2.72%ash, and 17.55-20.12%total carbohydrates.

Rheological characterization
Rheological measurements were carried out using a controlled stress rheometer, according to the method of, [27] with some modifications as are described below.GPI dispersions (30%, w/v) were prepared in distilled water, stirred for 3 h at 25°C and then allowed to stand for 24 h at 4°C.All rheological measurements were performed at 25°C using an equipment HR-1 (TA Instruments, Delaware, USA) provided with plate-cone geometry (60 mm diameter and 64 μm gap), and a Peltier temperature control system.The storage (G') and loss (G") moduli values were collected and analyzed with a TRIOS software, version 4.0 (TA Instruments, USA).The following sequence of measurements was done.
Strain sweep: The GPI dispersions were subjected to an amplitude strain in a range of 0.5 to 100% under a constant frequency of 1.0 Hz.Strain sweep analyses were performed in order to determine the linear viscoelastic region (LVR), where G' and G" are independent of amplitude strain. [28]requency sweep: Frequency sweep analyses of the GPI dispersions were performed at a range of 0.1 to 10 Hz.The used value for the frequency sweep depended on the strain sweep (Section 2.2.1).
Temperature sweep: Temperature sweep analyses of the GPI suspensions were performed in the temperature range of 25 to 95°C with a scan rate of 5°C/min.The loss tangent or phase angle (Tan δ) was recorded as a temperature function.
Apparent viscosity (η app ): η app of GPI dispersions was determined under a shear rate range (γ) of 0.1-100 s −1 for 4 min.Then, the experimental data of η app were fitted to Cross model to describe the flow behavior, as expressed in Equation 1. [29] η where η is the shear viscosity (Pa•s), η ∞ is the infinity shear viscosity (Pa•s), η 0 is the zero shear viscosity (Pa•s), C is the Cross time constant (s), m* is the Cross rate constant (dimensionless), and γ represent the shear rate (s −1 ).

Statistical analysis
All rheological measurements were performed in triplicate and the resulting values expressed as means ± standard deviations.The experiment results were subjected to one-way analysis of variance using the StatSoft Statistica version 7.1 (TIBCO Software, Inc. California, USA).Significant differences (p < .05) between treatments were determined using Fisher's test.

Strain sweep
Strain sweep provides information on linear viscoelastic region (LVR) and protein structure stability. [30]The G' indicates the amount of energy that is stored in the material in each deformation cycle, while G" is an indication of the amount of energy dissipated as heat. [31]Figure 1A-B shows the effect of USon on the G' and G" moduli of GPI dispersions in function of strain sweep.For all GPI dispersions, G' (Figure 1A) was higher than G" (Figure 1B) over the strain range tested, except for 0W, indicating that USon formed a more elastic dispersion (solid nature), while the control treatment (0W) formed more viscous dispersions (fluid character).The obtained results reflect that GPI treated by USon was stronger and more structured compared to non-treated GPI. [32]lso, for all USon treatments, G' and G" remained almost constant as strain increased and then suddenly decreased (Figure 1A-B).With basis in this trend the LVR was defined in the range of 0.5-2.5 strain (%) for sonicated GPI dispersions, while for the control treatment (0 W) was of 0.5-0.8strain (%).Therefore, USon increased the LVR range of the GPI, probably due to the enhancement of particle interactions, which produce better stabilization of the guamuchil proteins. [33]On the other hand, different trends on the G" was observed by effect of USon on the diverse treatments (Figure 1B).The limit of the LVR of 0 W, 200 W/15 min and 200 W/30 min treatments was 2%, while that for 400 W/ 15 min, 400 W/30 min, 600 W/15, and 600 W/30 min treatments the limit of LVR was 15%.
A previous study by, [34] in USon-treated chicken breast myofibrillar proteins had strain sweep changes similar to those observed for GPI dispersions from this study, where both LVR region range and G' values increased with ultrasonic treatment.On the contrary, a reduction on G' was observed on soy and porcine myofibrillar proteins by USon effect according with studies reported by [30,23] respectively.This could be a consequence of the partial denaturation of proteins, which decreases intermolecular interactions between protein molecules because of ultrasound.

Frequency sweep
Figure 2A-B shows the USon effect depending of the frequency on the G' and G" moduli of GPI dispersions.As can be observed, the increase of the frequency and the USon treatment provoked a higher augment of the G' and G" values in comparison than control treatment, same behavior that have been observed in protein dispersions of album [35] and quinoa. [36]urthermore, in this study, G' showed higher values in comparison with G" on studied frequency range, suggesting the dominance of elastic components over viscous components of GPI dispersions in the oscillation process (Figure 2A-B).In particular, 0 W, 200 W/15 min, and 400 W/30 min treatments had a cross-over point between G' and G" at frequencies of 20, 63, and 25 Hz, respectively.However, as the frequency increased, the values of G" also arose and exceeded the G' values, indicating that USon treatment modified the viscoelastic properties of the proteins, due probably to conformation changes of the molecules. [37]n this study, the values of G' and G" of the GPI dispersions treated with USon were higher than the control treatment (0 W), being the maximum and minimum values for both rheological parameters those corresponding to 600 W/15 min and 200 W/30 min treatments, respectively.The G' values at 0.1-63 Hz for 0 W ranged from 35.5 to 55.0 Pa, while for 200 W/30 min from 59.2 to 214.7 Pa, and for 600 W/15 min from 94.7 to 417.3 Pa (Figure 2A).In the case of G", its values ranged from 3.8 to 38.4 Pa for 0 W, 41.7 to 140.3 Pa for 200 W/30 min, and 78.9 to 236.6 Pa for 600 W/15 min (Figure 2B), which indicate that the USon treatment enhanced the elastic and viscous properties of the GPI dispersions.
In contrast [23] confirmed that the G' and G" values became smaller as USon [200 W; 15, 30, 60 and 120 min) treatment time increased, which may be because USon treatment induces protein unfolding and weakens non-covalent bonds, such as intermolecular hydrogen bonds.
In addition, [38] observed that the USon treatments (200, 400, 600, 800, and 1000 W for 15 min] induced a reduction in the G' and G" values of chicken myofibrillar proteins with the increase USon power, which suggests that the diminishing in both moduli depend on the protein type and sonication conditions.
The same behavior of the moduli where G' > G" after of the application of USon was observed in sunflower, [39] chickpea, [21] and date palm pollen [18] proteins.However, these authors not observed cross-over points between G' and G" in the frequency sweep.These studies determined that ultrasonic cavitation produces molecular unfolding of the proteins, resulting a smaller particle size and greater surface hydrophobicity, higher protein solubility, as well as secondary structure modification, leading to increase intermolecular interactions between functional groups of protein molecules. [27,40]n contrast, [41,38] observed that the USon treatment induced a reduction in the G' and G" values of the whey protein isolate and chicken myofibrillar proteins, respectively, which suggests that the diminishing in both moduli may depend on the protein type and sonication conditions.

Temperature sweep
Knowledge of rheological behavior of protein isolates with increase in temperature can be used to control the mouth feel and texture of various food products such as yogurts, pudding and sausages. [39]igure 3A-C shows the effect of USon on G', G" and Tan δ of the GPI as a function of temperature (from 25 to 95°C).In general, G' and G" of GPI augmented with the increase of the temperature (Figure 3A-B).First, G' and G" of GPI slightly increased up to approximately 65°C, and then both moduli increased sharply up to the maximum temperature of 95°C.In addition, G' was predominantly larger than G" throughout the whole measurement, exhibiting both moduli a similar pattern, which implies a system mainly elastic gel-like structure.These results are in agreement with those previously discussed in the frequency sweep section (3.2).
As is shown in Figure 3A-B In addition, the G' values in the temperature range of 25-95°C for 200 W/30 min and 600 W/ 15 min had increases of 27.6-96.7%and 200.8-196.4%,respectively, in comparison with 0 W, while in the case of G", its values augmented from 27.7% to 129.9% for 200 W/30 min, and 143.1-215.6%for 600 W/15 min, in comparison with 0 W (Figure 3B), which demonstrated that USon had a high impact in such rheological parameters.The augment in G' during heating could be explained by the effect of mechanical waves of USon, with the consequent unfolding of the proteins where more functional groups (hydrophilic and hydrophobic residues) were exposed, creating new bonds (hydrogen, hydrophobic, disulfide, and electrostatic) which led to protein aggregates and a network formation. [42,43]he results of this study with GPI about the gelation patterns and the impact of USon on the increase of G' and G" of GPI were similar to those obtained by [35,44,40] for albumin, beef myofibrillar and tuna myofibrillar proteins, respectively.
On the contrary, [45,18] observed different gelation patterns between G' and G" in field pea protein isolate and palm pollen protein concentrate, respectively.According to these studies were observed crossover points between G' and G" for field pea protein isolate at 84.0°C, while that for palm pollen protein concentrate at 53.1°C, indicating the specific temperatures of gel formation for each protein source.
According to, [27] proteins that form strong gels are due to the high percentage of β-sheet.β-sheets oriented in parallel or antiparallel direction have a larger contact surface compared to α-helix and random coil secondary structures and thus can enhance intermolecular hydrogen bonds, which could stabilize the gel network.However, no gel formation was observed in the GPI.In fact, according to, [45] a high β-sheet content induces greater thermal stability, and a higher temperature is then required to ensure the gelation of the protein.According to, [2] the β-sheet content of guamuchil proteins is the predominant secondary structure, which could favor its thermal stability as was observed in the temperature scanning test (25 to 95°C), resulting in a delay of the clear appreciation of the formation of a gel, probably because the protein denaturation temperature is above the temperature range studied (Figure 3A-B).
On the other hand, Tan δ is a parameter that is used to obtain information about the structure of gels, emulsions and dispersions.Tan δ indicates the ratio of G' and G", which determine the contribution of the viscous and elastic components to the final structure. [42]High values of Tan δ (>1) indicate the formation of weaker structures (more viscous) and low values of Tan δ (<1) indicate the formation of stronger structures (more elastic). [17]he impact of USon on Tan δ of GPI is shown in Figure 3C.All sonicated treatments showed a significant reduction (p < .05) in Tan δ with respect to 0 W, indicating the formation of strong and elastic dispersions.However, the largest reduction of Tan δ, in the temperature range of 25-95°C was 36.0-48.6%,respectively, for 600 W/15 min, while the smallest decrease of 16.0-28.6%was for 200 W/ 30 min, compared to 0 W (Figure 3C).According to, [35] the reduction in Tan δ values is due to the fact that USon generates a greater interaction between protein molecules, resulting in a larger capacity to form hydrogen bonds, disulfide bridges, hydrophobic and electrostatic interactions, which was observed in this study.
In a study with black bean protein isolate, [46] it has been reported that low (150 W) and high (450 W) power USon treatment augmented the strength of the formed gels (Tan δ < 1), because of cavitation produces unfolding of the proteins and the subsequent formation of new inter and intramolecular interactions as occurred in this study.

Apparent viscosity (η app )
η app of the non-Newtonian fluids is very important to understand their behavior in any process involving fluid flow such as extraction, purification, filtration, and extrusion.This property is defined as the ratio of shear stress to shear rate at a specified shear rate. [47,48]The viscosity of protein suspensions is related to the functional groups, molecular weight, shape, size, charge, molecular flexibility of said polymers, and their interaction with solvents. [49]s can be seen in Figure 4, the η app of GPI of all treatments decreased significantly (p < .05)with the increase of the shear rate, as well as with USon application, which is one of the typical shear thinning behavior of the non-Newtonian fluids. [50]The higher and lower differences between ultrasonicated treatments and control in η app at 0.1 s −1 were observed for 600 W/30 min (92.5%) and 200 W/15 min (6.7%), respectively, while that for these same treatments at shear rate of 100 s −1 the η app was reduced 95.5% and 18.3% (Figure 4).
Shear thinning behavior can be explained by a breakdown of intra and intermolecular interactions during shear rate, since as shear rate increases, chains of protein molecules become oriented and aligned in the direction of flow, resulting in a decrease in internal friction and resistance, such as was observed in this study. [51]According to, [52] USon increase the random coil content that led to an important shear thinning behavior.The augment in this structure type by the effect of USon was shown in our previous study. [2,15]reported that USon cavitation produces a certain degree of molecular unfolding of the proteins, which causes a better ordering of the proteins along the flow field, exhibiting less resistance to flow, which results in a lower viscosity as observed in Figure 4.
The behavior of the η app reduction by effect of USon on GPI is in agreement with results obtained by [20,53] for soybean protein isolates and [54] for a whey protein isolate [55] studied the rheology of soybean protein isolate solutions, finding that as the USon treatment [20 kHz for 30, 60, and 120 s) time increased, the η app significantly decreased with respect to the control sample, which could be due to the increase in the solubility of the protein. [56]also observed that the longer sonication time [450 W for 5, 10 and 15 min] reduced the η app of chicken breast meat protein isolate, which could be due to the diminishing in particle size and surface charge by effect of sound waves on proteins; this behavior is typical of shear thinning materials where the η app decreases as the shear rate increases.
On the contrary [57,58] confirmed that USon treatment increased the η app of whey protein isolates, which can be attributed to the cavitation effect that causes the unfolding of the proteins, making them less compact and more resistant to flow.On the other hand [49,59] did not found changes in η app by effect of USon on whey and lactoalbumin protein isolates, respectively, which may be due to the particular characteristics of each protein sources and the specific conditions of ultrasound treatment. [60]able 1 shows the effect of USon on the rheological parameters of GPI according to the Cross model.Each of these rheological parameters have a special means.η 0 provides information on the stability of suspensions and emulsions, η ∞ is related with aspects of high shear processing, m* measure the degree of dependency of viscosity on shear rate in the region of shear thinning, and 1/C determine the onset of shear thinning region. [17] significant reduction (p < .05) in η 0 ,η ∞ and 1/C parameters at higher values of γ was observed in the sonicated treatments with respect to 0 W. 600 W/30 min exhibited the greater decrease in η 0 (0.001 x10 6 Pa•s), η ∞ (0.011 Pa•s), and 1/C [0.010 s −1 ) with respect to 0 W. The reduction in the I/C parameter with respect γ is an indicator of onset of shear thinning behavior, which were observed earlier for the GPI treated with USon than control treatment.According to a report of [21] from chickpea protein isolate, USon cavitation (high power] accelerated the collision of aggregates and depolymerized the larger protein aggregates into smaller protein particles, exhibiting less resistance to flow, which resulted in lower values of η 0 ,η ∞ and 1/C as observed in this study (Table 1).On the contrary, the m* parameter augmented in the GPI by the application of USon in comparison with 0 W (0.900), being 600 W/30 min the treatment that exhibited the higher increment (1.335), which confirmed the shear thinning behavior (Table 1).Knowledge of these rheological properties in proteins systems could be of utility for improving or modifying the texture or during high shear processing operations such as pumping and filling. [61]

Conclusion
USon treatment changed the rheological properties of GPI, which altered its viscoelasticity properties according to G', G", and Tan δ parameters, as was revelated by the strain, frequency and temperature sweeps, as well as the flow behavior denoted by η app , especially at high sonication power (600 W).The USon provoked stronger and more elastic GPI dispersions.At both low (200 W) and high (600 W) sonication powers, primarily, the onset of the shear thinning behavior of GPI dispersions according to the Cross model.Modification on rheological properties of GPI could be the consequence of ultrasonic cavitation, possibly due to the alteration on the structure, conformation and particle size of guamuchil proteins derived from changes on the intra and intermolecular interactions and the subsequent unfolding of proteins by effect of ultrasonic waves.USon is a useful technology for modifying the rheological properties of GPI for the production of guamuchil proteins with improved rheology functionality for specific applications in foods.Studies on the sensory characteristics related to rheological properties of the GPI are need to diversify its potential applications in the food industry.

CRediT authorship contribution statement
Nitzia Thalía Flores-Jiménez: Investigation, Methodology, Writing-original draft preparation & editing.José Armando Ulloa: Funding acquisition, Investigation, Writing-original draft, Writing-review & editing, Conceptualization, Supervision, Data curation.Rosa Isela Ortiz-Basurto: Equipment, Software, Conceptualization, Supervision, Visualization.Judith Esmeralda Urías-Silvas: Software, Resources, Supervision, Validation.Results are expressed as the average of triplicates ± standard deviations.Different superscripts within the same column are significantly different (p < 0.05).0 W is the control treatment without ultrasound.For each treatment the numerator and denominator represent the power and time of exposure to ultrasound, respectively.

Figure 1 .
Figure 1.Impact of high-intensity ultrasound on the moduli of storage G' (A) and loss G" (B) of guamuchil protein isolate in function of strain sweep.For each treatment the numerator and denominator represent the power and time of exposure to ultrasound, respectively.

Figure 2 .
Figure 2. Impact of high-intensity ultrasound on the moduli of storage G' (A) and loss G" (B) of guamuchil protein isolate as function of frequency sweep.For each treatment the numerator and denominator represent the power and time of exposure to ultrasound, respectively.
, all USon conditions significantly increased (p < .05) the G' and G" values of GPI with respect to 0 W. The highest and lowest values of G' and G" of sonicated GPI were for the 600 W/15 min and 200 W/30 min treatments, respectively (Figure 3A-B).In addition, the G' values in the temperature range of 25-95°C for 0 W varied from 195.5 to 10,537.9Pa, while for 200 W/ 30 min from 249.4 to 20,731.6 Pa, and for 600 W/15 min from 588.1 to 31,235.6 Pa (Figure 3A), which means augments of 27.6-96.7%and 200.8-196.4%,respectively.

Figure 3 .
Figure 3. Impact of high-intensity ultrasound on the moduli of storage G' (A) and loss G" (B), and loss tangent of guamuchil protein isolate as function of temperature sweep.For each treatment the numerator and denominator represent the power and time of exposure to ultrasound, respectively.

Figure 4 .
Figure 4. Effect of ultrasound on the apparent viscosity (η app ) of guamuchil protein isolate as function of shear rate (γ).