Synergistic Effect of Heating pH and Transglutaminase on the Gelation Kinetics and Texture of Yak Skim Milk Gels

Textural defects (including syneresis and poor consistency) often occur in yogurt gels produced from yak milk. In this research, the synergistic effects of transglutaminase (TGase) and heating pH on the textural properties of acidified yak skim milk gels, as well as the related mechanism of action, were investigated. The pH values of yak skim milk were adjusted to 6.3, 6.7, and 7.1, respectively. The samples were heated at 80°C for 30 min and then cooled to 42°C. After treatment with different contents of TGase (0, 3, and 10 U TGase per gram proteins), the samples were acidified with glucono-delta-lactone. For a given TGase content, the final storage modulus (G′) of gels was positively related to the heating pH, whereas the opposite trend was observed for the gelation time. This effect was obvious between pH 6.3 and 6.7. At a definite heating pH value, the final G′ of the gels was highest at 3 U TGase per gram proteins. The highest water holding capacity and firmness value were observed in gels prepared using pH 7.1 and 3 U TGase per gram proteins. In the samples treated with 3 U TGase per gram proteins (preheating pH 7.1), more rigid network structures were seen in the gel than 0 or 10 U TGase per gram proteins. Therefore, adjusting the heating pH values and TGase contents is an effective way of improving the textural properties of yak milk gels.


Introduction
e yak (Bos grunniens) is a unique animal in the Qinghai-Tibet Plateau area which has an altitude of over 3000 m in western China [1,2]. Yak milk-based products have received considerable attention worldwide. Compared with cow milk, yak milk is richer in nutrients, easier to digest, and has lower allergenicity [2][3][4]. Set-type yogurt produced from yak milk has become one of the most popular dairy products in western China due to its unique sensory and flavor characteristics [5]. However, during storage and long-distance transport, textural defects often occur in set-type yogurt produced from yak milk. ese defects can be reflected in the syneresis and poor consistency of yogurt, which greatly reduce its acceptability. erefore, it is important to improve the quality of set-type yogurt produced from yak milk. e syneresis and consistency of set-type yogurt gels are closely related to the structural characteristics of gel networks [6][7][8]. e building blocks of the yogurt gel networks are milk proteins, including whey proteins and caseins [9]. In native milk, the phosphorylated serines in caseins are crosslinked by calcium phosphate, forming colloidal particles of approximately 200 nm in diameter [10,11]. ese colloids are named casein micelles. During acidification, caseins in the micelles are gradually liberated and then are rearranged into a weak network structure through noncovalent bonds at pH 4.6 [12,13]. To better improve the structural characteristics of gel networks, the introduction of covalent bonds into the yogurt gel network has been proved to be useful [14]. At present, two methods have been adapted to introduce covalent bonds: heat and transglutaminase (TGase) treatment of milk proteins.
Heat treatment before acidification has a significant impact on the gel properties [15,16]. After heat treatment, denaturation of whey proteins occurs [17,18]. It has been shown that by adjusting the pH of the milk before heat treatment, different amounts of denatured whey proteins are associated with the casein micelles [19]. When the milk pH before heating is lower than 6.7, most of the denatured whey proteins can interact with casein micelles. In this case, the denatured whey proteins are mainly in the micellar phase [20]. However, when the milk pH before heating is higher than 6.7, most of the denatured whey proteins interact covalently with the serum phase κ-casein. In this case, most of the denatured whey proteins exist in the soluble phase. It has been shown that both the final storage modulus (G′) and loss modulus (G″) are positively related to preheating pH in the range 6.2-6.9 [21]. TGase has been extensively used to produce covalent bonds among proteins. TGase catalyzes crosslinks between c-carboxyl groups and ε-amino groups in different protein molecules. Following treatment with TGase, the syneresis and poor consistency of yogurt can be greatly improved [22].
Heat treatment and TGase might have a synergistic effect on acid gels. Native whey proteins are difficult to catalyze by TGase due to their folding structure. After heat treatment, the whey proteins are denatured and become unfolded. is results in the complete unfolding of all amino acid residues, which might easily be crosslinked by TGase treatment. Numerous studies have been carried out on the use of TGase and heat treatment on the gelation kinetics and texture of acid milk gels. However, few studies have been carried out on the effect of TGase and heating pH of yak skim milk in terms of the gelation kinetics and texture of acid-induced milk gels.
In the present study, in order to improve the syneresis and poor consistency of acidified yak milk gels, we investigated the effects of TGase and heating pH of yak skim milk on the gelation kinetics and texture of acid-induced milk gels.

Sample Preparation and Characterization.
e experimental design in this study is shown in Figure 1. e sample treatment procedures and characterization methods are present in the following sections.

Preparation and Characterization of Heat-Treated Yak
Skim Milk. Yak milk was defatted by centrifugation (5-5N, Hunan Hengnuo Instrument Equipment Co., Ltd., Changsha, China) at 5000 g for 20 min at 25°C, followed by adjustment of the pH to 6.3, 6.7, or 7.1 with 2 mol/L HCl or NaOH. e samples were heated at 80°C for 30 min and then cooled to 42°C in a waterbath. e rennet and acid precipitation method in combination with reversed phase high performance liquid chromatography (RP-HPLC) was used to evaluate the distribution of whey proteins, as reported previously [19,24]. In this method, the heat-treated samples were first acidified with 2 mol/L HCl. irty minutes later, the dispersion was centrifuged at 5000g for 30 min at 25°C. e concentration of β-lactoglobulin in the supernatant was analyzed by RP-HPLC (Agilent 1100 Series, CA, USA) with a C4 column (4.6 × 250 mm, 300Å, Phenomenex, CA, USA). Before the RP-HPLC determination, 0.3 mL of the supernatants was first added to 2.5 mL of reducing agents. e reducing agents contained 0.1 mol/L Tris, 6 mol/L urea, and β-mercaptoethanol (0.4%, w/v), followed by adjusted to pH 7.0 with 2 mol/L HCl. Renneting of the heat-treated milk samples was carried out based on the reported literature [25]. Following the addition of rennet for 30 min, the mixture was centrifuged at 5000g for 30 min at 25°C. e concentration of β-lactoglobulin in the supernatant was analyzed by RP-HPLC. Before the RP-HPLC determination, 0.3 mL of the supernatants was first added to 2.5 mL of reducing agents. Solvent A was distilled water (with 0.1% (v/ v) trifluoroacetic acid); solvent B was acetonitrile (with 0.1% trifluoroacetic acid). A linear gradient from 35.0% to 55.0% of solvent B over 60 min was used.

Preparation and Characterization of TGase-Treated
Samples. After heat treatment, TGase was added to the samples of 0, 3, and 8 U/g milk proteins. e samples were magnetically stirred for 5 min and incubated at 42°C in a waterbath for 40 min. e TGase in the samples was inactivated at 75°C for 10 min. e pH values in the samples were readjusted to 6.7 before GDL addition, with 2 mol/L HCl or NaOH.
e degree of covalent crosslinking in the heat-and TGase-treated samples was measured by a spectrometer, as reported previously [26]. Briefly, 10 mL of the heat-and TGase-treated samples were freeze-dried (FD-1A-50, Hangzhou Chuanyi Experimental Instrument Co. Ltd., Hangzhou, China) to constant weight for 48 h. One milliliter of 4% NaHCO 3 solution and 1 mL of 0.2% (w/v) trinitrobenzene-sulfonic acid solution were added to 10 mg dried samples. After incubation at 42°C for 4 h, the samples were digested with 3 mL 6 mol/L at 65°C for 2 h. Following dilution with distilled water to 15 mL, the samples were added to cuvettes for absorbance determination (U-2900 spectrometer, Hitachi, Ltd., Tokyo, Japan). e crosslinking degree (DE) was calculated as follows: where A S and A NS is the absorbance of TGase-treated and non-TGase-treated samples, respectively. m s and m Ns is the mass of TGase-treated and non-TGase-treated samples, respectively.

Preparation and Characterization of Acidified Milk
Gels. e yak skim milk was acidified with 1.35% GDL at 37°C for 6 h. e final pH value of all the acidified skim milk gels was 4.4. e samples were stored at 4°C before use. e gelation kinetics of the samples were determined using a rheometer (AR 2000, TA Instruments, USA), equipped with a concentric cylinder. GDL was directly added to the TGasetreated samples to 1.35%. e samples were then added to the concentric cylinder after the addition of GDL for 2 min and were oscillated at 0.1 Hz and 37°C with a strain of 1%. Gelation time was defined as the point when G′ of the samples was more than 1 Pa. e firmness of the gels was measured by a texture analyzer (TMS-Pro, Sterling, USA). e gels were stored at 25°C for 120 min before texture measurement. A probe (with a diameter of 25 mm) was vertically moved into the samples to 15 mm at 25 mm/min for penetration measurement. e water holding capacity (WHC) of the acidified milk gels was measured according to a modified method [24]. Eight millimeters of the TGase-treated samples were acidified for 4 h at 37°C in 10 mL tubes. After acidification, the samples were centrifuged at 800g at 25°C for 10 min. e WHC was calculated as the percentage of the gel weight at the bottom of the centrifuge tubes compared to the initial weight. e structural characteristics of the gels were determined by cryoscanning electron microscopy (S-3000N, Hitachi Co., Tokyo, Japan). Samples were added into the specimen holder and sublimated at −90°C for 25 min before observation (Quorum PP 3000T, UK).

Statistical Analysis.
Independent experiments were repeated three times. Analysis of variance (ANOVA) was used to determine significant differences (P < 0.05). Statistical analyses were carried out using IBM SPSS 21 for Windows 10.0. Duncan's multiple range tests for differences were performed.

Extent of Denaturation of Whey Proteins and eir Distribution.
e distribution of whey proteins can be evaluated with the rennet and acid precipitation method.
is is because the casein micelles can be precipitated in the presence of rennet, while both casein micelles and denatured whey proteins can be precipitated at pH 4.6. e distribution of whey proteins including native whey proteins, whey protein aggregates in the soluble phase, and whey proteins associated with the casein micelles were determined, respectively. Table 1 gives the distribution of the highest content of whey protein in yak milk, β-lactoglobulin [27], after different heat treatments. No statistical differences were observed in the degree of undenatured β-lactoglobulin. is indicated that the degree of denaturation was little influenced by the heating pH of yak skim milk. However, the degree of denatured β-lactoglobulin present in the soluble or micellar phase was closely related to the heating pH of yak skim milk. When heated at pH 6.3, only 18.5% of β-lactoglobulin was present in the soluble phase; this value increased to 48.2% when the heating pH was 7.1.

Gelation
Kinetics. e evolution of the storage modulus (G′) after the addition of GDL to skim milk samples treated with different preheating pH values and TGase contents is presented in Figure 2. G′ and gelation time are summarized in Table 2. Preheating pH had a significant impact on the final G′ and gelation time (except for the case for 0 U/g protein). For a given TGase content, the final G′ was positively related to the preheating pH as the preheating pH increased from 6.3 to 7.1, whereas the opposite trend was observed for the gelation time. is effect was obvious between pH 6.3 and 6.7 but was smaller between preheating pH 6.7 and 7.1.
ese results are generally consistent with previously reported results [26].
It was observed at a definite preheating pH value that the TGase content had a significant impact on the final G′ and gelation time. Surprisingly, when the preheating pH was 6.7, the maximum G′ value was observed in the sample treated with 3 U TGase per gram proteins, followed by 10 U TGase per gram proteins. Similar results were also observed when the preheating pH was 6.3 or 7.1. In summary, at a definite pH value, the maximum G′ value was observed in the 3 U TGase per gram proteins sample, whereas the opposite trend was found for the gelation time. During acidification, caseins in the micelles are gradually liberated and rearranged into a weak network structure through noncovalent bonds at pH 4.6. Although the introduction of covalent bonds is a useful Skim milk

No heat
Heat treatment at pH 6.3, 6.7, and 7.1 TGase treatment (0, 3, and 10 U/g protein) Acid-induced skim milk gels Acidified Figure 1: e experimental design of the study. Native, soluble phase, and micellar phase are represented by native whey proteins, whey protein aggregates in the soluble phase, and whey proteins associated with casein micelles, respectively. Values are mean ± standard deviation; means with different superscript letters within the same column are significantly different (P < 0.05). method for improving the textural properties of the yogurt gel network, excess crosslinking of caseins in micelles can inhibit the adequate rearrangement of caseins during gelation [22]. is might explain why the samples treated with 3U TGase per gram proteins were higher than the samples treated with 10 U TGase per gram proteins.

Water Holding Capacity and Firmness.
e firmness of the gels is given in Table 3. For the samples treated with 0 or 3 U TGase per gram proteins, firmness of the final gels was higher at preheating pH values of 6.7 or 7.1 than at 6.3, whereas no significant difference was observed between pH 6.7 and 7.1. However, when the samples were treated with 10 U TGase per gram proteins, no significant differences were observed among different preheating pH values. When the preheating pH values were definite, the TGase contents also had a significant impact on gel firmness. When the heating pH was 6.7, the maximum firmness value was seen in the sample treated with 3 U TGase per gram proteins. Similar results were also observed when the preheating pH was 7.1. At a definite pH value, maximum firmness was observed at 3 U TGase per gram proteins. e WHC of the gels prepared with different preheating pH and TGase contents is given in Table 4. When the preheating pH values were definite, the WHC of samples treated with 0 and 10 U TGase per gram proteins showed no significant difference but were lower than the samples treated with 3 U TGase per gram proteins. When the TGase contents were 0 and 3 U TGase per gram proteins, the WHC of the samples treated with preheating pH 6.7 or 7.1 was higher than that with preheating pH 6.3, while the samples treated with pH 6.7 and 7.1 showed no significant difference. For the samples treated with 10 U TGase per gram proteins, the samples with different pH treatments showed no significant differences.

Microstructure and Crosslinking Degree.
e microstructure of acid-induced yak milk gels prepared from yak skim milk with preheating at pH 7.1 and different TGase contents are shown in Figure 3. In the samples treated with 3 U TGase per gram proteins, more rigid network structures were observed in the gel than 0 or 10 U TGase per gram proteins. is was consistent with the results of the textural properties. e crosslinking degrees of proteins (catalyzed by TGase) in milk samples heated at pH 6.3, 6.7, and 7.1 were 15.3 ± 3.7%, 33.2 ± 5.2%, and 38.1 ± 4.7%, respectively. is indicated that the number of covalent bonds introduced was positively related to the heating pH values. As mentioned above, more denatured whey proteins were present in the micellar phase when the heating pH of yak skim milk was lower, together with the fact that casein cannot be dissociated from the micelles [28]. is indicated that the attached whey proteins with casein micelles might hinder crosslinking between caseins, and thus, the crosslinking sites between proteins are limited. On the contrary, when the heating pH of yak skim milk was higher, dissociation of micelles may occur, and more sites catalyzed by TGase might be exposed, together with the fact that denatured whey proteins in the soluble phase cannot attach onto the surface of casein micelles which might hinder crosslinking. erefore, the crosslinking degree of milk proteins (catalyzed by TGase) was higher when the yak skim milk pH value was high. Values are mean ± standard deviation; means with different superscript letters within the same column are significantly different (P < 0.05). Table 3: Effects of preheating pH and TGase contents on the firmness of acid-induced yak milk gels.

Conclusions
In this study, we investigated the effects of TGase and heat treatment of yak skim milk on the gelation kinetics and texture of acid-induced milk gels, as well as the related mechanism of action. For a given TGase content, the final G′ was positively related to the preheating pH when the preheating pH increased from 6.3 to 7.1, whereas the opposite trend was observed for the gelation time. e yak skim milk treated with a preheating pH value of 7.1 and 3 U TGase per gram proteins demonstrated the highest WHC and firmness and more rigid network structures.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that there are no conflicts of interest.