Enzymatic hydrolysis of hydroxypropyl distarch phosphate (HPDSP) by α-amylase and its effects on the rheological properties of pasteurized yogurt

Abstract

Residual α-amylase activity in stored pasteurized yogurt hydrolyses hydroxypropyl distarch phosphate (HPDSP), threatening product quality and motivating this study to establish safe enzymatic thresholds for texture stability. Bacillus licheniformis α-amylase (HEa) demonstrated superior thermal stability with a static half-life 5–10 times longer than that of Aspergillus oryzae (ASa) at 85–100 °C. HEa effectively hydrolysed HPDSP, reducing starch residue to 77 % at ultra-trace concentrations of 1.0 × 10⁻⁴ U/L, respectively. Conversely, ASa exhibited negligible hydrolysis even at a concentration of 1 U/L. Rheological analysis revealed that elevated α-amylase concentrations significantly decreased yogurt viscosity and yield stress. However, controlled enzyme levels (ASa ≤ 1.0 × 10⁻¹ U/L or HEa ≤ 1.0 × 10⁻³ U/L) maintained optimal viscosity and shear-thinning behaviour throughout storage. These findings highlight the importance of stringent α-amylase level control to maintain consistent texture and quality in pasteurized yogurt products during storage.

Highlights

• •The type of α-amylase, temperature and pH, greatly affect hydroxypropyl distarch phosphate hydrolysis efficiency.
• •Heat-stable Bacillus licheniformis α-amylase needs higher temperature and longer times for inactivation.
• •Residual hydroxypropyl distarch phosphate in pasteurized yogurt depends on the type and concentration of α-amylase.
• •Even trace α-amylase during processing alters yogurt's rheological properties.

1. Introduction

Pasteurized yogurt, often referred to room-temperature yogurt, offers a practical and convenient alternative to refrigerated yogurt, owing to its extended shelf life of 4–6 months under ambient conditions. For healthy adults, pasteurized yogurt provides comparable immune benefits to fresh yogurt for healthy adults (Rivero-Pino et al., 2024). This prolonged shelf life is primarily attributed to the post-fermentation pasteurization process (Skryplonek et al., 2019).

Pasteurization effectively inactivates microorganisms while preserving residual enzymatic activity (Didouh et al., 2023). In donor milk, pasteurization completely eliminates lipase activity while retaining approximately 85 % of the initial activity of amylase (Henderson et al., 1998). Although extensive studies have demonstrated that pasteurization at 62.5 °C for 30 min effectively reduces the activity of both lipase and amylase in human milk (Robichaud et al., 2021), its effects on residual enzyme activity, particularly α-amylase [EC3.2.1.1] in yogurt, have not been systematically investigated.

α-Amylase is ubiquitously present in pasteurized yogurt additives (Dhital et al., 2017; Rebholz et al., 2021; Shad et al., 2023). It may originate from multiple sources such as intentional addition during process, endogenous production by plant or animal tissue, or microbial synthesis, notably from bacteria [e.g., Bacillus subtilis (Chunxiao et al., 2023)], fungi [e.g., Thermomyces lanuginosus (Dixit et al., 2023); Aspergillus oryzae (Lambré et al., 2023)], and yeast [e.g., Saccharomyces cerevisiae (Wang et al., 2022)]. In starch-based shelf-stable foods, residual α-amylase contamination can induce starch hydrolysis, leading to structural and functional degradation of the product. In starch-stabilized pasteurized yogurt, these enzymes catalyse the endohydrolysis of (1,4)-α-D-glucosidic linkages in starch during long-term storage, leading to reduced viscosity and textural stability. This degradation is often exacerbated by the lack of sensitive methods to detect trace α-amylase activity in dairy systems. Notably, this issue mirrors similar challenges in other matrices, such as eggs (da Silva et al., 2017), bread (Rebholz et al., 2021), and soymilk (Pascall et al., 2006). Overall, current α-amylase activity detection methods such as DNS and iodine colorimetric methods, lack sufficient sensitivity to quantify trace α-amylase in complex dairy systems (Wells et al., 2023). Furthermore, the inactivation kinetics of various α-amylases under varying pasteurization conditions remain uncharacterized (Stoforos et al., 2025; Tong & Ramaswamy, 2023), hindering the optimization of microbial safety and the maintenance of enzymatic stability.

Pasteurized yogurt additives include sugar, starch (Lee & Kang, 2024), gelatine, pectin (Zhang et al., 2025), alginate, whey protein (Nourmohammadi et al., 2020), and agar (Akshit et al., 2025). Pectin is primarily extracted from fruit peels, whereas starch derivatives are derived from crops such as corn, potato, and cassava (Li et al., 2023). In recent years, hydroxypropyl distarch phosphate (HPDSP) (Yuan et al., 2022), a starch derivative, has been widely adopted by many yogurt manufacturers. Casein particles in yogurt absorb on the surface of the HPDSP via electrostatic forces, contributing to the structural stability of HPDSP (Cui et al., 2014).

In the present study, we hypothesized that trace α-amylase contamination in HPDSP-stabilized yogurt significantly impacts its physicochemical stability during storage, and that conventional pasteurization conditions fail to fully inactivate α-amylases. This study provides the first quantitative assessment of HPDSP hydrolysis by trace α-amylases and its physicochemical consequences in pasteurized yogurt. This study aimed to establish a scientific basis for: (i) developing sensitive enzyme detection methods, (ii) revising ingredient specifications, and (iii) implementing contamination control measures in dairy processing.

2. Materials and methods

2.1. Materials

Human saliva α-amylase (Type XIII-A, lyophilized powder, 300–1500 units/mg protein), heat-stable Bacillus licheniformis α-amylase (TDF-100 A, solution, for use in Total Dietary Fiber Assay), and Aspergillus oryzae α-amylase (powder, approximately 30 U/mg) Starch Assay Kit were obtained from Sigma-Aldrich (St. Louis, MO, USA). HPDSP was provided by Ingredion Co., Ltd. The milk was supplied by Brightdairy Co., Ltd. (Shanghai, China). Commercially Streptococcus thermophilus and Lactobacillus delbrueckii subsp. Bulgaricus were purchased from DANISCO France SAS (Paris, France).

2.2. Evaluation of temperature- and pH- dependent activity of three α-amylases on HPDSP hydrolysis

Three α-amylase sources were investigated in this study: heat-stable α-amylase from B. licheniformis (HEa), α-amylase from A. oryzae (ASa), and human saliva α-amylase (HUa). The enzymatic hydrolysis of HPDSP was quantified using the DNS method, with maltose serving as the standard for calibration. One unit of α-amylase activity was defined as the amount of enzyme required to catalyse the disappearance of 1 μmol of reducing sugars per minute at 37 °C, pH 4.21, and 1.5 % HPDSP.

For pH-dependent activity analysis, reactions were performed using 1.5 % w/v HPDSP with 0.1 U α-amylase at 37 °C in the following buffer systems: 50 mM lactate buffer (pH 3.0–5.0), 50 mM phosphate buffer (pH 5.5–7.0), and 50 mM Tris-HCl buffer (pH 7.5–9.0). Temperature-dependent activity was assessed under identical substrate and enzyme concentrations in 50 mM lactate buffer (pH 4.21) across a temperature gradient (4, 10, 25, 37, 45, 55, 75, 95 °C).

2.3. Kinetic analysis

The kinetic parameters, including the Michaelis constant (K_m) and maximum velocity (V_max), were determined at varying HPDSP (0.1–3.2 % w/v) in 50 mM lactate buffer (pH 4.21) at 37 °C. Data analysis was performed using the Hanes–Woolf plot method implemented in Microsoft Excel (Microsoft Inc., Redmond, WA, USA). Solid line represents the best fit to the experimental data using the Michaelis–Menten equation as follows:

$$v=\frac{V_{\mathit{max}}\left[S\right]}{K_m+\left[S\right]}$$ (1)

2.4. Determination of thermal stability of three α-amylases

The thermal stability of α-amylase HEa and ASa was investigated by incubating 20 μL enzyme aliquots at temperatures ranging from 55 °C to 100 °C. Following incubation, residual α-amylase activity was determined. The reaction condition was as follows: 2.0 % w/v HPDSP with 0.1 U α-amylase in 50 mM lactate buffer (pH 4.21) at 45 °C. The percentage residual activity remaining was then plotted against time for different temperatures.

2.5. Effect of α-amylase concentration on hydrolysis of HPDSP

To evaluate concentration-dependent effects of α-amylase on HPDSP hydrolysis, the following enzyme solutions were prepared: ASa (1, 0.1, 0.01 U/L) or HEa (0.01, 0.001, 0.0001 U/L). Substrate solutions containing 1.5 % (w/v) HPDSP and 8 % (w/v) sucrose solution were prepared in water (100 mL each). Reaction mixtures were mixed at 55 °C for 20 min, then subjected to sterilization at 95 °C for 5 min. All samples were stored at 55 °C for 30 d.

Following the incubation period, reactions were terminated by adding 10 mL of iodine working solution in 1 mL aliquots of each sample. Absorbance measurements were conducted at 660 nm using a spectrophotometer. The residual starch content was quantified using an HPDSP standard curve. A decrease in colour intensity indicated the breakdown of the starch structure, correlating with the enzymatic hydrolysis activity.

2.6. Preparation of pasteurized yogurt with α-amylase

The study employed a two-tiered fermentation system: Laboratory-scale (100 g fresh milk in laboratory-scale) and pilot-scale (11 kg fresh milk in pilot-scale) batches. Milk was initially preheated to 55 °C, followed by the gradual incorporation of dry ingredients (1.5 % HPDSP and 8 % sucrose) under constant mixing. Subsequently, 1 mL of the enzyme solution (ASa: 1.0 × 10⁻⁴–50 U/L; HEa: 1.0 × 10⁻⁶–60 U/L) was added. The ingredients were incubated at approximately 50–55 °C for 25 min, followed by sterilization at approximately 90–95 °C for 5 min using the Water Bath Heating System. When the milk cooled to below 44 °C, commercial starter culture was added to each bottle with thorough mixing. The commercial starter culture used in this experiment was Streptococcus thermophilus and Lactobacillus delbrueckii subs. Bulgaricus. The samples were fermented at 43 °C for 5–6 h and subsequently stored at 4 °C. After fermentation to pH 4.2, the yogurt samples were subjected to secondary pasteurization using the Water Bath Heating System (laboratory-scale fermentation) or the HTST/UHT Pilot System (pilot-scale fermentation), then sealed with caps and cooled to 25 °C for storage.

2.7. Determination of starch and viscosity in pasteurized yogurt

Two scales of pasteurized yogurt samples prepared according to the method described in Section 2.6 were stored at 25 °C for 30 d prior to analysis.

To detect starch concentration in pasteurized yogurt, a Starch Assay Kit (using the amylase and amyloglucosidase method) was used. Each 10 g yogurt aliquot was mixed with 20 g of 80 % ethanol solution. The mixture was incubated at 85 °C for 5 min. The sample was centrifuged at 1000 ×g for 10 min. The pellet was resuspended in 20 g of 80 % ethanol solution and centrifuged at 1000 ×g for 10 min. Subsequently, 0.2 mL of 80 % ethanol solution, 3.0 mL of water, and 0.02 mL of α-amylase solution were added to each sample and incubated at 100 °C for 5 min. The volume of each tube was adjusted to 7 mL with water, and the samples were centrifuged at 5000 ×g for 3 min. The supernatants were collected for subsequent analysis.

A 1.0 mL aliquot of the supernatant was mixed with 1.0 mL of amyloglucosidase and incubated at 60 °C for 15 min, followed by the addition of 18 mL of water. For glucose determination, 1 mL of this diluted solution was combined with 2 mL of glucose assay reagent (containing glucose oxidase, peroxidase reagent, and o-dianisidine)– and incubated at 37 °C for 30 min, ensuring precise timing. Subsequently, 2.0 mL of 6 M sulfuric acid was added to each tube. Finally, absorbance at 420 nm was measured for each sample.

The viscosity measurements were performed using R180, ProRheo, viscometer (ProRheo, Bahnhofstr, Germany), with the following parameters: No.1 and 2 spindles at speeds of 64 or 1000 rpm. Three independent repetitions were conducted for each sample.

2.8. Determination of rheological properties of pasteurized yogurt

Dynamic oscillatory and flow measurements of yogurt were carried out using an ARES-G2 rheometer (TA Instruments, New Castle, DE, USA) at 25 °C, with a cone-plate geometry, in which the rotating cone was 50 mm in diameter, and a cone angle of 0.02 with a gap of 0.05 mm. Strain ranging from 0.05 % to 100 % was used at 1 Hz. Flow curves of the yogurts were obtained by varying the shear rate from 1 to 1000 s⁻¹, and the corresponding shear stress and apparent viscosity values were measured.

The flow curve data were fitted using the rheometer software based on Herschel–Bulkley model, which describes the behaviour of time-independent fluids as follows.

$$\text{Herschel}–\text{Bulkley}:\tau =\tau 0+k{r}^n$$ (2)

The Herschel–Bulkley model can be used to describe the rheological behaviour of certain non-Newtonian fluids. In this equation, τ0 is the yield stress. Generally, a higher τ0 (yield stress) value means that materials resist flow more strongly. The parameter k is the consistency factor (Pa·sn) and reflects the viscosity at a shear rate of zero. It correlates with the relaxation time of the material, which can be affected by moisture, temperature, and the type of material. n is the flow rate index. When n < 1, the material exhibits a pseudoplastic and a shear-thinning behaviour. Most materials are shear-thinning because the input shear energy tends to align anisotropic molecules or particles and disaggregate any large clumps of particles. This alignment and disaggregation reduce the overall hydrodynamic drag, which in turn reduces the dissipation of energy in the fluid and ultimately the viscosity.

2.9. Data analysis

Error bars indicate SD values, and the data represent the average of triplicate measurements. Data analysis was conducted using Microsoft Excel and Origin. The Origin Quick Fit Gadget was used to perform curve fitting. Four-way analysis of variance (ANOVA) was conducted in SPSS 30.0.0 software (SPSS Inc., Chicago, IL, USA) to assess significant (P < 0.05 and P < 0.01) differences among samples.

3. Results and discussion

3.1. Enzymatic hydrolysis characteristics of HPDSP by α-amylase

Starch derivatives (Wong et al., 2020) are widely incorporated into the milk base as thickeners to enhance textural firmness in pasteurized yogurt. The results of the four-way ANOVA revealed that the enzyme type (F = [10.156], P < 0.01), temperature (F = [3.741], P < 0.05), and pH (F = [1.922], P < 0.05) significantly affected enzyme activity. More importantly, we observed significant two-way interactions: enzyme type × temperature (F = [6.882], P < 0.05) and enzyme type × pH (F = [4.787], P < 0.05). Given that temperature and pH are the two most critical factors modulating enzyme activity (Krishnan et al., 2022), we systematically analysed their effects on three α-amylase variants.

As presented in Fig. 1A, all three α-amylases exhibited well-fitted activity curves within a pH range of 3–7. To further investigate the significant interactions, we generated fitted curves of enzyme activity against temperature and pH rather than pairwise significance tests. Fitting analysis (Table S1) revealed optimal pH values of 4.0, 5.0, and 6.0 for HEa, ASa, and HUa, respectively. Similarly, temperature-dependent activity curves (Fig. 1B, Table S2) revealed optimal temperatures of 94 °C, 73 °C, and 30 °C for HEa, ASa, and HUa, respectively. Notably, HEa displayed exceptional thermostability and acid tolerance, maintaining high activity at pH 4.0 and 94 °C. During yogurt fermentation, the pH decreases from 6.4 to approximately 4.0–4.6, and remains around 4.2 throughout storage (Krishnan et al., 2022). Therefore, these findings suggest that HEa may remain functionally active during the entire shelf life of the product.

Fig. 1.

Dependence of the reaction rate of three α-amylase on pH (A), temperature (B), substrate concentration (C), and Hanes–Woolf plot ([HPDSP]/V₀ vs. [HPDSP]) (D). The solid line represents the best curve fitting to the experimental data using OriginLab and Michaelis–Menten equation. HEa, ASa, and Hua referr to the α-amylase isolated from B. licheniformis, A. oryzae, and human saliva, respectively. HPDSP refers to hydroxypropyl distarch phosphate.

The interaction between enzyme type and substrate concentration was not significant (F = [0.683], P = [0.819]). As substrate concentration increased, the reaction velocity for all enzymes approached their maximum velocity (Vₘₐₓ) in a predictable manner, consistent with the classical Michaelis–Menten kinetic model (Fig. 1C). We investigated the substrate affinities of the three amylases for HPDSP (Fig. 1D). The K_m values for HEa, ASa, and HUa were 13.9, 3.4, and 7.2 mg/mL, respectively. The V_max values for HEa, ASa, and HUa were determined as 3.3, 3.9, and 5.8 mM/min, respectively. The lower K_m value of ASa indicates a strong affinity for the substrate, enabling efficient catalysis at low substrate concentrations, whereas HEa requires high substrate concentrations to achieve maximum catalytic activity. These results suggest that under low-substrate and low-temperature conditions, ASa is more efficient at hydrolyzing HPDSP.

Moreover, the tolerance of HEa to extreme pH and temperature conditions enables long-time starch hydrolysis, potentially periodically changing yogurt viscosity and gel properties. The activity of ASa driven by its high substrate affinity, could dominate in refrigerated products containing minimal starch content.

3.2. Effects of thermal stability of three α-amylases

We evaluated the thermal stability of α-amylase by measuring the static half-life of HEa and ASa. As depicted in Fig. 2A, we observed complete inactivation of ASa in 120 s when incubated at 70 to 100 °C in a substrate-free buffer. In contrast, HEa remained active for 800 s at 85 to 100 °C (Fig. 2B). Calculation of the static half-lives across the 55–100 °C range (derived from fitted equations in Tables S3–S4) revealed that HEa exhibited 5–10-fold greater thermal stability than ASa (Table 1).

Fig. 2.

Effect of temperature on the stability of α-amylase of ASa (A) and HEa (B).

Table 1.

Static half-life of α-amylases isolated from B. licheniformis, and A. oryzae.

Temperature (°C)	Static half-life (s)
	ASa	HEa
100	8.8	43.6
95	12.2	95.7
90	20.0	162.0
85	24.6	217.9
80	33.1	1738.6
75	39.8	none
70	60.6	none
65	97.5	none
60	152.2	none
55	774.2	none

Thermal stability studies revealed an inverse relationship between temperature and activation rate with HEa exhibiting remarkable thermal resistance at a temperature below 85 °C. According to a previous published study, charged peptides alter α-amylase activity via electrostatic-driven surface charge redistribution, conformational changes, and active-site microenvironment modulation (Krishnan et al., 2022). Moreover in vitro digestion conditions affect the kinetics of starch and protein digestion, but not final hydrolysis, with semi-dynamic methods accurately reflecting physiological digestion patterns (Duijsens et al., 2024). In the context of yogurt processing, the presence of starch and proteins may prolong the inactivation time of α-amylase, further influencing its residual activity. Our result suggests that a minimum treatment of 800 s at a temperature greater than 85 °C is required to achieve complete α-amylase inactivation in yogurt processing.

3.3. Effects of α-amylase concentration on HPDSP hydrolysis

We investigated the α-amylase concentration required to induce HPDSP hydrolysis under industrial processing conditions. Given the standard processing of yogurt involves mixing of raw materials at 55 °C followed by 95 °C sterilization, and considering the 6-month shelf life of pasteurized yogurt, we conducted accelerated stability testing at 55 °C for 30 d.

According to the results presented in Fig. 3, at ASa concentrations ranging from 1.0 × 10⁻² to 1.0 U/L, the starch residue remained 100 %, 109 %, and 108 %, respectively, indicating minimal enzymatic hydrolysis. These results suggest that ASa present in raw materials at a concentration ≤ 1 U/L would be effectively inactivated by standard sterilization protocols (95 °C, 5 min).

Fig. 3.

Effect of α-amylase concentration on the hydrolysis of HPDSP. HEa and ASa refer to the α-amylase isolated from B. licheniformis and A. oryzae, respectively. HPDSP refer to hydroxypropyl distarch phosphate. The symbols “*” and “**” indicate significant difference (P < 0.05) and extremely significant difference (P < 0.01), respectively. Error bars represent ± SD.

Under the same processing conditions, when the final concentration of HEa was 1.0 × 10⁻⁴ to 1.0 × 10⁻² U/L, the starch residue remained 77 %, 46 %, and 29 %, respectively. At the highest HEa concentration (H1: 1.0 × 10⁻² U/L), the starch residue decreased to less than 30 %, whereas at the lowest concentration (H3: 1.0 × 10⁻⁴ U/L), it remained close to 90 %. These findings indicate that the HEa concentration must be maintained below 1.0 × 10⁻⁴ U/L to preserve over 90 % starch content during storage.

3.4. Effects of α-amylase on HPDSP and viscosity in pasteurized yogurt

We investigated the correlation between enzyme concentration (ASa/HEa) and both starch (HPDSP) hydrolysis efficiency and yogurt viscosity. As depicted in Fig. 4A–B, we observed significant starch hydrolysis at concentrations of 50–0.1 U/L for ASa and 60–0.01 U/L for HEa in laboratory-scale fermentation trials. In contrast, when the enzyme concentrations were decreased below threshold levels (ASa < 1.0 × 10⁻² U/L; HEa < 1.0 × 10⁻³ U/L), starch hydrolysis was markedly attenuated, resulting in starch retention levels exceeding 90 %. As enzyme concentration decreased and starch hydrolysis efficiency dropped, yogurt viscosity progressively increased, reaching its maximum when minimal or no enzyme was added. These findings suggest that the activity of ASa and HEa directly impacts the breakdown of starch and consequently influences the textural properties of yogurt.

Fig. 4.

Effect of α-amylase (ASa and HEa) content on HPDSP and viscosity in laboratory-scale (A and B) and pilot-scale (C) fermented yogurts. The symbols “*” and “**” indicated significant difference (P < 0.05) and extremely significant difference (P < 0.01), respectively. Error bars represent ± SD.

To validate the impact of α-amylase content on yogurt fermentation and storage, we conducted pilot-scale fermentation alongside laboratory-scale trials. As presented in Fig. 4C, at concentrations of ≤1 U/L for ASa and ≤ 0.0001 U/L for HEa, the starch content and yogurt viscosity were comparable to those of the control group after 30 d of storage. This result is similar to the findings from the laboratory-scale fermentation, indicating that in the production of yogurt fermentation, the amylase content must be controlled at ASa ≤ 1 U/L or HEa ≤ 0.0001 U/L.

3.5. Effects of α-amylase on the rheological properties of pasteurized yogurt

We fitted the upward flow curves to the Herschel–Bulkley model, and the derived parameters are presented in Table 2, Table 3. Yield stress (τ0), which strongly correlated with the initial firmness of yogurt in sensory assessment (Grasso et al., 2020), gradually decreased with increasing concentrations of ASa and HEa. This finding indicates that α-amylase-mediated hydrolysis of HPDSP reduced the shear stress required to initiate flow, thereby lowering viscosity. The yield stress of yogurt from pilot-scale fermentation was lower than that from laboratory-scale fermentation, declining from 149.98 to 27.12 Pa. This decline is primarily attributed to the incorporation of a homogenizer during pilot-scale production, which increased shear forces during mixing. At the lowest enzyme concentrations (ASa: 1.0 × 10⁻¹ U/L; HEa: 1.0 × 10⁻⁴ U/L), the yield stress remained identical to the control (27 Pa). However, a 10-fold increase in α-amylase concentrations resulted in a 1.3- to 3.3-fold reduction in yield stress. Additionally, prolonged storage led to an increase in yield stress across all samples, likely owing to the temporary weakening of the yogurt matrix following secondary pasteurization.

Table 2.

Herschel–Bulkley parameters for laboratory-scale fermentation pasteurized yogurt.

α-amylase	Concentration (U/L)	K (Pa·sn)	n (−)	τ0 (Pa)
none	0.0	184.3 ± 68.2	−0.3 ± 0.0	152.5 ± 3.5
ASa	1.0 × 101	147.4 ± 66.8	−0.6 ± 0.3	48.8 ± 17.5b
	1.0 × 100	112.3 ± 26.5	−0.7 ± 0.2	28.5 ± 6.9b
	1.0 × 10−1	266.7 ± 93.0	−0.5 ± 0.2	127.3 ± 23.6
	1.0 × 10−2	700.1 ± 458.1a	−0.9 ± 0.4b	155.0 ± 14.2
HEa	1.0 × 10−2	192.0 ± 26.9	−0.6 ± 0.1	53.9 ± 25.1b
	1.0 × 10−3	1327.6 ± 754.5b	−0.7 ± 0.2 a	106.6 ± 39.7
	1.0 × 10−4	181.3 ± 79.0	−0.4 ± 0.3	113.2 ± 67.0
	1.0 × 10−5	203.5 ± 5.1	−0.5 ± 0.1	84.9 ± 52.8a
	1.0 × 10−6	345.6 ± 214.9	−0.7 ± 0.1a	134.8 ± 7.4

Note: The symbols “^a” and “^b” indicate significant difference (P < 0.05) and extremely significant difference (P < 0.01) compared to the control group, respectively.

Table 3.

Herschel–Bulkley parameters for pilot-scale fermentation of pasteurized yogurt.

α-amylase	Concentration (U/L)	1st week			4th week
		k(Pa·sn)	n (−)	τ0 (Pa)	k(Pa·sn)	n (−)	τ0 (Pa)
none	0.0	1.6 ± 0.6	0.6 ± 0.0	3.9 ± 1.0	1.7 ± 0.0	0.8 ± 0.0	27.1 ± 0.0
ASa	1.0 × 100	6.2 ± 0.3	0.5 ± 0.0	8.1 ± 0.8	0.7 ± 0.1	0.8 ± 0.0	21.6 ± 1.4
	1.0 × 10−1	4.8 ± 0.2	0.5 ± 0.0	10.2 ± 0.3	0.6 ± 0.0b	0.8 ± 0.0	27.7 ± 0.9
HEa	1.0 × 10−2	3.3 ± 0.5	0.5 ± 0.0	3.8 ± 1.0	2.3 ± 0.7	0.5 ± 0.0	5.4 ± 3.0
	1.0 × 10−3	7.1 ± 0.1	0.5 ± 0.0	8.1 ± 0.2	2.2 ± 0.2	0.6 ± 0.0	17.7 ± 0.8
	1.0 × 10−4	7.7 ± 0.2	0.4 ± 0.0	1.6 ± 0.3	2.4 ± 0.3	0.6 ± 0.0	15.9 ± 1.8

Note: The symbols “^a” and “^b” indicate significant difference (P < 0.05) and extremely significant difference (P < 0.01) compared to the control group, respectively.

Notably, samples containing 1.0 × 10⁻² U/L ASa and 1.0 × 10⁻³ U/L HEa (Figs. 5A) exhibited higher viscosities than the control samples. Similarly, yogurt containing 1.0 × 10⁻³ U/L HEa (Fig. 5B-C) exhibited elevated viscosities. These findings suggest that partial starch hydrolysis by α-amylase may enhance viscosity, contrary to the typical shear-thickening behaviour of starch in liquids. Starch granules can swell and increase in size through retrogradation, resulting in increased viscosity (Pang et al., 2019). However, during the process of rapid cooling after pasteurization, starch may not swell completely. After the addition of α-amylase, the enzyme hydrolyses starch into short-chain amylose and low-molecular-weight dextrins, which exhibit a pronounced propensity for rapid retrogradation. These fragments readily re-associate to protein through hydrogen bonding, forming a three-dimensional network that contributes to increased viscosity and a rigid gel-like structure (Lin et al., 2022).

Fig. 5.

Laboratory-scale (A) and pilot-scale (B and C) fermentation yogurt viscosity varying as a function of shear rate.

For pilot-scale fermented yogurt (sample 1a–6a: 1st week; samples 1b–6b: 4th weeks), only 1.0 × 10⁻² U/L HEa and 1.0 U/L ASa reduced viscosity, whereas 1.0 × 10⁻³ U/L HEa increased viscosity after storage (Fig. 5B-C). For accurate rheological measurements, samples must be in a homogenous state. Yogurts without amylase exhibited viscosity profiles similar to those containing ASa (1.0 × 10⁻¹ U/L) or HEa (1.0 × 10⁻⁴ U/L). In contrast, 1.0 × 10⁻² U/L HEa reduced viscosity, whereas 1.0 × 10⁻³ U/L HEa increased low-shear viscosity.

Notably, when the concentration of ASa was ≤1.0 × 10⁻¹ U/L and that of HEa ≤ 1.0 × 10⁻³ U/L, the viscosity of room-temperature-stored yogurt remained stable for one month, without excessive thinning or quality deterioration.

4. Conclusion

In the present study, we evaluated the effects of α-amylases derived from A. oryzae and B. licheniformis on starch hydrolysis in pasteurized yogurt during storage, focusing on their implications for viscosity and product quality. The results demonstrate that both enzymes significantly influence starch degradation and rheological properties. Notably, HEa exhibited stronger performance under acidic and high-temperature conditions than ASa, suggesting its greater suitability for hydrolyzing modified starch (HPDSP) in yogurt production. Consequently, strict quality control of raw materials and ingredients is warranted to mitigate potential adverse effects on product consistency. This study has limitations related to the variability in types and concentrations of raw materials, processing conditions, storage temperatures, and fermentation extent, which may affect the reproducibility and generalizability of the results.

Our findings also reveal that even trace amounts of α-amylase can substantially alter the viscosity and texture of the yogurt over extended storage periods, highlighting the importance of controlling enzyme levels for product quality. This study lays the groundwork for future research on the detection of low α-amylase concentrations in dairy systems. It also provides insights for optimizing yogurt production, improving shelf-life stability, and ensuring texture uniformity, particularly in room-temperature-stored products.

CRediT authorship contribution statement

Ying Qiao: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Junli Miao: Resources, Project administration, Methodology, Formal analysis, Data curation, Conceptualization. Yifan Hua: Writing – review & editing, Writing – original draft, Validation, Software, Methodology, Investigation, Funding acquisition, Formal analysis. Wei Li: Resources, Project administration, Methodology, Conceptualization. Xuehong Zhang: Project administration. Zhenmin Liu: Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

Funding

This work was supported by “The 2023 Shanghai White Magnolia Talent Program, Pujiang Project (Category D)” [grant number 23PJD017].

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.

Appendix A. Supplementary data

Supplementary data on fitting equation of enzyme activity.

Data availability

Data will be made available on request.