Glycosylation Remodeling and Thermal Denaturation Dictate the Functional Diversification of Protein Z

Abstract

Protein Z (PZ) derived from barley malt has been identified as one of the key proteins contributing to foam stability. Recently, PZ was also recognized as an effective carrier, a functionality attributed to its serpin-like activities. This study investigated key structural-functional changes in PZ during thermal processing (mashing and boiling). The structural modifications of PZ variants were analyzed using FT-IR (Fourier Transform Infrared Spectroscopy). The results indicated that the secondary structure of PZ, after mashing, did not change significantly, whereas the β-turn content of PZ after boiling increased to 24.08% ± 0.34%. Interfacial adsorption kinetics, coupled with structural analysis, revealed that PZ, after mashing, exhibited the highest foamability (41.4 ± 0.38%), which was associated with the highest diffusion rate constant (K_diff) (1.05 ± 0.03). In contrast, PZ after boiling demonstrated superior foam stability (68.54 ± 1.12%), which correlated with the highest rearrangement rate constant (K_R) (−6.13 ± 0.06). Moreover, PZ, after mashing, exhibited enhanced inhibitory activity, an effect associated with the removal of glycosylation at Thr344 and Thr350 (located in the reactive center loop of PZ) via enzymatic hydrolysis during the mashing process. In contrast, PZ after boiling resulted in a loss of thrombin inhibitory activity, consistent with protein denaturation at high temperatures. These findings elucidate how structural modifications affect the function of PZ during brewing, thereby providing a scientific foundation for its potential applications across multiple fields.

1. Introduction

Beer was the most widely consumed alcoholic beverage in the world, which was popular among consumers of all ages [1]. Foam represented one of the most crucial characteristics for beer, serving as a vital factor influencing consumer choices. The stability of beer foam was influenced by several compounds, such as proteins, polysaccharides, and iso-α-acids. Among these, proteins were primarily responsible for forming and stabilizing the interfacial layer of the foam [2]. In particular, protein Z (PZ) had been identified as one of the key proteins contributing to foam stability [3]. Initially, PZ was discovered in barley seeds, which were an important raw material in the process of beer fermentation by Hejgaard J in 1982 [4]. Although barley needs to withstand multiple procedures before fermentation, PZ possessed the ability to resist denaturation and proteolytic modification throughout these stages.

Generally, germination, kilning, mashing, and boiling were key processes for beer fermentation [5]. Mashing, as the first heating step, stimulated the activity of endogenous enzymes and adequate gelatinization of starch following the time-temperature curve [6,7]. During boiling, ingredients such as hops were added to enhance the aroma and flavor of the beer. The combined mashing and boiling processes modified wort protein conformations, which, in turn, could influence key beer quality attributes and protein functionality [8].

PZ was classified as a cereal serpin based on its sequence homology with human α1-antitrypsin, human α1-antichymotrypsin, and chicken ovalbumin [9]. The tertiary structure of serpins was highly conserved, with natural serpins comprising three β-sheets surrounded by eight to nine α-helices, along with an exposed reactive center loop (RCL) [10]. Serpins had been extensively investigated owing to their distinctive non-competitive inhibition mechanism [11] and their ability to interact with proteinases both as substrates and as suicide inhibitors, forming exceptionally stable inhibitor-proteinase complexes. Previous studies have demonstrated that PZ can form internal hydrophobic binding pockets capable of encapsulating small hydrophobic molecules, which might enhance their stability and bioavailability [12,13]. Current studies [14] on the brewing processes of mashing and boiling primarily focus on protein structural modifications, while their effects on functional properties remain largely unexplored.

This study investigated the structural changes of PZ during mashing and boiling processes in beer production, especially the effects of glycosylation. Foamability and thrombin inhibitory activity were critical characteristics of PZ. Therefore, the foamability of PZ was measured, and interfacial adsorption kinetics were employed to investigate the factors underlying the varying foaming performance among different samples. Furthermore, the mechanism underlying the altered thrombin inhibitory activity of PZ as a serpin after mashing was elucidated using molecular docking and glycosylation mass spectrometry. These findings are expected to enhance the understanding of PZ’s functional changes during brewing and provide a mechanistic basis for its potential application in functional beer or novel food development.

2. Materials and Methods

2.1. Chemical and Materials

The barley malt used is the Pilsner malt from COFCO’s COFCO PUREMALT series (Shijiazhuang, China). This malt has a moisture content of 5% and is characterized by a wort color of 2.5–4.0 EBC, a total protein content of 9.5–12.0%, a pH of 5.7–6.1, and a fine-grind extract (dry basis) of 80.0%. Ammonium sulphate, thrombin, and OPA were obtained from Macklin (Shanghai, China). Trypsin and fibrinogen were obtained from MedChemExpress (Princeton, NJ, USA). Tris was purchased from Biyuntian Biotechnology Co., Ltd. (Shanghai, China). Deionized water is produced by a laboratory deionizer (ZYWL-10B-UP, Beijing Zhongyang Yongkang Environmental Protection Technology Co., Ltd., Beijing, China). All other reagents used in this study were analytical grade.

2.2. Preparation of Protein Z After Different Processing

2.2.1. Mashing and Boiling Wort Preparation

The mashing and boiling procedures were performed according to a previous research [14]. Briefly, whole barley malt kernels were ground using a mechanical grinder. The crushed malt was mixed with deionized water at a 1: 5 (w/w) ratio and maintained at 48 °C for 30 min with stirring. Subsequently, the temperature was raised to 63 °C and held for 40 min, followed by a further increase to 72 °C for 20 min. After completing the above procedures, an iodine test was conducted to confirm completion of the mashing process, after which the temperature was raised to 78 °C and held for 10 min, marking the end of the mashing procedure. A consistent heating rate of 1 °C/min was employed for all temperature transitions throughout the mashing process. After the mashing procedure, the wort was boiled for 60 min to complete the boiling process. No additional substances were introduced, and no pH adjustment was performed. The samples were designated as natural wort (without process), mashing wort and boiling wort corresponding to the wort processing.

2.2.2. Protein Preparation

All the processed PZ samples were extracted using a modified method described by Xu [2]. The different worts, including natural wort, mashing wort, and boiling wort, were stirred and extracted for 2 h. Then the different worts were centrifuged at 5000 rpm for 15 min to obtain the supernatant. The supernatant was subjected to precipitation with 40–60% ammonium sulfate overnight. The collected sediment was dissolved in 50 mM Tris-HCl buffer (pH 8.0) and dialyzed for 48 h. All the fraction samples were then heated at 80 °C for 30 min, centrifuged at 10,000 rpm for 10 min to obtain crude PZ. The crude PZ was further purified using an anion exchange chromatography column (DEAE Sepharose FF, St. Louis, MO, USA). The PZ extracted from differently processed wort was termed as follows, corresponding to three successive processing stages: PZ, PZ after mashing, and PZ after boiling. The purity of processed PZ was verified by SDS-PAGE, and its assembly state was assessed by native-PAGE.

2.3. Composition Differences Between PZ Molecules with Different Processing

2.3.1. The Browning Intensity and the Content of Maillard Reaction Product

The browning intensity was determined by measuring the absorbance of a 0.4 mg/mL PZ solution (Tris-HCl, 50 mM, pH 8.0) and was measured at 420 nm using a UV-Vis spectrophotometer (INESA N4, INESA Analytical Instrument Co., Ltd., Shanghai, China). The absorbance values at 420 nm were used to evaluate the degree of browning in PZ [15]. The content of Maillard reaction products was measured by a Varian Cary 50 v 3.0 spectrophotometer (Varian, Palo Alto, CA, USA) with a 10 µM PZ solution (50 mM Tris, pH 8.0). The excitation wavelength was set to 347 nm, and emission spectra were collected from 350 to 600 nm [16].

2.3.2. Free Amino Loss Percentage

Free amino loss percentage values were measured using an o-phthaldialdehyde (OPA) reagent [17] with slight modifications. The assays were performed at a protein concentration of 2 mg/mL (Tris–HCl, 50 mM, pH 8.0). The modified method differed from approaches in that the free amino loss percentage was calculated from the ratio of OPA reagent volume consumed by different PZ solutions to that consumed by a bovine serum albumin (BSA) at 2 mg/mL, according to the following Equation (1):

$$\mathbf{F}\mathbf{r}\mathbf{e}\mathbf{e} \mathbf{a}\mathbf{m}\mathbf{i}\mathbf{n}\mathbf{o} \mathbf{l}\mathbf{o}\mathbf{s}\mathbf{s} \left(\mathbf{\%}\right)={\displaystyle \frac{{\mathit{A}}_{\mathit{B}}-{\mathit{A}}_{\mathit{P}}}{{\mathit{A}}_{\mathit{P}}}}\times 100\mathit{\%}$$

(1)

where, A_P was the absorbance value of the PZ, and A_B was the absorbance of the BSA.

2.4. Structural Differences Between PZ Molecules with Different Processing

2.4.1. The Secondary Structure of Protein

The secondary structure of different PZ was analyzed using an FT-IR spectrophotometer (Nicolet iS50, La Jolla, CA, USA). The lyophilized samples were mixed with KBr (1:100 w/w), ground, and pressed into thin slices, across 400–4000 cm⁻¹ with 32 averaged scans [18]. The amide I band (1600 cm⁻¹–1700 cm⁻¹) in the original spectra was analyzed using PEAKFIT 2018 software, combined with Fourier self-deconvolution, to determine the secondary structure composition.

2.4.2. The Tertiary Structure of Protein

The sample concentration was adjusted to 10 µM (Tris-HCl, 50 mM, pH 8.0) for fluorescence spectrum acquisition using a Cary Eclipse spectrophotometer (Cary, Varian, USA). The excitation wavelength was set to 280 nm, and fluorescence intensity was recorded from 300 to 450 nm. Both excitation and emission slit widths were maintained at 10 nm. An inner-filter correction was applied before the experiment.

2.5. The Thermal Stability of PZ Molecules with Different Processing

The thermal stability was analyzed using differential scanning calorimetry (DSC25, TA, New Castle, DE, USA). Approximately 5 mg of each lyophilized sample was weighed into an aluminum tray and heated from 20 to 200 °C at a rate of 10 °C/min under a nitrogen atmosphere [19]. An empty aluminum crucible served as the reference.

2.6. The Air/Water Interface Properties of PZ Molecules with Different Processing

2.6.1. The Foaming Performance

The foam performance for different PZ was measured according to the method of Xu’s at a concentration of 0.4 mg/mL (Tris-HCl, 50 mM, pH 8.0) [14]. The formulas for foamability and foamstability were expressed with Equations (2) and (3):

$$\mathit{F}\mathit{o}\mathit{a}\mathit{m}\mathit{a}\mathit{b}\mathit{i}\mathit{l}\mathit{i}\mathit{t}\mathit{y}\left(\mathit{\%}\right)={\displaystyle \frac{{\mathit{V}}_0}{15}}\times 100\mathit{\%}$$

(2)

V₀ was the initial foam volume, and 15 was the protein solution volume.

$$\mathit{F}\mathit{o}\mathit{a}\mathit{m}\mathit{s}\mathit{t}\mathit{a}\mathit{b}\mathit{l}\mathit{i}\mathit{t}\mathit{y}\left(\mathit{\%}\right)={\displaystyle \frac{{\mathit{V}}_0}{{\mathit{V}}_{30}}}\times 100\mathit{\%}$$

(3)

V₀ and V₃₀ represented foam volumes of initial and 30 min, respectively.

2.6.2. Interfacial Adsorption Kinetics Measurement of Protein

All samples were diluted to 0.4 mg/mL in Tris-HCl (50 mM, pH 8.0). Interfacial tension measurements were performed at the air/water interface using a Theta Flex tensiometer (Biolin Scientific, Vastra Frolunda, Sweden). A 10 μL droplet was formed at the needle tip, and the surface tension (γ) was immediately monitored for 2400 s. The equation to calculate interface pressure (π) was followed by the Young–Laplace Equation (4):

$$\mathit{\pi }={\mathit{\gamma }}_0-\mathit{\gamma }$$

(4)

γ₀ and γ denote the interfacial tension of the deionized water and the sample solutions, respectively.

In the initial diffusion-governed stage of adsorption, the surface adsorption behavior can be described by the Ward and Tordai equation. The diffusion rate constant (K_diff) for protein at the air/water interface could be obtained using Equation (5) [19]:

$$\mathit{\pi }={\mathit{C}}_0\mathit{K}\mathit{T}{\left({\displaystyle \frac{\mathit{D}\mathit{t}}{3.14}}\right)}^{{\displaystyle \frac{1}{2}}}$$

(5)

C₀ was the protein concentration of the samples, K was the Boltzmann constant, T was the absolute temperature, D was the diffusion coefficient, and t was time. The diffusion rate constant (K_diff) was calculated as the slope of π versus t_1/2.

After diffusion, the protein blend enters the penetration and rearrangement steps. To obtain the penetration rate constant (K_P) and rearrangement rate constant (K_R), Equation (6) was used as follows:

$$\mathit{l}\mathit{n}\left[{\displaystyle \frac{\left({\mathit{\pi }}_{\mathit{f}}-{\mathit{\pi }}_{\mathit{t}}\right)}{\left({\mathit{\pi }}_{\mathit{f}}-{\mathit{\pi }}_{\mathit{t}}\right)}}\right]=-{\mathit{k}}_{\mathit{i}}\mathit{t}$$

(6)

π_f, π_t, and π₀ were the final, real-time, and initial interface pressures at the air-water interface, respectively. k_i was the first-order rate constant.

2.7. Thrombin Inhibitory Activity of PZ Molecules with Different Processing

2.7.1. Determination of Thrombin Inhibition Kinetics

The thrombin inhibitory activity was determined according to a previously described method with slight modifications [20]. Briefly, the different PZ (10 µM, 40 µL) was mixed with fibrinogen (1 mg/mL, 140 µL) and incubated at 37 °C for 10 min. The reaction was initiated by adding 40 µL of thrombin (18 U/mL). The reaction was continuously monitored by measuring the absorbance at 405 nm at room temperature. All solutions were prepared in 50 mM Tris-HCl buffer (pH 8.0) and filtered through 0.45 µm membranes.

2.7.2. Protein-Protein Docking

The protein-protein docking between PZ and thrombin was performed by previously established methodology [21].

2.7.3. Glycosylation Mass Spectrometry Analysis

Protein bands corresponding to PZ and mashing PZ were excised from the SDS-PAGE gel for mass spectrometry analysis [22].

2.8. Statistical Analysis

All experimental data were expressed as mean ± standard deviation (SD) from at least three independent replicates. The data were analyzed using one-way analysis of variance (ANOVA), with the significance level set at p < 0.05. Statistical analysis was performed using one-way ANOVA, followed by Duncan’s multiple range test for post hoc analysis. Statistical significance was defined as p < 0.05.

3. Results and Discussion

3.1. Preparation of Protein Z After Different Processing

PZ samples obtained from different processing stages were analyzed by both SDS-PAGE and native PAGE (Figure 1). As expected, PZ from malt was confirmed to be a monomeric protein with a molecular weight of approximately 40 kDa, which was consistent with previous reports [4]. After the mashing and boiling process, the band of PZ did not change, suggesting the primary structure of PZ had not changed. The smearing observed in the native-PAGE band of PZ was likely associated with the protein’s dynamic aggregation behavior. Overall, differently processed PZ samples exhibited similar purity profiles and aggregation behavior [23], indicating their suitability for subsequent experiments.

Figure 1. Identification of protein Z with different processing. (A) SDS-PAGE, (B) Native-PAGE.

Figure 1. Identification of protein Z with different processing. (A) SDS-PAGE, (B) Native-PAGE.

3.2. Composition Differences Between PZ Molecules with Different Processing

The absorbance intensity at 420 nm indicated the presence of Maillard browning products. Mashing and boiling were key steps in beer production, during which proteins in the wort might undergo changes. Both processes involved heating-induced reactions that are known to promote the Maillard reaction, resulting in the formation of melanoidins [24]. As shown in Figure 2A,B, PZ after boiling exhibited the highest browning intensity and intermediate product content, which was associated with the extended boiling duration [14,25]. In contrast, the mashing temperature remained relatively low (<78 °C), which was attributed to limited Maillard reaction activity and consequently lower browning intensity. These results were consistent with previous reports indicating that temperature was one of the most important factors affecting Maillard reaction efficiency [26].

Meanwhile, the mashing and boiling processes also influenced changes in the free amino acid content. The free amino loss percentage, reflecting changes in free amino group content of the protein, was determined. As shown in Figure 2C, unheated PZ exhibited higher free amino loss percentage values compared with BSA, an observation that might be due to the modification of approximately 16% of its lysine residues following translation [2]. After mashing and boiling processes, the free amino loss percentage of PZ further increased, an effect associated with the binding of additional polysaccharides from the wort to PZ and their subsequent participation in Maillard reactions. The trends in free amino loss percentage observed across different PZ samples were consistent with both intermediate and final product formation patterns, indicating a uniform glycation mechanism throughout processing. The above results indicate that the structure of PZ was altered during heat treatment, an effect that is thought to be related to the Maillard reaction between PZ and polysaccharides in the wort. The Maillard reaction is recognized as a pathway to modulate protein function. Previous studies have shown that it can affect protein functionality, such as thermal stability and foaming properties, through structural modification [27].

3.3. Structural Differences Between PZ Molecules with Different Processing

FT-IR served as an effective analytical method for monitoring heat-induced conformational and structural changes in PZ. As shown in Figure 3A, the FT–IR spectra of all PZ samples exhibited two distinct absorption bands at 3000–3500 cm⁻¹ and 1100 cm⁻¹. The broad absorption band around 3200 cm⁻¹ was commonly associated with N–H stretching vibrations, primarily from amino groups on protein side chains. The gradual decrease in intensity of this band during heating indicated the consumption of free amino groups through their reaction with polysaccharides as processing progressed. The weakening of the peak at 1100–1000 cm⁻¹ in the FT–IR spectrum of PZ, after mashing, which corresponds to C–OH and C–C stretching vibrations of polysaccharides, suggested the hydrolysis of its glycan chains [13]. This effect was more pronounced in PZ after boiling than in PZ, after mashing, an observation that may be related to the partial degradation of polysaccharides via caramelization induced by prolonged boiling, consistent with the browning intensity results [25]. The intensity of the peak at 3600–3000 cm decreased after mashing, with a further reduction following boiling, further suggesting the occurrence of glycan chain hydrolysis during both processes.

The amide I band (1600–1700 cm⁻¹) second derivative spectrum was analyzed to investigate protein secondary structural changes. Figure 3B showed the percentage distribution of secondary structures for different PZ samples. The results indicated that the proportion of random coil increased during processing, while α–helix and β–sheet content decreased. For PZ after mashing, this structural alteration was associated with the loosening of protein molecules [26], likely resulting from the introduction of exogenous polysaccharide chains through Maillard reactions, as well as impaired hydrogen bonding under high-temperature conditions. Ultimately, an increase in β-sheet content was observed. PZ after boiling exhibited the highest content of random coil (21.42 ± 0.14%) and β-turn (24.08 ± 0.34%), consistent with previous similar studies [13]. This phenomenon was linked to the disruption of the hydrogen-bond networks essential for maintaining the α-helical conformation, which, in turn, contributed to further unfolding of PZ’s secondary structure. Prolonged heating had been shown to disrupt the dense structure of oat protein isolate, leading to a transition toward a more disordered conformation characterized by random coil formation, as documented in prior research [27]. Notably, prolonged boiling caused significant irreversible structural damage to proteins, ultimately leading to denaturation and loss of functional activity.

Similarly, the tertiary structure of PZ changed during mashing and boiling processes. As shown in Figure 3C, the fluorescence intensity gradually decreases during processing, an effect attributed to the oxidative degradation of tyrosine, resulting in the loss of fluorescence ability [28]. This phenomenon indicated that multiple chromophore groups, previously buried in the protein interior, became exposed to the exterior environment, leading to fluorescence quenching. However, the differential red-shift patterns observed among processed PZ samples were associated with the mashing process. This observation suggested that the mashing process exposed the β-sheet, resulting in the exposure of aromatic moieties (tryptophan and tyrosine residues) to polar environments and reflecting the unfolding of PZ’s tertiary structure [29]. In contrast, PZ, after boiling, underwent more extensive denaturation under severe heating conditions, with β-sheet structures that previously buried most tryptophan residues being converted into random coils. Consequently, PZ, after boiling, showed the most pronounced fluorescence quenching and red-shift effects.

3.4. The Thermal Stability of PZ Molecules with Different Processing

The thermal stability of PZ was highly relevant to its potential applications in the food industry. DSC measurements were used to evaluate protein thermal stability [28]. As observed in Figure 3D, the T_p (temperature peak) value of PZ was 84.74 ±1.42 °C, which suggested high thermal stability and was consistent with minimal structural alteration during the mashing stage. This result might be consistent with [29]. The T_p of PZ, after mashing, was 92.92 ± 0.32 °C, which was higher than that of unprocessed PZ. Similarly, the T_p value of PZ after boiling was 90.18 ± 0.84 °C. These results suggested that mashing and boiling processes could improve the thermal denaturation temperature of PZ. With the Maillard reaction in mind, PZ, after mashing and boiling, had more polysaccharide chains, which brought a large amount of hydroxyl groups, and effectively inhibits heat transfer and consequently enhances protein thermostability [30]. These results were consistent with the free amino loss percentage trends. In contrast to PZ after mashing, PZ after boiling exhibited a partial loss of α-helical content, which was associated with decreased thermal stability, as evidenced by FT-IR analysis.

3.5. The Air/Water Interface Properties of PZ Molecules with Different Processing

Foaming performance was a crucial characteristic of PZ in beer applications. During mashing and boiling, the foaming properties of PZ might change. Figure 4A,B showed the foaming properties of different PZ samples. As expected, PZ, after mashing, showed superior foamability (41.4 ± 0.38%) compared to unprocessed PZ, while PZ after boiling exhibited the highest foam stability (68.54 ± 1.12% retention over 30 min), which might be related to distinct protein adsorption behaviors at the water/air interface [31]. As shown in Figure 4C, all samples exhibited a time-dependent reduction in interfacial tension, with PZ after mashing achieving the lowest tension (45.90 mN/m). This observation was consistent with the superior foamability of PZ after mashing.

As the key stage for interface properties, K_diff, which was the slope of the linear relationship between surface pressure (π) and t_1/2, changed with different processing during the initial adsorption phase (Figure A1). As shown in

Table 1. It showed the K_diff, K_P and K_R.

	Kdiff	KP (×10−5)	KR (×10−3)
PZ	0.76 ± 0. 02 b	−79.77 ± 0.07 b	−4.94 ± 0.04 b
PZ after masing	1.05 ± 0.03 a	−74.87 ± 0.08 a	−4.42 ± 0.03 a
PZ after boiling	0.74 ± 0.01 c	−109.00 ± 0.08 c	−6.13 ± 0.06 c

Different lowercase letters denote statistically significant differences between groups (p < 0.05).

, PZ, after mashing, showed the highest K_diff value, which was consistent with its superior foamability. Previous studies had identified surface hydrophobicity as a key determinant of K_diff, with more hydrophobic proteins displaying stronger affinity for the air–water interface [32]. This correlation was particularly relevant for PZ, which contained >50% hydrophobic amino acids in its primary structure (Figure A2), a feature commonly associated with enhanced foam formation. The lower foamability observed for unprocessed PZ among all samples might be attributed to its extensive exogenous polysaccharide chains forming a hydrophilic layer, which likely reduced the surface hydrophobicity of the protein [27].

The observed K_diff difference between processed PZ with lower and higher values, respectively, reflected their distinct α–helix contents, as the helix conformation promotes faster interfacial migration [33]. The first and second slopes obtained from fitting Equation (6) corresponded to the first–order K_P and K_R, respectively (Figure A1). The fitted values of these constants were presented in Table 1. PZ after boiling exhibited elevated K_P and K_R values, an effect that was due to its abundant random coil structures. These disordered conformations were thought to contribute to greater molecular flexibility, which might facilitate faster protein rearrangement at the water/air interface in response to the new force field [34]. This enhanced rearrangement capability contributed to the formation of a stable viscoelastic interfacial layer, an effect associated with improved foam stability [35]. In contrast, PZ, after mashing, exposed more hydrophobic groups that might promote faster migration to the interface, but its predominant β-sheet and α-helix content resulted in more rigid structures. This structural rigidity appeared to hinder molecular rearrangement at the interface, which was associated with the formation of a less stable interfacial layer and thus reduced foam stability. These distinct interfacial adsorption behaviors among PZ variants were ultimately linked to their differences in both foamability and foam stability. A previous study reported that PZ contributed to improved beer foam stability through structural modification upon binding with hop humulinone [2]. The present study provides insights into how changes in protein structure during beer thermal processing may influence foam stability.

3.6. Thrombin Inhibitory Activity of PZ Molecules with Different Processing

As one of the serpin family members, native PZ has been suggested to possess potential for thrombosis inhibition in previous reports [21], which might be attributed to its thrombin-inhibitory activity. Expectedly, both PZ and PZ, after mashing, exhibited thrombin-inhibiting activity, while PZ after boiling completely lost its inhibitory function (Figure 5A). This loss of activity was likely associated with extensive structural denaturation that compromised its essential serpin conformation. The boiling process led to significant denaturation of PZ’s secondary and tertiary structures (Section 3.3). The complete loss of its thrombin inhibitory activity was attributed to this structural disruption, consistent with the fact that serpin function was critically dependent on precise conformation [11]. Notably, PZ, after mashing, demonstrated significantly increased inhibitory activity compared to native PZ, suggesting that controlled thermal processes might influence its functional properties.

To elucidate the molecular basis of this enhancement, molecular docking studies were performed to analyze the binding mode between PZ and thrombin. The results were shown in Figure 5D. PZ exhibited a binding pattern similar to that of other serpins. Thrombin (a serine protease) bound to the RCL of PZ, where conserved restriction enzyme sites functioned as a structural “trap” for thrombin inhibition. The inhibition of thrombin by PZ followed the canonical serpin mechanism, in which thrombin must first cleave the RCL of PZ before forming a stable PZ-thrombin complex that irreversibly inactivates the protease. Based on the glycosylation profile of PZ and the diverse glycosidase activities during mashing, we hypothesize that the observed changes were associated with alterations in glycan chain length in both PZ and PZ after mashing.

This hypothesis could be evidenced using glycosylation mass spectrometry, which enables the identification of glycosylation sites and the precise quantification of chain-length variations. Figure 5B,C showed the glycosylation sites of PZ and PZ, after mashing, respectively. The figures revealed a reduction in glycosylation sites in PZ, after mashing, an effect that was associated with complex enzyme reactions occurring during this period.

Table 2. It showed the glycosylation sites, which might influence the functional ability of PZ.

	PZ		PZ After Mashing
Residues Position	Mods (Variable)	Molecular Weight of Glycan Glycan Chain	Mods (Variable)	Molecular Weight of Glycan Glycan Chain
Ser50	OGlycan/1021.3598	HexNAc(2)Hex(1)Fuc(1)NeuGc(1)	-
Ser41	OGlycan/406.1587	HexNAc(2)	-
Ser44	OGlycan/609.2381	HexNAc(3)	-
Ser50	OGlycan/818.2804	HexNAc(1)Hex(1)Fuc(1)NeuGc(1)	-
Ser44	OGlycan/609.2381	HexNAc(3)	-
Ser44	OGlycan/818.2804	HexNAc(1)Hex(1)Fuc(1)NeuGc(1)	-
Ser50	OGlycan/1850.6926	HexNAc(6)Hex(3)Fuc(1)	-
Ser50	OGlycan/1313.4756	HexNAc(2)Hex(2)Fuc(2)NeuAc(1)	-
Thr53	OGlycan/1238.4184	HexNAc(1)Hex(1)NeuAc(3)	-
Thr344	OGlycan/947.323	HexNAc(1)Hex(1)NeuAc(2)	-
Thr350	OGlycan/672.2225	HexNAc(1)Hex(1)NeuGc(1)	-
Thr350	OGlycan/673.2429	HexNAc(1)Hex(2)Fuc(1)	-

presents the changes in glycosylation status that might influence the functional properties of PZ. Mass spectrometric analysis identified O-glycosylation modifications at Thr344 within the RCL (Figure 5E). The steric hindrance associated with these glycans might reduce thrombin-binding affinity, thereby decreasing PZ’s inhibitory activity. In contrast, PZ, after mashing, lacked glycosylation at Thr344 (Figure 5F), an alteration that was associated with reduced steric interference and enhanced thrombin binding, potentially contributing to its higher inhibitory potency compared to unmodified PZ. In addition, protease binding triggered cleavage of the exposed RCL, enabling its incorporation as an additional β-strand (s4A) into the serpin’s central β-sheet (sA) [36]. Previous studies have shown that oversized amino acid substitutions at positions P10, P12, and P14 in s4A reduced s4A insertion rates, consequently impairing the affected activity [37]. Figure A3 showed the P10-P18 sites in PZ. The glycosylation site at Thr344 of PZ corresponded to the position where previous studies demonstrated that oversized amino acid substitutions could affect inhibitory activity. Therefore, it was plausible that Thr344 glycosylation might compromise PZ’s inhibitory function through effects on this essential structural rearrangement. In summary, glycosylation at Thr350 and Thr344 of PZ was due to reduced thrombin inhibitory activity, potentially through distinct mechanisms: (i) the steric hindrance contributed by the glycan at Thr350 might interfere with thrombin recognition; (ii) glycosylation at Thr344 might affect PZ allostery, a conformational change that was essential for its inhibitory function.

4. Conclusions

This study systematically investigated the effects of different processing methods (native, mashing, and boiling) on the structural and functional properties of PZ. During beer brewing, PZ underwent structural modifications to varying degrees and was involved in Maillard reactions. Following heat treatment, the Maillard reaction was associated with browning of PZ and facilitated covalent conjugation with exogenous polysaccharide chains from wort. Most notably, the glycosylation of PZ was reduced, an effect associated with enzymatic hydrolysis during the mashing process. Moreover, PZ after boiling exhibited more disordered structures than either unprocessed PZ or PZ, after mashing, consistent with protein denaturation induced by prolonged boiling temperatures. These changes influenced interface rearrangement behaviors, enhancing foam stability. After mashing, PZ demonstrated superior foamability, attributable to glycan chain shortening during glycosylation. This modification was associated with increased protein hydrophobicity, potentially facilitating its migration from the aqueous phase to the air/water interface. Thrombin inhibitory activity, an important characteristic of serpins, was influenced by thermal processing. For PZ, after boiling, the activity was completely lost, owing to protein denaturation. In contrast, the activity for PZ after mashing increased, which was linked to the hydrolysis of glycan chains during the mashing process, potentially enhancing both thrombin affinity and allosteric efficiency. These findings illustrate how structural changes in PZ during beer production are related to the development of functionally distinct variants. Nevertheless, certain limitations of this study should be acknowledged. Our analysis did not capture exogenous glycan chains that may have been introduced via the Maillard reaction during the mashing process; therefore, further investigation is warranted to address this aspect.