Functional improvements in β-lactoglobulin by preparing edible conjugate with microbial transglutaminase

ABSTRACT

To improve the function of bovine β-lactoglobulin (ΒLG), BLG was conjugated with epsilon polylysine (PL) by using microbial transglutaminase (MTGase). The molar ratio of BLG to PL was 1:1.2 according to amino acid analysis. About 50% of retinol binding ability of BLG was maintained in the conjugate. Emulsifying property of BLG in acidic pH range and in the presence of salt was improved by conjugating with PL. Immunogenicity of BLG was reduced by this conjugation without inducing novel immunogenicity in BALB/c mice and C3H/He mice. Conjugation method in this study is valuable in that it is applicable to food processing.

Graphical Abstract

By conjugation with ε-polylysine, emulsifying property of β-lactoglobulin was improved and immunogenicity of β-lactoglobulin was reduced.

β-lactoglobulin, neoglycoconjugate, functional improvement, protein conjugation

Abbreviations

Abbreviations
 
• BLG:

  β-lactoglobulin

 
• PL:

  epsilon polylysine

 
• SDS-PAGE:

  sodium dodecyl sulfate poly acrylamide gel electrophoresis

 
• CBB:

  Coomassie brilliant blue

 
• IEF:

  isoelectric focusing

β-Lactoglobulin (BLG) is a major whey protein of Mw 18 kDa consisting of 162 amino acids and possesses 2 disulfide bridges and 1 free cysteine residue (McKenzie 1971). Although the physiological function of BLG still remains unclear, it is tentatively considered to be the binding and transportation of small hydrophobic ligands such as retinol and fatty acids, and the protein is categorized as a member of the lipocalin superfamily (Sawyer and Kontopidis 2000).

In terms of food science, BLG is considered to be a valuable protein with various functional properties such as gelling (Foegeding, Kuhn and Hardin 1992), foaming (Phillips, Hawks and German 1995) and emulsifying properties (Shimizu, Saito and Yamauchi 1985) as well as high content of essential amino acids (McKenzie 1971).

BLG has a β-barrel structure (Sawyer and Kontopidis 2000) which is a common feature among the lipocalins. This kind of molecule has high allergenic potential, and several allergens of animal origin belong to the lipocalin superfamily (Virtanen et al.  1999). In fact, BLG is known as a potent allergen of milk allergy and ∼82% of milk allergy patients are sensitive to this protein (Spies 1973). Therefore, it is strongly hoped to develop new methods that would reduce the allergenicity of BLG. Although attempts to reduce the allergenicity of proteins have been made by enzymatic digestion and denaturation, these methods destroy the physiological functions of the proteins and bring about problems with their taste. In contrast, protein conjugation is superior to other hypoallergenic methods in that it can simultaneously achieve reduced allergenicity and improved functional properties (thermal stability, solubility, emulsifying ability, etc.) while maintaining the physiological functions of proteins (Hattori 2002).

In the present study, we conjugated epsilon-polylysine (PL) to BLG to cover the epitopes of BLG which would lead to reduced allergenicity and improved emulsifying property of BLG. PL is a cationic, naturally occurring homopolyamide made of l-lysine, which has amide linkages between ε-amino and α-carboxyl groups (Shima and Sakai 1981). It was accidentally discovered as an extracellular material produced by Streptomyces albulus ssp. Lysinopolymerus strain 346 (Shima and Sakai 1977). PL is known as a natural, toxicologically safe, antibacterial food preservative. We used microbial transglutaminase (MTGase) (EC2.3.1.13) to conjugate PL to BLG (Ando et al.  1989). This enzyme catalyzes the acyl-transfer mechanism in which the γ-carboxamide group acts as an acyl donor and suitably unbranched primary amines act as acyl acceptors. γ-Carboxamide group of Gln residue in BLG and epsilon-amino residue of PL reacts. MTGase is an enzyme used to improve texture, and products catalyzed by this enzyme would be edible, as well as BLG–PL produced in this manner. In the present report, we will describe on the preparation of the BLG–PL conjugate and improvements in emulsifying property and immunogenicity of BLG.

Materials and methods

Materials

ε-Polylysine was a gift from Chisso Corporation (Tokyo Japan) (average polymerization degree; 30). Activa TG-K (E.C.2.3.2.13) was a gift from Ajinomoto Co. (Tokyo, Japan).

Preparation of BLG

BLG (genotype AA) isolated from fresh milk of a Holstein cow according to the method of Armstrong, McKenzie and Sawyer (1967) was further purified by ion-exchange chromatography in a DEAE-Sepharose Fast Flow column (2.5 ID × 50 cm; Amersham Biosciences, Buckinghamshire, UK). Crude BLG was applied to the column and eluted by a 0-500 m m NaCl linear gradient in a 0.05 m imidazole buffer at pH 6.7 and a flow rate of 1.0 mL/min. The eluted protein was detected by the absorbance at 280 nm. The major fraction was dialyzed against distilled water and lyophilized. The purity of BLG was confirmed by SDS-PAGE.

Preparation of the ΒLG–PL conjugate

0.2% solution of MTGase was used as MTGase for the enzymatic reaction. Activa TG-K (5 g) was dissolved in 0.1 m imidazole buffer (pH 7.0, 25 mL) and centrifuged at 4 °C at 18 000 rpm for 30 min. Enzyme titer of Activa TG-K was 100 units/g. filtrated with paper, and dialyzed overnight against the same buffer to remove calcium lactate and dextrin. MTGase-catalyzed reaction was carried out under the condition of BLG: PL = 1:3. BLG and PL were dissolved in 0.2 m imidazole buffer (pH 7.0, 20 mL). 200 mg of BLG and 126 mg of PL were used for conjugation. Enzymatic reaction was started by adding 20 mL of MTGase in this solution. The reaction mixture was incubated at 37 °C for 48 h.

Purification of the ΒLG–PL conjugate

After the enzyme reaction, pH was adjusted to 7.7 × 0.1 m NH₃, and applied to a CM Sepharose Fast Flow column (2.4 ID × 45 cm, Amersham Biosciences, Buckinghamshire, UK) to remove free ΒLG, free PL and MTGase. The column was equilibrated with 0.1 m Tris buffer (pH7.7) in advance, equilibrated with 200 mL of the same buffer, and then eluted with a linear gradient of 0-2.5 m NaCl. For detection of proteins and peptides, the absorbance was monitored at 230 nm. The purity of the conjugate was confirmed by SDS-PAGE, and then dialyzed against distilled water. Purified BLG–PL fractions were lyophilized, and fractions containing extra PL was purified again by the same procedure. This procedure was repeated until extra PL was removed.

Electrophoresis

SDS-PAGE was carried out in 4% stacking gel and 15% separating gel following the method of Laemmli (1970). Prior to electrophoresis, sample proteins were heated at 100 °C for 5 min in SDS-PAGE sample buffer. The gel electrophoresis was carried out at 20 mA constant current. The gels were stained with Coomassie brilliant blue (CBB).

Isoelectric focusing (IEF) was carried out with PhastSystem separation unit (Amersham Biosciences, Buckinghamshire, UK) (Kramlova, Printstoupil and Kraml 1986). The samples were applied to IEF gels (PhastGel IEF 3-9, Amersham Biosciences, Buckinghamshire, UK). The gel was stained with CBB R250.

Amino acid analysis

Amino acid analysis was carried out with LS 8800 High Speed Amino Acid Analyzer (Hitachi, Tokyo, Japan) to determine the ratio of BLG to PL in the BLG–PL conjugate. Samples were hydrolyzed in vacuo in 6 m at 110 °C for 24 h in advance. The hydrolysis was carried out in triplicate. Asp was used as an indicative amino acid to evaluate the amount of BLG in the BLG–PL conjugate. PL amount was evaluated by the amount of Lys.

CD spectrum measurement

CD spectrum of the BLG–PL conjugate was measured with a J-720 spectropolarimeter (Jasco, Tokyo, Japan), using a cell with 1.0 mm path length (Hattori et al.  1993). Each sample was dissolved in phosphate buffered saline (PBS; a 0.11 m phosphate buffer at pH 7.0 containing 0.04 m NaCl and 0.02% NaN₃) at a protein concentration of 0.1 mg/mL.

Fluorescence measurement

The intrinsic fluorescence of the BLG–PL conjugate dissolved in PBS at a protein concentration of 0.001% was measured under an excitation at 283 nm with an RF-5300PC instrument (Shimadzu, Kyoto, Japan).

Measurement of the retinol-binding activity of the BLG–PL conjugate

The retinol-binding ability of the BLG–PL conjugate was measured by emission of fluorescence titration (Futterman and Heller 1972; Cogan et al.  1976). Samples were dissolved in 2 mL of PBS (pH7.0) at a concentration of 0.01% (as a protein). Small increments (5 µL at a time) of retinol in ethanol at 3.33 × 10⁻⁵  m were added to the 2 mL BLG/BLG–PL solvent, and the fluorescence was measured with a Shimadzu RF-5300PC instrument (Kyoto, Japan) with excitation at 330 nm and emission at 470 nm.

Evaluation of the emulsifying property of the BLG–PL conjugate

Emulsifying property of BLG–PL was evaluated by the turbidity method by Pearce and Kinsella (1978). Samples were dissolved in a Mcllviane buffer at pH3.0, 5.0, 7.0 to evaluate the emulsifying property at different pHs, or in the buffer at pH5.0 and pH7.0 containing NaCl (0.1, 0.2, or 0.5 m). Protein concentration was 1 mg/mL), and PL concentration was adjusted to the amount of PL conjugated to the BLG–PL conjugate. To prepare an oil-in-water emulsion, 2 mL of a sample solution and 0.5 mL of corn oil were homogenized by a Polytron PTA-7 (Kinematica, Switzerland) at room temperature at 24 000 rpm for 1 min. Aliquot of 50 µL was taken after 0,10, 30, 60 and 120 min from the bottom of the homogenized emulsion, and 100-fold diluted with 0.1% SDS solution. The absorbance was read at 500 nm by a spectrophotometer. The emulsion stability was determined by the turbidity 30 min after homogenization. The emulsifying activity was calculated as follows.

$$\begin{equation*}{\rm{EAI}}( {{{\rm{m}}^2}/{\rm{g}}}) = 2\,T/\varphi C \end{equation*}$$

where T = 2.3A/L (A = A₅₀₀, L = 10⁻² m [light path], and φ = 0.2 [oil-phase volume fraction]).

Immunization

Female BALB/c and C57BL/6 mice (Clea Japan, Tokyo, Japan) at 6 weeks of age (7 animals per group) were immunized intraperitoneally with 100 µg (as a protein) of BLG or BLG–PL emulsified in Freund's complete adjuvant (Difco Laboratories, MI, USA). Fourteen days after the primary immunization, the mice were boosted with 100 µg of protein emulsified in Freund's incomplete adjuvant (Difco Laboratories). Blood samples were collected from mice 7 days after the secondary immunization and stored at 4 °C for 24 h to form a clot. Antisera were collected from each blood sample after clot formation. This study was performed in conformance with the guidelines for the care and use of experimental animals established by the ethics committee of the Tokyo University of Agriculture and Technology (R03-186, July 29, 2021).

Enzyme-linked immunosorbent assay (ELISA)

Noncompetitive ELISA was carried out as follows. BLG and the BLG–PL conjugate were dissolved in PBS at a protein concentration of 0.1 mg/mL (100 µL) and added to the wells of a polystyrene microtitration plate (Maxisorp, Nunc, Roskilde, Denmark), and the plate was incubated overnight at 4 °C to coat the wells with each antigen. After removal of the solution, each well was washed 3 times with 200 µL of PBS-Tween (PBS containing 0.05% Tween 20). A 125 µL amount of a 1% ovalbumin/PBS solution was added to each well, the plate was incubated for 2 h, and then the plate was washed 3 times with 200 µL of PBS-Tween. A 100 µL amount of an antiserum diluted with PBS was added to each well, and the plate was incubated for 2 h. After 3 washing, 100 µL of alkaline phosphatase-labeled rabbit antimouse immunoglobulin diluted with PBS-Tween was added to each well. The plate was incubated for 2 h, and then the wells were washed 3 times. 100 µL amount of 0.1% p-nitrophenyl phosphate disodium salt dissolved in a 1 m diethanolamine hydrochloride buffer (pH 9.8) was added to each well, and the plate was incubated at 2 °C. After the addition of a 5 m sodium hydroxide solution (20 µL) to each well to stop the reaction, the absorbance at 405 nm was measured with an MPR-A4i microplate reader (MPR-A4i, Tosoh, Tokyo, Japan).

Competitive ELISA was carried out to investigate the local conformational changes in BLG after conjugating with PL by using anti-BLG mAbs (mAbs 21B3, 31A4, 61B4 and 62B6) (Hattori et al.  1993, 1994). The equilibrium constants (K_AS) of the mAbs with the BLG and BLG–PL conjugate were calculated (Hogg et al.  1987).

Results and discussion

Isolation and characterization of the BLG–PL conjugate

The BLG–PL conjugate was obtained after CM chromatography (Figure 1a). About 40 mg of the BLG–PL conjugate was obtained from the enzymatic reaction using 200 mg of BLG and 126 mg of PL. According to SDS-PAGE, the BLG–PL conjugate showed 2 different molecular weights (Figure 1b). Fraction numbers 65-68 were collected and 69-78 were subjected to rechromatography and collected. These fractions were used as the BLG–PL conjugate in subsequent experiments. Molecular weight of the BLG–PL was approximately 30 kDa which is corresponded to 3-4 PLs conjugated to BLG. To evaluate the accurate pI of BLG–PL, isoelectric focusing was carried out after removing the free PL by ion-exchange chromatography. Migrated band showed broad band in the pI range of approximately 6.5-8.3. The pI value of the BLG–PL conjugate was about 7.5 which was determined by the band of deepest color. By amino acid analysis, the ratio of BLG to PL in the conjugate was clarified to be 1:1.2.

Figure 1.

CM-chromatographic pattern and SDS-PAGE pattern of the BLG–PL conjugate. (a) CM-chromatographic pattern: column, CM-Sepharose Fast Flow (2.5 ID × 45 cm, GE Healthcare, Buckinghamshire, UK); flow rate, 3 mL/min; elution, 0-1.0 m NaCl linear gradient elution in 0.1 m imidazole buffer (pH7.7); fraction, 10 mL; detection, absorbance at 280 nm. (b) SDS-PAGE pattern: SDS-PAGE was conducted by Laemmli's method. The concentration of acrylamide was 4% for the stacking gel and 15% for the running gel. Marker and samples were run at a constant current of 20 mA per gel for 90 min. After electrophoresis, it was attained with 0.1% CBB/40% CH₃OH/10% CH₃COOH and decolorized with 10% CH₃OH/7% CH₃COOH. M, molecular weight marker (Mw, 97 000, 66 000, 45 000, 30 000, 20 100, and 14 400).

Structural features of the BLG–PL conjugate

Structural changes in BLG after conjugation with PL was evaluated by 4 methods: CD spectra measurement, intrinsic fluorescence measurement, ELISA with mAb, and evaluation of retinol-binding activity.

CD spectra of the BLG–PL conjugate and the mixture of BLG and PL (Figure 2a) showed a negative maximum at 214 nm, which indicated that BLG is rich in β-sheet structure. CD spectrum of the conjugate was similar to that of the mixture of BLG and PL, which indicated that the secondary structure of BLG was almost maintained after conjugation with PL.

Figure 2.

Structural analysis of the BLG–PL conjugate. (a) CD spectrum of the BLG–PL conjugate: CD spectra of BLG and BLG–PL conjugate were measured with Jasco J-720 WI Spectropolarimeter (Jasco, Tokyo, Japan). BLG and PL mixture (broken line), BLG–PL (solid line). (b) Intrinsic fluorescence of the BLG–PL conjugate: the intrinsic fluorescence of the BLG–PL conjugate dissolved in PBS was measured under an excitation wavelength of 283 nm by using RF 5300PC fluorescence spectrophotometer (Shimadzu, Kyoto, Japan). BLG (dotted line), BLG–PL (solid line). (c) Equilibrium constants (K_AS) of the BLG–PL conjugate in binding to anti-BLG mAbs. Competitive ELISA was carried out to investigate local conformational changes in the BLG–PL conjugate by using anti-BLG mAbs (mAbs 21B3, 31A4, 61B4, and 62A6) as probes. K_AS values were calculated from the results of competitive ELISA according to the procedure of Hogg et al. BLG (●), BLG–PL (◯). (d) Retinol-binding activity of the BLG–PL conjugate. Retinol-binding ability was measured by fluorescence titration. Retinol in ethanol was added to 2.0 mL solution in a cuvette containing 0.01% of BLG in PBS. The fluorescence was measured with excitation at 470 nm. BLG (●), BLG–PL (◯), PL (▲).

Intrinsic fluorescence of the BLG–PL conjugate is shown in Figure 2b. As the conformation of BLG changes, the fluorescence intensity increases with red shift of the wavelength for maximum emission. The BLG–PL conjugate showed similar wavelength for maximum emission (334 nm) to that of BLG. Hence, the conformation around Trp residue (¹⁹Trp and ⁶¹Trp) of BLG–PL was considered to maintain native form. On the other hand, fluorescence intensity for the BLG–PL was lower than that for BLG, which indicates that bound PL shielded the fluorescence from Trp in the conjugate.

The local conformational changes in BLG after conjugation with PL was evaluated by competitive ELISA with 4 anti-BLG mAbs (21B3, 31A4, 61B4, and 62A6) (Figure 2c). These mAbs can be used to detect subtle conformational changes in local areas within the BLG molecule during unfolding and refolding and after conjugation with other substances by determining the affinity change. The epitope region for mAbs 21B3 and 31A4, 61B4 are ¹⁵Val-²⁹Ile (β-sheet), ⁸Lys-¹⁹Trp (random coil, short helix), ¹²⁵Thr-¹³⁵Lys (α-helix). The epitope region for mAb 62A6 is adjacent to that for 61B4. MAb 61B4 and 62A6 bind preferentially to native BLG, whereas mAb 21B3 and 31A4 bind more strongly to denatured form of BLG. The equilibrium constants (K_AS) for binding 21B3 and 31A4 to the BLG–PL were similar to those to BLG. However, K_AS for binding 61B4 and 62A6 were smaller than those to BLG. The conformation around ¹⁵Val-²⁹Ile (β-sheet) and ⁸Lys-¹⁹Trp (random coil, short helix) was considered to have maintained the native structure, whereas the conformation around ¹²⁵Thr-¹³⁵Lys (α-helix) was considered to have slightly changed (Hattori et al.  1993).

Retinol binding activity of the BLG–PL conjugate was evaluated by fluorescence titration. The titration curve is shown in Figure 2d. The maximum retinol binding ability of the BLG–PL conjugate was about 54% of that of BLG. This result indicates that the rigidity of BLG has changed by conjugating with PL (Katakura et al.  1994).

Although a little collapse in the conformational changes occurred, the BLG–PL conjugate almost maintained the native-like structure of BLG.

Improvement in emulsifying property of BLG by conjugation with PL

Emulsifying property of the BLG–PL conjugate was evaluated by turbidity method at various pHs and in the presence of salt. The effect of pH on the emulsifying ability of BLG and the BLG–PL conjugate was evaluated on the basis of the emulsifying activity index (EAI). EAI value increased as the pH value decreased and reached the highest EAI value at pH3.0 (the lowest pH that has been conducted on this experiment) (Figure 3a). The emulsion stability of BLG–PL at pH 7.0 was as high as that of BLG, but higher at low pH range (Figure 3b). The addition of hydrophilicity by conjugation with PL is considered to have enhanced the emulsifying property of BLG. Nagasawa et al. (1996) revealed that increase in polysaccharide content by conjugation with acidic polysaccharides was more effective to improve the emulsifying property of BLG. Their findings indicate that addition of hydrophilicity is important for the emulsifying property of BLG bioconjugates. In the case of this study, addition of hydrophilicity by conjugation with PL is considered to be important for improved emulsifying property of BLG.

Figure 3.

Emulsifying properties of the BLG–PL conjugate at various pH values. Emulsifying properties of the BLG–PL conjugate at various pH values. The emulsifying properties of the BLG–PL conjugate at various pH values were evaluated by the turbidimetric method (Pearce and Kinsella 1978). (a) The emulsifying activity was evaluated by the EAI. (b) The emulsion stability was evaluated by the absorbance at 500 nm 30 min after emulsification. Values at the same pH were compared by Tukey–Kramer test. Different alphabet indicates a significant difference (p < .01). BLG (●), BLG–PL (◯), BLG + PL (△).

The emulsifying property of the BLG–PL conjugate in the presence of salt was also evaluated at pH 5.0 (Figure 4). Emulsifying ability generally degrades when salt exists because salt binds to charged emulsifiers and decreases its net charge, resulting to unstabilize oil droplets by the degradation of electrostatic repulsion. In the presence of 0.5 m NaCl, the BLG–PL conjugate maintained its emulsifying activity. There was little influence of salt concentration on the emulsifying property of BLG but EAI value of BLG was very low at all salt concentration. As for emulsion stability, values for BLG were low at all NaCl concentrations investigated. Although the emulsion stability of the emulsion prepared with the conjugate decreased as NaCl concentration increased, the emulsion stability of the emulsion prepared with the conjugate was higher than the emulsion prepared with BLG. Emulsifying property of BLG was much improved by conjugation with PL. Nagasawa et al. (1996) also revealed that addition of net charge by conjugation with acidic polysaccharides was more effective to improve the emulsifying property of BLG in the presence of salt. In the case of this study, addition of net charge by conjugation with PL is considered to be important for improved emulsifying property of BLG in the presence of salt.

Figure 4.

Emulsifying properties of the BLG–PL conjugate in the presence of 0.2 m NaCl. The emulsifying properties of the BLG–PL conjugate in the presence of 0.2 m NaCl were evaluated by the turbidimetric method. (a) The emulsifying activity was evaluated by the EAI. (b) The emulsion stability was evaluated by the absorbance at 500 nm 30 min after emulsification. Values at the same NaCl concentration were compared by Tukey–Kramer test. Different alphabet indicates a significant difference (p < .01). BLG (●), BLG–PL (◯), and BLG + PL (△).

Reduced immunogenicity of BLG by conjugation with PL

The immunogenicity of the BLG–PL conjugate in BALB/c and C57BL/6 mice was evaluated by noncompetitive ELISA (Figure 5). The average anti-BLG antibody response was significantly lowered after conjugation with PL in both strains of mice (Figure 5a and c). In both strains, novel immunogenicity of BLG was not observed after conjugation with PL (Figure 5b and d). Conjugation with PL was considered to be an effective method to reduce immunogenicity of BLG without inducing novel immunogenicity.

Figure 5.

Immunogenicity of the BLG–PL conjugate. Seven heads of BALB/c and C57BL/6 mice were immunized with BLG or BLG–PL. (a, c) The anti-BLG response, (b, d) anti-BLG–PL response after the secondary immunization were evaluated by noncompetitive ELISA. A significant difference as determined by Student's t-test is indicated by a single asterisk (p < .05). In the panels of (a) and (c), the results using BLG as coating antigens are shown. In the panels of (b) and (d), the results using BLG–PL as coating antigens are shown.

Conclusion

In this study, we prepared BLG–PL conjugate with modified functionality by using MTGase. By conjugation with PL, the emulsifying properties of BLG in the acidic pH region and in the presence of NaCl were much improved. Because acidic pHs are frequently used in food, the BLG–PL conjugate is considered to be useful for food applications. Immunogenicity of BLG was reduced by this conjugation. Since the conjugation method used in this study is a safe method, this method is very valuable in that it would be applicable for food processing.

Acknowledgments

We are very grateful to Emeritus Professor Shuichi Kaminogawa of The University of Tokyo for his help in the ELISA experiments. We thank CHISSO CORPORATION (Tokyo, Japan) for presenting ε-polylysine. We thank Ajinomoto Co. (Tokyo, Japan) for presenting Activa TG-K (E.C.2.3.2.13). This work was supported in part by JSPS KAKENHI Grant Number JP22580126.

Data availability

The data underlying this article are available in the article and also from the corresponding author upon request.

Author contribution

T.Y. assisted in conceptualization, data curation, and writing manuscript; M.T., A.S., O, H.T., S.T. assisted in data curation, and writing manuscript; K.T. assisted in conceptualization and writing manuscript; MH assisted in conceptualization, writing manuscript, and funding acquisition.

Funding

None declared.

Disclosure statement

No potential conflict of interest was reported by the authors.

Author notes

These authors equally contributed to this work.