Self-association of casein studied using enzymatic …...Please cite this article as: Raak N, Brehm...

Please cite this article as: Raak N, Brehm L, Abbate RA, Henle T, Lederer A, Rohm H, Jaros D (2019) Self-association of casein studied using enzymatic cross-linking at different temperatures. Food Bioscience 28, 89-98.The final publication is available at https://doi.org/10.1016/j.fbio.2019.01.016.

1

Original Manuscript

Self-association of casein studied using enzymatic cross-linking at

different temperatures

Running title: Enzymatic casein cross-linking at different temperatures

Norbert Raak1,*, Lena Brehm1, Raffaele Andrea Abbate2,3, Thomas Henle4, Albena Lederer2,3,

Harald Rohm1, Doris Jaros1

1 Chair of Food Engineering, Institute of Natural Materials Technology, Technische Universität Dresden, 01062 Dresden,

Germany

2 Leibniz-Institut für Polymerforschung Dresden e.V., 01069 Dresden, Germany

3 School of Science, Technische Universität Dresden, 01062 Dresden, Germany

4 Institute of Food Chemistry, Technische Universität Dresden, 01062 Dresden, Germany

*Corresponding author: [email protected], phone: +49 351 463 34219, fax: +49 351 463 37761 (N. Raak)

© 2019. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/

https://doi.org/10.1016/j.fbio.2019.01.016

http://creativecommons.org/licenses/by-nc-nd/4.0/


2

Abstract

Sodium caseinate and β-casein self-associate to casein particles in solution because of

hydrophobic interactions. Microbial transglutaminase (mTGase) presumably preferentially cross-

links molecules that are located within the same particles, meaning that polymer size is limited to

the number of monomers initially present in the particles. The aim of this study was to affect the

self-association of casein by varying the temperature and thereby controlling the maximum size of

casein polymers formed using mTGase. Activity and stability of mTGase were determined at

different temperatures, showing that about 10 times more enzyme had to be added at 10 than at

40°C to compensate for differences in enzyme activity. Analysis with gel electrophoresis showed

that incubation temperature had no effect on maximum polymer size at the selected protein

concentration (27 g/kg), resulting in similar stiffness of acid-induced gels after incubation for 24 hr.

For a more detailed characterization of casein polymers, size exclusion chromatography was

coupled to multi-angle light scattering (MALS). Estimated molar mass distributions of casein

polymers were similar at both incubation temperatures and the increase in molar mass leveled off

after moderate incubation time. This underlines the idea of a maximum polymer size and suggests

no cross-linking between existing polymers. Besides that, the MALS detector showed a

contaminant of low concentration but large size which co-eluted with casein polymers and possibly

led to overestimation of their molar masses.

Highlights

• Sodium caseinate and β-casein were cross-linked using microbial transglutaminase

(mTGase)

• Activity and stability of mTGase were determined as a function of temperature

• Incubation temperature had no effect on maximum polymer size

• Size exclusion chromatography was coupled to multi-angle light scattering

• Molar mass determination confirmed the occurrence of maximum polymer sizes

Keywords: Milk, Casein, Transglutaminase, Gelation





3

1. Introduction

Enzymatic cross-linking of food proteins remains of interest to scientific research and industrial

applications. Microbial transglutaminase (mTGase; EC 2.3.2.13) predominantly catalyses acyl

transfer reactions between γ-carboxamide groups of protein-bound glutamine residues and primary

amines such as ε-amino groups of protein-bound lysine residues (Buchert et al., 2010; Gaspar and

de Góes-Favoni, 2015), and its application in yoghurt (Jaros et al., 2006a; Gharibzahedi and

Chronakis, 2018a; Loveday et al., 2013) and other dairy products (Gharibzahedi et al., 2018b;

Romeih and Walker, 2017) has been reviewed. Because of its simple cross-linking reaction

mTGase is also suitable to create well-defined structures in model substrates such as sodium

caseinate (NaCn) which, in turn, allows studying structure-function-interrelations. Dissolved NaCn

self-associates to particles consisting of 9 – 11 monomers at ambient temperature because of

interactions between hydrophobic side chains (HadjSadok et al., 2008; Huppertz et al., 2017). Due

to the repulsive forces between individual casein particles it is likely that cross-linking with mTGase

occurs preferably between molecules that are located within the same particle, as has been shown

for the supramolecular casein micelles in milk (Mounsey et al., 2005). In this case, the maximum

achievable polymer size would depend on the aggregation number of the casein particles and thus

might be adjustable through controlling the self-association, e.g., changing temperature, pH, or

ionic strength (HadjSadok et al., 2008). In a previous study, caseinate was cross-linked using

mTGase in different ionic milieus and larger polymers were observed at a higher ionic strength and

in the presence of Ca2+ ions, although N-ε-(γ-glutamyl)-lysine isopeptide content and

polymerization degree were comparable (Raak et al., 2019). This was ascribed to an increased

casein self-association triggered by the ions (HadjSadok et al., 2008) since cross-linking of

molecules within the same casein particle would result in polymers with higher monomer numbers

when self-association is more pronounced. The ionic milieu, however, affected the gelation

behavior of casein. Therefore, from these experiments alone no distinct conclusion could be drawn

on the effect of polymer size on gel stiffness.

With regard to the effect of temperature on casein self-association, results are rather

contradictory. While HadjSadok et al. (2008) observed a reversible increase in molar mass of

casein particles by a factor of 2 when temperature was raised from 10 to 60°C, Ruis et al. (2007)

found no influence. On the other hand, pure β-casein showed a pronounced temperature-

dependent association behavior; some studies point to the presence of individual monomers at

temperatures below 15°C (Dauphas et al., 2005; de Kruif and Grinberg, 2002; O'Connell et al.,

2003a). O'Connell and de Kruif (2003b) concluded that mTGase forms intramolecular isopeptide

bonds in monomeric β-casein at ~0°C, whereas intermolecular cross-links were formed between

associated β-casein molecules at elevated temperatures. Other studies, however, indicated that





4

β-casein can associate even at low temperature if the concentration is sufficiently high (Moitzi et

al., 2008; Portnaya et al., 2006), and Cragnell et al. (2017) recently reported that β-casein has a

polydisperse size distribution at ambient temperature, ranging from monomers to particles

consisting of more than 100 molecules.

The main objective of this study was to explore the temperature-dependent self-association of

NaCn and β-casein by cross-linking with mTGase. Assuming that predominantly molecules within

the same casein particle are cross-linked, different maximum polymer sizes will be observed when

temperature affects the number of monomers in the particles. With this approach, the significance

of polymer size on stiffness of acid-induced gels might be evaluated. However, common methods

for size determination, i.e., sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-

PAGE) and size exclusion chromatography (SEC) allow a rather limited discrimination between

casein polymers while monomers, dimers and trimers are frequently the only fractions

distinguishable (Raak et al., 2018). Size standards for SEC were previously used to estimate molar

mass ranges (Hiller and Lorenzen, 2008, 2009; Moeckel et al., 2016), but it is still not clear which

polymers represented the hydrodynamic volume and thus the elution behavior of caseins and

especially casein polymers to enable molar mass calibration. Therefore, establishing the absolute

molar mass of the eluted fractions using on-line multi-angle light scattering (MALS) detection was

used to obtain reliable differentiation of the polymeric fractions. Such an approach had been used

previously to characterize the composition of different NaCn preparations (Lucey et al., 2000) and

to compare β-casein cross-linking using tyrosinase and mTGase (Monogioudi et al., 2009). In the

latter study, however, average values were calculated instead of the molar mass distributions, and

the molar mass of β-casein monomers was not determined.

2. Materials and methods

2.1. Chemicals and reagents

mTGase “Activa MP” from Streptomyces mobaraensis was provided by Ajinomoto Foods

Europe SAS (Hamburg, Germany). Acid casein powder from bovine milk with 879 g/kg crude

protein (determined using Kjeldahl method, N × 6.38; IDF, 1979) was purchased from Sigma-

Aldrich GmbH (Steinheim, Germany). Skim milk powder (fat content <10 g/kg according to the

manufacturer) was purchased from Sachsenmilch Leppersdorf GmbH (Leppersdorf, Germany).

Glucono-δ-lactone was provided by Kampffmeyer Nachf. GmbH (Ratzeburg, Germany). All other

chemicals were of analytical grade; a detailed list can be found in Tab. 1. Demineralized water was





5

prepared from tap water using an ELGA Purelab Option system (ELGA LabWater Veolia Water

Technologies Deutschland GmbH, Celle, Germany).

2.2. Activity and stability of microbial transglutaminase as a function of temperature

Enzyme activity was determined using the method of Folk and Cole (1965; 1966) with slight

modifications. mTGase powder was dissolved in 0.2 M Tris acetate buffer (pH 6.0) at 5 mg/ml and

blended 1:2 with substrate reagent (0.01 M L-glutathione, 0.1 M hydroxylamine hydrochloride,

0.03 M N-carboxybenzyl-L-glutaminyl-glycine (Z-Gln-Gly) in Tris acetate buffer; pH 6.0). The

samples were incubated in a water bath at 10 – 60°C at 10°C intervals for 10 min, followed by

blending 1:1 with FeCl3 reagent (50 mg/ml FeCl3 in 0.1 M HCl, 120 ml/l HCl, and 120 g/l

trichloroacetic acid in a volumetric ratio of 1:1:1) to stop the enzymatic reaction and to induce a

color reaction with hydroxamate. The concentration of hydroxamate was determined

photometrically at λ = 525 nm (Ultrospec 8000, GE Healthcare Europe GmbH, Freiburg, Germany)

using L-glutamic acid γ-monohydroxamate as the calibration substance (concentration range

0.5 – 2.5 mM, 5 points, linear regression, R² = 0.99). Enzyme activity is expressed as Units (U)/g

mTGase powder, where 1 U corresponds to 1 µmol hydroxamate formed during 1 min of enzymatic

reaction (Ando et al., 1989; Jin et al., 2016). For determination of mTGase stability, the same

method was used after storing mTGase solutions for up to 24 hr at the respective temperature. All

results shown are mean values from triplicate experiments.

2.3. Sample preparation

A β-casein rich powder with a crude protein content of 952 g/kg and ~80% β-casein (using SDS-

PAGE; see section 2.5) in the protein fraction was prepared as described previously (Raak et al.,

2017a). Briefly, refrigerated reconstituted powdered skim milk (2.5°C) was microfiltrated (0.1 µm

polyethersulfone; Sartorius AG, Göttingen, Germany), and the permeate was adjusted to pH 4.6

using 6 M HCl to precipitate the β-casein. The precipitate was washed with demineralized water

and freeze-dried (Alpha 1-4, Martin Christ GmbH, Osterode am Harz, Germany). Commercial acid

casein and β-casein rich powders were suspended in demineralized water and dissolved by raising

the pH to 6.6 with 1 M NaOH, resulting in NaCn solutions. Target protein concentration was

27 g/kg, and 0.3 g/kg sodium azide was added to prevent microbial growth.

In the preceeding tests, NaCn was treated with 10 or 100 µl/ml Lipolase 100 L (Novo Nordisk

A/S, Bagsværd, Denmark) for 24 hr at 30°C, or centrifuged twice for 4 hr at 20,000 x g and 4°C

(Sigma Laborzentrifugen GmbH, Osterode am Harz, Germany) to remove a lipid-containing

contaminant that was reported in previous light scattering studies (e.g., HadjSadok et al., 2008;

Lucey et al., 2000).





6

Cross-linking of casein using mTGase was done at different incubation temperatures (Tinc =

10 – 40°C at 10°C intervals) for 0 – 48 hr with the enzyme dosages adjusted to 3 U/g protein by

considering the enzyme activity at the respective Tinc; additional experiments at Tinc = 30°C were

carried out with 1.5 or 6 U/g protein. The required amount of mTGase powder was dispersed in

10 ml demineralized water and added to 1 kg of temperature equilibrated casein solution that was

subsequently split into separate containers for each individual incubation time. Unlike previous

studies (e.g., Jaros et al., 2010; Rohm et al., 2014; Raak et al., 2017b), 0 hr refers to samples with

the enzyme added, but immediately inactivated. Enzyme inactivation in samples for gelation

experiments was through heating at 85°C for 15 min and subsequent cooling in ice water, and the

solutions were stored frozen (-18°C) until usage, a maximum of 6 wk. Samples for chemical

analyses were diluted with urea-containing buffers (see corresponding sections) for protein

unfolding and thus enzyme inactivation without any heat treatment and measured within one day.

Unless otherwise stated, all samples refer to NaCn prepared from the commercial acid casein

powder.

2.4. Gelation experiments

Acid-induced gelation was investigated using the method of previous studies (e.g., Jaros et al.,

2006b; Rohm et al., 2014) using a strain-controlled ARES RFS3 rheometer (TA Instruments,

Eschborn, Germany) equipped with a cup and bob geometry (di = 32 mm, do = 34 mm,

h = 33.5 mm). Temperature equilibrated NaCn solutions were treated with 40 mg/g glucono-δ-

lactone and transferred to the rheometer; the sample surface was covered with paraffin oil to

prevent evaporation. Gelation was monitored using time-based small amplitude oscillatory shear

experiments at ω = 1 rad/s and γ = 0.003, and temperature was maintained at 30°C using a

circulator surrounding the outer cylinder. Maximum storage modulus G'MAX and loss factor tan δ at

G'MAX were taken from the gelation curves for sample characterization. All measurements were

carried out in duplicate; half range was always smaller than 5% of the mean value.

2.5. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)

Linear SDS-PAGE with 125 and 40 mg/ml polyacrylamide in separating and stacking gel (Jaros

et al., 2010), respectively, was carried out using a vertical apparatus from C.B.S. Scientific Co. Inc.

(Del Mar, CA, USA). mTGase-treated NaCn solutions were diluted with a 1:1 mixture of 8 M urea

and sample buffer (0.8 M Tris, 2 mM ethylenediaminetetraacetic acid (EDTA), 4 M glycerin,

20 mg/ml SDS, 0.2 g/l Orange G; pH 8.0) after defined incubation periods at 10 or 30°C, causing

enzyme inhibition with no prior heat treatment. Finally, samples were treated with 150 mg/ml

dithiothreitol and boiled for 5 min to reduce disulfide bonds. SDS-PAGE was run at 120 V using an





7

electrode buffer that contained 50 mM Tris, 380 mM glycine, and 2 g/l SDS. Protein fractions were

stained with Coomassie Brilliant Blue R250 (0.6 g/L in ethanol, acetic acid, and demineralized

water in a volumetric ratio of 4:1:5) for 45 min, and subsequently destained with methanol, acetic

acid, and demineralized water in a volumetric ratio of 5:1:14 for 4 hr.

2.6. Size exclusion chromatography coupled with multi-angle light scattering (SEC-MALS)

For SEC-MALS, a Superdex 200 Increase 10/300 GL column (GE Healthcare Europe GmbH)

was coupled to an HPLC-system from Agilent Technologies, Inc. (Santa Clara, CA, USA). Casein

solutions were diluted in an elution buffer (6 M urea, 0.1 M NaCl, 0.1 M Na2HPO4 ∙ 2 H2O, 1 g/l

CHAPS; pH 6.8) to inhibit mTGase and resulted in protein unfolding, treated with 100 mg/ml

dithiothreitol to cleave disulfide bonds, and filtered through syringe filters (0.45 µm, regenerated

cellulose, Analytik-Zubehör GmbH, Langen, Germany). The experiments were done at ambient

temperature (~22°C) with an injection volume of 100 µl and a flow rate of 0.5 ml/min. The

separated casein was detected using a UV/Vis detector (λ = 280 nm; Agilent Technologies, Inc.), a

differential refractive index detector (dRI; λ = 633 nm; Optilab T-rEx, Wyatt Technology Europe

GmbH, Dernbach, Germany), and a MALS detector (λ = 663.8 nm, 18 angles; Dawn Heleos II,

Wyatt Technology Europe GmbH). The data was acquired and evaluated using Astra software V6.1

(Wyatt Technology Europe GmbH), and molar mass was calculated with a 1st order Berry fit using

either dRI (dn/dc = 0.186 ml/g; de Kruif and Grinberg, 2002; Monogioudi et al., 2009) or a UV/Vis

signal (ε = 0.85 ml/mg; Thomar et al., 2013).

2.7. Statistical analysis

One-factor analysis of variance with Tukey's post-hoc test was done using the Systat 12

software package (Systat GmbH, Erkrath, Germany). The statistical acceptence level was p<0.05.

3. Results and discussion

3.1. Enzymatic cross-linking of NaCn at different temperatures

3.1.1. Temperature-dependent mTGase activity

The activity of mTGase in Tris acetate buffer at 10 – 60°C using Z-Gln-Gly as acyl donor and

hydroxylamine as acyl acceptor is shown in Fig. 1a. Previous studies showed similar patterns for

the temperature-dependent activity of mTGase from S. mobaraensis, including an optimum at

~50°C (e.g., Ando et al., 1989; Jin et al., 2016; Zhang et al., 2012). Nevertheless, it was important

to ascertain the exact activities of the mTGase preparation to adjust enzyme dosage for casein





8

cross-linking at different temperatures. For instance, the enzyme activity was 9.1 0.7 and

87 6 U/g at 10 and 40°C, respectively, meaning that about 10 times more enzyme had to be

added at 10°C. Fig. 1b shows mTGase activity after storing the enzyme solution for up to 24 hr.

Activity was retained completely at 30°C, but decreased considerably during storage at 40 and

50°C: after 24 hr, 67 and 10% of the initial activity was obtained, respectively. Eissa et al. (2004)

monitored the storage stability of mTGase (pH 6.0) at 50°C and found a higher residual activity

after 6 hr (~50%). In other studies, the stability was determined after much shorter storage

(≤30 min) (e.g., Ando et al., 1989; Menendez et al., 2006; Zhang et al., 2012). Casein was

supposed to be cross-linked using mTGase for up to 48 hr in the present study. Therefore, 50°C

was not selected for the experiments as the rapidly decreasing enzyme activity would limit the

comparability of samples with the same incubation time. On the other hand, incubation at 40°C

was included since it has been a standard condition in several previous studies (e.g., Anema et al.,

2005; Ercili-Cura et al., 2010; Jaros et al., 2010; Lauber et al., 2000; Macierzanka et al., 2011).

3.1.2. Rheological properties of acid-induced gels

Stiffness (G'MAX) of NaCn gels increased with increasing casein cross-linking at Tinc = 40°C for up

to 48 hr (Fig. 2a, top). Decreasing Tinc resulted in considerably higher G'MAX in case of short

incubation periods (≤2 hr), whereas no differences were observed after cross-linking for ≥3 hr. The

working hypothesis suggests that mTGase predominantly acts on molecules located within the

same particle formed from self-associated caseins (Raak et al., 2017b; 2019), and that particle

conformation (i.e., monomer number, size, density) can be influenced by ionic strength, pH, and

temperature of the solution (HadjSadok et al., 2008). During acidification, these particles

aggregate, with internal isopeptide bonds contributing to the stiffness of the gel network. Therefore,

differences in G'MAX were expected after excessive cross-linking of temperature-dependent casein

particles that had been fixed in their particular conformation. The results, however, indicated that

temperature had no noticeable effect at the conditions used (i.e., 27 g/kg casein, pH 6.6), resulting

in gel networks with equal stiffness. The differences after short incubation (≤2 hr) probably resulted

from experimental difficulties. Higher enzyme dosages were added at lower Tinc to compensate for

the differences in enzyme activity. During the heat treatment for mTGase inactivation, however,

samples passed the temperature region where mTGase is most active (40 – 50°C) so that enzyme

activity increased strongly for a short time and considerable cross-linking occurred, making the

preceding polymerization less significant. This is most apparent for Tinc = 10°C where even the 0 hr

sample showed the same G'MAX as 1 and 2 hr incubated samples. After longer incubation,

susceptible glutamine and lysine residues are cross-linked which slows down the reaction because

of substrate limitation and hence diminishes the effect of the temperature increase during heat

treatment. This is supported by previous studies showing a rapid formation of





9

N-ε-(γ-glutamyl)-lysine isopeptide bonds during the early stages of incubation which levels off over

time (Jaros et al., 2014a,b; Lauber et al., 2000; Raak et al., 2017b).

For short incubation periods (≤3 hr), G'MAX was inversely linearly related to the corresponding

loss factor tan δ (R² = 0.88; Fig. 2a, bottom) as was reported in earlier studies (Jaros et al., 2010;

Rohm et al., 2014). As cross-linking proceeded (≥5 h), tan δ increased again, and, although no

differences in G'MAX were noticeable, slightly higher values were observed at elevated Tinc. Similar

gel stiffness might indicate the same amount of isopeptide bonds (Raak et al., 2017b), while higher

tan δ was recently attributed to a higher compactness of cross-linked casein particles (Raak et al.,

2019). Perhaps, temperature may not have affected the monomer number of the casein particles

(Ruis et al., 2007), but could have had an impact on their density. When increased temperature

forces casein particles to shrink because of reduced solvent quality (de Kruif et al., 2015),

extensive cross-linking may fix the given conformation and lead to more compact particles.

3.1.3. Molecular characteristics of cross-linked NaCn

To confirm the conclusions drawn from the rheological data, NaCn cross-linked at Tinc = 10 or

30°C for 0 – 24 hr were analyzed using SDS-PAGE after inactivation of mTGase by adding urea

and SDS containing sample buffer (Fig. 3). Even though mTGase concentration was adjusted to

equal enzyme activities at different Tinc, polymer formation occurred faster at Tinc = 30°C. This might

be because the activation energy of casein for the enzymatic reaction is different from that of the

artificial substrates used in the enzyme activity assay (Folk and Cole, 1965, 1966), leading to a

lower reactivity than assumed and thus to a slower polymerization at 10°C. At both incubation

temperatures caseins were cross-linked in the order β-casein > αS-caseins > κ-casein as is typical

for caseinates (Jaros et al., 2010; Ercili Cura et al., 2010). Assuming that cross-linking by mTGase

occurs mainly between molecules located within the same particle, different maximum polymer

sizes would be expected if temperature affected the monomer number of casein particles, but no

differences regarding oligomer and polymer formation could be observed. Additionally, a shift of

monomer bands towards smaller polymers would indicate intramolecular cross-linking at low

temperatures (O'Connell and de Kruif, 2003b; Partanen et al., 2013), but such a shift was not

observed. An additional band between monomers and dimers in samples incubated at 10°C

corresponded to mTGase (~40 kg/mol) (Partanen et al., 2013) since the enzyme was added at a

much higher dosage at this temperature. The results therefore indicated that monomer number of

casein particles was not affected by temperature, resulting in similar polymer sizes. In case of

NaCn cross-linked for 24 hr at 30°C, however, trimers and larger fractions migrated further

compared to cross-linking at 10°C, indicating a more compact structure of the polymers. Although





10

larger polymers cannot be evaluated because they did not penetrate the electrophoresis gel, the

findings are in agreement with the conclusions that were drawn from the rheological data.

Besides temperature, pH and ionic strength, casein concentration is an important factor driving

casein association (HadjSadok et al., 2008; Pitkowski et al., 2008), which was rather high in this

study (27 g/kg). Previous studies investigated casein association as a function of temperature

using 1 – 5 g/kg (e.g., Dauphas et al., 2005; HadjSadok et al., 2008; O'Connell et al., 2003a).

However, additional experiments with 5 g/kg did not result in temperature-dependent cross-linking

of NaCn as well (Fig. S1). It was decided not to test lower concentrations since this would

complicate gelation experiments that were a substantial part of the research.

3.1.4. Effect of enzyme dosage on casein cross-linking

In a second trial, enzyme concentrations of 1.5, 3, or 6 U/g protein were used at Tinc = 30°C (Fig.

2b). In this case, incubation time was adjusted and rheological data were plotted against the

theoretical amount of reaction product nth (mmol/gprotein) that could have been formed during this

period (Equation 1):

nth = cE ∙ tinc ∙ (60/1000) (1)

where cE is the enzyme concentration (U/gprotein), tinc is the incubation time (hr), and 60/1000 is a

conversion factor. Since 1 U corresponds to 1 µmol/min of reaction product released, nth represents

the mass-related amount of isopeptide bonds that would be formed by mTGase during a defined

incubation period assuming infinite availability and susceptibility of glutamine and lysine residues.

For instance, 3 U/g protein result in nth = 0.18, 0.72, and 4.32 mmol/gprotein after tinc = 1, 4, and 24 hr,

respectively. A comparison of these values with experimentally determined isopeptide contents of a

previous study (Raak et al., 2019) shows considerable discrepancies: 0.03 (1 hr) and

0.12 mmol/gprotein (24 hr) are lower by factors of 6 and 36, respectively. This suggests that glutamine

and lysine residues of casein are less suitable substrates for mTGase than Z-Gln-Gly and

hydroxylamine, and underlines that they become increasingly unavailable with ongoing cross-

linking. This is supported by Dinnella et al. (2002) who observed a lower extent of casein

polymerization by mTGase in the presence of Z-Gln-Gly despite a higher number of blocked lysine

residues. The effect of solvent components on enzyme activity may also be considered: Kütemeyer

et al. (2005) determined a higher mTGase activity in Tris acetate buffer and in NaCl solution than in

demineralized water, which was the solvent for NaCn. This could be confirmed in the present

study, where differences were more pronounced at higher temperature (Fig. 1a). On the other

hand, isopeptide contents of mTGase treated NaCn and casein in 0.1 M phosphate buffer were





11

recently shown to be similar (Raak et al., 2019), suggesting that the reaction velocity of casein

cross-linking using mTGase is dominated by substrate availability.

Gel stiffness of NaCn cross-linked at Tinc = 30 and 40°C was in good agreement (Fig. 2, top).

Nevertheless, slight differences in G'MAX were observed after short incubation periods

(nth = 0.18 mmol/gprotein) when varying mTGase concentration: higher enzyme dosages resulted in

higher gel stiffness, indicating again that cross-linking was facilitated during heat treatment when

more enzyme was added. These differences diminished with ongoing cross-linking, resulting in a

trend that is comparable to the results from varying Tinc. On the other hand, gels showed no

differences in tan δ after longer incubation times (nth ≥0.72 mmol/gprotein) (Fig. 2b, bottom),

underlining the effect of incubation temperature on particle compactness.

3.2. Characterization of casein using denaturing SEC-MALS

3.2.1. Identification of a contaminant fraction by light scattering

MALS chromatograms showed a fraction eluting at ~14 – 17 min which did not show a

pronounced UV-signal (Fig. 4). Light scattering is sensitive to concentration but in particular to

molar mass, hence large analytes cause high intensities even at a low concentration. Such a

fraction was previously found in batch light scattering studies (e.g., HadjSadok et al., 2008;

Panouillé et al., 2004) and using SEC-MALS using an imidazole-containing elution buffer (Lucey et

al., 2000). In contrast, Monogioudi et al. (2009) used a similar urea-containing buffer but did not

report the occurrence of such a fraction, possibly because they used purified β-casein for their

experiments. Lucey et al. (2000) reported that average molar masses of NaCn preparations as

determined in batch light scattering experiments decreased from 1228 – 4746 to 335 – 575 kg/mol

after removal of a cloudy supernatant using ultracentrifugation. This suggests that the presence of

this contaminant might also affect the molar mass determination of casein polymers using SEC-

MALS in case of co-elution.

It was previously hypothesized that the contaminant is composed mainly of lipid residues with

some adsorbed protein (HadjSadok et al., 2008; Panouillé et al., 2004). Therefore, the amount of

this fraction was decreased by hydrolysis with lipase “Lipolase 100 L”. The peak area of the

contaminant was decreased by ~33 and ~50% during incubation with 10 and 100 µl/ml lipase,

respectively (Fig. 4a), but casein was also degraded as indicated by decreased peak height and

increased peak width of the UV signal (~23 – 27 min). The question whether lipases are able to

hydrolyze peptide bonds was raised earlier: Maruyama et al. (2003) reported that, although lipase

has a similar active centre as a serine protease, proteolytic activity is unlikely and possible

proteolysis is probably because most commercial lipase preparations are contaminated with





12

residual protease. Since this side effect was highly unfavorable, lipase treatment was not

considered for further experiments.

After centrifugating twice a sediment and a cloudy supernatant were found and the actual NaCn

solution became clearer as was also reported by HadjSadok et al. (2008) and Lucey et al. (2000).

After careful separation of NaCn from sediment and supernatant, peak area of the contaminant

was found to be lower by ~46% with no change in the casein fraction (Fig. 4b). [Please note that

the sample shown is after 0 hr mTGase treatment (dotted lines) so that minimal polymerisation

occurred. The area under the curve of the UV signal was identical to the reference (full lines).]

Lucey et al. (2000) also observed a decreased peak in SEC-MALS but could not completely

remove the contaminant, and HadjSadok et al. (2008) stated that it might contain a fraction with a

density similar to water, making it impossible to eliminate using centrifugation alone. Experiments

were continued with double centrifuged NaCn as this was currently the best way to decrease the

amount of contaminant.

3.2.2. Molar mass distribution of cross-linked NaCn

Fig. 5a shows typical chromatograms of cross-linked NaCn from MALS and dRI detectors.

Concentration sensitive UV (not shown) and dRI detectors were in a good agreement, providing

the usual pattern of polymers eluting at first (~17.5 min) followed by less resolved shoulders

commonly considered as trimers (~19 min) and dimers (~21 min) and finally a distinct peak for

monomeric casein (~25 min). Further UV and dRI chromatograms that show decreasing monomer

and increasing polymer peaks with ongoing cross-linking of NaCn are shown in Fig. S2. The MALS

detector again showed a peak for the contaminant (~15 min) that overlapped with casein polymers

(~17 min); smaller fractions were barely detected because of the limit of static light scattering

towards smaller molecules. Monogioudi et al. (2009) who used SEC-MALS with comparable

conditions did not show MALS chromatograms, and Lucey et al. (2000) used a different elution

buffer which allowed casein association so that peaks in MALS chromatograms corresponded

mainly to casein particles and aggregates of larger sizes. Therefore, a contaminant peak in

denaturing SEC-MALS was observed for the first time.

Fig. 5a also shows the molar mass distribution of the NaCn sample as estimated from MALS

and dRI signals. The plot shows a molar mass decrease with elution time, which is interrupted by

deviations from this tendency at low concentration signals between the main polymeric fractions at

~20, ~23 and >26 min. Fig. 6a shows molar masses of casein monomers (~25 min in Fig. 5a) as

estimated on the basis of UV or dRI signals plotted against the corresponding peak maximum of

the chromatogram obtained after different incubation times (see Fig. S2). Estimated molar mass of

monomers was in a good agreement with the theoretical value (~24 kg/mol) when concentration





13

was sufficiently high as it was for 0 hr incubated NaCn. With ongoing casein cross-linking and

decreasing monomer peak intensity, however, calculated molar masses increased considerably.

Incubation with mTGase did not change the appearance of the monomer fraction in SDS-PAGE

(Fig. 3) and SEC (Fig. S2), so that the formation of compact high molar mass casein polymers that

co-elute with monomers is unlikely. This suggested that low concentration signals misrepresent

molar mass determinations of small molecules that hardly provide light scattering signals.

Overestimation of molar mass of dimers (i.e., ~200 kg/mol instead of ~40 – 50 kg/mol) resulted

from co-elution with larger polymers.

3.2.3. Molar mass of sodium caseinate polymers

Fig. 7a compares peak shapes and estimated molar mass distributions of polymer fractions

after cross-linking of NaCn for 3, 5 and 24 hr. Similar to SDS-PAGE (see section 3.1), SEC showed

a faster polymerization at Tinc = 30°C than at 10°C. Cross-linking for ≥3 hr at 30°C resulted in only

slight changes of the polymer peak shape. Additionally, cross-linking for 24 hr resulted in a similar

shape of the polymer peak at both Tinc = 10 and 30°C (see insert to Fig. 7a), indicating similar

maximum polymer sizes. On the other hand, the polymer peak of NaCn incubated at 30°C was

broader, confirming the conclusions from SDS-PAGE that casein particles are more compact and

appear smaller after extensive cross-linking at elevated temperatures. Qualitative evaluations are

supported by the similar molar mass distributions of all samples. Regardless of Tinc, estimated

molar mass of polymers that eluted at the peak maximum (~17.2 min) was ~400 and ~450 kg/mol

after 5 and 24 hr of cross-linking, respectively, indicating that prolonged incubation resulted in an

increase in polymer concentration and an incorporation of remaining monomers rather than in

intermolecular cross-linking between polymers. The calculated molar masses of the polymers

suggest a monomer number of ~16 – 20, which is higher than expected from previous studies on

casein association in NaCn. For instance, molecular modelling resulted in casein particles

consisting of 4 β-casein,4 αS1-casein, and one κ-casein or αS2-casein molecule (Farrell Jr. et al.,

2013; Kumosinski et al., 1994), which is in good agreement with the αS1:αS2:β:κ ratio found in

bovine milk (~4:1:4:1). Huppertz et al. (2017) later suggested that αS2-casein containing casein

particles may be observed in much higher numbers since κ-casein tends to self-association and

particles might therefore contain more than one molecule. These findings imply a monomer

number of 9 – 11 and a molar mass of ~200 – 250 kg/mol for casein particles, which is in line with

the experimental data of HadjSadok et al. (2008). Considering that molar masses of 103 up to

>105 kg/mol were calculated for the contaminant (Fig. 5a), co-elution likely resulted in

overestimation of casein polymer molar mass. This is additionally supported by the fact that dimers

and small oligomers were still present after 24 hr of cross-linking (Fig. 3), indicating that maximum

polymer size is actually lower than the monomer number of the casein particles.





14

3.2.4. SEC-MALS of cross-linked β-casein

Additional cross-linking experiments were done with β-casein, which was expected to undergo a

substantial temperature-dependent association (Dauphas et al., 2005; de Kruif and Grinberg, 2002;

O'Connell et al., 2003a; O'Connell and de Kruif, 2003b). As shown by the dRI chromatograms in

Figs. 5a and 5b, isolated β-casein showed a more rapid polymer formation than NaCn at the same

conditions because it is more susceptible to mTGase than the other casein types (see section 3.1).

Interestingly, the contaminant was also present in β-casein samples, and the MALS chromatogram

showed the peak overlapping with the polymeric fraction. Concerning the molar mass

determination, the same problems arose as for NaCn: higher molar masses were calculated from

the low signals of the concentration detectors as can be seen in particular for the monomer fraction

(Fig. 6b).

The polymer fraction of cross-linked β-casein (Fig. 7b) was comparable to that of NaCn

(Fig. 7a): both eluted at ~16.4 min with a peak maximum at ~17 min, however, the peak of NaCn

polymers was broader. As shown by SDS-PAGE (Fig. 3), dimers, trimers and small oligomers were

still present in NaCn after cross-linking for 24 hr, and this diversity led to peak broadening. In

contrast, cross-linked β-casein showed a higher content of high molar mass polymers in

SDS-PAGE (not shown) and therefore a sharper peak in SEC. Similar to NaCn at Tinc = 30°C,

maximum polymer size was reached after short incubation times, but the peak decreased and was

shifted to higher elution times by longer cross-linking, indicating a transition into a more compact

conformation. The shape of the polymer peaks after 24 hr was similar for cross-linking at Tinc = 10

and 30°C, suggesting similar maximum polymer sizes, but again, polymers formed at 30°C eluted

slightly later as they might be more compact. Molar masses as estimated from MALS and dRI

signals were very similar to that of NaCn: ~450 kg/mol were calculated at the peak maximum of

β-casein cross-linked for 5 and 24 hr at both temperatures. Monogioudi et al. (2009) reported

average molar masses of 500 – 1600 kg/mol for β-casein cross-linked using mTGase, but they

took only SEC-fractions larger than 500 kg/mol into account. Estimated molar masses appear

rather low considering previous studies that reported aggregation numbers ranging from 25 to

>100 for β-casein particles at ambient temperature (Cragnell et al., 2017; de Kruif and Grinberg,

2002; Moitzi et al., 2008). Additionally, a polydisperse size distribution was observed (Cragnell et

al., 2017; de Kruif and Grinberg, 2002). Complete cross-linking of the particles would thus result in

polymers with molar masses much higher than 600 kg/mol and higher heterogeneity, indicating

again that maximum polymer size is lower than the aggregation number.





15

3.2.5. Potential improvements of molar mass determination

Reliability of molar mass determinations might be increased once the contaminant is removed.

(Ultra-)centrifugation was previously discussed to be of limited suitability because of a fraction with

a density similar to water (HadjSadok et al., 2008). Assuming that this fraction consists of fat

residues, extraction with organic solvents might work. Another possibility is the separation of

contaminant and casein monomers using preparative SEC, but this requires a proper recovery

from the elution buffer prior to cross-linking. If the contaminant cannot be removed, a SEC column

with broader fractionation range could enable its separation from casein polymers, but this would

decrease the separation efficiency for smaller oligomers. The application of field flow fractionation

complementary to SEC might be the most promising way to investigate cross-linked casein as it

permits the separation of casein and contaminant even with native conditions, i.e., without urea

(Abbate et al., 2018; Raak et al., 2018).

4. Conclusions

Cross-linking with mTGase was used to study casein self-association as a function of

temperature (10 – 40°C). Based on the assumption that mTGase acts predominantly on molecules

that are located within the same casein particle, different polymer sizes would have been expected

if temperature affected the aggregation number. Polymer size was, however, not temperature-

dependent, suggesting that casein association was also not affected. A rather high protein

concentration of 27 g/kg was selected to study acid-induced gelation and might have had a greater

impact on self-association than temperature. The results suggested that incubation temperature

can be selected as required by the processing steps for concentrations relevant for acid-induced

gelation. On the other hand, enzyme dosages should be chosen carefully to avoid uncontrolled

enzyme activity during the heat treatment for inactivation. Additionally, the potential of SEC-MALS

for molar mass determination of cross-linked casein was evaluated. Estimated molar masses of

polymeric fractions were consistent with the idea of a maximum polymer size, but results

suggested that it might be lower than the aggregation number of casein particles. Casein polymers

co-eluted with a contaminant inherent to casein powders which possibly led to overestimation of

molar masses. The removal of this contaminant prior to cross-linking is therefore important for

molar mass determination of casein polymers. Furthermore, molecular characterization of cross-

linked casein particles with native conditions is important to draw definite conclusions on the

mechanism of action of mTGase and the properties of the resulting casein polymers. This

knowledge is also important for the combined application of mTGase and innovative processing

techniques such as high pressure treatment, ultrasonication or microwave irradiation that have





16

been recently reported to affect enzymatic cross-linking due to changes in secondary and tertiary

structure of the target proteins (Gharibzahedi et al., 2018c).

Acknowledgements

Financial support was received from the Deutsche Forschungsgemeinschaft (Bonn, Germany)

under grant numbers RO3454/5-1 and LE1424/9-1. mTGase was kindly provided by Ajinomoto

Foods Europe SAS (Hamburg, Germany), and glucono-δ-lactone by Kampffmeyer Nachf. GmbH

(Ratzeburg, Germany).

Declaration of interest

The authors declare no conflict of interest.

References

Abbate, R.A., Raak, N., Boye, S., Janke, A., Rohm, H., Jaros, D., & Lederer, A. (2019). Asymmetric flow

field flow fractionation for the investigation of caseins cross-linked by microbial transglutaminase.

Food Hydrocolloids. doi:10.1016/j.foodhyd.2019.01.043 [Link]

Ando, H., Adachi, M., Umeda, K., Matsuura, A., Nonaka, M., Uchio, R., Tanaka, H., & Motoki, M. (1989).

Purification and characteristics of a novel transglutaminase derived from microorganisms.

Agricultural and Biological Chemistry, 53, 2613–2617. [Link]

Anema, S.G., Lauber, S., Lee, S.K., Henle, T., & Klostermeyer, H. (2005). Rheological properties of acid

gels prepared from pressure- and transglutaminase-treated skim milk. Food Hydrocolloids, 19,

879–887. [Link]

Buchert, J., Ercili Cura, D., Ma, H., Gasparetti, C., Monogioudi, E., Faccio, G., Mattinen, E., Boer, H.,

Partanen, R., Selinheimo, E., Lantto, R., & Kruus, K. (2010). Crosslinking food proteins for improved

functionality. Annual Review of Food Science and Technology, 1, 113–138. [Link]

Cragnell, C., Choi, J., Segad, M., Lee, S., Nilsson, L., & Skepö, M. (2017). Bovine β-casein has a

polydisperse distribution of equilibrium micelles. Food Hydrocolloids, 70, 65–68. [Link]

Dauphas, S., Mouhousriou, N., Metro, B., Mackie, A., Wilde, P., Anton, M., & Riaublanc, A. (2005). The

supramolecular organisation of β-casein: Effect on interfacial properties. Food Hydrocolloids, 19,

387–393. [Link]



https://doi.org/10.1016/j.foodhyd.2004.10.005


https://doi.org/10.1146/annurev.food.080708.100841


https://doi.org/10.1080/00021369.1989.10869735




17

de Kruif, C.G., Bhatt, H., Anema, S.G., & Coker, C. (2015). Rheology of caseinate fractions in relation to

their water holding capacity. Food Hydrocolloids, 51, 503–511. [Link]

de Kruif, C.G., & Grinberg, V.Y. (2002). Micellisation of β-casein. Colloids and Surfaces A:

Physicochemical and Engineering Aspects, 210, 183–190. [Link]

Dinnella, C., Gargaro, M.T., Rossano, R., & Monteleone, E. (2002). Spectrophotometric assay using o-

phtaldialdehyde for the determination of transglutaminase activity on casein. Food Chemistry, 78,

363–368. [Link]

Eissa, A.S., Bisram, S., & Khan, S.A. (2004). Polymerization and gelation of whey protein isolates at

low pH using transglutaminase enzyme. Journal of Agricultural and Food Chemistry, 52,

4456–4464. [Link]

Ercili Cura, D., Lille, M., Partanen, R., Kruus, K., Buchert, J., & Lantto, R. (2010). Effect of Trichoderma

reesei tyrosinase on rheology and microstructure of acidified milk gels. International Dairy Journal,

20, 830–837. [Link]

Farrell Jr., H.M., Brown, E.M., & Malin, E.L. (2013). Higher order structures of the caseins: A paradox?

In P.L.H. McSweeney & P.F. Fox (Eds.), Advanced Dairy Chemistry, 4th ed. Boston: Springer US.

p 161–184. [Link]

Folk, J.E., & Cole, P.W. (1965). Structural requirements of specific substrates for guinea pig liver

transglutaminase. The Journal of Biological Chemistry, 240, 2951–2960. [Link]

Folk, J.E., & Cole, P.W. (1966). Mechanism of action of guinea pig liver transglutaminase. I. Purification

and properties of the enzyme: Identification of a functional cysteine essential for activity. The Journal

of Biological Chemistry, 241, 5518–5525. [Link]

Gaspar, A.L.C., & de Góes-Favoni, S.P. (2015). Action of microbial transglutaminase (MTGase) in the

modification of food proteins: A review. Food Chemistry, 171, 315–322. [Link]

Gharibzahedi, S.M.T., & Chronakis, I.S. (2018a). Crosslinking of milk proteins by microbial

transglutaminase: Utilization in functional yogurt products. Food Chemistry, 245, 620–632. [Link]

Gharibzahedi, S.M.T., Koubaa, M., Barba, F.J., Greiner, R., George, S., & Roohinejad, S. (2018b).

Recent advances in the application of microbial transglutaminase crosslinking in cheese and ice

cream products: A review. International Journal of Biological Macromolecules, 107, 2364–2374. [Link]

Gharibzahedi, S.M.T., Roohinejad, S., George, S., Barba, F. J., Greiner, R., Barbosa-Cánovas, G.V., &

Mallikarjunan, K. (2018c). Innovative food processing technologies on the transglutaminase

functionality in protein-based food products: Trends, opportunities and drawbacks. Trends in Food

Science & Technology, 75, 194–205. [Link]



https://doi.org/10.1016/j.tifs.2018.03.014

https://doi.org/10.1016/j.ijbiomac.2017.10.115

https://doi.org/10.1016/j.foodchem.2017.10.138

https://doi.org/10.1016/j.foodchem.2014.09.019

http://www.jbc.org/content/241/23/5518.short

https://pdfs.semanticscholar.org/51bf/7ca49c7660d591f95a25984baa14deec0e18.pdf

https://doi.org/10.1007/978-1-4614-4714-6_5%20

https://doi.org/10.1016/j.idairyj.2010.06.008

https://doi.org/10.1021/jf0355304

https://doi.org/10.1016/S0308-8146(02)00109-7

https://doi.org/10.1016/S0927-7757(02)00371-0




18

HadjSadok, A., Pitkowski, A., Nicolai, T., Benyahia, L., & Moulai-Mostefa, N. (2008). Characterisation of

sodium caseinate as a function of ionic strength, pH and temperature using static and dynamic light

scattering. Food Hydrocolloids, 22, 1460–1466. [Link]

Hiller, B., & Lorenzen, P.-C. (2008). Effect of buffer systems on the extent of enzymatic oligomerisation

of milk proteins. LWT - Food Science and Technology, 41, 1140–1144. [Link]

Hiller, B., & Lorenzen, P.C. (2009). Functional properties of milk proteins as affected by enzymatic

oligomerisation. Food Research International, 42, 899–908. [Link]

Huppertz, T., Gazi, I., Luyten, H., Nieuwenhuijse, H., Alting, A., & Schokker, E. (2017). Hydration of

casein micelles and caseinates: Implications for casein micelle structure. International Dairy Journal,

74, 1–11. [Link]

IDF (1979) Caseins and caseinates – Determination of protein content (Reference method). IDF

Standard 92. Brussels, Belgium: International Dairy Federation.

Jaros, D., Partschefeld, C., Henle, T., & Rohm, H. (2006a). Transglutaminase in dairy products:

Chemistry, physics, applications. Journal of Texture Studies, 37, 113–155. [Link]

Jaros, D., Pätzold, J., Schwarzenbolz, U., & Rohm, H. (2006b). Small and large deformation rheology of

acid gels from transglutaminase treated milks. Food Biophysics, 1, 124–132. [Link]

Jaros, D., Jacob, M., Otto, C., & Rohm, H. (2010). Excessive cross-linking of caseins by microbial

transglutaminase and its impact on physical properties of acidified milk gels. International Dairy

Journal, 20, 321–327. [Link]

Jaros, D., Schwarzenbolz, U., Raak, N., Löbner, J., Henle, T., & Rohm, H. (2014a). Cross-linking with

microbial transglutaminase: Relationship between polymerisation degree and stiffness of acid casein

gels. International Dairy Journal, 38, 174–178. [Link]

Jaros, D., Schwarzenbolz, U., Raak, N., Löbner, J., Henle, T., & Rohm, H. (2014b). Corrigendum to

“Cross-linking with microbial transglutaminase: Relationship between polymerisation degree and

stiffness of acid casein gels” [Int Dairy J 38 (2014) 174–178]. International Dairy Journal, 39,

345–347. [Link]

Jin, M., Huang, J., Pei, Z., Huang, J., Gao, H., & Chang, Z. (2016). Purification and characterization of a

high-salt-resistant microbial transglutaminase from Streptomyces mobaraensis. Journal of Molecular

Catalysis B: Enzymatic, 133, 6–11. [Link]

Kumosinski, T.F., King, G., & Farrell Jr., H.M. (1994). An energy-minimized casein submicelle working

model. Journal of Protein Chemistry, 13, 681–700. [Link]

Kütemeyer, C., Froeck, M., Werlein, H.-D., & Watkinson, B.M. (2005). The influence of salts and

temperature on enzymatic activity of microbial transglutaminase. Food Control, 16, 735–737. [Link]



https://doi.org/10.1016/j.foodcont.2004.06.012

https://doi.org/10.1007/BF01886952

https://doi.org/10.1016/j.molcatb.2016.07.005




https://doi.org/10.1007/s11483-006-9016-1

https://doi.org/10.1111/j.1745-4603.2006.00042.x


https://doi.org/10.1016/j.foodres.2009.04.022

https://doi.org/10.1016/j.lwt.2007.07.003




19

Lauber, S., Henle, T., & Klostermeyer, H. (2000). Relationship between the crosslinking of caseins by

transglutaminase and the gel strength of yoghurt. European Food Research and Technology, 210,

305–309. [Link]

Loveday, S.M., Sarkar, A., & Singh, H. (2013). Innovative yoghurts: Novel processing technologies for

improving acid milk gel texture. Trends in Food Science & Technology, 33, 5–20. [Link]

Lucey, J.A., Srinivasan, M., Singh, H., & Munro, P.A. (2000). Characterization of commercial and

experimental sodium caseinates by multiangle laser light scattering and size-exclusion

chromatography. Journal of Agricultural and Food Chemistry, 48, 1610–1616. [Link]

Macierzanka, A., Bordron, F., Rigby, N.M., Mills, E.N.C., Lille, M., Poutanen, K., & Mackie, A.R. (2011).

Transglutaminase cross-linking kinetics of sodium caseinate is changed after emulsification. Food

Hydrocolloids, 25, 843–850. [Link]

Maruyama, T., Nakajima, M., Kondo, H., Kawasaki, K., Seki, M., & Goto, M. (2003). Can lipases

hydrolyze a peptide bond? Enzyme and Microbial Technology, 32, 655–657. [Link]

Menéndez, O., Rawel, H., Schwarzenbolz, U., & Henle, T. (2006). Structural changes of microbial

transglutaminase during thermal and high-pressure treatment. Journal of Agricultural and Food

Chemistry, 54, 1716–1721. [Link]

Moeckel, U., Duerasch, A., Weiz, A., Ruck, M., & Henle, T. (2016). Glycation reactions of casein

micelles. Journal of Agricultural and Food Chemistry, 64, 2953–2961. [Link]

Moitzi, C., Portnaya, I., Glatter, O., Ramon, O., & Danino, D. (2008). Effect of temperature on self-

assembly of bovine β-casein above and below isoelectric pH. Structural analysis by cryogenic-

transmission electron microscopy and small-angle X-ray scattering. Langmuir, 24, 3020–3029. [Link]

Monogioudi, E., Creusot, N., Kruus, K., Gruppen, H., Buchert, J., & Mattinen, M.-L. (2009). Cross-

linking of β-casein by Trichoderma reesei tyrosinase and Streptoverticillium mobaraense

transglutaminase followed by SEC–MALLS. Food Hydrocolloids, 23, 2008–2015. [Link]

Mounsey, J.S., O’Kennedy, B.T., & Kelly, P.M. (2005). Influence of transglutaminase treatment on

properties of micellar casein and products made therefrom. Le Lait, 85, 405–418. [Link]

O’Connell, J.E., Grinberg, V.Y., & de Kruif, C.G. (2003a). Association behavior of β-casein. Journal of

Colloid and Interface Science, 258, 33–39. [Link]

O’Connell, J.E., & de Kruif, C.G. (2003b). β-Casein micelles; cross-linking with transglutaminase.

Colloids and Surfaces A: Physicochemical and Engineering Aspects, 216, 75–81. [Link]

Panouillé, M., Nicolai, T., & Durand, D. (2004). Heat induced aggregation and gelation of casein

submicelles. International Dairy Journal, 14, 297–303. [Link]




https://doi.org/10.1016/S0927-7757(02)00503-4

https://doi.org/10.1016/S0021-9797(02)00066-8

https://doi.org/10.1051/lait:2005028


https://doi.org/10.1021/la702802a

https://doi.org/10.1021/acs.jafc.6b00472

https://doi.org/10.1021/jf0522863

https://doi.org/10.1016/S0141-0229(03)00053-X


https://doi.org/10.1021/jf990769z


https://doi.org/10.1007/s002170050554



20

Partanen, R., Forssell, P., Mackie, A., & Blomberg, E. (2013). Interfacial cross-linking of β-casein

changes the structure of the adsorbed layer. Food Hydrocolloids, 32, 271–277. [Link]

Pitkowski, A., Durand, D., & Nicolai, T. (2008). Structure and dynamical mechanical properties of

suspensions of sodium caseinate. Journal of Colloid and Interface Science, 326, 96–102. [Link]

Portnaya, I., Cogan, U., Livney, Y.D., Ramon, O., Shimoni, K., Rosenberg, M., & Danino, D. (2006).

Micellization of bovine β-casein studied by isothermal titration microcalorimetry and cryogenic

transmission electron microscopy. Journal of Agricultural and Food Chemistry, 54, 5555–5561. [Link]

Raak, N., Rohm, H., & Jaros, D. (2017a). Enzymatic cross-linking of casein facilitates gel structure

weakening induced by overacidification. Food Biophysics, 12, 261–268. [Link]

Raak, N., Rohm, H., & Jaros, D. (2017b). Cross-linking with microbial transglutaminase: Isopeptide

bonds and polymer size as drivers for acid casein gel stiffness. International Dairy Journal, 66,

49–55. [Link]

Raak, N., Abbate, R.A., Lederer, A., Rohm, H., & Jaros, D. (2018). Size separation techniques for the

characterisation of cross-linked casein: A review of methods and their applications. Separations, 5,

14. [Link]

Raak, N., Schöne, C., Rohm, H., & Jaros, D. (2019). Acid-induced gelation of enzymatically cross-

linked caseinate in different ionic milieus. Food Hydrocolloids, 86, 43–49. [Link]

Rohm, H., Ullrich, F., Schmidt, C., Löbner, J., & Jaros, D. (2014). Gelation of cross-linked casein under

small and large shear strain. Journal of Texture Studies, 45, 130–137. [Link]

Romeih, E., & Walker, G. (2017). Recent advances on microbial transglutaminase and dairy application.

Trends in Food Science & Technology, 62, 133–140. [Link]

Ruis, H.G.M., Venema, P., & van der Linden, E. (2007). Relation between pH-induced stickiness and

gelation behaviour of sodium caseinate aggregates as determined by light scattering and rheology.

Food Hydrocolloids, 21, 545–554. [Link]

Thomar, P., Nicolai, T., Benyahia, L., & Durand, D. (2013). Comparative study of the rheology and the

structure of sodium and calcium caseinate solutions. International Dairy Journal, 31, 100–106. [Link]

Zhang, L., Zhang, L., Yi, H., Du, M., Ma, C., Han, X., Feng, Z., Jiao, Y., & Zhang, Y. (2012). Enzymatic

characterization of transglutaminase from Streptomyces mobaraensis DSM 40587 in high salt and

effect of enzymatic cross-linking of yak milk proteins on functional properties of stirred yogurt.

Journal of Dairy Science, 95, 3559–3568. [Link]



https://doi.org/10.3168/jds.2011-5125




https://doi.org/10.1111/jtxs.12056


https://doi.org/10.3390/separations5010014


https://doi.org/10.1007/s11483-017-9483-6

https://doi.org/10.1021/jf060119c

https://doi.org/10.1016/j.jcis.2008.07.003




21

Figures:

Fig. 1: (a) Activity of microbial transglutaminase (mTGase) at different temperatures in

demineralized water (white), and Tris acetate buffer (black), and (b) storage stability of mTGase in

Tris acetate buffer at 30 (white), 40 (grey), and 50°C (black) expressed as residual activity in

relation to initial activity at the respective temperature. Different small letters indicate significant

differences along x-axis (p<0.05); different capital letters indicate significant differences along

y-axis (p<0.05).





22

Fig. 2: Maximum storage modulus G'MAX (top) and loss factor tan δ at G'MAX (bottom) obtained from

gelation curves (40 mg/g glucono-δ-lactone, 30°C) of cross-linked sodium caseinate solutions

(27 g/kg). (a) Incubation temperature varied between 10 (white), 20 (grey), and 40°C (black) using

microbial transglutaminase at 3 U/g protein; 30°C data were left out to give a clearer figure.

(b) Enzyme concentration was varied between 6 (white), 3 (grey), and 1.5 U/g protein (black) at an

incubation temperature of 30°C; numbers in brackets refer to incubation times (hr) with 3 U/gprotein.

All samples were heat treated (85°C, 15 min) for enzyme inactivation after pre-defined incubation

times.





23

Fig. 3: Electropherogram of sodium caseinate (27 g/kg) treated with microbial transglutaminase

(mTGase; 3 U/g protein) at different incubation temperatures (Tinc) for different incubation times

(tinc).





24

Fig. 4: Size exclusion chromatograms (grey: UV signal at λ = 280 nm; black: MALS signal at

θ = 90°) of untreated sodium caseinate (full lines) and sodium caseinate after different treatments:

(a) Treatment with 10 (dotted line) and 100 µl/ml (dashed line) lipase preparation for 24 hr at 30°C.

(b) Centrifugation twice for 4 hr at 20,000 x g and 4°C, and treatment with microbial

transglutaminase (3 U per g protein) for 0 hr at 30°C (dotted line).





25

Fig. 5: MALS (θ = 90°; black lines) and dRI chromatograms (λ = 633 nm; grey lines) and estimated

molar mass distributions (symbols) of (a) sodium caseinate and (b) β-casein cross-linked with

microbial transglutaminase (3 U/g protein) at 30°C for 1 hr.





26

Fig. 6: Estimated molar mass of the monomer fractions of (a) sodium caseinate and (b) β-casein

incubated at 10 (open symbols) or 30°C (closed symbols) for tinc = 0 (blue), 1 (black), 3 (orange),

5 (green), or 24 hr (violet) with microbial transglutaminase (3 U/g protein) as a function of

corresponding UV (λ = 280 nm; circles) or dRI (λ = 633 nm; rhombuses) peak maximum of the

chromatograms (see Fig. S2). Dotted lines are a guide for the eye, dashed lines indicate

theoretical molar mass of β-casein monomer (24 kg/mol).





27

Fig. 7: Sections of size exclusion chromatograms (dRI signal at λ = 633 nm; lines) showing the

polymer fraction and the corresponding estimated molar mass (symbols) of (a) sodium caseinate

and (b) β-casein incubated with 3 U microbial transglutaminase per g protein at 10 (top) or 30°C

(bottom) for 3 (dotted lines; white symbols), 5 (dashed lines; grey symbols), or 24 hr (full lines;

black symbols). Inserts compare polymer peaks after 24 hr of cross-linking at 10 (grey lines) and

30°C (black lines).





28

Fig. S1: Electropherogram of sodium caseinate (5 g/kg) treated with microbial transglutaminase

(mTGase; 3 U/g protein) at different incubation temperatures (Tinc) for different incubation times

(tinc).





29

Fig. S2: Size exclusion chromatograms (UV signal at λ = 280 nm, top; dRI signal at λ = 633 nm;

bottom) of sodium caseinate incubated with microbial transglutaminase (3 U/g protein) at 10°C for

0 (blue), 1 (black), 3 (orange), 5 (green), or 24 hr (violet). Symbols indicate maximum of monomer

peak and correspond to those in Fig. 6a.





30

Tables

Tab. 1: List of the chemicals and reagents used.

Chemical/Reagent Manufacturer

3-[(3-Cholamidopropyl)dimethylammonio]-1-

propansulphonate (CHAPS)

Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Acetic acid VWR International GmbH, Darmstadt, Germany

Acid casein from bovine milk Sigma-Aldrich GmbH, Steinheim, Germany

Acrylamide Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Coomassie brilliant blue R250 Applichem GmbH, Darmstadt, Germany

Disodium monohydrogen phosphate dihydrate

(Na2HPO4 ∙ 2 H2O)

Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Dithiothreitol Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Ethylenediaminetetraacetic acid (EDTA) Sigma-Aldrich GmbH, Steinheim, Germany

Glucono-δ-lactone Kampffmeyer Nachf. GmbH, Ratzeburg, Germany

Glycerin Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Glycine Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Hydrochloric acid (HCl) Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Hydroxylamine hydrochloride Sigma-Aldrich GmbH, Steinheim, Germany

Iron trichloride hexahydrate (FeCl3 ∙ 6 H2O) VWR International GmbH, Darmstadt, Germany

L-Glutamic acid γ-monohydroxamate Sigma-Aldrich GmbH, Steinheim, Germany

L-Glutathione Sigma-Aldrich GmbH, Steinheim, Germany

Lipolase 100 L Novo Nordisk A/S, Bagsværd, Denmark

Microbial transglutaminase “Activa MP”

from Streptomyces mobaraensis

(Lactose, Maltodextrin, ≥ 1% Transglutaminase)

Ajinomoto Foods Europe SAS, Hamburg, Germany

N-carboxybenzyl-L-glutaminyl-glycine (Z-Gln-Gly) Bachem Holding AG, Bubendorf, Switzerland

N,N-Methylenebisacrylamide Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Orange G Merck KGaA, Darmstadt, Germany

Paraffin oil Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Skim milk powder Sachsenmilch Leppersdorf GmbH, Leppersdorf, Germany

Sodium azide Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Sodium chloride (NaCl) Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Sodium dodecyl sulfate (SDS) Merck KGaA, Darmstadt, Germany

Sodium hydroxide (NaOH) Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Trichloroacetic acid Carl Roth GmbH & Co. KG, Karlsruhe, Germany

Tris(hydroxymethyl)aminomethan (Tris) VWR International GmbH, Darmstadt, Germany

Urea Carl Roth GmbH & Co. KG, Karlsruhe, Germany




Self-association of casein studied using enzymatic …...Please cite this article as: Raak N, Brehm...

Documents

Transcript of Self-association of casein studied using enzymatic …...Please cite this article as: Raak N, Brehm...