Byproduct utilization is an important consideration in the development of sustainable processes. Whey protein isolate (WPI), a byproduct of the cheese industry, and gelatin, a byproduct of the leather industry, were reacted individually and in blends with microbial transglutaminase (mTGase) at pH 7.5 and 45 C. When a WPI (10% w/w) solution was treated with mTGase (10 U/g) under reducing conditions, the viscosity increased four-fold and the storage modulus (G0) from 0 to 300 Pa over 20 h. Similar treatment of dilute gelatin solutions (0.5–3%) had little effect. Addition of gelatin to 10% WPI caused a synergistic increase in both viscosity and G0 , with the formation of gels at concentrations greater than 1.5% added gelatin. These results suggest that new biopolymers, with improved functionality, could be developed by mTGase treatment of protein blends containing small amounts of gelatin with the less expensive whey protein.

1. Bioresource Technology 100 (2009) 3638–3643
Contents lists available at ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
Properties of biopolymers produced by transglutaminase treatment of whey
protein isolate and gelatin q
Eduard Hernàndez-Balada a,b, Maryann M. Taylor a, John G. Phillips a, William N. Marmer a,
Eleanor M. Brown a,*
a
US Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center, 600 E Mermaid Lane, Wyndmoor, PA 19038, USA
b
Department of Chemical Engineering, University of Barcelona, c/ Martí i Franquès 1, 08028 Barcelona, Spain
a r t i c l e i n f o a b s t r a c t
Article history: Byproduct utilization is an important consideration in the development of sustainable processes. Whey
Received 12 November 2007 protein isolate (WPI), a byproduct of the cheese industry, and gelatin, a byproduct of the leather industry,
Received in revised form 11 June 2008 were reacted individually and in blends with microbial transglutaminase (mTGase) at pH 7.5 and 45 °C.
Accepted 18 August 2008
When a WPI (10% w/w) solution was treated with mTGase (10 U/g) under reducing conditions, the vis-
Available online 25 March 2009
cosity increased four-fold and the storage modulus (G0 ) from 0 to 300 Pa over 20 h. Similar treatment
of dilute gelatin solutions (0.5–3%) had little effect. Addition of gelatin to 10% WPI caused a synergistic
Keywords:
increase in both viscosity and G0 , with the formation of gels at concentrations greater than 1.5% added
Whey protein isolate
Gelatin
gelatin. These results suggest that new biopolymers, with improved functionality, could be developed
Microbial transglutaminase by mTGase treatment of protein blends containing small amounts of gelatin with the less expensive whey
Crosslink protein.
Dithiothreitol Published by Elsevier Ltd.
1. Introduction bundle for access by tanning chemicals. The hide may then be split;
the upper layer would be tanned to make leather, and the lower
Whey and gelatin are both proteinaceous agricultural byprod- layer would provide a rich source for the extraction of type B gel-
ucts. Whey is an abundant, inexpensive and readily available atin. Gelatin extracted from limed hides is edible, and is used in
byproduct of the cheese industry. The United States is the world’s sausage casings, as an ingredient in a variety of food products, as
largest whey exporter; latest available figures from the US Dairy well as in pharmaceuticals. Inedible gelatin may be surplus lower
Export Council (USDEC, 2006) show an increase in exportation by grade material from production of edible gelatins or may be from
104% between 2001 and 2006. Despite the exports, a considerable an inedible source such as leather waste (Taylor et al., 1994). The
amount of whey is wasted through disposal, and would be avail- highest value market for inedible gelatin traditionally was the pho-
able for potential new uses. These uses could be for food or non- tographic industry, where the demand for gelatin has declined
food products. markedly in recent years. The Bloom value, a measure of gel
Animal hides, skins and bones, major byproducts of the meat strength of a 6.67% (w/w) gelatin, is often used to predict the
industry, are rich sources of collagen, the structural protein of con- behavior of a gelatin in a particular application. Gelatin suppliers
nective tissue. Gelatin, produced by partial hydrolysis of collagen, typically give a nominal Bloom value for the gelatins they market,
is a polydisperse mixture of varying sized collagen fragments. Gel- in this case 75 g. High Bloom (> 225 g) gelatins are taken from the
atins are classified by type, edibility, and gel strength. Type A gel- highest molecular weight fraction of crude gelatin and are rela-
atin is obtained by acid treatment usually of pigskins, while type B tively uniform. High and low molecular weight gelatin fractions
is obtained by alkaline treatment of cattle hides or bones (Rose, are blended to produce low Bloom (<125 g) gelatins, and variation
1992). An early step in leather manufacturing is the liming of the between lots may be expected.
hide to assist in hair removal and opening up of the collagen fiber Transglutaminase (TGase, EC 2.3.2.13) is a calcium-dependent
enzyme that catalyzes an acyl transfer reaction between the c-car-
boxyamide of a protein or peptide bound glutamine and a primary
amine (Folk and Chung, 1973). An e(c-glutamyl)lysine bond is
q
Mention of trade names or commercial products in this article is solely for the formed when the primary amine is the e-amino group of a lysine
purpose of providing specific information and does not imply recommendation or
endorsement by the US Department of Agriculture.
residue. In 1989, Ando et al. used Streptoverticillium mobaraense
* Corresponding author. Tel.: +1 215 233 6481; fax: +1 215 233 6795. to produce a calcium-independent form of the enzyme, namely
E-mail address: eleanor.brown@ars.usda.gov (E.M. Brown). microbial transglutaminase, mTGase.
0960-8524/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.biortech.2009.02.039

2. E. Hernàndez-Balada et al. / Bioresource Technology 100 (2009) 3638–3643 3639
A variety of individual food proteins, including gelatin, casein der mild agitation. Reaction products were then heated at 90 °C for
and whey proteins have been polymerized by the formation of 10 min to deactivate the mTGase. Control samples, without mTG-
mTGase-mediated intermolecular crosslinks (Sakamoto et al., ase, were subjected to the same thermal treatment.
1994; Rodríguez-Nogales, 2006). Biopolymers formed by the enzy- Gelatin and WPI solutions (1–10%) were treated with mTGase at
matic crosslinking of dissimilar proteins have the potential for gen- 10 U/g of protein. The effect of enzyme concentration (0–10 U/g of
erating novel products (Motoki and Nio, 1983; Yildirim and protein) was also investigated for gelatin and WPI solutions at 10%.
Hettiarachchy, 1997; Oh et al., 2004). The most abundant proteins Finally, the effect of gelatin (0.5–3%) was tested by adding gelatin
in whey, b-lactoglobulin and a-lactalbumin, are globular and con- to a 10% WPI sample containing mTGase at either 0 or 2 U/g of total
tain two and four disulfide bonds, respectively (Farrell et al., 2004). protein.
The reduction of disulfide bonds in globular proteins, prior to mTG-
ase treatment, has been shown (Frgemand et al., 1997) to in- 2.3. Analyses
crease access to primary amino groups. Dithiothreitol (DTT) is
generally the reductant of choice for cleaving disulfide bonds, gi- The gelatin used for this study was characterized by the stan-
ven that it is needed in much lower concentration than other dard Bloom test (AOAC Method 948.21, 1990). The strength of a
reductants to complete the reaction (Grazú et al., 2003). gel formed in a standard 59 mm Bloom jar from a 6.67% (w/w)
In this study, biopolymers produced by treating gelatin or whey solution of gelatin at pH 6.5 was measured on a TA.XT2 Texture
protein isolate (WPI) and blends of WPI and gelatin with mTGase Analyzer (Stable Micro Systems, Surrey, UK). Gel strengths of
under reducing or nonreducing conditions are characterized. In experimental samples were determined by the modified small
blends, WPI, the less expensive of the protein sources, is the pri- sample Bloom test (Wainewright, 1977). Samples were transferred
mary component and the focus is on the effect of including small to 39 mm weighing bottles, to a height of 40 mm, cooled to room
amounts of gelatin with the WPI in these biopolymers. temperature and then chilled for 17 h at 10 °C in a constant tem-
perature bath. Each sample was then placed under a 0.5 in. diam-
eter analytical probe, which then was driven into the sample to a
2. Experimental depth of 4 mm at 1 mm/s. The measured force, using the correction
factor (1.389) previously determined for small samples (Taylor
2.1. Materials et al., 1994), was expressed in grams. After the determination of
gel strength, samples were melted at 60 °C for viscosity
mTGase, Activa TG-TI (Ajinomoto USA Inc., Paramus, NJ) with an measurements.
active range of pH 4.0–9.0 at 0–70 °C was used without further Viscosity was determined using a Model LV 2000 Rotary Vis-
purification. The mTGase activity, in the presence of the malto- cometer (Cannon, State College, PA) equipped with a low Centi-
dextrose carrier, under the assay conditions of Folk and Chung poise Adapter and a jacketed sample chamber connected to a
(1985) was approximately 100 U/g. Type B gelatin, alkaline ex- refrigerated bath circulator, Model RTE-8 (Neslab, Portsmouth,
tracted from bovine skin, was obtained from Sigma (St. Louis, NH). An 18 ml aliquot of each sample was added to the sample
MO), and characterized in this laboratory as 115 g Bloom. WPI, Ala- chamber and equilibrated for at least 10 min. Viscosity was mea-
cen 895, containing 93.2% protein (manufacturer’s data), was gen- sured at a spindle speed of 60 rpm, corresponding to a shear rate
erously supplied by NZMP (formerly New Zealand Milk Products; of 73.42 sÀ1. After the viscometer had been stabilized for one min-
Lemoyne, PA). DTT was obtained from Calbiochem (San Diego, ute, readings were taken at 60 °C for gelatin as recommended by
CA). All other chemicals were reagent grade and used as received. Rose (1992) and at 25 °C for WPI and blended biopolymer
solutions.
2.2. Sample preparation Dynamic oscillatory rheometric measurements were carried out
in a controlled stress AR-2000 Rheometer (TA Instruments, New
Solutions of gelatin or WPI ranging from 1% to 10% (w/w) were Castle, DE) at room temperature (21 ± 2 °C) after enzyme deactiva-
prepared by suspending protein powder in the required weight of tion and equilibration of samples for 1 h. The sample was placed
deionized water. WPI–gelatin solutions (10% WPI with 0.5–3% gel- between parallel 25 mm diameter plates and the gap between
atin) were prepared by suspending the required amounts of WPI them was set to 2.5 mm. Excess sample was trimmed off and a thin
and gelatin powders in deionized water and stirring for about layer of mineral oil applied to the exposed free edges of the sample
30 min to ensure a uniform suspension. To test the effect of a to prevent moisture loss. Time sweep measurements were used to
reducing environment, a 10% DTT (w/v) solution was prepared study the evolution of the storage modulus G0 as a function of time.
and the volume necessary to give a concentration of 10 mg DTT A strain sweep test was performed at a constant frequency of
per g protein was included in the preparation of selected samples. 0.1 Hz to find an oscillation stress value that was within the linear
Samples were allowed to swell at room temperature for 4 h, and viscoelastic range. This value, 0.5 Pa, was then used to perform
then adjusted to pH 7.5 with 1 N NaOH or 1 N HCl. Samples were time sweep measurements at a constant 0.1 Hz frequency. Storage
then heated at 38 °C for 1 h, cooled to room temperature and modulus G0 was recorded and analyzed over 20 h, at a rate of two
stored overnight at 4 °C. points per hour.
Typical conditions for mTGase-mediated crosslinking of gelatin Inter-protein crosslinking was evaluated by polyacrylamide gel
are pH 6.5 and 50 °C for 4 h (Taylor et al., 2001). For whey proteins, electrophoresis in sodium dodecyl sulfate (SDS–PAGE) (Laemmli,
the optimum conditions are pH 7.5 and 40 °C for 8 h (Truong et al., 1970) using precast gradient gels (4–15%). Gels were calibrated
2004). To provide a basis for comparison, all reactions in this study using the broad range (BRM) SDS calibration standard (Bio-Rad,
were performed at pH 7.5 and 45 °C for 5 h, considering that WPI Hercules, CA) that contains a mixture of nine proteins ranging in
was the major protein component in the mixtures. size from 6.5 to 200 kDa. Samples (approximately 0.5 mg) of lyoph-
For crosslinking experiments, protein samples were prepared in ilized protein dissolved in sample buffer (10 mM Tris–HCl at pH 8.0
10 ml less than the required volume. The appropriate concentra- containing 1 mM EDTA, 25 mg/ml SDS, 50 ll/ml b-mercap-
tion of mTGase was then prepared in 10 ml of water and this solu- toethanol and 0.1 ll/ml bromphenol blue) were heated at 40 °C
tion added with stirring to gelatin, WPI, or WPI–gelatin solutions to for 4 h. Separation was achieved using a Phast System (Pharmacia
give the desired final protein concentration. Samples were read- Biotech Inc., Piscataway, NJ). Gels were stained with Coomassie
justed to pH 7.5 and incubated for 5 h at 45 °C in a shaker bath un- Blue.

3. 3640 E. Hernàndez-Balada et al. / Bioresource Technology 100 (2009) 3638–3643
2.4. Statistical modeling
Because x, y plots of the data points always showed curvature,
second order models in %gelatin or %WPI were generated by
regression analysis of viscosity and gel strength data for products
from the treatment of combinations of gelatin, WPI, and blends
with mTGase and DTT. These models are of the form:
Response = a + b*Gelatin + c*Gelatin*Gelatin, where a, b, and c are
regression coefficients. Analysis of covariance (ancova) was per-
formed on the models to determine whether the coefficients were
significantly different between treatments. Comparisons between
treatments of the slopes and second order coefficients were per-
formed using orthogonal contrasts (Littell et al., 2002). All tests
of significance were performed at the P 0.05 level.
3. Results and discussion
3.1. Gel strength
The effects on gel strength of increasing concentrations of gela-
tin (1–10%) in the presence or absence of mTGase and DTT were
investigated. The second order model showed that gel strengths
of all samples significantly increased (P 0.05) with increasing gel-
atin concentration and were greatest at each concentration for
samples containing gelatin alone (Fig. 1a). Because type I collagen,
the primary source of gelatin, has no disulfide bonds, the addition
of DTT to gelatin was not expected to have a noticeable effect on
gel strength, and at concentrations below 6% gelatin, the addition
of DTT did not impact the gel strength. At higher concentrations
of gelatin (6–10%), however, gel strengths were significantly
(P 0.05) reduced by up to 15%. One possible explanation is the
presence of type III collagen, which is generally co-located with
type I collagen in connective tissue (van der Rest et al., 1990).
The individual chains of type III collagen are disulfide-linked (Bou-
dko and Engel, 2004), and separation of these chains may have
interfered with the partial renaturation that normally accompanies
gelation. Inclusion of mTGase in dilute (1–5%) gelatin solutions,
either with or without DTT, resulted in gel strengths significantly
(P 0.05) lower than for gelatin alone (Fig. 1a). The formation of
intramolecular crosslinks, which would inhibit the attainment of Fig. 1. Gel strengths of gels formed from (a) bovine type B gelatin, 1–10% (w/w),
collagen-like structure, is favored over intermolecular ones in di- and (b) WPI–gelatin blends, 10% WPI with 0–3% added gelatin. Gelatin gels were
lute solutions (Sakamoto et al., 1994). At higher (6–10%) gelatin prepared in the presence or absence of mTGase (10 U/g protein) and DTT (10 mg/g
protein), WPI–gelatin gels in the presence or absence of mTGase (2 U/g protein) and
concentrations where mTGase-mediated intermolecular crosslink-
DTT (10 mg/g protein). Each data point represents a separate measurement.
ing is more favorable, the resulting gel strengths were about 90% of Abbreviations are: G-gelatin, W-WPI, R-reducing conditions (DTT), E-enzyme
those for gelatin alone. (mTGase).
When the effect of mTGase (1–10 U/g gelatin) was determined,
no trend could be observed in the gel strength of a 10% gelatin
(data not shown). The range of values was between 188 and Under reducing conditions, a dramatic increase in gel strength
225 g, with an average gel strength of 207 ± 12 g, significantly was seen for mTGase-treated WPI–gelatin blends, the significance
(P 0.05) lower than the 254 ± 1 g determined for 10% gelatin of which was confirmed by the second order statistical model
without enzyme. In the presence of the reducing agent, DTT, the (P 0.05). Because reducing conditions and gelatin were essential
activity of mTGase was enhanced, as reported by Kolodziejska for enhancing the gel strength, it is likely that crosslinking was be-
et al. (2006), and gel strengths (263 ± 14 g) were comparable to tween whey proteins and gelatin chains. The reduction of disulfide
those for gelatin alone. bonds in the globular whey proteins by the action of DTT would ex-
The formation of mTGase-mediated WPI gels has been previ- pose reactive groups to the action of the mTGase. In the absence of
ously reported at WPI concentrations greater than 10% (Frge- DTT, the WPI proteins were not sufficiently unfolded to serve as
mand et al., 1997). Under the conditions of these experiments, no effective substrates for mTGase.
gel formation was observed for WPI alone at concentrations up
to 10%, with or without mTGase, under reducing or nonreducing 3.2. Viscosity
conditions. Gels did form when small amounts of gelatin (0.5–
3%) were included in 10% WPI solutions (Fig. 1b). Gelatin alone The viscosities at 60 °C, of gelatin solutions (1–10%), with and
at these concentrations forms only very weak gels, and so long as without DTT and mTGase, were determined (Fig. 2a). Neither DTT
the reaction conditions were not reducing, the gel strength was nor mTGase alone had a significant effect on gelatin viscosity at
not significantly altered whether the gelatin was alone or in a this temperature. Similar to the case for gel strength, the effect of
10% WPI solution, with or without mTGase at 2 U/g WPI–gelatin. mTGase under reducing conditions was not significant at gelatin

4. E. Hernàndez-Balada et al. / Bioresource Technology 100 (2009) 3638–3643 3641
centrations less than 7%, neither DTT nor mTGase, alone or to-
gether, had any significant effect on the viscosity of the solution.
Between 7% and 10% WPI, mTGase alone did not affect the viscosity
(data not shown), and by inference had little ability to crosslink the
proteins. In this concentration range, the viscosity was decreased
slightly when disulfide bonds were reduced and the protein con-
formation was less well defined. A highly significant effect on the
viscosity (P 0.001) was observed as a result of the treatment that
combined the action of DTT and mTGase on a 10% WPI solution; a
four-fold increase in viscosity was attained.
Viscosities measured at 25 °C for blends of gelatin (0.5–3%) with
10% WPI were always higher than those for gelatin alone (Fig. 2c).
When mTGase was included in blends of 10% WPI with gelatin, the
increase in observed viscosity was not significant. When both DTT
and mTGase were included in the blend, their effect on viscosity
became more significant (P 0.001). In fact, viscosities for blends
of 10% WPI with more than 1.5% gelatin could not be measured be-
cause the solution formed a gel that did not melt below 60 °C.
3.3. Rheological properties
The storage modulus (G0 ) was determined for 10% WPI, blends
of 10% WPI with 3% gelatin, and samples treated with mTGase un-
der reducing or nonreducing conditions (Fig. 3). A 10% WPI solu-
tion, which remained liquid at 20 °C, exhibited a constant value
less than 1 for G0 . In contrast, for a 10% WPI solution treated with
mTGase under reducing conditions, G0 remained nearly constant
and low for the first 10 h and then increased to 300 Pa between
10 and 20 h, suggesting the formation of a stable and permanent
gel structure (Comfort and Howell, 2002). Because G0 can be re-
lated to the amount of crosslinking, the similar and flat response
of the WPI sample in a reducing environment and the WPI sample
with mTGase under nonreducing conditions suggest that no signif-
icant crosslinking occurred under these conditions. WPI–gelatin
blends under reducing conditions without mTGase or with mTGase
in a nonreducing environment showed a small increase in G0 be-
tween 10 and 20 h, possibly due to noncovalent interactions be-
tween gelatin and whey protein molecules. Chen and Dickinson
(1999) suggested that electrostatic interactions, hydrogen bonding
or hydrophobic interactions that have a physical nature and are
temperature-dependent might contribute to the appearance of
weak gel-like behavior. A similar study of 3% gelatin samples with
or without mTGase (2 U/g gelatin) produced a constant G0 value
near zero (data not shown). Blends of 10% WPI with 3% gelatin after
Fig. 2. Viscosities of solutions of (a) gelatin, (b) WPI and (c) WPI–gelatin blends
prepared as described for Fig. 1. Viscosities were measured at 60 °C for gelatin
solutions and at 25 °C for WPI or WPI–gelatin blends. Abbreviations are as reported
for Fig. 1.
concentrations in the 1–6% range. Above 7% gelatin, the effect was
a significantly dramatic increase (P 0.001) in viscosity to values
greater than 1800 mPa s, beyond the limit of the viscometer. These
results are consistent with the formation of intramolecular cross-
links at low gelatin concentrations and both intra- and intermolec-
ular crosslinks at higher concentrations (Clark and Courts, 1977; Yi
Fig. 3. Time sweep analysis of 10% (w/w) WPI and WPI–gelatin blends. Blends
et al., 2006). contain 10% (w/w) WPI and 3% (w/w) gelatin, mTGase was 10 U/g protein for WPI
The viscosities at 25 °C of solutions of WPI (1–10%) with and and 2 U/g protein for WPI-gelatin blends, DTT was 10 mg/g protein. Abbreviations
without DTT and mTGase were determined (Fig. 2b). At WPI con- are as designated for Fig. 1.

5. 3642 E. Hernàndez-Balada et al. / Bioresource Technology 100 (2009) 3638–3643
incubation with mTGase (2 U/g protein) under reducing conditions
showed an exponential increase of G0 from approximately 70 Pa to
approximately 5000 Pa within the first 10 h. This remarkable gain
suggests the existence of an initial gelled network that became
more reticulated and stable with time (Eissa et al., 2004). Similar
enhancements in storage modulus were reported for blends of gel-
atin with small amounts of chitosan, a larger polymer, and mTGase
(Chen et al., 2003).
3.4. SDS–PAGE
Gelatin, WPI, and blended biopolymers were analyzed by
SDS–PAGE. Ten percent gelatin (Fig. 4, lane 2) reflects a polydis-
perse mixture of different sized collagen fragments with some
aggregated material in the stacking gel at the top, and faint
bands visible through the general smear in the separating gel. Fig. 4. SDS–PAGE gels: lane 1, molecular weight markers, ranging in size from
Although more distinct bands can be seen in SDS–PAGE patterns 200 kDa at the top to 6.5 kDa at the bottom; lane 2, gelatin 10% (w/w); lane 3, 10%
for gelatins isolated from a specific source (Taylor et al., 1994) gelatin after treatment with mTGase (10 U/g) under reducing conditions; lane 4,
WPI 10% (w/w); lane 5, 10% WPI after treatment with mTGase (5 U/g); lane 6, 10%
or high Bloom (250 g) commercial gelatins (Tosh et al., 2003),
WPI after treatment with mTGase (5 U/g) under reducing conditions; lane 7, WPI
the pattern seen here is typical of low Bloom (115 g) commer- 10% (w/w) with gelatin 1.5% (w/w); lane 8, WPI 10% (w/w) with gelatin 1.5% after
cial gelatins (Taylor et al., 2004), which are sold on the basis treatment with mTGase (2 U/g); and lane 9 WPI 10% (w/w) with gelatin 1.5% after
of gel strength, and are blends of gelatins from different prepa- treatment with mTGase (2 U/g) under reducing conditions. This is a composite
figure produced from two gels that were run simultaneously.
rations. The pattern for 10% gelatin treated with mTGase under
reducing conditions (Fig. 4, lane 3) shows only a faint smear,
representing small collagen fragments that most likely lack the
lysine or glutamine sidechains necessary for mTGase-mediated and rheological data that the inclusion of 1.5% gelatin in 10% WPI
crosslinking. Although samples prepared for lanes 2 and 3 were induces gel formation and enhances physical properties, the bio-
of the same weight, the faintness of the pattern in lane 3 and polymers formed range in size from those formed by WPI when
the lack of material in the stacking gel and the upper range treated with mTGase under reducing conditions to the much lar-
of the separating gel suggests the formation of aggregates so ger aggregates that do not penetrate the gel.
large they could not penetrate the stacking gel (Taylor et al.,
2004). A similar observation (Sharma et al., 2002) was inter- 4. Conclusions
preted to suggest that the protein had become so highly poly-
merized that it could not penetrate the stacking gel; they also Both gelatin (ICIS, 2006) and whey (NASS, 2006) are relatively
demonstrated that although the highly polymerized products low value byproducts of the American food industry. While the
did not appear on the gels, their presence could be confirmed actual prices vary over time the ratio is relatively constant at
by chromatography. approximately 5:1 (gelatin:whey). The addition of minor
The SDS–PAGE patterns for 10% WPI alone (Fig. 4, lane 4) or amounts of relatively low quality gelatin to whey protein im-
after incubation with mTGase 5 U/g under nonreducing conditions proves the strength and stability of gels formed by the action
(Fig. 4, lane 5) show the major whey proteins, b-lactoglobulin (MW of mTGase in a reducing environment. As a byproduct of the
approximately 18 kDa) and a-lactalbumin (MW approximately meat industry, and a breakdown product of collagen, gelatin
14 kDa) as well as faint bands for bovine serum albumin (MW comes in a range of qualities. The higher quality gelatins find
approximately 66 kDa) and lactoferrin (MW approximately a variety of uses in food, pharmaceutical and industrial products
86 kDa). Incubation of 10% WPI with mTGase 5 U/g under reducing and are heavily traded. Whey is a lower value byproduct that
conditions (Fig. 4, lane 6) resulted in bands, above both the sepa- can be recovered from the waste stream of the cheese industry.
rating gel and the stacking gel, showing formation of high molecu- The use of mTGase to catalyze the formation of crosslinks in and
lar weight polymers in addition to the typical WPI bands. An earlier between protein molecules is well established, as is the efficacy
study (Frgemand et al., 1997) produced similar results. of its usefulness with either gelatin or WPI. When a small
The addition of 1.5% gelatin to 10% WPI had little effect on the amount of gelatin was added to WPI, before mTGase treatment
gel pattern (Fig. 4, lane 7). In an earlier study, when small under reducing conditions, a dramatic rise in viscosity, higher
amounts of WPI were added to 10% gelatin samples (Taylor gel strengths, and the appearance of high molecular weight
et al., 2006), the whey component was clearly visible in the gel bands due to inter-protein crosslinking in SDS–PAGE gel patterns
patterns, and its participation in the formation of crosslinked bio- than for either gelatin or WPI treated separately were observed.
polymers could easily be monitored. Here, the lack of defined These results suggest that the reducing environment partially
molecular weight bands in patterns for gelatin and the low con- unfolds the whey proteins, increasing access to glutamine and
centration of gelatin in the blends contribute to its invisibility. lysine side chains, and that the gelatin chains crosslink the whey
Incubation of the WPI–gelatin blend with mTGase under nonre- proteins to form a network. The improvement in physical prop-
ducing conditions resulted in the appearance of a high molecular erties over either protein component, given the same treatment,
weight band, possibly representing crosslinked gelatin fragments, suggests the possibility of greater utilization and new products
but had little effect on the WPI pattern (Fig. 4, lane 8). The gel from these byproducts.
pattern (Fig. 4, lane 9) for a similar sample incubated under
reducing conditions shows a decrease in intensity of the WPI Acknowledgements
bands, an increase in high molecular weight aggregates and a de-
crease in total protein. The pattern is consistent with the forma- The authors would like to thank Lorelie Bumanlag, Paul Pierlott,
tion of WPI–gelatin biopolymers in addition to aggregates of the Michael Tunick and Carlos Carvalho for their technical support in
individual biopolymers. Although it is clear from the gel strengths this work.

6. E. Hernàndez-Balada et al. / Bioresource Technology 100 (2009) 3638–3643 3643
References NASS, National Agricultural Statistics Service, 2006, Dairy product prices, June
30, 2006. http://usda.mannlib.cornell.edu/MannUsda/viewDocumentInfo.do?
documentID=1450.
Ando, H., Adachi, M., Umeda, K., Matsuura, A., Nonaka, M., Uchio, R., Tanaka, H.,
Oh, J.H., Wang, B., Field, P.D., Aglan, H.A., 2004. Characteristics of edible films made
Motoki, M., 1989. Purification and characteristics of a novel transglutaminase
from dairy proteins and zein hydrolysate cross-linked with transglutaminase.
derived from microorganisms. Agric. Biol. Chem. 53, 2613–2617.
Int. J. Food Sci. Tech. 39, 287–294.
Association of Official Analytical Chemists, 1990. Jelly strength of gelatine, final
Rodríguez-Nogales, J.M., 2006. Effect of preheat treatment on the transglutaminase-
action (948.21). In: Helrich, K. (Ed.), Official Methods of Analysis of the
catalyzed cross-linking of goat milk proteins. Process Biochem. 41, 430–437.
Association of Official Analytical Chemists, 15th ed. AOAC, Arlington, VA, p. 929.
Rose, P.I., 1992. Inedible gelatin and glue. In: Pearson, A.M., Dutson, T.R. (Eds.),
Boudko, S.P., Engel, J., 2004. Structure formation in the C terminus of type III
Inedible Meat By-Products, Advances in Meat Research. Elsevier Applied
collagen guides disulfide cross-linking. J. Mol. Biol. 335, 1289–1297.
Science, New York, pp. 217–263.
Chen, J., Dickinson, E., 1999. Interfacial ageing effect on the rheology of a heat-set
Sakamoto, H., Kumazawa, Y., Motoki, M., 1994. Strength of protein gels prepared
protein emulsion gel. Food Hydrocoll. 13, 363–369.
with microbial transglutaminase as related to reaction conditions. J. Food Sci.
Chen, T., Embree, H.D., Brown, E.M., Taylor, M.M., Payne, F.P., 2003. Enzyme-
59, 866–871.
catalyzed gel formation of gelatin and chitosan: potential for in situ
Sharma, R., Zakora, M., Qvist, K.B., 2002. Susceptibility of an industrial
applications. Biomaterials 24, 2831–2841.
a-lactalbumin concentrate to cross-linking by microbial transglutaminase. Int.
Clark, R.C., Courts, A., 1977. The chemical reactivity of gelatin. In: Ward, A.G., Courts,
Dairy J. 12, 1005–1012.
A. (Eds.), The Science and Technology of Gelatin. Academic Press, New York, pp.
Taylor, M.M., Diefendorf, E.J., Marmer, W.N., Brown, E.M., 1994. Effect of various
209–247.
alkalinity-inducing agents on chemical and physical properties of protein
Comfort, S., Howell, N.K., 2002. Gelation properties of soya and whey protein isolate
products isolated from chromium-containing leather waste. J. Am. Leather
mixtures. Food Hydrocoll. 16, 661–672.
Chem. Assoc. 89, 221–228.
Eissa, A.S., Bisram, S., Khan, S.A., 2004. Polymerization and gelation of whey protein
Taylor, M.M., Cabeza, L.F., Marmer, W.N., Brown, E.M., 2001. Enzymatic modification
isolates at low pH using transglutaminase enzyme. J. Agric. Food Chem. 52,
of hydrolysis products from collagen using a microbial transglutaminase I.
4456–4464.
Physical properties. J. Am. Leather Chem. Assoc. 96, 319–332.
Frgemand, M., Otte, J., Qvist, K.B., 1997. Enzymatic cross-linking of whey proteins
Taylor, M.M., Marmer, W.N., Brown, E.M., 2004. Molecular weight distribution and
by Ca2+-independent microbial transglutaminase from Streptomyces lydicus.
functional properties of enzymatically modified commercial and experimental
Food Hydrocoll. 11, 19–25.
gelatins. J. Am. Leather Chem. Assoc. 99, 129–141.
Farrell Jr., H.M., Jiménez-Flores, R., Bleck, G.T., Brown, E.M., Butler, J.E., Creamer, L.K.,
Taylor, M.M., Marmer, W.N., Brown, E.M., 2006. Preparation and characterization of
Hicks, C.L., Hollar, C.M., Ng-Kwai-Hang, K.F., Swaisgood, H.E., 2004.
biopolymers derived from enzymatically modified gelatin and whey. J. Am.
Nomenclature of the proteins of cows’ milk-sixth revision. J. Dairy Sci. 87,
Leather Chem. Assoc. 101, 235–248.
1641–1674.
Tosh, S.M., Marangoni, A.G., Hallett, F.R., Britt, I.J., 2003. Aging dynamics in gelatin
Folk, J.E., Chung, S.I., 1973. Molecular and catalytic properties of transglutaminases.
gel microstructure. Food Hydrocoll. 17, 503–513.
Adv. Enzymol. Relat. Areas Mol. Biol. 38, 109–191.
Truong, V.D., Clare, D.A., Catignani, G.L., Swaisgood, H.E., 2004. Cross-linking and
Folk, J.E., Chung, S.I., 1985. Transglutaminases. Method. Enzymol. 113, 358–375.
rheological changes of whey proteins treated with microbial transglutaminase.
Grazú, V., Ovsejevi, K., Cuadra, K., Betancor, L., Manta, C., Batista-Viera, F., 2003.
J. Agric. Food Chem. 52, 1170–1176.
Solid-phase reducing agents as alternative for reducing disulfide bonds in
US Dairy Export Council’s, 2006. Annual Report at http://usdec.org (accessed
proteins. Appl. Biochem. Biotech. 110, 23–32.
31.08.07.).
ICIS News, Chemical Industry News, Chemical Prices and Chemical Suppliers from
van der Rest, M., Dublet, B., Champliaud, M.-F., 1990. Fibril-associated collagens.
ICIS, 2006 at http://www.icis.com/Articles/2006/08/28/2015785/Chemical-
Biomaterials 11, 28–31.
Prices-F-J.html.
Wainewright, F.W., 1977. Physical tests for gelatin and gelatin products. In: Ward,
Kolodziejska, I., Piotrowska, B., Bulge, M., Tylingo, R., 2006. Effect of
G.A., Courts, A. (Eds.), The Science and Technology of Gelatin. Academic Press,
transglutaminase and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide on
New York, pp. 507–534.
the solubility of fish gelatin-chitosan films. Carbohydr. Polym. 65, 404–409.
Yi, J.B., Kim, Y.T., Bae, H.J., Whiteside, W.S., Park, H.J., 2006. Influence of
Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the
transglutaminase-induced cross-linking on properties of fish gelatin films. J.
head of bacteriophage T4. Nature 227, 680–685.
Food Sci. 71, E376–E383.
Littell, R.C., Stroup, W.W., Freund, R.J., 2002. SAS for Linear Models, fourth ed. SAS
Yildirim, M., Hettiarachchy, N.S., 1997. Biopolymers produced by cross-linking
Institute Inc., Cary, NC.
soybean 11S globulin with whey proteins using transglutaminase. J. Food Sci.
Motoki, M., Nio, N., 1983. Crosslinking between different food proteins by
62, 270–274.
transglutaminase. J. Food Sci. 48, 561–566.