LWT - Food Science and Technology 53 (2013) 346e354

Contents lists available at SciVerse ScienceDirect

LWT - Food Science and Technology

journal homepage: www.elsevier.com/locate/lwt

Protein enrichment and its effects on gluten-free bread characteristics

Cátia Regina Storck a, *, Elessandra da Rosa Zavareze a, Marcia Arocha Gularte a, Moacir Cardoso Elias a,
Cristina Molina Rosell b, Alvaro Renato Guerra Dias a
a
Department of Agroindustrial Science and Technology, Federal University of Pelotas, 96010-900, Pelotas, Brazil
b
Cereal Group, Institute of Agrochemistry and Food Technology (IATA-CSIC), P.O. Box 73, 46100 Burjassot, Valencia, Spain

a r t i c l e i n f o a b s t r a c t

Article history: Protein-enriched, rice-based flours, composed of casein and albumin protein isolates as well as trans-
Received 1 September 2010 glutaminase for reinforcing protein networks, were designed for optimising a gluten-free formulation
Received in revised form suitable for breadmaking. Experimental design resulted in composite protein-enriched blends with
30 January 2013
different pasting, mechanical and surface-related textural properties. The presence of transglutaminase
Accepted 4 February 2013
resulted in a significant decrease of all the protein fractions, which suggested protein crosslinking.
Protein isolates significantly (p < 0.05) modified the gelatinisation and gelling behaviour of the flour,
Keywords:
reducing both its peak viscosity and final viscosity. The textural properties were primarily affected by the
Rice
Rheology
addition of casein and transglutaminase. The design allowed the determination of the optimum
Bread formulation for obtaining the highest specific volume of gluten-free baked product with the lowest
Proteins crumb hardness by combining transglutaminase (1.35 U of enzyme/g of rice flour protein), egg albumin
Transglutaminase (0.67 g/100 g of flour) and casein (0.67 g/100 g of flour). The use of albumin and casein protein isolates
and transglutaminase constitutes a promising approach to producing protein-enriched blends for
making fermented, gluten-free products.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Arendt, 2010); however, gluten-free baked products are only

acceptable when their sensory characteristics are similar to those of
Celiac disease is one of the most common lifelong disorders wheat flour yeast bread (Gallagher, Gormley, & Arendt, 2004;
worldwide, with an estimated mean prevalence of 1% in the general Marco & Rosell, 2008a). Gluten-free breads are usually charac-
population (Catassi & Fasano, 2008). Individuals who have this terised by low quality and exhibit dry, crumbling crumb, poor
disease (celiacs) are unable to consume some of the most common mouthfeel and dry flavour (Gallagher, Gormley, & Arendt, 2003a,
products on the market, including breads and other food products 2004).
made with wheat flour. Bread is a highly consumed product and To improve the quality of gluten-free breads, ingredients like
usually contributes significantly to caloric intake; however, gluten- modified starch, hydrocolloids, enzymes and proteins of different
free bread suitable for consumption by coeliacs is difficult to find sources have been proposed (Gujral et al., 2003; Gujral & Rosell,
because its supply is restricted to speciality shops. 2004a,2004b; Moore, Heinbockel, Dockery, Ulmer, & Arendt,
Recently, rice flour has been widely proposed as an alternative 2006). A common practice in food processing is the incorporation
for making gluten-free breads due to its hypoallergenic proteins, of protein ingredients in product formulation to increase product
soft taste and white colour (Clerici, Airoldi, & El-Dash, 2009; Rosell, qualities such as flavour, texture and storage stability (Bonet, Bonet,
Brites, Pérez, & Gularte, 2007); however, rice-based bread has low Blaszczak, & Rosell, 2006). The addition of dairy and egg ingredients
volume and hard crumb (Gujral, Guardiola, Carbonell, & Rosell, in breadmaking has been widely used because their proteins are
2003; Gujral & Rosell, 2004a,2004b). The use of other cereal sour- highly functional and can be readily incorporated into the dough.
ces in breadmaking can increase the variability of products and can These ingredients can be used in bread for both nutritional and
also improve nutritional characteristics (Flander, Salmenkallio- functional benefits. The use of dairy powder in gluten-free baked
Marttila, Suortti, & Autio, 2007; Renzetti, Courtin, Delcour, & product formulations has resulted in improved volume as well as
better appearance and sensory aspects of the loaves (Gallagher,
Kundel, Gormley, & Arendt, 2003b). According to Stathopoulos
(2008), the most used ingredients in gluten-free baked product
* Corresponding author. Tel./fax: þ55 53 32757258.
E-mail address: [email protected] (C.R. Storck). formulations are caseinates, skim milk powder, dry milk, whey

0023-6438/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.lwt.2013.02.005
 C.R. Storck et al. / LWT - Food Science and Technology 53 (2013) 346e354 347

protein concentrate and milk protein isolate. Demirkesen, Mert, 17  C  3 and 65 g/100 g  5 relative humidity. The rice flour
Sumnu & Serpil Sahin (2010) also reported that it is necessary to chemical composition were 12.9 g/100 g moisture, 7.3 g/100 g
use a gum, emulsifier, enzyme or dairy product together with rice protein (by nitrogen determination using the Kjeldahl method e
flour for the desired viscoelastic mixture. N  5.95), 0.8 g/100 g ash (550  C/5 h), 0.6 g/100 g crude fat
Nevertheless, proteins often do not meet the requirements for (Soxhlet machine e solvent ether) using methods of AOAC (1995)
food processing, and additional modifications are necessary. The and 31.6g/100 g amylase (Gilbert & Spragg, 1964).
use of enzymes as processing aids in breadmaking is a frequent The food grade microbial transglutaminase was Activa STG-M
method for improving functional properties to promote networks (27 U/g) from Ajinomoto Co. Bread. The improver was Soft’r
and improve dough baking characteristics (Bonet et al., 2006; Alpaga obtained from Puratos e São Paulo, Brazil, composed by
Gujral & Rosell, 2004a,2004b; Marco & Rosell, 2008a; Renzetti, corn starch, stearoyl-2-lactil sodium lactate, ascorbic acid and alpha
Bello, & Arendt, 2008). An enzyme that has received extensive amylase. The casein was from Synth (São Paulo, Brazil), and egg
attention for its ability to crosslink proteins is transglutaminase. albumin was from Neo Nutri (Poços de Caldas, Brazil). Xanthan gum
This enzyme catalyses an acyletransfer reaction between the g- 200 (water relative density ¼ 1/solution at 25 of 1.0068/20 ; pH at
carboxyamide group of peptide-bound glutamine residues (acyl 1% ¼ 6e8; particle size 92% pass in a 0.075 mm) was from Arinos
donors) and a variety of primary amines (acyl acceptors), química, Brazil. The other ingredients for baked product production
including the ε-amino group of lysine residues in certain proteins were obtained in a local market.
(Motoki & Seguro, 1998). Compared to other cereals, rice has high
lysine content, so it potentially has more substrate for enzyme
2.1. Experimental design
action. According to Renzetti et al. (2008), rice and wheat are the
best substrates for this enzyme action when comparing different
Rice flour was replaced by combinations of transglutaminase,
flours.
albumin and casein following a central composite design of three
Gerrard and Sutton (2005) and Cabrera-Chávez and Calderón de
factors (transglutaminase, albumin and casein), and a five-level
la Barca (2008) published a research into the molecular mechanism
pattern was used for sampling (Table 1). Design factors (quanti-
of coeliac disease that suggests the disturbing possibility that
tative independent factors) that were tested at five levels ( a, 1,
transglutaminase in baked products may act upon gliadin proteins
0, þ1, þa) included transglutaminase (0e12 U/g of protein), al-
in dough to generate the epitope associated with the coeliac
bumin (0e6 g/100 g of flour) and casein (0e6 g/100 g of flour).
response and considered not safe the use of transglutaminase in
The model resulted in 16 different combinations of composite
cereal processing. But, the researchers have not found anything
flours.
about this mechanism in rice-based products, only in wheat, barley,
rye and oats.
A major task of modern technology is to generate new structures 2.2. Protein quantification
with characteristics that please the consumer while using a limited
range of ingredients. Proteins are one of the main classes of mol- Protein fractions were extracted following a sequential extrac-
ecules available for conferring textural attributes, and the cross- tion using various solvents according to a method described by
linking and aggregation of protein molecules has been cited as one Marco et al. (2008) with modifications. Dough was prepared by
of the most important mechanisms for engineering food structures mixing 2 g of solid sample (according to the experimental design)
with desirable mechanical properties (Gerrard, 2002). with 2 ml of water and placed in a centrifuge tube. An albumin-
Several studies have evaluated the influence of trans- globulin extraction was conducted by adding 10 ml of 5 ml/
glutaminase in gluten-free baked product formulations (Collar & 100 ml (w/v) NaCl to dough. The suspension was homogenised for
Bollaín, 2004; Gujral & Rosell, 2004a,2004b; Marco, Pérez, León, 5 min in a shaker (Shaker Certomat MO e B. Braun Biotech Inter-
& Rosell, 2008; Marco & Rosell, 2008c; Moore et al., 2006; national, Melsungen, Germany) and was centrifuged at 5500 g for
Renzetti et al., 2008). However, relatively little work has been done 10 min. After albumin-globulin extraction, the alcohol-soluble
to investigate the effect of interactions between transglutaminase, fraction was extracted from the residue by adding 10 ml of 50 ml/
albumin and casein on the technological and textural properties of 100 ml (v/v) 1-propanol, following the same procedure as
bread with rice flour. transglutaminase reacts differently with described for the albumin-globulins. Insoluble proteins were
various protein sources and at different addition levels; therefore, extracted with 10 ml of 0.1 mol equi/L NaOH containing 0.5 ml/
the aim of this study was to design a protein-enriched, rice-based, 100 ml (w/v) SDS and 0.6 ml/100 ml (v/v) b-mercaptoethanol (ME).
gluten-free baked product in the presence of transglutaminase to Each extraction was repeated twice to increase the protein
favour the formation of a protein network that improves crumb extraction. The supernatants were collected, and the protein con-
textural properties. With this goal, an experimental design was tents in both the mixed flour and the fractions were determined
used with different levels of transglutaminase and two different following the micro-Kjeldahl method (n 46-13 of AACC, 1995)
sources of proteins: albumin and casein. using N  5.95 as a protein conversion factor. The protein in the
final residue was determined by subtracting the albumin-globulin,
2. Materials and methods alcohol-soluble and insoluble fractions from 100%.

A rice cultivar (IRGA 417) was grown in southern Brazil in 2007.

Samples were harvested at a 20 g/100 g moisture content, dried to Table 1
Central composite design for sampling.
12 g/100 g in a pilot dryer (at 38  C) (Stationary Model, MV-
Máquinas Vitória, Pelotas, Brazil) and stored in a closed bag in a Factors Levels
cold room maintained at 17  C  3 and 65 g/100 g  5 relative a 1 0 þ1 þa
humidity prior to milling. Rice kernels were dehusked and polished Transglutaminase (U/g of rice 0 1.35 6 10.65 12
using a Zaccaria rice test-machine (Industrias Machina Zaccaria S/A, flour protein)
São Paulo, Brazil). Polished samples were then ground (Hammer Albumin (g/100 g of rice flour) 0 0.67 3 5.32 6
Mill, Model 3100, Perten Instruments, Hägersten, Sweden) and Casein (g/100 g of rice flour) 0 0.67 3 5.32 6

sieved (sieve openings 70 mm). Flour was stored in plastic bags at Concentrations are based on 100 g of flour.
348 C.R. Storck et al. / LWT - Food Science and Technology 53 (2013) 346e354

Table 2
Variables studied and the response for protein fractions and pasting properties.

Treatments Variables Analysis

Transglutaminase Albumin Casein Albumin/globulin Prolamin Glutelin Final residue Peak Break Final Setback (RVU)
(U/g rice flour (g/100 g of (g/100 g of (g/100 g of rice (g/100 g of (g/100 g of (g/100 g of viscosity down viscosity
protein) rice flour) rice flour) flour) rice flour) rice flour) rice flour) (RVU) (RVU) (RVU)
Control 0 0 0 7.74 1.02 71.47 20.02 462.7 144.3 708.2 389.8
1 1.35 0.67 0.67 11.49 2.46 62.53 23.51 292.4 31.2 519.8 227.3
2 1.35 0.67 5.32 8.02 6.51 52.63 31.69 258.5 26.9 425.7 167.2
3 1.35 5.32 0.67 35.96 2.00 46.92 15.24 192.2 8.0 323.5 131.3
4 1.35 5.32 5.32 27.85 6.62 39.10 26.89 184.3 5.0 312.9 128.6
5 10.65 0.67 0.67 10.64 2.01 47.66 39.70 265.8 26.3 467.3 201.6
6 10.65 0.67 5.32 7.48 5.87 48.58 38.06 236.5 20.7 390.0 153.5
7 10.65 5.32 0.67 31.88 4.68 39.45 22.83 174.4 2.2 288.8 114.4
8 10.65 5.32 5.32 20.08 5.34 31.82 42.45 163.9 0.9 287.9 124.0
9 6 3 3 19.66 3.96 43.28 32.87 204.5 10.7 340.3 135.8
10 6 3 3 19.04 3.64 43.55 36.67 214.4 11.7 357.0 142.6
11 0 3 3 24.89 5.50 48.61 21.40 229.8 18.5 399.7 169.8
12 12 3 3 15.06 2.65 34.06 48.23 206.4 9.0 360.1 153.7
13 6 0 3 5.19 3.72 51.91 39.18 288.0 39.3 473.8 185.7
14 6 6 3 28.89 3.41 36.23 33.10 171.0 4.8 288.5 117.5
15 6 3 0 21.90 0.96 48.03 30.71 232.1 17.7 394.4 162.3
16 6 3 6 14.79 5.04 39.13 41.04 221.1 9.5 381.8 160.7

RVU: Rapid Visco Unit.

2.3. Pasting properties rate of 12  C/minute, held at 95  C for 2.5 min, cooled to 50  C at a
rate of 12  C/minute, and held at 50  C for 2.5 min. The rotational
Pasting properties were determined in a Rapid-Visco Analyser speed of the paddle was maintained at 160 rpm throughout the run,
(RVA) (RVA-4, Newport Scientific model, version 2.0, Australia). To except during the first 10 s, when a 960-rpm speed was used. Peak
avoid any interference concerning total volume or moisture con- viscosity, breakdown, final viscosity and setback (difference be-
tent, a preliminary equilibration of all blends to 12 g/100 g moisture tween final viscosity and peak viscosity) were evaluated.
content was carried out. Three grams of powdered samples
(Table 1) was weighed directly into the aluminium RVA canister, 2.4. Dough texture
and 25 mL of distilled water was added and mixed with the rice
flour. The RVA test profile used was the Newport Scientific Rice The dough texture of the samples derived from the experi-
Profile. The sample was held at 50  C for 1 min, heated to 95  C at a mental design (Table 1) was evaluated by assessing the texture

Fig. 1. Response surface plots of the significant protein fractions.

 C.R. Storck et al. / LWT - Food Science and Technology 53 (2013) 346e354 349

profile analysis (TPA) in a TA.XTplus texturometer (Stable Micro

Systems Ltd., Godalming, UK) according to a method described by
Marco and Rosell (2008b), with modifications. Dough was placed in
Petri dishes (50-mm diameter) and kept resting for 10 min. Dough
was compressed with a cylinder probe (Acrylic compression disc,
45-mm diameter) at 60% compression with a test speed of
2 mm s 1 and a 75-s waiting period between compressions. Texture
determinations were carried out using a plastic film on the dough
surface to avoid the distortion induced by the negative peak of
adhesiveness and without using the plastic film to measure the
adhesiveness alone. Were evaluated five replicates of each
treatment.

2.5. Breadmaking

Breads were made following the experimental design (Table 1).

Control bread was made without transglutaminase and proteins.
Preliminary assays were carried out to optimise the amount of
water necessary for the breadmaking process based on the specific
volume and crumb hardness of the resulting breads. The addition of
albumin, casein and transglutaminase did not change the water
absorption of the dough. The baked product recipe based on 100 g
of rice flour: 115 mL/100 g water, 5 g/100 g sugar, 3 g/100 g salt, 2 g/
100 g compressed yeast, 3 g/100 g bread improver, 3 g/100 g soy-
bean oil and 1 g/100 g xanthan gum. Ingredients for six breads were
mixed for 10 min in a planetary mixer (Model K5SS, KitchenAid, St.
Joseph, USA) at medium speed. Dough (approximately 200 g) was
placed in individual pans using a spoon (7 cm width, 14 cm length
and 4.3 cm height) and was proofed for 90 min at 38  C in a
proofing cabinet. Subsequently, proofed dough was baked for
40 min at 200  C (electric oven, Fischer, Brusque, Brazil). Loaves
were removed from the pans and were cooled to room temperature
for 1 h before evaluation.

2.6. Baked product evaluation

In order to determine the baked product quality, volume (seed

displacement), loaf weight and crumb texture were quantified. The
bake loss (%) and specific volume (mL g 1) were also calculated
(Pizzinatto; Magno; Campagnolli, 1993).
Bread was sliced manually and crumb texture analysis was
carried out in two slices (1.25-cm thickness each) from the centre of
the loaves. Texture profile analysis (TPA) was performed using a
universal machine (TA.XTplus texturometer e Stable Micro Systems
Ltd., Godalming, UK) equipped with a stainless steel cylinder probe
(P/20, 20-mm diameter). The pre-test, test and post-test speeds
were 5, 2.5 and 5 mm s 1, respectively. The compression used was
40%, with a trigger force of 20 g, and the time between compression
cycles was 5 s.
Fig. 2. Response surface plots of the significant pasting responses obtained from the
regression equations.
2.7. Statistical analysis

For each dough/baked product characteristic (response),

measured analysis of variance was conducted using SAEG (version 3. Results and discussion
9.1, Brazil, 2007) to determine significant differences among the
factor combinations. Response surface plots were generated from Experimental data from the central composite design of rice
the regression equations using the SAEG program (version 9.1, composite dough, as well as baked product characteristics (re-
Brazil, 2007). sponses), were statistically analysed to determine the significance
Response surface plots were obtained by holding the indepen- of design factors. Multiple regression equations were constructed
dent variable with the least significant effect on the particular to estimate the effect of transglutaminase, casein and albumin
response at a constant value while changing the other two vari- (independent variables) on dough and baked product responses.
ables. Response surface were done only for significant interactions Regression equations for all responses were statistically signif-
obtained by analysis of variance. icant, but only some interactions between variables could generate
350 C.R. Storck et al. / LWT - Food Science and Technology 53 (2013) 346e354

a surface response with a satisfactory adjustment that will be dough significantly decreased prolamin extraction, which sug-
presented by figures. gested the formation of protein aggregates of low solubility. The
Table 2 shows the measure data for protein fractions and pasting same effect occurred in rice flour dough enriched with soybean
properties for the 16 treatments. proteins (Marco et al., 2008).
The addition of transglutaminase, albumin and casein had a
3.1. Quantification of protein fractions significant (p < 0.05), negative effect on the glutelin fraction. Ac-
cording to Mujoo and Ng (2003), the transglutaminase added to
Protein quantification allowed for the determination of the wheat flour promoted protein crosslinks, mainly within the glutelin
extent of transglutaminase activity as well as the protein fraction fraction. The individual addition of transglutaminase and casein
affected by the crosslinking activity of the enzyme. Table 2 presents also had a positive linear effect and a negative quadratic effect on
the results for protein fractions for all 16 treatments from experi- the amount of final residue. The increase in the amount of final
mental design and control bread. Results showed that the fraction residue in rice flour dough has been previously attributed to an
albumin/globulin was significantly affected by transglutaminase, increase in the amount of proteins with high-molecular weight and
albumin and casein addition. The addition of transglutaminase and low solubility, which resulted from crosslinking promoted by
albumin simultaneously produced a significant antagonistic effect transglutaminase (Marco et al., 2008).
(Fig. 1), because when albumin was at maximum levels of addition
and transglutaminase was added, the amount of albumin/globulin 3.2. Pasting properties
decreased. The same effect occurred with the combination of al-
bumin and casein. Experimental data from the pasting properties were submitted
It has been reported that transglutaminase crosslinks lysine and to analysis of variance to determine the main effects of the added
glutamic acid residues when added to rice flour, primarily affecting proteins and transglutaminase as well as their interaction. The
the albumin fraction, followed by glutelins, globulins and pro- values found are in Table 2.
lamins (Rosell, Brites, Pérez, Gularte, & Arroz, 2007). The protein- Peak viscosity was not significantly influenced by trans-
enriched composite blends showed the same trend, and a glutaminase addition, but albumin and casein showed significant
decrease of the water-soluble protein fraction was obtained when effects, as observed in Fig. 2A. Albumin had a higher influence in
the level of transglutaminase was increased, which was likely due peak viscosity than casein. The interaction between those proteins
to insoluble structures created by crosslinking. These founds can be had a significant negative effect on peak viscosity. In consequence,
observed in Table 2 when the addition of albumin and casein in the protein-enriched dough at the highest protein concentration tested
presence of higher amounts of transglutaminase, decreased the resulted in a reduction of the peak viscosity. All treatments resulted
albumin fraction and increased final residue (treatments 4 and 8). in decrease in all pasting properties when compared to bread
Similar results were observed by Marco et al. (2008); when 1% of control.
transglutaminase was added to rice flour, the albumin/globulin Marco and Rosell (2008c) observed the same tendency when
fraction decreased with a simultaneous increase in the final residue testing the effect of proteins from different sources on the pasting
due to the formation of high-molecular-weight polymers. In the properties of rice. The effect likely resulted from a dilution effect
same way, Bonet et al. (2006) evaluated the effect of trans- because a negative correlation has been established between the
glutaminase on a wheat flour/egg albumin blend and observed a protein content and the peak viscosity in rice flour (Tan & Corke,
decrease of the water-soluble protein fraction. 2002) as well as between the protein content and the specific
The prolamin fraction was also affected by all the factors tested. gelation process of each protein (Ngarize, Adams, & Howell, 2004).
Fig. 1 shows the surface response for the prolamin fraction. trans- Viscosity breakdown was significantly influenced only by albu-
glutaminase and albumin had a quadratic effect; the response min addition. Final viscosity was significantly (p < 0.05) reduced by
surface graph showed a minimum in the presence of 6 U of all factors studied, and a significant interaction was only observed
transglutaminase/g of protein and 3 g/100 g of albumin, where the with albumin and casein (Fig. 2B). Marco and Rosell (2008c) also
highest impact of those factors was detected. Bonet et al. (2006) observed a decrease in final viscosity with the addition of pea,
showed that the presence of different proteins sources in wheat soybean, egg albumen and whey protein; however, they did not

Table 3
Variables studied and the response for dough texture properties.

Treatments Variables Analysis

Transglutaminase Albumin (g/100 g Casein (g/100 g Hardness Adhesiveness Springiness Cohesiveness

(U/g rice flour protein) of rice flour) of rice flour) (g) (gs 1) (g) (g)
Control 0 0 0 842.7 3370.1 0.99 0.11
1 1.35 0.67 0.67 521.5 1842.8 0.90 0.15
2 1.35 0.67 5.32 743.0 2193.3 0.91 0.13
3 1.35 5.32 0.67 536.9 1718.1 0.81 0.17
4 1.35 5.32 5.32 891.0 3086.0 1.00 0.14
5 10.65 0.67 0.67 528.2 2048.5 0.85 0.14
6 10.65 0.67 5.32 767.2 2784.0 0.94 0.13
7 10.65 5.32 0.67 559.7 2249.1 0.90 0.13
8 10.65 5.32 5.32 1011.7 3609.9 0.98 0.11
9 6 3 3 629.5 2165.6 0.85 0.12
10 6 3 3 649.9 2544.9 0.81 0.12
11 0 3 3 644.1 2166.2 0.95 0.15
12 12 3 3 937.1 3057.1 0.86 0.12
13 6 0 3 665.8 2321.0 0.88 0.12
14 6 6 3 733.2 2353.2 0.89 0.12
15 6 3 0 483.6 1543.2 0.90 0.15
16 6 3 6 849.7 3143.1 0.96 0.12
 C.R. Storck et al. / LWT - Food Science and Technology 53 (2013) 346e354 351

find any effect with the addition of transglutaminase. It is likely that

this discrepancy was due to the different enzyme levels used.
Nevertheless, Collar and Bollaín (2004) described a progressive
decrease in the final viscosity of wheat flour when trans-
glutaminase was added.
The setback, or the increase of viscosity during cooling, was
significantly influenced by all factors studied, but only the inter-
action between albumin and casein significantly affected the
setback (Fig. 2C). The individual addition of transglutaminase, al-
bumin or casein significantly decreased the setback. The setback is
usually related to crystallisation of the amylose chains, which could
be affected by reorganisation of the denatured proteins from the
protein isolates.

3.3. Texture profile analysis of rice composite dough

Bollaín, Angioloni, and Collar (2006) successfully used the

assessment of wheat dough properties as a good predictor of
breadmaking dough performance because those properties provide
information about gluten and starch molecular structures. Despite
the fact that gluten-free matrixes are structurally different than
gluten dough, rheological assessment of gluten-free matrixes might
give an indication of its further functionality.
The textural properties of rice composite doughs are show in
Table 3 and were significantly dependent on the factors studied.
Dough hardness was significantly influenced by transglutaminase
and casein addition, and the albumin effect was only significant
when interacting with casein, which indicates a synergistic effect
(Fig. 3A). Hardness increase when higher amounts (up to 10 g/100 g
of flour) of protein were added.
Marco and Rosell (2008b) found that rice dough hardness was
significantly influenced by protein isolate addition and trans-
glutaminase. The individual addition of pea protein or trans-
glutaminase produced an increase in hardness, and in contrast, the
addition of soybean decreased hardness. In the present study, the
effect of transglutaminase on hardness may be explained by an
increase in the molecular weight of the proteins resulting from the
crosslinking action of this enzyme, which results in larger polymers
(Marco et al., 2008).
Adhesiveness was significantly influenced by the individual
addition of transglutaminase and casein. A synergistic effect was
observed between transglutaminase and casein (Fig. 3B) as well as
the combination of albumin and casein, which induced a reduction
of the adhesiveness (Fig. 3C). These results agree with previous pea
and soybean protein findings (Marco & Rosell, 2008b). Adhesive-
ness is an important factor that affects handling convenience in
dough processing (Bonet et al., 2006).
The single addition of transglutaminase and casein resulted in a
significant decrease in cohesiveness. In wheat dough, cohesiveness
has been reported as a good predictive parameter of fresh-baked
product quality because more-cohesive wheat doughs produce
softer breads with higher specific volumes (Armero & Collar, 1997). Fig. 3. Response surface plots of the mechanical and surface-related textural proper-
However, rice composite doughs behave differently than gluten ties obtained from the regression equations.
doughs (Marco & Rosell, 2008c) because they do not produce
cohesive doughs. transglutaminase has the ability to trap water and it thus causes an
increase in water-holding capacity and, consequently, a decrease in
3.4. Baked product evaluation the percentage of bake loss (Moore et al., 2006; Renzetti et al.,
2008). Ingredients such as soy flour and dairy ingredients also in-
Experimental data from the technological quality assessment of crease water-holding capacity in baked goods (Moore et al., 2006).
gluten-free breads were submitted to analysis of variance to The specific volume of breads presented a mean value of
determine the main effects of the proteins and transglutaminase. 1.49 cm3 g 1, which was negatively affected by the individual
The data found are presented in Table 4. addition of the factors studied. The highest specific volume
The bake loss of breads was influenced by all factors studied. (1.62 cm3 g 1) was obtained when 1.35 U of transglutaminase/g of
transglutaminase, albumin and casein had a significant negative protein, 0.67 g/100 g of albumin and 0.67 g/100 g of casein were
effect on bake loss. Presumably, the protein network formed by added (Table 4, treatment 1). Conversely, the lowest specific
352 C.R. Storck et al. / LWT - Food Science and Technology 53 (2013) 346e354

Table 4
Variables studied and the response for gluten-free baked product characteristics and texture.

Treatments Variables Analysis

Transglutaminase Albumin (g/100 g Casein (g/100 g Specific Bake loss Hardness Adhesiveness Springiness Cohesiveness Chewiness
(U/g rice flour protein) of rice flour) of rice flour) volume (g/100 g) (g) (gs 1) (g) (g) (g)
(cm3.g 1)
Control 0 0 0 1.43 18.80 1492.2 5.40 0.92 0.41 505.9
1 1.35 0.67 0.67 1.62 17.42 1228.6 4.08 0.94 0.51 583.6
2 1.35 0.67 5.32 1.51 17.17 1462.4 5.51 0.92 0.61 862.9
3 1.35 5.32 0.67 1.59 14.25 2313.2 4.90 0.95 0.56 1223.7
4 1.35 5.32 5.32 1.54 14.17 2695.1 4.40 0.93 0.60 1517.6
5 10.65 0.67 0.67 1.50 15.80 1599.9 6.29 0.93 0.47 698.1
6 10.65 0.67 5.32 1.35 17.68 2175.2 12.38 0.94 0.50 1044.8
7 10.65 5.32 0.67 1.45 14.70 2814.3 9.47 0.95 0.54 1337.8
8 10.65 5.32 5.32 1.40 14.11 3377.2 10.79 0.95 0.59 1967.2
9 6 3 3 1.47 14.42 2161.9 6.16 0.96 0.54 1119.7
10 6 3 3 1.46 13.83 2142.6 10.75 0.94 0.56 1110.6
11 0 3 3 1.57 15.72 2005.7 3.88 0.95 0.52 915.1
12 12 3 3 1.45 14.01 2428.4 19.47 0.94 0.56 1292.0
13 6 0 3 1.54 15.75 1302.9 4.39 0.92 0.54 651.3
14 6 6 3 1.54 12.91 2593.8 8.85 0.94 0.59 1362.0
15 6 3 0 1.47 14.59 1990.2 6.67 0.94 0.52 999.1
16 6 3 6 1.45 14.17 2404.2 7.28 0.95 0.56 1247.8

volume (1.35 cm3.g 1) resulted with the addition of 10.65 U of Rosell (2004b) added up to 1.5% of transglutaminase (16 U/g of
transglutaminase/g of protein, 0.67 g/100 g of albumin and 5.32 g/ protein) for rice baked product production, obtaining the optimum
100 g of casein (Table 4, treatment 6), which agrees with the bread volume with 1% of enzyme (11 U/g of protein). Renzetti et al.
findings of Marco and Rosell (2008a), who reported a decrease in (2008) observed a decrease in the specific volume of baked product
the specific volume of rice-based bread when adding high levels (from 1.83 to 1.79 cm3 g 1) when adding 1 U of transglutaminase/g
(13%) of soybean protein isolate to rice flour. of protein to brown rice flour.
In this research, when the higher level of protein (10.64 g e Gluten-free breads have a unique texture structure that is
treatments 4 and 8) were added, the specific volume of breads only responsible for their gritty mouthfeel. Due to this fact, there have
decreased when higher amounts of transglutaminase were added, been major efforts to improve the texture of these products by
when compared to control bread. obtaining wheat-like structures (Gujral et al., 2003; Gallagher et al.,
Moore, Schober, Dockery, and Arendt (2004) showed higher 2004; Marco & Rosell, 2008a). The textural properties of rice baked
specific volume in the presence of skin milk powder (2.08 cm3 g 1) products were also affected for all the factors studied (Table 4), and
when comparing the effect of skin milk powder on the specific baked product hardness was significantly influenced by all factors
volume of gluten-free baked product; however, the extent of the studied. transglutaminase and albumin had an effect on hardness.
effect was highly dependent on the protein source (Moore et al., transglutaminase and casein had a negative effect on this param-
2006). Gluten-free baked product containing egg powder pre- eter, and albumin had a positive effect. The design allowed for the
sented a higher specific volume (2.0 cm3 g 1) than breads with determination of the optimum formulation for obtaining the lowest
soybean flour (1.7 cm3 g 1) or milk powder (1.7 cm3 g 1). Gallagher crumb hardness, which was obtained in the presence of 1.35 U of
et al. (2004) also showed that the addition of dairy proteins to the enzyme/g of protein, 0.67 g/100 g of albumin and 0.67 g/100 g of
gluten-free formulation improves the specific volume. casein. Gallagher et al. (2003b) obtained better bread shape and
As observed by Moore et al. (2006), crosslinking induced by volume and firmer crumb texture when testing the addition of
transglutaminase addition does not always result in better specific seven types of dairy products used in gluten-free baked product
volume in breads. These authors added transglutaminase (0.1e formulation. It seems that the effect of transglutaminase is highly
10 U/g of protein) to protein-enriched, composite gluten-free dependent on the level of enzyme added, inducing an increase of
blends without noticing any effect on specific volume in the pres- crumb hardness when adding up to 1 U of transglutaminase/g of
ence of egg powder or soybean flour, and they observed a decrease protein; however, the opposite effect is observed when using 10 U
in the specific volume in the presence of milk powder. Gujral and of transglutaminase/g of protein (Moore et al., 2006). In addition,

Fig. 4. Cross section of gluten-free baked good containing 1.35 U transglutaminase/g of protein, 0.67 g of albumin/100 g and 0.67 g of casein/100 g (A) and cross section of control
gluten-free baked good (B).
 C.R. Storck et al. / LWT - Food Science and Technology 53 (2013) 346e354 353

References

AACC. (1995). Approved methods of the American Association of Cereal Chemists.

American Association of Cereal Chemists.
AOAC. (1995). Official methods of analysis of the AOAC International (16th ed.)..
supplement 1998, Washington, DC. Chapter 32: pp. 25e28.
Armero, E., & Collar, C. (1997). Texture properties of formulated wheat doughs:
relationships with dough and bread technological quality. Zeitschrift für Leb-
ensmitteluntersuchung und -Forschung A, 204, 136e145.
Bollaín, C., Angioloni, A., & Collar, C. (2006). Relationships between dough and bread
viscoelastic properties in enzyme supplemented wheat samples. Journal of Food
Engineering, 77, 665e671.
Bonet, A., Bonet, A., Blaszczak, W., & Rosell, C. M. (2006). Formation of homopoly-
mers and heteropolymers between wheat flour and several protein sources by
transglutaminase catalysed crosslinking. Cereal Chemistry, 83, 655e662.
Cabrera-Chávez, F., & Calderón de la Barca, A. M. (2008). Letter to the editor. Journal
of Cereal Science, 48, 278.
Catassi, C., & Fasano, A. (2008). Celiac disease. In E. K. Arendt, & F. Dal Bello (Eds.),
Gluten-free cereal products and beverages). Elsevier, Academic Press. pp. 464.
Clerici, M. T. P. S., Airoldi, C., & El-Dash, A. A. (2009). Production of acidic extruded
rice flour and its influence on the qualities of gluten-free bread. LWT e Food
Science and Technology, 42, 618e623.
Collar, C., & Bollaín, C. (2004). Impact of microbial transglutaminase on the visco-
Fig. 5. Response surface plots of chewiness. Effect of adding different levels of albumin
elastic profile of formulated bread doughs. European Food Research Technology,
and casein. 218, 139e146.
Demirkesen, I., Mert, B., Sumnu, G., & Sahin, S. (2010). Rheological properties of
gluten-free bread formulations. Journal of Food Engineering, 96, 295e303.
Flander, L., Salmenkallio-Marttila, M., Suortti, T., & Autio, K. (2007). Optimization of
the combination of transglutaminase with different protein sources ingredients and baking process for improved wholemeal oat bread quality. LWT
in gluten-free baked product formulations usually increases crumb e Food Science and Technology, 40, 860e870.
Gallagher, E., Gormley, T. R., & Arendt, E. K. (2003a). Crust and crumb characteristics
hardness, an effect that has been associated with the crosslinking of gluten free breads. Journal of Food Engineering, 56, 53e161.
action of the enzyme (Marco & Rosell, 2008a; Renzetti et al., Gallagher, E., Gormley, T. R., & Arendt, E. K. (2004). Recent advances in the
2008).Fig. 4. formulation of gluten-free cereal-based products. Trends in Food Science and
Technology, 15, 143e152.
Baked product chewiness was influenced by all factors, and
Gallagher, E., Kundel, A., Gormley, T. R., & Arendt, E. K. (2003b). The effect of dairy
there was a significant interaction between albumin and casein and rice powder addition on loaf and crumb characteristics, and on shelf life
(Fig. 5). Increasing albumin and casein levels led to an increase in (intermediate and long-term) of gluten-free breads in a modified atmosphere.
European Food Research Technology, 218, 44e48.
chewiness. The highest chewiness was found when blends of 10.65
Gerrard, J. A. (2002). Protein-protein crosslinking in food: methods, consequences,
U of transglutaminase/g of protein, 5.32 g/100 g of albumin and applications. Trends in Food Science & Technology, 13, 391e399.
5.32 g/100 g of casein were used for making gluten-free composite Gerrard, J. A., & Sutton, K. H. (2005). Addition of transglutaminase to cereal products
formulations (Table 4, treatment 8). This result agrees with previ- may generate the epitope responsible for coeliac disease. Trends in Food &
Technology, 16, 510e512.
ous findings in gluten-free bread formulated in the presence of Gilbert, G. A., & Spragg, S. P. (1964). Iodine sorption: “Blue value”. In (Series Ed.) &
transglutaminase (Marco & Rosell, 2008a; Renzetti et al., 2008). R. L. Whistler, et al. (Vol. Eds.). Methods in carbohydrate chemistry: Vol. IV. Starch
(pp. 168e169). London: Academic Press.
Gujral, H. S., Guardiola, I., Carbonell, J. V., & Rosell, C. M. (2003). Effect of cyclo-
dextrin glycoxyl transferase on dough rheology and bread quality from rice
4. Conclusions flour. Journal of Agricultural and Food Chemistry, 51, 3814e3818.
Gujral, H. S., & Rosell, C. M. (2004a). Functionality of rice flour modified with a
A protein-enriched, gluten-free formulation based on rice flour microbial transglutaminase. Journal of Cereal Science, 39, 225e230.
Gujral, H. S., & Rosell, C. M. (2004b). Improvement of the breadmaking quality of
can be obtained with a suitable combination of egg albumin and rice flour by glucose oxidase. Food Research International, 37(1), 75e81.
casein using transglutaminase as a processing aid for creating a Marco, C., Pérez, G., León, A. E., & Rosell, C. M. (2008). Effect of transglutaminase on
protein network. Protein-enriched composite blends are reinforced protein eletrophoretic pattern of rice, soybean, and rice-soybean blends. Cereal
Chemistry, 85, 59e64.
with the addition of transglutaminase. Protein isolates significantly
Marco, C., & Rosell, C. M. (2008a). Breadmaking performance of protein enriched,
(p < 0.05) modified both the gelatinisation and gelling behaviour of gluten-free breads. European Food Research Technology, 227, 1205e1213.
the rice starch, reducing the peak viscosity and final viscosity. Marco, C., & Rosell, C. M. (2008b). Functional and rheological properties of
protein enriched gluten free composite flours. Journal of Food Engineering, 88,
Dough hardness and adhesiveness were significantly affected by
94e103.
transglutaminase and casein addition. Marco, C., & Rosell, C. M. (2008c). Effect of different protein isolates and trans-
Protein-enriched, gluten-free baked goods with improved glutaminase on rice flour properties. Journal of Food Engineering, 84, 132e139.
crumb texture and better specific volume could be obtained with Moore, M. M., Heinbockel, M., Dockery, P., Ulmer, H. M., & Arendt, E. K. (2006).
Network formation in gluten-free bread with application of transglutaminase.
the combination of transglutaminase (1.35 U/g of protein), albumin Cereal Chemistry, 83, 28e36.
(0.67 g/100 g) and casein (0.67 g/100 g). Overall, lower amounts of Moore, M. M., Schober, T. J., Dockery, P., & Arendt, E. K. (2004). Textural comparisons
transglutaminase and a combination of proteins resulted in prod- of gluten-free and wheat-based doughs, batters, and breads. Cereal Chemistry,
81, 567e575.
ucts with technological characteristics that may be more desirable Motoki, M., & Seguro, K. (1998). Transglutaminase and its use for food processing.
for consumers. However, more studies are needed about the use of Trends Food Science & Technology, 9, 204e210.
transglutaminase in other food systems and if it presents any risk to Mujoo, R., & Ng, P. K. W. (2003). Identification of wheat protein components
involved in polymer formation on incubation with transglutaminase. Cereal
celiac disease. Chemistry, 80, 703e706.
Ngarize, S., Adams, A., & Howell, N. K. (2004). Studies on egg albúmen and whey
protein interactions by FT-Raman spectroscopy and rheology. Food Hydrocol-
Acknowledgements loids, 18, 49e59.
Pizzinatto, A., & Campagnolli, D. M. F. (1993). Avaliação tecnológica de produtos
derivados da farinha de trigo (pão, macarrão, biscoitos). In (Technology assessment
This work was financially supported by CAPES (Coordenação de of products derived from wheat flour (bread, pasta, biscuits)): . Campinas: Boletim
Aperfeiçoamento de Pessoal de Nível Superior), Consejo Superior ITAL.
Renzetti, S., Bello, F. D., & Arendt, E. K. (2008). Microstructure, fundamental
de Investigaciones Científicas (CSIC) (AGL2008-00092/ALI) and a
rheology and baking characteristics of batters and breads from different gluten-
joint CYTED project (PANXTODOS P106AC0301).
354 C.R. Storck et al. / LWT - Food Science and Technology 53 (2013) 346e354

free flours treated with a microbial transglutaminase. Journal of Cereal Science, panes: granos, harinas y productos de panificación en iberoamérica (pp. 123e160).
48, 33e45. Córdoba: Hugo Baéz Editor.
Renzetti, S., Courtin, C. M., Delcour, J. A., & Arendt, E. K. (2010). Oxidative and SAEG- Sistema para Análises Estatísticas. (2007). (System for Statistical Analyses).
proteolytic enzyme preparations as promising improvers for oat bread formu- Versão 9.1. Viçosa: Fundação Arthur Bernardes e UFV.
lations: rheological, biochemical and microstructural background. Food Chem- Stathopoulos, C. E. (2008). Dairy-based ingredients. In E. Arendt, & F. Dal Bello (Eds.),
istry, 119, 1465e1473. Gluten-free cereal products and beverages (pp. 464). Elsevier, Academic Press.
Rosell, C. M., Brites, C. M., Pérez, E., & Gularte, M. Arroz (2007). In such flours, such Tan, Y., & Corke, H. (2002). Factor analysis of physicochemical properties of 63 rice
breads: Grain, flour and bread products in Latin America. De tales harinas, tales varieties. Journal of the Science of Food and Agriculture, 82, 745e752.