Role of gluten and starch in crumb structure and texture of

fresh and stored straight-dough bread

a

Bert Lagrain , Edith Wilderjans , Christ Glorieux , Jan A. Delcour

Laboratory of Food Chemistry and Biochemistry and LFoRCe, KU Leuven, Kasteelpark Arenberg 20 box 2463, 3001
Leuven, Belgium
b
Laboratory for Acoustics and Thermal Physics, KU Leuven, Celestijnenlaan 200D, 3001 Leuven, Belgium

ABSTRACT
Bread quality decreases during storage mainly due to crumb firming. The objective of this study was to
establish the relation of the main flour components, gluten and starch, to bread crumb cellular structure
and elasticity after baking and during further storage. While the elastic modulus of bread crumb strongly
increased upon storage, image and ultrasonic analyses demonstrated that its cellular structure did not
change. Changing gluten properties in dough had a profound impact on bread density and its foam
structure without affecting the rheological properties of the crumb cell walls. Bread of lower density
initially had a lower elastic modulus, which remained low during storage, while bread of higher density
showed the opposite. Inclusion of an antifirming amylase in the recipe altered neither bread density, nor
its cellular structure, but strongly affected the impact of storage on crumb texture. It was concluded that
bread density is a major determinant of bread crumb structure and texture during storage. In breads
with a similar density and crumb structure, the evolution of crumb stiffness during storage is determined
by the changes in the starch population.

Introduction

Bread crumb is a cellular solid or sponge with a high volume fraction ( 0.8) of dispersed gas in a matrix
of mainly open cells (Scanlon and Zghal, 2001). The cell walls themselves consist of swollen starch granules
interlinked by a mixture of aggregated starch and gluten proteins. Bread quality decreases during storage,
because bread crumb firms and loses resilience (Goesaert et al., 2009). Since mainly starch, protein and water
contribute to the structural architecture and the mechanical strength of bread, they are assumed to play an
important role in the change in cell wall properties during bread storage. However, there is still controversy
about the exact role of the different flour components in bread firming (Gray and Bemiller, 2003). While the
role of gluten proteins in bread crumb firming is still under debate, both the rate and extent of starch
(re)crystallization with the formation of supermolecular structures have been related to the texture and
acceptability of bread. Therefore, research to date has focussed on the changes in starch during heating, cooling
and storage. Besides the molecular changes that determine crumb firming, bread firmness itself depends on its
density (Fearn and Russell, 1982). Most anti-firming additives influence not only cell wall constituents, such as
starch, but often they also affect structural properties like cell size and density. Hence, it is necessary to
understand the macroscopic behavior and complexity of bread crumb during storage and to relate them to the
biopolymer building blocks of bread.
The evaluation of the mechanical properties of crumb is of particular importance, because they determine
the consumers perception of bread quality (Scanlon and Zghal, 2001). The most common method to measure
crumb firmness is based on indentation (Liu and Scanlon, 2002). At low stresses, bread crumb behaves as a
linear elastic solid and its deformation behaviour is independent of the time scales of observation. Within these
boundaries, the linear zone of a load-displacement curve from a classical indentation can be used for extracting
bread crumb elastic properties (Liu and Scanlon, 2002).
This study aimed at determining crumbs textural and structural properties during storage and the role of
gluten and starch therein. The role of these main flour biopolymers was studied by specifically targeting gluten
functionality with redox agents and changing starch properties with a maltogenic exo-amylase. The linearelastic properties of the bread sponge were then determined using indentation, while the cellular structure of
bread during storage was monitored with digital image analysis and by means of non-contact ultrasound as
optimized for bread crumb (Lagrain et al., 2006).

InsideFood Symposium, 9-12 April 2013, Leuven, Belgium

1|P a g e

2
2.1

Materials and Methods

Materials

Flour [cultivar Legat, moisture content (mc): 13.9%; protein content: 10.5% on dry basis (db)] was
milled on a Bhler MLU-202 laboratory mill (Uzwil, Switzerland). Bacillus stearothermophilus maltogenic
amylase (BStA, E.C. 3.2.1.133) from Novozymes (Novamyl , Bagsvaerd, Denmark) was used as anti-firming
enzyme. All other reagents were from Sigma-Aldrich (Steinheim, Germany) unless otherwise specified.
2.2

Bread making

Control bread of ca. 400 g contained flour (1000 parts), compressed yeast (53 parts), water (590 parts),
sucrose (60 parts) and salt (15 parts). To change bread properties, KBrO3 (40 mg/kg flour), glutathione, (GSH,
40 mg/kg flour), or BStA [5.06 enzyme units (EU)/g flour] were added before mixing. The ingredients were
mixed for 5.5 min at 20 C to obtain 4.0 kg of dough. The dough was divided into 8 pieces of 450 g , fermented
for 90 min at 30 C and proofed for 36 min in the baking pan. The dough was then baked for 40 min at 210 C
in a rotary oven (National Manufacturing, Lincoln, NE, USA). The breads were weighed and loaf volume was
measured by rapeseed displacement. Thereafter, breads were packed in sealed plastic bags and stored for 168 h
at 23 C. After 5, 48, 96 and 168 h two breads were used for further analysis. From the centre of the crumb mc
was determined according to AACC method 44-19.01(AACCI, 2011).
2.3

Differential scanning calorimetry (DSC)

Amylopectin melting enthalpy, a measure for amylopectin retrogradation, was measured with a
differential scanning calorimeter (DSC Q1000, TA instruments, New Castle, NH, USA). After 5 h and 168 h of
storage, crumb from the bread centre (35-45 mg) was weighed into aluminium pans. The pans were sealed and
heated from 25 to 200 C at a rate of 4 C/min. The enthalpies corresponding to the melting of recrystallized
amylopectin were measured using Universal Analysis (TA instruments) and expressed on crumb db. Analyses
were performed in triplicate on samples of one bread.
2.4

Elasticity measurements

For texture analysis, two breads of each type (control bread, bread with KBrO3, GSH or BStA) were
analyzed. Three slices of 25 mm thick were cut from the bread centre and the crust was removed. Loaddisplacement curves were obtained by pressing a 25-mm diameter cylindrical indenter into the centre of the
slices using a Texture Analyser (TA-XT2i, Stable Micro Systems, Surrey, UK) equipped with a 5 kg load cell.
The compressive Youngs modulus E (N/m2) of bread crumb was calculated from the slope (N/m) of the loaddisplacement curves in the linear elastic regime according to Liu and Scanlon (2002) by

(1 2 )

(1)
D
where D (m) denotes the intender diameter and the Poissons ratio. For a value of 0.103 was taken as
described in Lagrain et al. (2012)
2.5

Image analysis

Images (40 x 40 mm) the centre of bread slices were captured using a flatbed scanner (Vuego Scan Brisa
610S, Acer Peripherals, Shizou, China) and supporting software (iPhoto Plus 4.0, Ulead Systems, Inc., Taipei,
Taiwan). Images were scanned full scale in 256 gray levels at 300 dots per inch (dpi) and further processed
using the image processing toolbox of Matlab 6.1 (Mathworks, Natick, MA, USA). Otsu thresholding was used
for conversion to binary images. For each bread type, the mean cell area, the number of cells/cm, and the cell to
total area ratio were extracted from the images.
2.6

Acoustic analysis

A transmission setup was used to determine the propagation of ultrasonic waves through bread crumb
and to calculate the tortuosity, a form factor describing the path of connected pores, and the viscous
characteristic length (m), a measure for the size of the intersections in the crumb cell walls (Lagrain et al.,
2006). A reflection setup was used to determine the open porosity of bread. For both setups, a pulser-receiver
(Panametrics, PR5085, Waltham, MA, USA) was used as a function generator and amplifier unit. Time signals
were registered using a 9310M oscilloscope (LeCroy, New York, NY, USA). In the transmission configuration,
a pair of air coupled transducers (NCT-202, Ultran Group, State College, PA, USA), centred at 200 kHz, were
used. The tortuosity was determined by measuring the phase velocity of an ultrasonic wave travelling through a
P a g e |2

bread slice of 8.0 mm thickness, while was calculated from the attenuation of the transmitted wave. To
estimate the open porosity, the reflection coefficient was determined by using the same transducer as emitter and
detector. From the ratio of the frequency spectra of the incident and reflected waves, the reflection coefficient
was deduced and, with knowledge of the tortuosity from the transmission measurement, the open porosity was
estimated (Lagrain et al., 2006). Two breads of each type were analyzed and measurements were on three slices
of each bread after 3, 48, 96 and 168 h.
2.7

Data analysis

The cellular structure variables, acoustic variables and retrogradation enthalpies were normally
distributed (P > 0.05, the Shapiro-Wilk test) (SAS system for Windows V8, SAS Institute, Cary, NC, USA).
Significant differences (P < 0.05) in the variables were determined by the ANOVA procedure using the SAS
package and were based on at least three individual measurements. Linear regression was performed in Excel
(Microsoft, Redmond, WA, USA).

Results and discussion

During storage of control bread, E steadily increased (Fig. 1). It was further demonstrated that cell
geometry or the original crumb structure remained unaffected during bread storage (Lagrain et al., 2012). The
origin of the observed crumb stiffening and the factors contributing to it will be further elucidated in the next
sections by specifically targeting the major flour constituents, gluten and starch.

Fig. 1 Evolution of Youngs modulus (E) during 168 h storage of control bread crumb and bread crumb with glutathione
(GSH), KBrO3 and maltogenic exo-amylase (BStA). Reprinted from Lagrain et al. (2012) with permission from Springer
Science+Business Media.

3.1

Gluten modification with redox agents

To affect gluten properties in dough and bread, redox additives were added. Redox agents, such as GSH
and KBrO3, affect the levels of disulfide bonds in and between gluten proteins. This, in turn, impacts dough
rheology and consequently also bread volume. The addition of GSH reduces the average molecular weight of
the gluten aggregates, which results in increased dough extensibility (Lagrain et al., 2012). In our study, this led
to bread with lower density as shown in Table 1. Bread with added GSH in the recipe had a significantly (P <
0.05) smaller mean cell area than control bread and, hence, also significantly more cells/cm 2 (Table 1). Both
observations indicate an increased gas cell stability during fermentation and the first stages of baking. The cell
to total area ratio of bread prepared with GSH was significantly (P < 0.05) lower than that of the other breads
(Table 1). Ultrasonic inspection further revealed that GSH resulted in bread with a higher crumb tortuosity and a
lower crumb porosity, a further indication of a fine grain bread crumb (Lagrain et al., 2006). In contrast, KBrO3
stiffens dough during fermentation (Lagrain et al., 2012). This prevented dough expansion and increased bread
density compared to the control bread (Table 1). For both bread types with redox additives, the viscous

3|P a g e

characteristic length , a measure for the voids or intersections in the cell walls, differed significantly from the
control bread.
Table 1 Density and crumb grain characteristics determined by digital image and ultrasonic analysis of fresh control
bread and fresh bread with 2.1 mol KBrO3/g protein, 1.2 mol GSH/g protein and 5.06 EU maltogenic exo-amylase
(BStA)/g flour. Standard deviations are shown between brackets. Values in the same row followed by a different letter
differ significantly (P < 0.05).

Density (g/cm3)
Mean cell area (mm2)
Cells/cm2
Cell to total area ratio
Tortuosity
(m)
Open porosity

Control
0.34 (0.00)b
0.32 (0.04)a
83.0 (4.4)bc
0.29 (0.03)a
1.07 (0.03)b
15 (2)b
0.88 (0.04)a

KbrO3
0.44 (0.01)a
0.32 (0.03)a
86.6 (4.3)ab
0.32 (0.03)a
1.09 (0.04)b
11 (1)c
0.88 (0.04)a

GSH
0.26 (0.01)c
0.26 (0.02)b
89.7 (4.8)a
0.26 (0.01)b
1.23 (0.05)a
12 (2)c
0.81 (0.03)b

BStA
0.34 (0.01)b
0.33 (0.04)a
76.3 (8.2)c
0.28 (0.02)a
1.08 (0.03)b
19 (2)a
0.89 (0.04)a

Figure 1 also shows the influence of storage time on E for breads baked with the redox additives and the
maltogenic exo-amylase. Bread prepared with GSH remained much softer than control bread. In contrast, bread
made with KBrO3 firmed faster and, hence, was also significantly firmer after 168 h than control bread. The
relative density is the dominant physical parameter representing the three-dimensional structure of cellular
solids (Scanlon and Zghal, 2001). When bread is considered as open-cell foam, it can be modeled as arrays of
cubic cells that are staggered. The relation between the overall Youngs modulus E and density is then
described by the Equation of Gibson and Ashby for open-cell foams, using the cell-wall modulus Es and the cell
wall density s (Gibson and Ashby, 1988):

E s s

(2)

or

Es

s2

(3)

Fig. 2 Evolution of normalized Youngs modulus (E/2 in m/s) during 168 h storage of control bread crumb and bread
crumb with glutathione (GSH), KBrO3 and maltogenic exo-amylase (BStA). Reprinted from Lagrain et al. (2012) with
permission from Springer Science+Business Media.

P a g e |4

Figure 2 depicts the evolution of the normalized modulus (E/2) during bread storage. Given the limited
changes in crumb moisture content and the conservation of cell geometry during storage, it can be assumed that
remained quasi constant for each bread type. Moreover, the composition of all bread types was the same,
except for the neglible levels of additive. Hence, s can also be considered constant and identical for all bread
types. If then Equation 3 is used, the curve in Figure 2 essentially conveys the increase in cell wall modulus Es
during storage, which was independent of crumb density . The similar shape of the curves for control bread and
the breads with redox additives in Figure 2 demonstrate that the molecular changes in the gluten protein
network, as induced by the redox agents, had little if any effect on crumb cell wall stiffening. The latter was
related to starch properties and will be further discussed in the next section.When presuming a linear relation
between E and Es, (/s)2 determines the slope of the linear function. In practice, this means that the difference
between the E moduli of breads with the same composition, but different density will increase with increasing Es
or storage time.
Table 2 Retrogradation enthalpies (H in J/g of crumb) for crumbs from fresh bread and breads stored during 168 h at 23
C. Enthalpies are for control bread and bread with addition of 1.2 mol GSH/g protein, 2.1 mol KBrO 3/g protein and
5.06 EU BStA/g flour in the dough recipe. Values in the same row followed by a different letter differ significantly (P <
0.05). Relative standard deviations were less than 18%.

Storage time (h)

5
168
3.2

Control
0.26 a
2.41 b

KbrO3
0.29 a
2.00 b

GSH
0.28 a
2.10 b

BStA
0.25 a
0.63 a

Starch modification with maltogenic amylase

Starch behavior during bread aging was determined with DSC. Amylopectin recrystallization, such as
judged from amylopectin crystal melting enthalpies, increased during bread storage (Table 2). Moreover, there
was a linear relation between retrogradation enthalpies and E moduli of breads with a similar density (Fig. 3). At
the same stage of aging, no significant differences in amylopectin retrogradation enthalpies were observed for
control bread and breads with a different density due to redox agents. This confirms that crumb cell wall
stiffening is related to amylopectin retrogradation and independent of crumb density.

Fig. 3 Relation between Youngs modulus (E) and starch retrogradation enthalpy of bread crumb with similar density from
control bread (dots) and bread with maltogenic exo-amylase (diamonds). Reprinted from Lagrain et al. (2012) with
permission from Springer Science+Business Media.

To modify starch properties during storage, bread was made with the maltogenic amylase BStA. BStA is
an effective anti-firming amylase that reduces amylopectin retrogradation (Goesaert et al., 2009). Table 2 shows
that BStA strongly decreased amylopectin retrogradation after 168 h of storage. Moreover, BStA had a limited
impact on bread volume and structure. Addition of BStA to the bread recipe indeed resulted only in small
changes in crumb structure as shown in Table 1. When compared to control bread, no significant differences
were observed in the cellular structure parameters, except for a higher value for crumb cell wall intersection
when the enzyme had been used (Table 1). Figure 1 also shows the evolution of E for breads baked with and
without BStA during storage. The enzyme strongly reduced the rate crumb stiffening during aging with only a
slight increase in E after 168 h. No significant differences in moisture content of the crumb centre or crumb

5|P a g e

structure parameters during storage were observed between control bread and bread with BStA (results not
shown). Hence, the different evolution of E in control bread and bread with BStA did not originate from a
different crumb density or structure such as was the case for bread with the redox agents, but can be entirely
contributed to a different evolution of the cell wall modulus Es as illustrated in Figure 2. Since BStA has no
reported impact on other flour constituent properties, such as gluten proteins, its effect on E modulus is caused
by its impact on starch (re)crystallization. This further supports the important role of amylopectin retrogradation
in the stiffening of the crumb cell walls.

Conclusions

The mechanical properties of bread crumb structure at low stresses and their changes during storage can
be understood by considering it to be a linear-elastic, cellular solid with open cells. Within this approximation,
the role of the main flour constituents, gluten and starch, can be explained. The Youngs modulus of bread
crumb (E) is then function of the relative crumb density /s and the cell-wall modulus Es. Although the
presence of a gluten network as such, may well impact Es of bread, changing gluten properties has little if any
effect on the changes in Es during aging. Gluten rheology strongly determines bread density . More extensible
dough leads to more stable gas cells and higher bread volume, while the opposite is true for stiff dough. Starch
properties, such as amylose and mainly amylopectin crystallization contribute strongly to Es of bread crumb
during storage. Since the organoleptic assessment of firmness is related to the E modulus of bread (Fearn and
Russell, 1982), both low bread density and limited starch (re)crystallization will contribute to the consumers
appreciation of bread freshness
Acknowledgements
This work is part of the Methusalem program Food for the Future at the K.U.Leuven. B. L. wishes to
acknowledge the Research Foundation Flanders (FWO, Brussels, Belgium) for his position as postdoctoral
fellow. Jan A. Delcour is W.K. Kellogg Chair in Cereal Science and Nutrition at the K.U.Leuven.
References
AACCI, 2011. Approved Methods of Analysis AACC International, St. Paul, MN, USA.
Fearn, T., Russell, P.L., 1982. A kinetic-study of bread staling by differential scanning calorimetry - the effect of loaf
specific volume. Journal of the Science of Food and Agriculture 33, 537-548.
Gibson, L.J., Ashby, M.F., 1988. The mechanics of foams: basic results Cellular Solids: Structure and Properties. Pergamon
Press, Oxford, U.K., pp. 120-168.
Goesaert, H., Leman, P., Bijttebier, A., Delcour, J.A., 2009. Antifirming effects of starch degrading enzymes in bread
crumb. Journal of Agricultural and Food Chemistry 57, 2346-2355.
Gray, J.A., Bemiller, J.N., 2003. Bread staling: molecular basis and control. Comprehensive Reviews in Food Science and
Food Safety 2, 1-21.
Lagrain, B., Boeckx, L., Wilderjans, E., Delcour, J.A., Lauriks, W., 2006. Non-contact ultrasound characterization of bread
crumb: Application of the Biot-Allard model. Food Research International 39, 1067-1075.
Lagrain, B., Wilderjans, E., Glorieux, C., Delcour, J.A., 2012. Importance of Gluten and Starch for Structural and Textural
Properties of Crumb from Fresh and Stored Bread. Food Biophysics 7, 173-181.
Liu, Z.Q., Scanlon, M.G., 2002. Understanding and modeling the processing-mechanical property relationship of bread
crumb assessed by indentation. Cereal Chemistry 79, 763-767.
Scanlon, M.G., Zghal, M.C., 2001. Bread properties and crumb structure. Food Research International 34, 841-864.

P a g e |6