Lebensm.-Wiss. u.-Technol.

, 255 } 260 (1999)

Changes in Starch Microstructure on Baking and Staling of Wheat Bread

Susanna Hug-Iten, Stephan Handschin, Be H atrice Conde-Petit* and Felix Escher Institute of Food Science, Swiss Federal Institute of Technology (ETH), CH-8092 Zurich (Switzerland) (Received October 27, 1998; accepted February 23, 1999)

The microstructure of starch in dough and in fresh and aged bread crumb was studied by light microscopy. The samples were cryosectioned and stained with Light Green and iodine to localize protein and starch, respectively. In dough a partial segregation of starch from the protein phase is observed. On baking, starch was gelatinized and led to the formation of a continuous starch network. The starch fraction itself was inhomogeneous and consisted of swollen and interconnected starch granules. The two starch polymers, amylose and amylopectin, were found to phase separate and amylose was accumulated in the centre of starch granules. Polarized-light microscopy of fresh bread crumb showed that starch gelatinization was accompanied by the loss of birefringence. On ageing the bread crumb regained birefringence. The combination of light microscopy in the bright-xeld and polarized mode allowed identixcation of the birefringent structures. The most intense birefringence was observed in the amylose rich centre of starch granules, whereas the outer amylopectin rich zones showed slight birefringence. It is concluded that the ordered structures result from the reordering of amylose and amylopectin. It is hypothesized that the reorganization of the intra-granular amylose fraction enhances the rigidity of starch granules on bread staling. 1999 Academic Press Keywords: microstructure; bread crumb; starch; light microscopy

Introduction Starch is the main component of bread and its gelatinization induces major structural changes during baking of wheat bread. The swollen granules and partially solubilized starch act as essential structural elements of bread (1). The other important component of wheat is protein, which is responsible for the formation of the viscoelastic gluten in dough. On heating, the gluten transforms from a gel to a coagel by polymerization, which means that the gel looses its cohesiveness (2). Thus, the transformation from dough to bread involves changes both in the starch and the protein fraction. At the macroscopic level, baking induces the solidi"cation of dough and a change from a foam type system with gas cells to an open pore system, i.e. a sponge (2). On cooling and ageing of bread, rearrangements in the starch fraction lead to a series of changes including gelation and crystallization. This transformation is called starch retrogradation and is thought to be the major cause of bread crumb "rming on ageing, commonly referred to as bread staling (3). Light microscopy presents a valuable method for the study of the microstructural changes of starch. Several authors described the microstructure of dough and bread as shown by light microscopy (4}10). Nevertheless, there

*To whom correspondence should be addressed 0023-6438/99/050255 # 06 $30.00/0 1999 Academic Press

is still a lack of information on the localization of amylose and amylopectin in bread crumb, on the starch protein interfaces and on the changes upon staling. A classical method to monitor starch gelatinization is the loss of starch granule birefringence during heating. Native starch granules show strong birefringence in the form of Maltese crosses, which disappear during heating due to the loss of long range molecular order (11). On baking the starch granules of bread crumb also lose the typical Maltese cross pattern of native starch but still retain slight birefringence (4). Several authors (6, 8, 12, 13) have pointed out the di$culty involved in preparing dough and bread samples for microscopy. The disruption of the protein network and a distorted image of the bread crumb due to hydration during "xation (8) and staining (6) are common artefacts. The shrinkage of starch and protein as consequence of dehydration was also described (12}14). The objective of this study were to characterize starch and protein distribution in wheat dough and bread crumb and to localize the two starch polymers, amylose and amylopectin in the bread crumb. Finally, the microstructural changes upon staling of bread were of interest with emphasis on the amylose fraction. The microstructure of dough and bread crumb was assessed by bright"eld and polarized-light microscopy. As few preparatory steps as possible were introduced in order to minimize the risk of artefacts such as structural distortion and swelling of starch and protein.

Article No. fstl.1999.0544 All articles available online at http://www.idealibrary.com on

255

lwt/vol. 00 (1999) No. 5

Experimental Preparation of bread White pan bread was used for the investigations. Bread doughs were prepared from low-extraction wheat #our (Coop Mu K hle Zu K rich CMZ, CH-Zurich). According to the manufacturer, the #our contained 11.8 to 12.1 g/100g (dry base) protein and 0.35 to 0.38 g/100 g (db) ash. Yeast (42 g) was dispersed in a small amount of water and added to the #our (1 kg). The optimal amount of water for dough preparation was determined by recording farinograms (15) and varied between 64 to 67 g water/100 g #our (wet base). The dough was kneaded in a mixer (Artofex AG, CH-Gra K nichen) for 3 min at low speed, followed by 4 min at high speed. Salt (20 g) was added after 2 min mixing at low speed. Dough pieces of 600 g each were placed in greased (Boeson Trennwax, Boehringer, D-Ingelheim) moulds. Proo"ng was accomplished in a ventilated oven (UWS 880, Wyrsch Ing., CH-Meggen) during 90 min at 25 3C, the moulds being covered with a wet cloth. The loaves were baked at 220 3C for 35 min in a ventilated oven (Blodgett Co. Inc., USA-Burlington). The breads were cooled 4.5 h to room temperature, sealed in plastic bags and stored for 0 to 7 d at 20 3C.

were examined in an Axioplan photomicroscope (Zeiss Ltd., D-Oberkochen)

Results and Discussion The methods of sample preparation for microscopy were developed based on preliminary experiments (data not shown) and information from the literature (16, 17). The preparation of specimens from dough was similar to a method used by Cunin (18) for pasta and consisted of freezing the sample without prior "xation. Bread crumb was soaked in diluted O.C.T. compound for 15 min before freezing. The aim was to stabilize the porous structure and to prevent a distortion of the pores during cryosectioning. A rather short soaking time was selected, since the artefacts as described by Moss (8) are most probably promoted by long hydration times (18 to 24 h). However, in spite of the short hydration time, the swelling of bread crumb could not be inhibited completely. On the other hand, shrinkage of protein and starch due to dehydration was avoided by using the cryosectioning technique. Carbon dioxide was used as cryogen to limit the temperature gradient within the sample which prevented the piece of bread from stress cracking. The risk of ice crystal formation was considered to be negligible since O.C.T. acts as a cryoprotectant. Iodine was selected for staining starch because it allows di!erentiation between amylose and amylopectin. However, staining intensity is known to vary between the starch granules. Staining with aqueous solution presents a potential source of artefacts. Samples which had been stained with aqueous solutions, that is, Light Green and Lugol's solution, were compared to samples which had only been exposed to iodine vapour. The aqueous staining solutions were found to slightly increase the swelling of starch. Overall, the changes caused by hydration during stabilization and staining did not induce major changes in the starch fraction. The starch granules maintained their shape. Neither the localization of starch in the protein matrix nor the distribution of the two starch polymers were a!ected. It is therefore concluded that the sample preparation techniques were suitable for the purpose of the investigation.

Light microscopy Small pieces of dough and crumb (approx. 7;7;7 mm) were cut from the centre of the dough and bread, respectively. Dough samples were immediately frozen with carbon dioxide. The bread crumb samples were soaked for 15 min under vacuum in Tissue-Tek O.C.T. compound (Miles, USA-Kankakee) diluted 1 : 4 with 20% sucrose solution, and then frozen with carbon dioxide. The samples were cut in a Reichert-Jung cryostat (Leica, A-Vienna) at !20 3C. Cryostat was displayed for sections of 10 m thickness. The samples were then mounted in the frozen state to microscope slides, which had been covered with a solution of glycerol/gelatine to improve the adhesiveness of the cryosections during staining. No glycerol/gelatine coating of the microscope slides was used for samples which were determined for polarized microscopy. The thin sections were either directly observed under polarized light to assess the degree of starch birefringence or "rst stained in order to localize the protein and the two starch fractions, amylose and amylopectin, as well as to improve contrast. Protein was stained with aqueous Light Green solution (1 g/L, Fluka Chemie AG, CH-Buchs) for 30 min, starch with iodine in a diluted 1 : 10 (v/v) Lugol's solution (stock solution I "14 mM, KI"44 mM, Fluka Chemie AG, CH Buchs) for 10 to 20 s. After each staining step the slides were rinsed in water. Before examining the samples under the microscope they were covered by a droplet of glycerol/water solution (1 : 1 v/v) and a cover glass. When only starch had to be stained the thin sections were exposed to iodine vapour (Lugol's solution I "14 mM,  KI"44 mM, Fluka Chemie AG, CH-Buchs) for 1 to 2 min. Then the sections were covered with the glycerol/water solutions as described above. All specimens

Microstructure of dough Figure 1 presents a light micrograph of a cryosection of proofed wheat dough stained with iodine and Light Green. The native starch granules stain slightly violet
Fig. 1 Light micrograph of a cryosection of proofed dough stained with Light Green and Lugol's solution (bar"25 m) Fig. 2 a, b Light micrographs of cryosections of bread crumb at two di!erent magni"cations. Samples are stained with Light Green and Lugol's solution (a) bar"50 m (b) bar"10 m Fig. 3a, b Polarized-light micrographs of cryosections of fresh and aged bread crumb. Samples are not stained (bar"50 m) (a) fresh, 0 d (b) aged, 7 d Fig. 4a, b Light micrographs of cryosections of bread crumb. Samples are stained with iodine vapour (bar"10 m) (a) bright-"eld mode (b) polarized-light mode

256

lwt/vol. 00 (1999) No. 5

257

lwt/vol. 00 (1999) No. 5

with iodine while the protein fraction appears green as coloured with Light Green. The starch granules show the characteristic shape of native wheat starch granules. The typical bimodal size distribution of wheat starch granules is also recognizable. Protein and starch are not evenly distributed in the dough and regions are found where several starch granules are accumulated. The literature data regarding the distribution of starch and protein in bread dough is controversial. Based on microscopy, several authors (6, 7) concluded that each starch granule is surrounded by protein. A more recent structural model described dough as a bicontinuous starch-protein system (2). The starch granules have a surface coat of &free' water and tend to fuse into a continuous phase. The gluten gel "lls the space between the water-fused starch granules. The results of the present investigation on the microstructure of dough are consistent with the latter model of dough. However, it remains to be investigated how the processing steps (mixing, proo"ng and reshaping) and the protein quality of wheat in#uence the microstructure of dough. According to Kie!er and Stein (19), starch-protein segregation occurs during the reshaping of the dough after a rest period. This is thought to be the reason for the large increase of resistance in uniaxial extension, called strain hardening. Although, starch is by far the largest fraction of low-extraction wheat #our by weight, with a concentration of approx. 70 g/100 g, it only "lls about 60% of the dough volume due to the high swelling capacity of gluten (20). The latter fraction determines the rheological properties of dough whereas starch is considered to act as a "ller (20).

Microstructure of bread crumb Figures 2a and b show cross sections of pore walls of fresh bread crumb at two di!erent magni"cations. Again, starch and protein are stained in order to localize the two fractions. At low magni"cation (Fig. 2a) the starch granules appear swollen and elongated, but still retain their granular identity. Most granules are aligned parallel to the pore surface. The protein phase is distinguishable as green areas throughout the bread crumb. Details of the starch fraction are recognizable at higher magni"cation (Fig. 2b). Clearly, the starch granules are oriented and partly fused with neighbouring granules. The iodine staining reveals that amylose and amylopectin, which stain blue and brown/violet, respectively, are not homogeneously distributed in the granules. The blue, elongated areas in the inner zone of the large starch granules correspond to accumulations of amylose whereas the brown/violet surrounding phase corresponds to amylopectin rich structures. In contrast, the small granule fraction of starch does not show this phase separation. Outside the starch granules, amylose rich zones can be observed as dark lines along the starch protein interface. In Fig. 2b, concentric growth rings for one particular aspect of the internal structure of starch granules are visible for one highly swollen starch granule. At the microscopic resolution level studied, the pore walls of bread crumb may be described as a bicontinuous

structure, which is built up by starch and protein, respectively. The starch fraction itself is inhomogeneous and consists of swollen &phase separated' granules and leached starch, mainly amylose. The accumulation of amylose in the starch granule centre is also documented for rye bread, but not further commented by the authors (9). Phase separation of starch is explained by the thermodynamic incompatibility of amylose and amylopectin (21) and was also observed in wheat starch pastes (22). The elongated form of starch granules and their orientation in bread crumb was reported in previous studies (4}6). This is thought to arise from the extension of the lamellae due to the growing gas cells in the initial stage of baking. The fact that growth rings of starch are recognizable in bread crumb proves that in spite of the morphological changes and the partial amylose-amylopectin phase separation the native organization of starch granules, which is mainly determined by the packing density and the radial orientation of amylopectin (23), is largely preserved. The transformation of starch during baking leads to a continuous starch network. It is therefore not surprising to "nd that the macroscopic properties are determined by the gelatinized starch. As shown by Sandstedt (24), a typical bread crumb structure cannot be obtained when starch is replaced by an inert "ller such as glass beads. This experiment led to the conclusion that starch does not simply act as a "ller that dilutes gluten to a suitable consistency. On the other hand, it is possible to generate starch systems with a sponge structure by combining starch with other polymers (e.g. xanthan, pregelatinised starch) which replace gluten (1, 25). The resulting mechanical properties are similar to bread crumb and are determined by the swelling state of starch granules and by the distribution of the thickness of the pore walls (1). The investigations with regular bright-"eld microscopy were complemented by polarized-light microscopy. Figures 3a and b show fresh and 7 d old bread crumb at low magni"cation. The samples were not stained. As expected, fresh bread crumb exhibited almost no birefringence, since starch granule swelling upon gelatinization is accompanied by a loss of order of starch molecules (11). Interestingly, aged bread crumb showed a marked increase of birefringence. At low magni"cation, the birefringence is visible as bright irregular longish structures, and no recurrence of the typical Maltese cross pattern of native starch is observed. The circular spots should be disregarded as they may be the result of birefringent dust laying in planes out of focus. Figures 4a and b show the same "eld of stored bread crumb (7 d) at high magni"cation in bright-"eld and polarized-light mode, respectively. The protein staining was omitted, and starch was stained by exposing the cryosection to iodine vapour. This technique was adopted to prevent the washing out of amylose and swelling of the starch granules by aqueous staining solutions. The bright-"eld micrograph (Fig. 4a) clearly reveals accumulations of amylose (blue coloured) inside the large starch granules and an amylopectin rich phase (violet/brown coloured) at the outer zone of starch granules. On the corresponding polarized-light micrograph (Fig. 4b) the amylose within

258

lwt/vol. 00 (1999) No. 5

the phase separated granules is birefringent. A less intense birefringence is observed at the outer amylopectin rich zones. This congruence of bright-"eld and polarizedlight microscopy allows the localization and identi"cation of birefringent structural elements. It con"rms that reorganizations in the starch fractions, that is, starch retrogradation, is responsible for the observed birefringence. Birefringent zones as detected with polarized light microscopy correspond to anisotropic structures but not necessarily to crystallinity. The slight birefringence at the outer zones of starch granules can be attributed to retrograded amylopectin which forms organized anisotropic regions. This observation is consistent with the results of Jacobson et al. (26). They describe that granule remnants of low-concentrated starch pastes regained slight birefringence upon storage near the outer edges of intact swollen granules. The fact that the intra-granular amylose becomes birefringent implies that amylose is in an ordered state in aged bread crumb. From studies on pure amylose systems and on starch gels with rheological measurements and X-ray di!raction (27}29) it was concluded that the initial stages of starch retrogradation are dominated by the gelation of the solubilized amylose. The development of a three-dimensional network results from a phase separation into polymer-rich and polymerde"cient phases (27). This is followed by a slow crystallization presumed to occur in the polymer-rich phase. Thus, in connection with the observed birefringence of starch granules in bread crumb it is conceivable that the amylose in the starch granule centre rearranges on ageing and eventually partly crystallizes by lateral chain association. The possibility that the reorganization of amylose is induced by iodine complexation can be excluded, since the samples of Fig. 3 were not stained. However, the ordering of the amylose fraction may have been promoted by complexation with endogenous wheat starch lipids. Finally, as the gelatinized granules cannot be simply considered as amylopectin rich structures, the question of how the intra-granular amylose in#uences the mechanical properties of starch granules arises. Studies on pure amylose solutions showed that their recrystallization did not in#uence the complex modulus (27). Conversely, Conde-Petit (30) concluded that the rheological behaviour of wheat starch dispersions over a period of 30 d is dominated by changes in the amylose fraction, as determined by the loss of iodine binding capacity. Although none of the model systems may be fully transferable to bread, it is hypothesized that the changes in the intra-granular amylose fraction enhance the rigidity of starch granules on bread staling.

References
1 KEETELS, C. J. A. M., VISSER, K. A., VAN VLIET, T., JURGENS, A. AND WALSTRA, P. Structure and mechanics of starch bread. Journal of Cereal Science, 24, 15 } 26 (1996) 2 ELIASSON, A.-C. AND LARSSON, K. Cereals in breadmaking 2 A molecular colloidal approach. New York: Marcel Dekker Inc. (1993) 3 ZOBEL, H. F. AND KULP, K. The staling mechanism. In: HEBEDA, R. E. AND ZOBEL, H. F. (Eds), Baked Goods Freshness: echnology, Evaluation and Inhibition of Staling. New York: Marcel Dekker, Inc., pp. 1 } 64 (1996) 4 VARRIANO-MARSTON, E., KE, V., HUANG, G. AND PONTE, J. Comparison of methods to determine starch gelatinization in bakery foods. Cereal Chemistry, 57, 242 } 248 (1980) 5 BURHANS, M. E. AND CLAPP, J. A microscopic study of bread and dough. Cereal Chemistry, 19, 196 } 216 (1942) 6 SANDSTEDT, R. M., SCHAUMBURG, L. AND FLEMING, J. The microscopic structure of bread and dough. Cereal Chemistry, 31, 43 } 49 (1954) 7 BECHTEL, D. B., POMERANZ, Y. AND DE FRANCISCO, A. Breadmaking studied by light and transmission electron microscopy. Cereal Chemistry, 55, 392 } 401 (1978) 8 MOSS, R. Bread microstructure as a!ected by cysteine, potassium bromate and ascorbic acid. Cereal Foods =orld, 20, 289 } 292 (1975) 9 AUTO, K., HA G RKO G NEN, H., PARKKONEN, T., FRIGARD, T., POUTANEN, K., SUKA-AHO, M. AND AMAN, P. E!ects of puri"ed endo-beta-xylanase and endo-beta-glucanase on the structural and baking characteristic of rye doughs. ebensmittel-=issenschaft und-echnologie, 29, 18 } 27 (1996) 10 POMERANZ, Y. AND MEYER, D. Light and scanning electron microscopy of wheat- and rye-bread crumb. Interpretation of specimens prepared by various methods. Food Microstructure, 3, 159 } 164 (1984) 11 SEIDEMAN, J. Sta K rke Atlas. Grudlagen der Sta K rkemikroskopie und Beschreibung der wichtigsten Sta K rken. Berlin: Paul Parey (1966) 12 CHABOT, J. F. Preparation of food science samples for SEM. Scanning Electron Microscopy, 3, 279 } 286 (1979) 13 VERSCHAFFELT, E. AND VAN TEUTEM, F. E. Die A G nderung der mikroskopischen Struktur des Brotes beim Altbackenwerden. Hoppe-Seyler1s Zeitschrift fu K r Physiologische Chemie, 95, 130 } 135 (1915) 14 LEE, R. M. K. W. A critical appraisal of the e!ects of "xation, dehydration and embedding on cell volume. In: REVEL, J.-P., BARNARD, T. AND HAGGIS, G. H. (Eds), Proceedings of the 2nd Pfe+erkorn conference. Chicago: SEM Inc., pp. 61 } 70 (1983) 15 International Association for Cereal Science and Technology. ICC-Standards, No. 115/1, Scha K fer, Detmold (1992) 16 FLINT, O. Food microscopy: a manual of practical methods, using optical microscopy. Oxford: BIOS Scienti"c Publishers Ltd, pp. 17 (1994) 17 BARTHEL, L. K. AND RAYMOND, P. A. Improved method for obtaining 3- m cryosections for immunocytochemistry. he Journal of Histochemistry and Cytochemistry, 38, 1383 } 1388 (1990) 18 CUNIN, C., HANDSCHIN, S., WALTHER, P. AND ESCHER, F. Structural changes of starch during cooking of durum wheat pasta. ebensmittel-=issenschaft und-echnologie, 28, 323 }328 (1995) 19 KIEFFER, R. AND STEIN, N. Investigations on the phenomenon of dough hardening during the processing of wheat doughs. In: WINDHAB, E. J. AND WOLF, B. (Eds), Proceedings of the 1st International Symposium on Food Rheology and Structure. Hannover: Vincentz Verlag, pp. 292 } 296 (1997) 20 BLOKSMA, A. H. Dough structure, dough rheology, and baking quality. Cereal Foods =orld, 35, 237 } 244 (1990) 21 KALICHEVSKY, M. T. AND RING, S. G. Incompatibility of amylose and amylopectin in aqueous solution. Carbohydrate Research, 162, 323 } 328 (1987)

Acknowledgements Thanks are due to Claudia Meyer (Department of Anatomy, Prof. M. Mu K ntener, University of Zurich) for her technical assistance, and to Chantal Bussmann (Institute of Food Science, ETH Zurich) for preparing micrographs.

259

lwt/vol. 00 (1999) No. 5

22 LANGTON, M. AND HERMANSSON, A.-M. Microstructural changes in wheat starch dispersion during heating and cooling. Food Microstructure, 8, 29 } 39 (1989) 23 GALLANT, D. J., BOUCHET, B. AND BALDWIN, P. M. Microscopy of starch: evidence of a new level of granule orgnization. Carbohydrate Polymers, 32, 177 } 191 (1997) 24 SANDSTEDT, R. M. The function of starch in the baking of bread. he Bakers Digest, 35, 36 } 44 (1961) 25 MORGAN, K. R., HUTT, L., GERRARD, J., EVERY, D., ROSS, M. AND GILPIN, M. Staling in starch breads: the e!ect of antistaling alpha-amylase. Starch/Sta K rke, 49, 54 } 59 (1997) 26 JACOBSON, M. R., OBANNI M. AND BEMILLER, J. N. Retrogradation of starches from di!erent botanical sources. Cereal Chemistry, 74, 511 } 518 (1997)

27 MILES, M. J., MORRIS, V. J. AND RING, S. G. Gelation of amylose. Carbohydrate Research, 135, 257 } 269 (1985) 28 MILES, M, J., MORRIS, V. J., ORFORD, P. D. AND RING, S. G. The roles of amylose and amylopectin in the gelation and retrogradation of starch. Carbohydrate Research, 135, 271 } 281 (1985) 29 CLARK, A. H., GIDLEY, M. J., RICHARDSON, R. K. AND ROSS-MURPHY, S. B. Rheological studies of aqueous amylose gels: the e!ect of chain length and concentration on gel modulus. Macromolecules, 22, 346 } 351 (1989) 30 CONDE-PETIT, B. Interaktionen von Sta K rke mit Emulgatoren in wasserhaltigen Lebensmittel-Modellen, Ph.D. Thesis No. 9785, Swiss Federal Institue of Technology Zurich (1992)

260