Exploring the surface morphology of developing wheat starch granules by using Atomic Force Microscopy

RESEARCH ARTICLE

Exploring the surface morphology of developing wheatstarch granules by using Atomic Force Microscopy

Renuka N. Waduge1, Song Xu2, Eric Bertoft1 and Koushik Seetharaman1

1 Department of Food Science, University of Guelph, Guelph, ON, N1G 2W1, Canada2 Application Scientist, Agilent Technologies, Detroit, MI, USA

Wheat starch granules from 7, 14, 21, 28, 35, 42, and 49 days after anthesis (DAA)

were imaged using AFM. Starches were scanned in their native form as well as following

exposure to iodine vapour in-situ at 100% humidity. Starch at 7 DAA exhibited only one

granule population, while starches at other maturities had two size populations. Starches

from all stages of maturation exhibited blocklet structures. Larger fuzzy blocklets were

observed at early stages while they became smaller and less fuzzy at subsequent stages

of maturity. Furthermore, at all stages, small granules had larger fuzzy blocklets and

higher surface roughness than the large granule counterpart; however the surface rough-

ness of both small and large granule fractions decreased gradually with maturation.

Amorphous growth ring of starch granules with �1 � 1 mm to �3 � 6 mm size islands

were also observed. In-situ iodine exposure demonstrated that iodine interacted with

glucan polymers in the amorphous background first and then spread over to polymers

on top of blocklets. Iodine-glucan polymer interaction further increased the surface

roughness.

Received: August 12, 2012

Revised: September 13, 2012

Accepted: September 19, 2012

Keywords:

Atomic force microscopy / Developing starch granules / Granule architecture / Iodine-starch interaction / Wheat starch

1 Introduction

Atomic force microscopy (AFM) is a powerful technique

which allows surface imaging of non- conducting samples

in nanometer scale. In this technique, samples are imaged

under ambient conditions, either in air or under liquids, and

with minimal sample preparations [1]. Since AFM does not

rely on conductive sample preparations as in scanning

electron microscopy (SEM), it is widely used to study

the surface [2–4], internal structure of starch granules

[5–7], as well as starch polymers [8–9]. Moreover, the

AFM imaging technique has also been utilized to investi-

gate enzyme hydrolysis [10–11], freezing and thawing

effects [12–13], retrogradation [14], and heating effects

[15] of starches.

Starch is deposited in higher plants as semi-crystalline

granules and comprises two homoglucans called amylose

(AM) and amylopectin (AP). In the starch granule, AM and

AP molecules are organized in an array of alternating

amorphous and semi-crystalline growth rings. According

to the model described by Gallant and coworkers [16],

both amorphous and semi-crystalline rings are composed

of blocklets; however, blocklets in the amorphous rings

are smaller than those in semi-crystalline rings. These

blocklets are thought to be composed of alternating

crystalline and amorphous lamellae. AP forms double

helices and occupies the crystalline lamellae of the

granule, while AP branch points and AM are located in

the amorphous lamellae of the granule. However, Tang

and coworkers [17] suggested some modifications to

the original blocklet model and proposed that the semi-

Colour online: See the article online to view Figs. 2–8 in colour.

Correspondence: Koushik Seetharaman, Department of FoodScience, University of Guelph, Guelph, ON, N1G 2W1, CanadaE-mail: [email protected]: þ1-519-8246631

Abbreviation: DAA, days after anthesis

DOI 10.1002/star.201200172398 Starch/Starke 2013, 65, 398–409

� 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com

crystalline growth rings are composed of normal blocklets

while the amorphous growth rings are composed of

‘‘defective’’ blocklets. They further stated that the blocklets

could be fused together as well. Accordingly, depending

on the type of blocklets present, growth rings can be

divided into two different types - homogeneous and

heterogeneous.

In mature wheat starch, there are two distinct

populations of large and small granules [18]. Large

granules are mostly disc-shaped, while small granules

are spherical [18]. These two granule populations are

initiated in the endosperm of the wheat grain during

different stages of grain development: large granules

are initiated at about 4–7 DAA (days after anthesis),

whereas small granules appear at about 10–14 DAA

[19–21]. In addition, some researchers have observed

a third population of granules in the wheat endosperm

which is initiated at around 21 DAA [22]. Therefore,

the small granule population observed at the final maturity

is probably the combination of granules initiated at

about 10–14 DAA and those initiated at 21 DAA. At final

maturity, the large granule population have average

diameter of 10 to 35 mm, while the small granules range

from about 1 to 10 mm.

Iodine forms V-type inclusion complexes with linear

polymers or linear segments of branched polymers

of starch. This has been extensively studied in aqueous

suspensions and widely used in AM content determi-

nation of starches [23–30]. Recently, we demonstrated

that the linear polymer-iodine vapour interaction can

be utilized to elucidate the three dimensional structure

of starch granules. The extent of interaction that

takes place when granular starch is exposed to iodine

vapour depends essentially on the molecular organiz-

ation of the granule, i.e., starch granule architecture

[4, 31–34].

While several researchers have studied the surface of

mature starch granules by using AFM [35–38], there is no

information related to the microstructure of granules at

different stages of kernel development. Potentially, the

visualization of the microstructure of starch granules from

immature kernels is a way to explore the interior of starch

granules, since granules grow by apposition [39–40].

Therefore, the objective of this research was to visualize

the surface of the wheat starch granules during different

stages of development using AFM, and consequently

to understand the architectural development during

maturation. A real time AFM scanning protocol was

also carried out in the presence of iodine vapor in a

humid environment to explore the molecular organization

of the starch granules. Even though it is recognized

that AFM only visualizes a selected region of a granule

surface, and it may not be representative of all granules

in a seed, much less the granules in a panicle of wheat;

in this paper we analyzed several granules in a population

and observed differences within and between granule

surfaces at different stages of maturity. These obser-

vations contribute significantly to the understanding

of the evolving surface morphology and architectural

organization of starch granules.

2 Material and methods

2.1 Materials

2.1.1 Sampling Protocol

Hard red spring wheat (cv. Hobson) at seven different

stages of kernel developments – 7, 14, 21, 28, 35, 42,

and 49 DAA – were harvested in Ailsa Craig, Ontario,

Canada in 2009. Heads were tagged when 50% of

the spikelets within the head were pollinated. Grains were

sampled and stored at �208C to prevent any enzyme

activity. Seeds were cleaned to get rid of debris and stored

again at �208C until the starch was isolated. Care

was taken to minimize the time that the sample stayed

at room temperature during cleaning. Starch was then

isolated according to the method described in a previous

publication [33].

2.2 Atomic Force Microscopy (AFM)

A model 5500 AFM (Agilent Technologies, Chandler,

AZ, USA) was used for the experiment. Imaging was

performed in the tapping mode with silicon cantilevers

with nominal spring constant of 48 N/m and resonance

frequency of around 300 kHz (Nanoworld AG,

Switzerland). Scanning rates and resolution were

1�2 Hz and 512 pixels per line, respectively. The obser-

vation was performed inside a chamber at room tempera-

ture. Starch granules were sprinkled on a double-sided

tape attached to a microscope slide. Then the microscope

slide was placed on the sample holder and the exposed

surface of starch granules were scanned directly without

any preliminary preparations. The real time scanning was

performed as explained by Park et al. [4], but with a

modified experimental set up (Fig. 1). The surface of the

starch granules was scanned after introducing humidity

first and then again after exposing to iodine vapour in the

presence of humidity. At least three granules each of large

or small granule fraction at each maturity were scanned

and granules larger than 10 mm were considered as large

granules. The dimensions of blocklets and the surface

roughness were determined using PicoView (version

1.4; Agilent Technologies, Chandler, AZ, USA) and the

Gwyddion (version 2.24; http://gwyddion.net) software,

respectively. Surface roughness was measured by using

Starch/Starke 2013, 65, 398–409 399


Ra, the arithmetic average of absolute values, and is

defined by:

Ra ¼ 1

n

Xn

i�1

yij j

Where, yi is the height of the surface at ith location n is

the number of data points in the image, and using hmax,

which is the maximum height difference of the area

scanned.

3 Results

3.1 Wheat starch granules at different stages ofmaturity

3.1.1 Starch granules at 7 DAA

Topographic image and the cross section analysis of 7

DAA starch granule are shown in Fig. 2. As seen in the

topographic image (Fig. 2a)�, blocklets are clearly visible

already at this stage of maturity. The top-view of these oval

shaped blocklets had dimensions of �68 � 97 nm by

�98 � 154 nm, and appeared fuzzy. i.e., the blocklets

did not have a clear definition. These blocklets were clus-

tered and/or fused together with different heights, forming

nodules (encircled in Fig. 2a). Depressions were also

observed on the surface of the granule.

The phase image in AFM highlights the stiffness of the

sample surface. It records the phase lagging between the

cantilever oscillation and the phase of driving signal giving

an indication of different degrees of stiffness at different

locations of the surface of the sample [4]. They can provide

textural information of surfaces inside depressions which

are not visualized in the topographic image. The surface of

the 7 DAA starch granule was mostly stiff (white coloring

on the scale in the phase image), although there were

randomly distributed areas with lower stiffness (dark color-

ing) (Fig. 2b). An additional feature of the top of the block-

lets and nodules was stiff with a grainy texture (like the

Figure 1. The schematic representation of the experi-mental set up for in-situ AFM scanning.

Figure 2. Topographic (a) and phase(b) images and their cross section ana-lysis (c) of a 7 DAA starch granule.Magnification is 1 � 1 mm. The loca-tion of cross section analysis is indi-cated by the white line in figures (a)and (b). The top graph in (c) is the crosssection analysis of the topography, andthe bottom graph is that of the phaseimage. Scale of the Y-axis of the crosssection analysis of topograhy is in mmwhereas that in phase image is indegrees. Circles indicate nodule (N intop graph) and blocklet (B). The arrowin (c) shows the grainy texture on top ofa nodule.

400 R. N. Waduge et al. Starch/Starke 2013, 65, 398–409


surface of a sand-paper) as visualized in the cross sections

analysis of the phase image (Fig. 2c bottom shown by an

arrow). The value of Ra and hmax for 7 DAA starch was

�7.5 nm and �67 nm, respectively.


Topographic images of large and small granules from

14 DAA starch and their cross section analysis are

shown in Fig. 3. Blocklets were visible on the surface of

both large (Fig. 3a) and small (Fig. 3b) granules. The size

of blocklets in small granules was larger (�32 � 36 nm

to �49 � 82 nm) and were fuzzier compared to that in

large granules (�20 � 25 nm to �45 � 55 nm). However,

blocklets in both small and large granules of 14 DAA

starches were less fuzzy than those in 7 DAA starch granule.

Depressions and gaps (sharp changes in topography as

seen in the cross section analysis; shown by arrows in

Fig. 3e) were also observed on the surface of both large

and small granules (shown by arrows in Fig. 3a).

According to phase images, the stiffer top of blockets

and softer valleys were observed (Fig. 3c and d) like in 7

DAA sample. The stiffer areas were larger in small gran-

ules compared to those in large granules. However, these

structures in 7 DAA starch granules were much larger than

those in both large and small granules of 14 DAA starch.

The Ra of small and large granules were �7.3 nm and

� 3.4 nm, respectively, and the hmax of small and large

granules were �40 nm and � 65 nm, respectively.


At 28 DAA, wheat kernels are close to their physiological

maturity. Fig. 4 shows the topographic images of large

(Fig. 4a) and small (Fig. 4b) granules, and their phase images

(Fig. 4c and d) and cross sectional analysis (Fig. 4e and f).

Similar to starches at earlier maturity, depressions, blocklets,

and nodules were visible. Blocklets in small granules were

larger (�25 � 42 nm to �45 � 84 nm) and fuzzier than

those in large granules (�20 � 36 nm to �25 � 51 nm),

and had similar trends as was observed at 14 DAA.

Phase images demonstrated the presence of stiffer

structures in both large and small granules, but, again,

the stiffer areas were larger in size in small granules

Figure 3. Topography and phaseimages of 14 DAA large (a and c,respectively) and small (b and d,respectively) starch granules, and theircross section analysis (e and f, respec-tively). Magnification is 1 � 1 mm. Thelocation of cross section analysis areindicated by white lines in (a) to (d).Top graphs in (e) and (f) are the crosssection analysis of the topography oflarge and small granules, respectively,while the bottom graphs are that of thephase images, respectively. Scale ofthe Y-axis of the cross section analysisof topograhy is in mm, whereas that inphase image is in degrees. The arrowsin (a) show a depression (D) and a gap(G) and the corresponding areas in thecross section analysis is shown in (e).

Starch/Starke 2013, 65, 398–409 401


compared to its large granule counterpart. The cross sec-

tion analysis of large and small granules (Fig. 4e and f top

graphs) supported the observation of fuzzier blocklets on

small granules compared to large granules. The Ra of

small and large granules was � 2.5 nm and � 2.4 nm,

respectively, and hmax of small and large granules was

�42 nm and �32 nm, respectively.


The surface features of granules from 35 and 42 DAA, both

at post-physiological maturity, had similar trends as that

seen for granules at 49 DAA; and therefore the figures are

not shown here. Fig. 5 shows AFM images at 49 DAA.

Similar observations as at previous stages of maturity

were observed, i.e., depressions on the surface of both

large and small granules, larger and fuzzier blocklets in

small granules compared to those in large granules, and

formation of nodules by clustering blocklets together.

However, the difference in blocklet sizes between large

and small granules at this stage of maturity was less than

that at younger maturities. The size of blocklets in large

granules were�20 � 40 nm to�26 � 60 nm and in small

granules�21 � 38 nm to�36 � 56 nm. The Ra and hmax

values were�2.91 nm and�33 nm for large granules and

�3.5 nm and �42 nm for small granules, respectively.

Phase images also demonstrated the stiffer areas like with

granules at other maturities.

3.2 Other observations in surface morphology

Some large starch granules from 21, 35, 42 and 49 DAA

exhibited unique attributes compared to the general

observations reported above. Fig. 6 shows the topological

images of the surface of two large granules from 35 DAA

starch; one with the typical blocklets and stiffer surface,

like those described above (Fig. 6a–c) and the other

with different surface attributes (Fig. 6d–f). The other

surface was softer (Fig. 6h bottom graph) and did not show

blocklets even at 500 � 500 nm magnification (Fig. 6f).

Instead, island-like structures of different shapes and sizes

(�1 � 1 mm to �3 � 6 mm) embedded in a softer back-

Figure 4. Topography and phaseimages of 28 DAA large (a and c,respectively) and small (b and d,respectively) starch granules, and theircross section analysis (e and f, respec-tively). Magnification is 1 � 1 mm. Thelocation of cross section analysis areindicated by white lines in (a) to (d).Top graphs in (e) and (f) are the crosssection analysis of the topography oflarge and small granules, respectively,while the bottom graphs are that ofthe phase images, respectively. Scaleof the Y-axis of the cross section analy-sis of topograh is in mm, whereas that ofphase images is in degrees.



ground material (Fig. 6d) containing randomly distributed

stiff depressions (Fig. 6e, f, and h) were seen. Although,

those islands were stiffer than the background, they were

softer than depressions. Furthermore, depressions in

the background are much stiffer than rest of the area in

the granule and around 5–7 nm deep (Fig. 6h). There were

no depressions seen on islands. The hmax of this surface

was comparatively small.

3.3 Effect of iodine on developing wheat starchgranules

Iodine-starch linear polymer interaction in an intact

starch granule under humid condition was investigated

by exposing starch granules from 7, 28, and 49 DAA to

iodine vapor in-situ and images are shown in Fig. 7. The

control and changes following exposure to iodine are from

the same scan area. In all three stages of maturity, block-

lets swelled and merged when the granules were humidi-

fied, as has been reported earlier for corn and potato

starches [4]. In some areas, swelling was not large enough

to diminish the gaps between blocklets or nodules, while

in other locations the gaps disappeared. When these

starches were exposed to iodine vapor, they interacted

with iodine to different extents. Different level of inter-

actions of starch with iodine at different stages of maturity

was clearly visible even to the naked eye because different

shades of color were observed.

Fig. 7(a–d) shows the same location of the 7 DAA starch

granule before (Fig. 7a and c) and after (Fig. 7b and d)

iodine exposure. However, no significant differences

were observed at this stage of maturity, except for an

increase in hmax in the topographic image and

some changes in the stiffness as observed in the phase

image.

At 28 DAA, in the topographic image, the blocklets in

valleys of granule surface which were less detectable

before iodine exposure became more visible (arrows in

Fig. 7e and f). The hmax of the overall scanned area was

increased following iodine exposure. New spike-like pro-

trusions appeared in the phase image following exposure

to iodine (Fig. 7g–h) because of the complexation of glucan

Figure 5. Topography and phaseimages of 49 DAA large (a and c,respectively) and small (b and d,respectively) starch granules, and theircross section analysis (e and f, respec-tively). Magnification is 1 � 1 mm. Thelocation of cross section analysis areindicated by white lines in (a) to (d).Top graphs in (e) and (f) are the crosssection analysis of the topography oflarge and small granules, respectively,while the bottom graphs are that ofthe phase images, respectively. Scaleof the Y-axis of the cross section analy-sis of topograh is in mm, whereas thatin phase image is in degrees.

Starch/Starke 2013, 65, 398–409 403


polymers with iodine. These spike-like structures were

initiated in gaps between blocklets/nodules (Fig. 8d

and h) and then started to appear, about 15 min later,

on top of the blocklets in repetitive scans (Fig. 8f and i).

The behaviour of 49 DAA starch following exposure to

iodine was similar to that of 28 DAA starch, although iodine

complexation was not that pronounced (Fig. 7i–l).

However, the appearance of randomly distributed �2 nm

long spike-like structures on gaps between grain-like

structures together with increased hmax in the topology

and the reduced degree in the phase image is indicative

of complex formation.

4 Discussion

It is known that granules grow by apposition. Therefore, in

this study, we are conceptually interpreting the data as

surface features of the developing granule as they are laid

down during biosynthesis. In wheat starch, large granules

Figure 6. Topographic images of a semi-crystalline surface (a to c) and an amorphous surface (d to f) of two different 35 DAAstarch granules and their cross section analysis (g and h, respectively). Same location of each granule was scanned as5 � 5 mm surface area (a and d) and then magnified into 1 � 1 mm (b and e) and 500 � 500 nm (c and f). The locations ofcross section analysis are indicated by white lines in figures (b) and (e). Top graphs in (e) and (f) are the cross section analysisof the topography of semicrystalline and amorphous surfaces, respectively, while the bottom graphs are that of the phaseimages. Scale of the Y-axis of the cross section analysis of topograh is in mm whereas that in phase image is in degrees.BG – background and Is – islands.



initiate at around 4–7 DAA, whereas small granules initiate

at around 12–14 DAA [19–21]. Therefore, it is reasonable

to assume that the starch granule population at 7 DAA

are younger large granules, and the small granule fraction

at 14 DAA is younger small granules. Since there are no

clear boundaries for separating immature large and small

granule populations by their size, i.e., the small granule

fraction at early stages might have a smaller proportion of

large granules and vice versa, to minimize the error in this

study, granules around 4–7 mm were considered as small

Figure 7. AFM images of before andafter in situ exposure of 7 (a to d), 28 (eto h), and 49 (i to l) DAA starch granulesto iodine vapor at 100% humidity. (a), (e)and (i): Topographic image of the starchgranule in the presence of 100% humidityonly. (b), (f) and (j): Topographic image ofstarch granule in the presence of 100%humidity and iodine vapor. (c), (g) and (k):Phase image of starch granules in thepresence of 100% humidity only. (d),(h) and (l): Phase image of starch gran-ules in the presence of 100% humidityand iodine vapor. Magnification is500 � 500 nm. Arrows in (e) and (f)show the appearance of blocklets in val-leys because of the exposure to iodinevapor.

Starch/Starke 2013, 65, 398–409 405


granules and those bigger than 15 mm were considered as

large granules from 14 DAA onwards.

4.1 Surface of wheat starch granules at differentstages of maturity

The fact that blocklets were observed as early as 7 DAA

shows that blocklets are early features in granule organ-

ization, where the starch molecules appear to have the

propensity to organize. However, these blocklets were

larger and fuzzier than those at later stages. This obser-

vation holds true for the small granules as well (14 DAA

small granule vs 49 DAA small granule). In addition, these

larger and fuzzier blocklets at younger stages and in small

granules had organized into different heights making the

surface of the starch granule rough. The surface of imma-

ture granules is representative of a boundary between two

layers that develop over two consecutive days, i.e., by

apposition [39]. Higher surface roughness at younger

stages suggests the initiation of the next layer at different

heights, which results in a less clear separation between

layers at the center of the granule. This would explain

microscopic images of the cross section of starch granules

with unclear growth ring separations [39]. Furthermore, the

Figure 8. Repetitive scans of the in situ iodine exposure of 28 DAA starch granules in the presence of 100% humiditydemonstrating the appearance of spikes in gaps first and then spreading over to top of blocklets. (a) and (b): Topographicand phase images of the granule surface in the presence of 100% humidity only. (c) and (d): Topographic and phase imagesof the first scan of the surface in the presence of iodine vapor and 100% humidity. (e) and (f): Topographic and phase images ofthe second scan of the surface in the presence of iodine vapor and 100% humidity. (g), (h) and (i): Cross section analysisof the topography and phase images of the granule surface before exposing to iodine, first scan after exposing to iodine,and the second scan after exposing to iodine, respectively. The locations of cross section analysis are indicated by white lines in (a)to (f). Top graphs in (g) to (i) are the cross section analysis of topography, while the bottom graphs are of the phase images. Scaleof the Y-axis of the cross section analysis of topography is in mm, whereas that in phase image is in degrees. Arrows in (h) showsthe formation of the complex in gaps, whereas those in (i) shows formation of the complex on top of blocklets.



differences between large and small granules suggest that

the biosynthesis of small granules lags behind the large

granule throughout the development.

Phase image of an AFM scanning gives information

about the stiffness of the scanned surface. It further

visualizes the structure of surface depressions, which

were hidden in the topography. As such, there were micro-

structures, possibly blocklets, which have similar texture

as other areas in those valleys. Phase images of starch

granules further demonstrated the presence of deep gaps

dividing bundles of nodules/blocklets from each other.

4.2 Amorphous growth ring structure

At several stages of maturity, large starch granules pos-

sessed different surface properties (Fig. 6) in contrast

to the surface with blocklets commonly observed and

reported in literature [3, 35, 41]. The fuzzy, soft, non-block-

let structure seen here probably represents the amorphous

growth ring. The structure of the amorphous growth ring

is still in dispute. Although assumed as non-structured,

Gallant et al. [16] argued that it was also composed of

blocklets and Tang et al. [17] suggested that these block-

lets are structurally defect. Recent data [42] showed the

presence of amorphous blocklets (�80 nm) in potato and

corn lintners. The presence of a well-defined conformation

in amorphous component is also previously reported [43].

However, the presence of large islands with a compara-

tively smooth surface embedded in a softer background in

immature wheat starch suggests a microstructure larger

than and different from the blocklets previously observed.

It is possible that these islands consist of coalesced glucan

polymers giving rise to a stiffer texture than the surround-

ings. Furthermore, the presence of � 5–7 nm deep

depressions with higher stiffness similar to regular granule

surface in the background of the amorphous surface

suggests that these depressions likely extend to the under-

neath semi-crystalline growth ring. These depressions

might also be a part of internal channels.

4.3 Visualization of the interaction of glucanpolymers with iodine vapor

The real time scanning, which was carried out while expos-

ing the surface of 7, 28, and 49 DAA starch granules to

iodine vapor at humid conditions, demonstrated different

levels of interactions with iodine at different maturations.

Hardly any effect of iodine was seen in 7 DAA starch,

except an increase in surface roughness. However, the

28 DAA starch clearly visualized the location of iodine

interaction on the surface of the starch granule. The com-

plexation of iodine with glucan polymer segments in gaps

between blocklets suggests the presence of complexes

about 2 nm long in these gaps. These appear to have a

higher affinity to iodine than those on top of blocklets. In a

starch granule, blocklets are embedded in an amorphous

matrix, which contains non-crystalline glucan polymers of

starch, i.e., mainly amylose and non-crystalline amylopec-

tin. Therefore, the gaps between blocklets, which possibly

are occupied by the non-crystalline material, have linear

glucan polymers that can bind with iodine faster. However,

molecules on the top of blocklets apparently also have

linear segments long enough to form inclusion complexes

with iodine thus, by time, extending the area that form

complexes with iodine. Similar observations were seen

also in the 49 DAA starch, although to a lesser degree.

In-situ iodine exposure of potato and corn starch was

reported by Park et al. [4]. They found that, the location of

iodine binding in potato starch was on the top of the block-

let while that of corn starch was both on top and around the

blocklet. Hence, the iodine binding of wheat starch was

similar to corn; however it was initiated within gaps and

then spread over to the top of blocklets suggesting a

variation in the location of glucan polymers with different

affinity for iodine within the starch granule.

Because of the complex formation with iodine, the other-

wise hidden microstructures in valley areas became visible.

This would happen if soft and flexible glucan polymers,

which were laid down on the surface of the starch granule

at the native state, were transformed into rigid molecules

and stood up as a result of complex formation. Furthermore,

depending on the lengths of the glucan chains that inter-

acted with iodine, the surface will obtain a new pattern as

observed; as this change would be just few nanometers, it

will not affect the blocklet and nodule structures.

The increased surface roughness of starch granules by

the iodine-glucan polymer interaction is likely to be due to

the transformation of glucan polymers of different lengths

and heights into rigid structure with variable heights.

In conclusion, imaging of large and small wheat starch

granules at different stages of maturation showed that

blocklets at younger stages, which are assumed to be at

the interior of the starch granule at harvest, are larger and

fuzzier than those at the periphery. In addition, small starch

granules had both higher surface roughness and larger and

fuzzier blocklets than their large granule counterpart at all

stages of maturation. Surfaces with different properties,

presumably amorphous growth rings, were observed in

starch granules from several maturities. Interaction of gran-

ular wheat starch with iodine vapor was initiated within the

amorphous matrix and then spread over to top of blocklets.

Authors would like to thank Mark Etienne from Dow

Agrosciences-Narin Research Lab, Ailsa Craig, Ontario

for providing seeds and to Dr. Hyuksang Park for useful

discussions.

The authors have declared no conflict of interest.

Starch/Starke 2013, 65, 398–409 407


5 References

[1] Perez, S., Baldwin, P. M., Gallant, D. J., in: Starch: Chemistryand Technology (Eds., J. BeMiller, R. Whistler), AcademicPress, USA 2009, pp. 149–192.

[2] Baldwin, P. M., Adler, J., Davies, M. C., Melia, C. D., Highresolution imaging of starch granule surfaces by atomic forcemicroscopy. J. Cereal Sci. 1998, 27, 255–265.

[3] Szymonska, J., Krok, F., Potato starch granule nanostructurestudied by high resolution non-contact AFM. Intl. J. Biol.Macromol. 2003, 33, 1–7.

[4] Park, H., Xu, S., Seetharaman, K., A novel in situ atomic forcemicroscopy imaging technique to probe surface morphologi-cal features of starch granules. Carbohydr. Res. 2011, 346,847–853.

[5] Baker, A. A., Miles, M. J., Helbert, W., Internal structure of thestarch granule revealed by AFM. Carbohydr. Res. 2001, 330,249–256.

[6] Ridout, M. J., Gunning, A. P., Parker, M. L., Wilson, R. H.,Morris, V. J., Using AFM to image the internal structure ofstarch granules. Carbohydr. Polym. 2002, 50, 123–132.

[7] Parker, M. L., Kirby, A. R., Morris, V. J., In-situ imaging of peastarch in seeds. Food Biophys. 2008, 3, 66–76.

[8] Dang, J. M. C., Braet, F., Copeland, L., Nanostructuralanalysis of starch components by atomic force microscopy.J. Microsc. 2006, 224, 181–186.

[9] Gunning, A. P., Giardina, T. P., Faulds, C. B., Juge, N., Ring,S. G., Williamson, G., Morris, V. J., Surfactant-mediatedsolubilisation of amylose and visualisation by atomic forcemicroscopy. Carbohydr. Polym. 2003, 51, 177–182.

[10] Morris, V. J., Gunning, A. P., Faulds, C. B., Williamson, G.,Svensson, B., AFM images of complexes between amyloseand Aspergillus niger glucoamylase mutants, native, andmutant starch binding domains: A model for the action ofglucoamylase. Starch/Starke 2005, 57, 1–7.

[11] Thomson, N. H., Miles, M. J., Ring, S. G., Shewry, P. R.,Tatham, A. S., Real-time imaging of enzymatic degradationof starch granules by atomic-force microscopy. J. Vac. Sci.Technol. 1994, B 12, 1565–1568.

[12] Krok, F., Szymonska, J., Tomasik, P., Szymonski, M., Non-contact AFM investigation of influence of freezing process onthe surface structure of potato starch granule. Appl. SurfaceSci. 2000, 157, 382–386.

[13] Szymonska, J., Krok, F., Tomasik, P., Deep freezing of potatostarch. Intl. J. Biol. Macromol. 2000, 27, 307–314.

[14] Tang, C. M., Copeland, L., Investigation of starch retrograda-tion using atomic force microscopy. Carbohydr. Polym. 2007,70, 1–7.

[15] An, H., Yang, H., Liu, Z., Zhang, Z., Effects of heating modesand sources on nanostructure of gelatinized starch mol-ecules using atomic force microscopy. LWT-Food Sci.Technol. 2008, 41, 1466–1471.

[16] Gallant, D. J., Bouchet, B., Baldwin, P. M., Microscopy ofstarch: Evidence of a new level of granule organization.Carbohydr. Polym. 1997, 32, 177–191.

[17] Tang, H., Mitsunaga, T., Kawamura, Y., Molecular arrange-ment in blocklets and starch granule architecture. Carbohydr.Polym. 2006, 63, 555–560.

[18] Peng, M., Gao, M., Abdel-Aal, E.-SM, Hucl, P., Chibbar, R. N.,Separation and characterization of A- and B-type starchgranules in wheat endosperm. Cereal Chem. 1999, 76,375–379.

[19] Bechtel, D. B., Wilson, J. D., Amyloplast formation and starchgranule development in hard red winter wheat. Cereal Chem.2003, 80, 175–183.

[20] Bechtel, D. B., Zayas, I., Kaleikau, L., Pomeranz, Y., Size-Distribution of wheat-starch granules during endospermdevelopment. Cereal Chem. 1990, 67, 59–63.

[21] Langeveld, S. M. J., Van Wijk, R., Stuurman, N., Kijne, J. W.,de Pater, S., B-type granule containing protrusions and inter-connections between amyloplasts in developing wheat endo-sperm revealed by transmission electron microscopy andGFP expression. J. Exp. Bot. 2000, 51, 1357–1361.

[22] Bechtel, D. B., Wilson, J. D., Variability in a starch isolationmethod and automated digital image analysis system usedfor the study of starch size distributions in wheat flour. CerealChem. 2000, 77, 401–405.

[23] Bates, L., French, D., Rundle, R. E., Amylose and amylopec-tin content of starches determined by their iodine complexformation. J. Am. Chem. Soci. 1943, 65, 142–148.

[24] Knutson, C. A., Jr., Cluskey, J. E., Dintzis, F. R., Properties ofamylose-iodine complexes prepared in the presence ofexcess iodine. Carbohydr. Res. 1982, 101, 117–128.

[25] Murdoch, K. A., The amylose-iodine complex. Carbohydr.Res. 1992, 233, 161–174.

[26] Nimz, O., GeOler, K., Uson, I., Laettig, S., Welfle, H., Sheldrick,G. M., Saenger, W., X-ray structure of the cyclomaltohexaico-saose triiodide inclusion complex provides a model for amylose-iodine at atomic resolution. Carbohydr. Res. 2003, 338, 977–986.

[27] Rundle, R. E., Baldwin, R. R., The configuration of starch andthe starch-iodine complex. The dichroism of flow of starch-iodine solutions. J. Am. Chem.Soci. 1943, 65, 554–558.

[28] Rundle, R. E., French, D., The configuration of starch and thestarch-iodine complex. III. X-ray diffraction studies of the starch-iodine complex. J. Am. Chem.Soci. 1943, 65, 1707–1710.

[29] Teitelbaum, R. C., Ruby, S. L., Marks, T. J., A resonanceRaman/iodine Mossbauer investigation of starch-iodinestructure. Aqueous solution and iodine vapour preparations.J. Am. Chem.Soci. 1980, 102, 3322–3328.

[30] Thoma, J. A., French, D., The starch-iodine-iodide inter-action. Part I. Spectrophotometric investigations. J. Am.Chem. Soci. 1960, 82, 4144–4147.

[31] Saibene, D., Seetharaman, K., Segmental mobility of poly-mers in starch granules at low moisture contents. Carbohydr.Polym. 2006, 64, 539–547.

[32] Saibene, D., Zobel, H. F., Thompson, D. B., Seetharaman, K.,Iodine-binding in granular starch: Different effects of moisturecontent for corn and potato starch. Starch/Starke 2008, 60,165–173.

[33] Waduge, R. N., Xu, S., Seetharaman, K., Iodine absorptionproperties and its effect on the crystallinity of developing wheatstarch granules. Carbohydr. Polym. 2010, 82, 786–794.

[34] Marti, A., Pagani, M. A., Seetharaman, K., Characterizingstarch structure in a gluten-free pasta by using iodine vapouras a tool. Starch/Starke 2011, 63, 241–244.

[35] Baldwin, P. M., Adler, J., Davies, M. C., Melia, C. D., Holes instarch granules: Confocal, SEM and light microscopy studiesof starch granule structure. Starch/Starke 1994, 46, 341–346.

[36] Juszcak, L., Surface of triticale starch granules-NC-AFMobservations. Elect. J. Polish Agri. Uni. 2003, 6, 8.

[37] Juszczak, L., Fortuna, T., Krok, F., Non-contact atomic forcemicroscopy of starch granules surface. Part I. Potato andtapioca starches. Starch/Starke 2003, 55, 1–7.



[38] Juszczak, L., Fortuna, T., Krok, F., Non-contact atomic forcemicroscopy of starch granules surface. Part II. Selectedcereal starches. Starch/Starke 2003, 55, 8–16.

[39] Buttrose, M. S., Submicroscopic development and structure ofstarch granules in cereal endosperm. J. Ultrastr. Res. 1960, 4,231–257.

[40] Mukerjea, R., Mukerjea, R., Robyt, J. F., Starch biosynthesis:experiments on how starch granules grow in vivo. Carbohydr.Res. 2009, 344, 67–73.

[41] Baldwin, P. M., Davies, M. C., Melia, C. D., Starch granule surfaceimaging using low-voltage scanning electron microscopy andatomic force microscopy. Intl. J. Biol. Macromol. 1997, 21,103–107.

[42] Chauhan, F., Seetharaman, K., On the organization of chains inamylopectin. Starch/Starke 2012, DOI: 10.1002/star.201200132.

[43] Gidley, M. J., Bociek, S. M., Molecular organization in starches:A 13C CP/MAS NMR study. J. Am. Chem. Soci. 1985, 107,7040–7044.

Starch/Starke 2013, 65, 398–409 409


Documents

Exploring the surface morphology of developing wheat starch granules by using Atomic Force Microscopy