International Journal of Biological Macromolecules 89 (2016) 305318

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

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

Impact of full range of amylose contents on the architecture of starch

granules*
Avi Goldstein a, , George Annor b , Jean-Luc Putaux c,d , Kim H. Hebelstrup e ,
Andreas Blennow f , Eric Bertoft a
a

Department of Food Science and Nutrition, University of Minnesota, USA

Department of Nutrition and Food Science, University of Ghana, Legon-Accra, Ghana
c
Universit Grenoble Alpes, Centre de Recherches sur les Macromolcules Vgtales (CERMAV), F-38000 Grenoble, France
d
CNRS, CERMAV, F-38000 Grenoble, France
e
Department of Molecular Biology and Genetics, Aarhus University, Denmark
f
Department of Plant and Environmental Sciences, University of Copenhagen, Denmark
b

a r t i c l e

i n f o

Article history:
Received 3 March 2016
Received in revised form 13 April 2016
Accepted 17 April 2016
Available online 22 April 2016
Keywords:
Starch
Acid hydrolysis
Amylose
Barley
Lintnerization
Crystalline polysaccharides

a b s t r a c t
The effects of amylose deposition on crystalline regions of barley starch granules were studied in granules
containing zero to 99.1% amylose using waxy (WBS, 0% amylose), normal (NBS, 18% amylose) and
amylose-only barley lines (AOS, 99.1% amylose). The effects were probed after hydrolysis of amorphous
regions of starch granules in dilute HCl generating lintners, which typically represent the crystalline
lamella of starch granules. Compared to NBS and WBS, AOS granules exhibited an irregular, multilobular
morphology with a rough surface texture. AOS displayed lower rates of acid hydrolysis than WBS, and
AOS reached a plateau at 45 wt% acid hydrolysis. High-performance anion-exchange chromatography
of lintners at equivalent levels of hydrolysis (45 wt%) revealed the average degree of polymerization (DP)
of AOS lintners was 21, substantially smaller than that of NBS and WBS (DP 42). AOS lintners contained
the lowest number of chains (NC) per molecule (1.1) compared to NBS (2.8) and WBS (3.3) and the
average chain length of AOS, NBS and WBS lintners was 19, 15 and 13, respectively. Hence, both NC and
the average chain length correlated with amylose content. The size distribution prole of AOS lintners
revealed a repeat motif in the molecules corresponding to 56 glucose residues.
2016 Elsevier B.V. All rights reserved.

1. Introduction
Starch is the main storage polysaccharide in higher plants and
is a globally important food and industrial material. Starch is
cheap, renewable, biodegradable and composed of two main constituents, namely, amylose and amylopectin. Amylose is a mostly
linear macromolecule containing -(1,4)-linked d-glucosyl units,
whereas amylopectin is a branched macromolecule composed of (1,4)-linked d-glucosyl chains connected by -(1,6) branch points
[1]. Despite this rather simple chemical composition, the structure and architecture of native starch granules varies signicantly
between plant species due to differences in the genes that encode
starch biosynthetic enzymes, along with environmental factors [2].

This publication is dedicated to the memory of Dr. Koushik Seetharaman

(19662014).
Corresponding author.
E-mail address: golds333@umn.edu (A. Goldstein).
http://dx.doi.org/10.1016/j.ijbiomac.2016.04.053
0141-8130/ 2016 Elsevier B.V. All rights reserved.

Variability in starch structure drives numerous food and industrial

uses of native starch, however, native starches do have natural
limitations and modication can greatly enhance their desired
functionality [3].
It was previously believed that the amylopectin fraction of
starch was mandatory for the generation of starch granules [4],
because double helices in amylopectin were shown to associate into
crystallites of native starch granules [5]. However, it has recently
been shown that semicrystalline starch granules can be synthesized with only the amylose component through the silencing
of all known genes coding for starch branching enzymes (SBE I,
SBE IIa, SBE IIb) in barley using a single RNAi hairpin [6]. Just
like normal amylose [7], amylose-only starch (AOS) is slightly
branched and has a clustered chain structure [8] permitting higher
organized structures to be formed. However, AOS starch granules display an unorganized morphology [6]. Also, starch synthesis
was slightly inhibited in the endosperm of grains with silencing
of the SBE genes. When AOS was gelatinized, it contained very
high contents of resistant starch (65%) compared to the control

306

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

starch (29%) [6]. The high content of resistant starch is typically

explained by decreased accessibility of amylose chains to digestive enzymes due to retrogradation, i.e., the formation of stable
double-helical segments during cooling following gelatinization
[9]. AOS has great potential to be incorporated into food products
designed to have a low glycemic impact, coinciding with a slow and
sustained glucose release. Similarly, the grain endosperm starch
showed increased resistance towards degradation by endogenous amylolytic enzymes during germination and early seedling
establishment, which was retarded in amylose-only grains. This
indicates that amylopectin is necessary for an apparent structural
adaptation of starch for efcient remobilization of the energy and
biomass stored in the grain starch [10].
Semicrystalline AOS granules exhibit an allomorphic composition corresponding to a mixture of Vh - and B-types [6] as opposed
to the NBS and WBS granules that are of A-type [5,6]. However,
the crystalline and molecular structures of AOS have not yet been
fully studied. A common approach used to investigate the crystalline architecture of starch granules is based on the idea that the
crystalline regions are more resistant towards mild acid hydrolysis
than amorphous ones [11]. The insoluble residue after acid hydrolysis is referred to as lintners or Ngeli amylodextrins, depending on
the type of acid that is used: 2.2 N HCl or 15% H2 SO4 , respectively
[2]. The dextrins that compose the residue after acid hydrolysis
are typically either linear (average degree of polymerization (DP)
1317) or single-branched (average DP 2430) [12]. The singlebranched dextrins consist of chains of almost equal length [13]. In
addition, a small amount of multiple-branched dextrins may also
be present [14]. Investigating the kinetics of acid hydrolysis, as well
as the structure of the resulting insoluble residues provides insightful information pertaining to the molecular structure, architecture,
and dynamics of the crystalline components of starch in granules
[15].
In this study, the impact of amylose on starch granule morphology and crystalline and molecular structure was studied using
lintners prepared from WBS, NBS and AOS. The results provide
novel insights into how amylose, in the widest range of amylose
contents, affects hydrolytic stability of the starch granule in terms
of its crystalline sections. Plausible architectural explanations for
the effects are given.
2. Materials and methods
2.1. Barley
Waxy barley (cv. Cinnamon, kindly provided by Lantmnnen SW
Seed, Sweden), normal barley (cv. Golden Promise) and amyloseonly [6] barley were cultivated in a greenhouse with natural
sunlight and articial supplementing light using mercury lamps
from 4 a.m.8 p.m. at the University of Copenhagen, Denmark.
2.2. Starch extraction and composition
Starch was extracted from ground barley our according to
the method of Carcio et al. [16] with modications. Briey, 5 g
of milled barley was mixed with 25 mL of 5 mM dithiothreitol
containing 1% sodium dodecyl sulphate for 30 min at room temperature, and subsequently centrifuged at 3300g for 15 min. The
pellet was then washed twice with water and ltered through a
70 m mesh cloth. The ltrate was then centrifuged and 50 mL of
20 mM NaH2 PO4 buffer (pH 6.5) was added. The mixture was incubated in a 50 C water bath for 5 min before the addition of 100 L
lichenase (to remove trace amounts of cell wall -glucan), after
which the sample was incubated for 1 h, with stirring every 15 min.
After centrifugation (3300g for 15 min) the pellet was washed twice

with distilled water, once with ethanol, followed by air-drying

overnight. The content of amylose and amylopectin in normal and
waxy barley starch were determined with chromatography following debranching with isoamylase and pullulanase. The debranched
sample was applied on a column (1.6 90 cm) of Sepharose CL 6B
(GE Healthcare, Uppsala, Sweden) and eluted with 0.5 M NaOH at
1 mL min1 . Fractions (1 mL) were analysed for carbohydrate content according to Dubois et al. [17]. By dividing the chromatograms
at the lowest points between the two peaks, the relative content of
amylose and amylopectin were determined according to Sargeant
[18].
2.3. Lintnerization
Linterization was conducted according to two methods. In
method 1, 200 mg starch granules were suspended in 8 mL 2.2 N
HCl and incubated at 35 C or 40 C according to Robin [19]. Starch
suspensions were gently mixed daily. The rate of acid hydrolysis
was determined by taking small aliquots (15 L) at regular intervals, diluting to 1.5 mL with distilled water, centrifuging at 2000g,
and analyzing the supernatant for carbohydrate content using the
phenol-sulphuric acid reagent [17]. For structural analysis, large
aliquots of sample prepared at 40 C were taken at 45% hydrolysis
(via determination of solubilized carbohydrate content), neutralized with 0.1 M NaOAc, washed twice with distilled water, and the
insoluble residue was recovered by lyophilization. In method 2, the
native starch powders were also incubated at 35 C in 2.2 N HCl but
all resistant residues were collected after 35 days and extensively
washed to neutrality by centrifugation in distilled water. These
samples were kept in water at 4 C.
2.4. Enzyme treatment
Freeze-dried lintners obtained by method 1 were enzymatically
treated in three different manners. Firstly, lintners were treated
with -amylase (from barley) to obtain their -limit dextrins [19].
Secondly, lintners were debranched with the addition of isoamylase (from Pseudomonas amyloderamosa) and pullulanase (from
Klebsiella pneumoniae) [15]. Lastly, following debranching, lintners were treated with -amylase to obtain the -limit dextrin
of debranched components [15]. All enzyme preparations were
kind gifts from Megazyme International Ireland (Bray, Wicklow,
Ireland).
2.5. Molecular size-distribution analysis
The molecular size distribution of lintners (obtained by method
1) and enzymatically treated lintners at 45% acid hydrolysis was
determined by gel-permeation chromatography (GPC) using a column (1.6 90 cm) of Sepharose CL 6B as described by Bertoft [21].
The GPC column was calibrated with glucose, maltose, and maltoheptaose (Sigma-Aldrich, St. Louis, MO, USA), and larger -glucans
using debranched WBS, as quantied by high-performance anionexchange chromatography with pulsed amperometric detection
(HPAEC-PAD). A standard curve for the GPC column was obtained
by comparing the elution of dextrins from debranched WBS with
HPAEC analysis on a weight basis, and the standard curve was
extended linearly past the last clearly resolved DP-peak of 60 by
HPAEC analysis to cover the remaining volume of the GPC column.
2.6. Anion-exchange chromatography
Lintners and enzymatically treated lintners (obtained by
method 1) at 45% hydrolysis were analyzed by HPAEC-PAD
(Dionex-ICS 5000+ , Sunnyvale, CA, USA) equipped with a CarboPac PA-100 column and analogous guard column [15]. Areas under

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

307

Fig. 1. (A) Acid hydrolysis of WBS (), NBS () and AOS () incubated at 35 C (lled symbols) and 40 C (open symbols), including enlarged proles for (B) NBS and (C) WBS.
Scatter plot markers represent experimental data, whereas hydrolysis curves are tted.

peaks were corrected to carbohydrate content [22], and dextrins

with DP > 35, which were not resolved as peaks, were quantitatively
approximated by a continuous area division of the chromatograms
[23].
2.7. Optical microscopy
Droplets of aqueous suspensions of native starch granules and
lintnerized samples (prepared by method 2) were deposited on
glass slides and observed with an Axiophot II (Leica Microsystems,
Wetzlar, Germany) optical microscope equipped with polarized
light illumination. Digital images were recorded using a ColorView
12 (Olympus-SIS, Mnster, Germany) CCD camera.
2.8. Electron microscopy
Native starch granule powders were sputter-coated with Au
and observed in secondary electron mode using a Hitachi 4700
(Hitachi High Technologies, Tokyo, Japan) and a Quanta 250 (FEI,
Hillsboro, OR, USA) scanning electron microscope equipped with a
eld-emission gun (FEG-SEMs). Transmission electron microscopy
(TEM) images of barley starch lintners prepared by methods 1
and 2 were recorded [24]. Dilute suspensions of lintnerized starch
were briey mechanically homogenized with an UltraTurrax and
droplets were deposited on glow-discharged carbon-coated copper grids. The specimens were negatively stained with 2 wt% uranyl
acetate and observed using a Philips CM200 microscope (FEI, Hillsboro, OR, USA) operating at 200 kV. The images were recorded with
a F216 TemCam camera (TVIPS GmbH, Gauting, Germany).

2.9. Wide-angle X-ray scattering (WAXS)

Native starch granules and lintners (prepared according to
method 1) were allowed to equilibrate in 95% relative humidity
for 5 days and poured into 1 mm outer diameter glass capillaries.
The capillaries were ame-sealed and X-rayed for 2 h in vacuum
using a Philips PW383 generator operating at 30 kV and 20 mA (Niltered CuK radiation, = 0.1542 nm). Two-dimensional powder
diffraction patterns were recorded on Fujilm imaging plates (Fujilm, Tokyo, Japan), read with a Fujilm BAS 1800-II bio-imaging
analyzer.

2.10. Small-angle X-ray scattering (SAXS)

SAXS analyses were carried out on the BM02 (D2AM) beamline
at the European Synchrotron Radiation Facility (Grenoble, France).
Concentrated suspensions of native starch granules were poured
into 3 mm outer diameter glass tubes that were attached to an
automatic heating sample changer. Two-dimensional scattering
patterns were recorded at room temperature for a duration of 20 s,
at an energy of 16 keV ( = 0.0775 nm) with a CCD detector placed
at a distance of 1.65 m. SAXS proles were calculated by radially
averaging the two-dimensional patterns. The peak positions were
calibrated using a silver behenate standard.

2.11. Statistical analysis

All analyses were conducted in duplicate and analyzed using
SPSS (IBM Corporation, Armonk, NY, USA). Signicant differences

308

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

Table 1
Molecular composition of WBS, NBS, and AOS.
Sample

Amylose (%)

Amylopectin (%)

Waxy barley starch

Normal barley starch
Amylose-only starcha

0%
18%
99.1%

100%
82%
0.9%

Reported in Carcio et al. [6].

were determined by comparing means by Tukeys test at a signicance level of p < 0.05.
3. Results
3.1. Morphology and structure of native starch granules
FEG-SEM images of native NBS and WBS powders show that
they contained a mixture of large disc-shaped A-type granules
and small spheroidal B-type granules (Fig. 2A and B), as previously described by several authors [2830]. AOS granules were very
different (Fig. 2C), exhibiting a variety of irregular shapes, from
elongated to multilobular, with some granules having a very rough
surface, in agreement with a previous report [6].
When observed under polarized light, native NBS and WBS granules displayed the typical Maltese crosses (Fig. 3AC and DF,
respectively), whereas native AOS granules displayed an extremely
weak birefringence with no recognizable pattern (Fig. 3GI), which
indicated that there was no clear preferential orientation of the
glucan chains, in agreement with observations reported by Carcio
et al. [6].
The proportions of amylose and amylopectin in NBS and WBS
were determined by GPC following enzymatic debranching with
isoamylase and pullulanase (Table 1). NBS contained 18% amylose,
which was similar to the value (21.6%) obtained previously for the
same variety with a similar method [25]. WBS was amylose-free
in accordance with an earlier report for the variety Cinnamon [26].
The amylose content of AOS was previously determined to be 99.1%
using size-exclusion chromatography coupled with iodine binding
[6].
The SAXS proles recorded from hydrated granules also
revealed a difference between the AOS granules and the two other
samples (Fig. 4). While the proles of NBS and WBS granules
exhibited a scattering peak at about q = 0.62 and 0.61 nm1 , which
correspond to lamellar repeats of 10.1 and 10.3 nm, respectively,
in semicrystalline amylopectin [31], there was no such peak in the
prole of AOS granules. This conrms that in the absence of amylopectin, no regular structure exists in the granule at a length scale
of a few nanometers.
The WAXS patterns of the three samples showed that those from
WBS and NBS granules (Fig. 5A and C, respectively) predominantly
correspond to allomorph A, generally found in cereal starches [5].
However, a weak diffraction ring could be observed in both patterns
that corresponds to the characteristic 100 reection of allomorph
B [5]. In agreement with a previous analysis by Carcio et al. [6],
the pattern from AOS native granules contained weak rings corresponding to allomorph B (Fig. 5E) but the strongest ring reveals the
presence of a signicant fraction of Vh allomorph that most probably corresponds to complexes between single helical amylose and
lipids [32].
3.2. Kinetics of lintnerization
The kinetics of lintnerization of the three starch samples was
determined under two different incubation temperatures (Fig. 1).
Typically, acid hydrolysis patterns of starch exhibit two stages, an
initial fast hydrolysis of the amorphous regions, followed by slower

Fig. 2. Secondary electron SEM images of initial WBS (A), NBS (B) and AOS (C)
granules. Bars: 20 m.

hydrolysis of the amorphous and crystalline regions [27]. The more

rapid hydrolysis of the amorphous regions can be explained by the
looser packing of amorphous components compared to their crystalline counterparts [11]. The two-stage pattern was more evident
at the lower incubation temperature (35 C). Increased amylose
content was related to retarded acid hydrolytic solubilization of
the starch granules. The incubation temperature during hydrolysis
strongly inuenced the rate of solubilization of WBS and NBS, with
WBS reaching 80% solubilization in 3 days at 40 C and 6 days at
35 C, whereas NBS reached 80% solubilization in 5 days at 40 C
compared to 14 days at 35 C incubation. Interestingly, the incubation temperature did not inuence the kinetics of lintnerization
of AOS. In addition, AOS was highly resistant to acid hydrolysis
plateauing at 45% hydrolysis.

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

309

Fig. 3. Optical micrographs of WBS (AC), NBS (DF) and AOS (GI) granules dispersed in water. Images A, D and G were recorded in bright eld mode while images B, C, E,
F, H and I are polarized light micrographs. For images C, F and I, a plate was used to generate polarization colors. Bars: 20 m.

Fig. 4. SAXS proles of native WBS, NBS and AOS granules displayed in the q-range
corresponding to the lamellar repeat of semicrystalline amylopectin.

3.3. Morphology and structure of lintnerized starch granules

At 45% acid hydrolysis, complete granules could still be observed
by optical microscopy, provided that they were gently handled
when preparing the glass slides. When imaged between crossed

nicols, and although growth rings were more clearly marked than
in native granules, lintnerized NBS and WBS granules still exhibited
Maltese crosses (Fig. 6AC and DF, respectively), demonstrating
that the general chain orientation was preserved. However, as soon
as the pressure of the cover slip increased due to liquid evaporation, the granules crumbled into smaller fragments, showing that
they became extremely porous and fragile after extensive hydrolysis. This effect was also reported by Wikman et al. [33] after acid
hydrolysis of waxy potato starch granules. The behavior of AOS lintners was more complex. While the native granules did not show
any clear sign of molecular orientation, the lintnerized granules
exhibited a strong birefringence (Fig. 6H and I) suggesting that
molecular re-orientation was permitted following acid hydrolysis
of the amorphous parts.
The WAXS patterns of the native and lintnerized specimens
identied some specic dynamic effects of the AOS sample. While
the lintners from WBS and NBS granules (Fig. 5B and D, respectively) retained the predominant A-type of the parent samples
(Fig. 5A and C, respectively), those from AOS exhibited a clear Btype with the strongest diffraction rings corresponding to the 100
and 121 reections (Fig. 5F). There was thus a signicant contrast
with the initial allomorphic composition dominated by the Vh -type
that completely disappeared after acid hydrolysis while the signal
of allomorph B increased.
TEM images of the residues from lintnerized starch prepared
with the two different protocols generating different degrees of
hydrolysis (Fig. 7) demonstrates that at 45 wt% acid hydrolysis at
40 C (method 1), the elongated fragments of NBS and WBS lintners exhibited a lamellar ultrastructure (Fig. 7A and B, respectively),
being composed of stacks of 57 nm-thickness. These elements are

310

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

Fig. 5. WAXS patterns of hydrated native granules (A,C,E) and corresponding lintners (B,D,F) from WBS (A,B), NBS (C,D) and AOS (E,F) samples. The lintners have been
prepared using method 1. The allomorphic types corresponding to the main diffraction rings have been indicated.

Fig. 6. Optical micrographs of WBS (AC), NBS (D-F) and AOS (G-I) granules after a 45 wt% hydrolysis in HCl. Images A, D and G were recorded in bright eld mode while
images B, C, E, F, H and I are polarized light micrographs. For images C, F and I, a plate was used to generate polarization colors. Bars: 20 m.

believed to correspond to crystallites seen edge-on and indicates

that the lamellar organization of amylopectin had not been completely disrupted by the hydrolysis. After 35-days of hydrolysis
(method 2), with acid hydrolysis exceeding 80% (Fig. 1), the fragments from lintnerized NBS and WBS could be described as loose
lacy networks (Fig. 7D and E, respectively). In particular, angular platelets could be recognized in WBS lintners, that were very
similar to those observed in the resistant residue from waxy maize
starch granules hydrolyzed with HCl [24] or H2 SO4 [34]. Fig. 8A is a
higher magnication TEM image of WBS lintners. As explained by
Putaux et al. [24] and Hansen et al. [35], the platelets would correspond to crystallites lying at on the supporting carbon lm and
in which the constituting double helices would thus be oriented
nearly perpendicularly to the crystal plane.

The microstructure of the AOS lintners obtained with both

protocols was very similar (Figs. 7C and F) and was composed
of strongly textured aggregates. The similarity of AOS lintners
obtained with the two protocols is likely related to the resistant
nature of the resulting lintners. Their constituting units can be more
clearly seen in Fig. 8B. They were slightly elongated with a length
typically ranging from 20 to 30 nm and a width around 10 nm. Their
overall shape was less well dened than that of WBS nanocrystals (Fig. 8A) and the staining pattern suggests that they were not
platelets but more bulky particles. Considering the strong B-type
WAXS patterns, it can be assumed that the particles were B-type
nanocrystals.

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

311

Fig. 7. TEM images of negatively stained preparations of lintners from WBS (A,D), NBS (B,E) and AOS (C,F) granules. Images AC correspond to lintners obtained at 45 wt%
hydrolysis whereas images D-F correspond to specimens taken after a 35-day hydrolysis. Bars: 100 nm.

Fig. 8. Groups of nanocrystal aggregates observed in the lintners from WBS (A) and AOS (B) granules (35-day hydrolysis TEM images of negatively stained specimens).
Bars: 50 nm.

3.4. Molecular composition of lintners

The size distribution of acid-hydrolyzed barley starches prepared at 40 C was determined by GPC using Sepharose CL 6B. The
size distributions of lintners at 45% hydrolysis (Fig. 9) showed
that NBS and WBS displayed similar proles with WBS containing
a slightly higher concentration of smaller components with DP < 40.
The average size of NBS and WBS lintners were DP 54 and DP 52,
respectively. AOS exhibited a narrower size distribution than WBS
and NBS, with an average DP of 31.
The molecular size distributions of the lintners were determined in greater detail by HPAEC-PAD analysis with representative
chromatograms shown in Fig. 10. NBS and WBS exhibited similar
proles (NBS not shown). A small quantity of small dextrins (DP 6)
in NBS and WBS was detected. However, this population of dextrins
was not considered to be part of the granular structure as it was
previously observed that these chains were easily removed following extensive washing [34]. The rst population of dextrins in WBS

lintners with a peak at DP 12 represents linear dextrins, the second

population with a peak at DP 24 corresponds to single branched
dextrins, and the third population starting at DP 37 is composed
of multiple branched dextrins [12,14]. Therefore, the dextrins contained in WBS and NBS lintners were divided into three populations
consisting of small dextrins of DP 618 (Fraction I), intermediatesized dextrins of DP 1936 (Fraction II), and larger dextrins with
DP 37 (Fraction III).
The AOS lintners did not display an elution prole comparable to those of NBS and WBS. Rather, it displayed a prole with
peaks and shoulders with an apparent periodicity of 56 glucose
residues (Fig. 10). Although the three dextrin populations found
in NBS and WBS lintners were not apparent in the AOS lintners,
the chromatogram was divided into similar fractions to allow for
a quantitative comparison between all three samples. AOS lintners contained signicantly greater relative molar quantities of
Fractions I and II (together 89%), and therefore signicantly lower
quantities of Fraction III (11%) than NBS and WBS (Table 2). NBS

312

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

Fig. 9. Size distribution of WBS (), NBS () and AOS () lintners at 45 wt% acid hydrolysis prepared at 40 C determined by gel permeation chromatography on Sepharose
CL 6B.

Fig. 10. Size distribution of WBS (A) and AOS (B) lintners at 45 wt% solubilization prepared at 40 C determined by HPAEC. Numbers indicate degrees of polymerization.

Table 2
Relative molar composition (%) of lintnerized starches at 45% acid hydrolysis prepared at 40 C.
Fraction
1

I
II
III

WBS
b

34
29a
36b

NBS
a

32
29a
37b

AOS
48c
41b
11a

1
Fraction I includes dextrins of DP 618, fraction II DP 1936, and fraction III
DP 37
Values in rows with different letter are signicantly different at p < 0.05.

and WBS lintners displayed very similar relative molar compositions, although WBS lintners statistically (p < 0.05) maintained a
signicantly greater quantity of Fraction I compared to NBS. The
relative molar distribution of NBS and WBS lintners were much

more evenly distributed across the three fractions than the AOS
lintners (Table 2).
The molecular size distribution of lintners analyzed by HPAEC
provides resolution of peaks considerably higher than GPC analysis [23]. Using HPAEC analysis, Bertoft [23] determined the average
chain length of a debranched waxy maize starch by continuously
dividing the HPAEC chromatograms into successive areas past the
last clearly resolved peak for each chain, and reported that the
average chain lengths were comparable between GPC and HPAEC
analysis. In the HPAEC analysis of lintners, dextrins larger than DP
35 were not clearly resolved. However, using an approach in line
with that utilized by Bertoft [23], the amount of dextrins past the
last clearly resolved HPAEC peak was estimated by a continuous
division of the chromatogram into successive areas. The average
size of NBS (DP 42) and WBS (DP 42) lintners (Table 3) determined

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

Table 3
Structure of lintnerized starches at 45% acid hydrolysis prepared at 40 C.

DPn (GPC)1
DPn (HPAEC)2
CL3
NC4
Fraction I (DPn )5
Fraction II (DPn )
Fraction III (DPn )

WBS

NBS

AOS

52b
42b
13b
3.3b
12a
25a
83c

54b
42b
15a
2.8a
13b
27b
79b

31a
21a
18c
1.1c
12a
25a
44a

Average DPn of lintner determined by GPC on a column of Sepharose CL 6B.

Average DPn of lintner determined by HPAEC.
3
Average chain length of lintners determined by HPAEC following debranching.
4
Number of chains = DPn whole lintner (HPAEC)/chain length.
5
Fraction I includes dextrins DP 618, fraction II DP 1936, and fraction III DP 37.
Values in rows with different letter are signicantly different (p < 0.05).
2

by HPAEC were nearly identical and they were double the average DP of the AOS lintner (21), which supported the substantially
smaller size distribution of the AOS lintner determined by GPC.
The average DP values for NBS and WBS lintners were comparable
between GPC and HPAEC analysis. However, the average DP of the
AOS lintner was considerably smaller when estimated by HPAEC
analysis as compared to GPC. Apparently, this was due to considerable differences in the average DP of Fraction III containing larger
dextrins (Fig. 10), with AOS possessing an average DP of 44, compared to average DP values of 79 and 83 for NBS and WBS lintners,
respectively, whereas similar average DP values for Fraction I and
II were observed across the three lintner samples (Table 3).
The -limit dextrins of lintners were obtained following the
addition of -amylase, which hydrolyzes successive -maltose
units from the non-reducing end of glucosidic chains until one or
two glucose units from a branch point is reached which cannot be
bypassed. The remaining dextrins were present in very low quantity and did not allow for proper quantication (Fig. 11). However,
it was readily apparent that the majority of the remaining dextrins from NBS (not shown) and WBS lintners had branches located
near the reducing end of the dextrins, as the majority of the -limit
dextrins were of low DP, eluting shortly after maltose and maltotriose (Fig. 11). Remaining dextrins from AOS lintners following
treatment with -amylase were present in trace amounts. These
resistant dextrins maintained a larger size distribution with greater
DP values compared to the resistant dextrins from NBS and WBS
lintners. This suggested that branch points were not located close
to the reducing end as virtually no peaks were observed near the
peaks of maltose and maltotriose.
3.5. Chain proles of lintners
The structure of the lintners was further characterized following debranching with isoamylase and pullulanase, which hydrolyze
the -(1,6)-linkages present in the remaining residues. It is important to note, however, that the debranching enzymes used in this
study (isoamylase and pullulanase) are unable to hydrolyze single
-(1,6)-linked glucosyl residues attached to an -(1,4)-backbone
chain [36]. The prole of the chromatograms of the debranched
lintners analyzed by HPAEC (Fig. 12) showed some specic differences between the barley starches. WBS lintners displayed a
population of short dextrins with DP 27, and a second population of larger dextrins with a maximum at DP 12 with gradually
smaller peaks eluting as the DP increased. As was the case with the
proles of the chromatograms prior to debranching, WBS and NBS
displayed very comparable debranched size distribution proles
(NBS not shown) and conrmed similar distributions previously
found for barley samples [37]. In contrast, the chromatogram of
the debranched AOS lintner was unique, exhibiting a broad distribution of peaks with no major populations. However, within the

313

broad distribution of peaks a periodicity of 56 glucose residues

was observed.
The molar distributions of the debranched lintners are compared with the distribution of the lintners prior to debranching in
Fig. 13. The debranched AOS lintners displayed short chains with
DP < 6 but otherwise exhibited a prole very similar to the lintners prior to debranching, indicating low levels of branching in
the lintners. When comparing the molar distribution of NBS and
WBS lintners prior to and following debranching, it was apparent
that the position of the glucan population with a peak at DP 12
remained essentially the same in both samples, however, the population with a peak at DP 24, believed to represent single branched
glucose chains [2] largely disappeared in the debranched sample.
The average chain length (CL) of the debranched lintners
appeared to be related with the amylose content of the samples,
as WBS had the lowest CL (13), followed by NBS (15) and AOS (19)
(Table 3). With knowledge of the CL of debranched lintners and the
average DP of lintners, the average number of chains per molecule
(NC) can be approximated. The apparent NC of the AOS lintners
was 1.1, compared to 2.8 in NBS and 3.2 in WBS, again implying a
relationship with the amylose content.

4. Discussion
In this study, the structure and composition of granular remnants of barley starch spanning 099% amylose was probed using
lintnerization. It is known that amylose delays the kinetics of acid
hydrolysis [15,20]. This study conrmed this effect for an amyloseonly system. A similar hydrolysis plateau found for the AOS was also
observed in acid-hydrolyzed high-amylose maize starch, reaching
60% hydrolysis after 102 days of incubation in 15.3% v/v sulfuric
acid [38]. AOS granules, with a relative crystallinity of 25%, initially
contain a mixture of B-type (55%) and V-type (45%) allomorphs
[6], i.e. an allomorphic composition different from that of WBS and
NBS (A- and V-types) [39]. A similar combination of B- and V-type
allomorphs has previously been observed in high-amylose maize
starches [40], although the high-amylose maize also contained the
A-type allomorph possibly originating from residual amylopectin
in this starch type. For maize starch, allomorph V is preferentially
degraded in the initial stages of acid hydrolysis, coinciding with an
increase in the B-type fraction [40]. This effect was also observed
in AOS and it coincided with an increase in the relative crystallinity
of the B-type allomorph (Fig. 5), in agreement with the effects
observed in the maize system [40].
Beside the increase in the amount of B-type allomorph, an
increase in birefringence of the resistant AOS granules was also
observed. The treatment thus seems to concomitantly promote the
crystallization (or recrystallization) of amylose chains into B-type
by formation of double helices at the nanoscale, and redirecting the
preferential orientation of the molecules over larger regions. The
increase in chain mobility due to the degradation of amorphous
domains and a decrease of the chain length by acid hydrolysis may
be responsible for both effects [2].
It has been reported that the B-type crystalline fraction from
high-amylose maize starch exhibited greater resistance to hydrolysis than A-type crystals, perhaps due to the ability of longer chains
from B-type crystals to form more stable structures [40], or the
ability of amylose chains to intertwine and form B-type crystals
[41]. Bertoft [20] investigated the susceptibility of waxy maize and
waxy potato starch containing A- and B-type allomorphs, respectively, and could not support the observation of differences in
susceptibility of differing crystal type, reporting similar kinetics of
hydrolysis between A- and B- type crystals from granules made
of pure amylopectin. Therefore, the resistance of B-type crystals
from high-amylose maize starch may not be representative of B-

314

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

Fig. 11. Size distribution of -limit dextrins of WBS (A) and AOS (B) determined by HPAEC. Number indicates the degree of polymerization, and B refers to branched
dextrins.

Fig. 12. Size distribution of debranched WBS (A) and AOS (B) lintners prepared at 40 C determined by HPAEC. Numbers indicate degrees of polymerization.

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

315

Fig. 13. Molar distribution of lintners (bar graphs) prepared at 45% acid hydrolysis and their components after debranching (scatter plot) for WBS (A), NBS (B), and AOS (C)
determined by HPAEC prepared at 40 C.

type crystals starch in general [21], and the plateau observed in

AOS cannot be attributed to its B-type crystal structure per se, but
the resistance appears to be inherent to amylose possibly forming longer range B-type crystals. As the acid hydrolysis of AOS
plateaued at 45 wt%, and the relative crystallinity of the initial
granules was 25% [6], the resistance of AOS to further hydrolysis
implied that a fraction of amorphous material was present in the
materials resistant to further hydrolysis. The plateau may indicate
an association between the crystalline and amorphous components
of amylose preventing further degradation. Alternatively, some
amorphous material formed new B-type crystals during the lintnerization process. In comparison, the relative crystallinity of NBS
and WBS has been reported to be 1823% and 2830%, respectively [25,42]. Hence, acid hydrolysis progressing past 80% (Fig. 1)
indicates that the majority of amorphous material was hydrolyzed.

The incubation temperature has been shown to have a significant impact on the rate of acid hydrolysis [43]. Branch-structure
differences in A- and B-type starches were revealed by their Ngeli
dextrins [43] or lintners [15,20]. The fact that the incubation temperature did not inuence the kinetics of acid hydrolysis of AOS
was therefore unexpected and appeared to reect the nature of the
molecules in the amorphous areas, as it is that part of the granules which the acid preferentially hydrolyzes [20]. For moderately
increased amylose contents (from 2 to 36%) in the potato starch system, there is no inuence on the lamellar thickness even though
amylose at these concentrations introduces defects in the crystalline lamella [44]. However, the impact of high amylose contents
on the semicrystalline structure of barley starch granules is more
dramatic and induces an increase in the thickness of the crystalline
lamellae, and subsequently a decrease in that of the amorphous
lamellae, as amylose disrupts the packing of amylopectin double

316

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

helices by two proposed mechanisms [45]. The rst mechanism

is the co-crystallization of amylose with amylopectin, which pulls
amylopectin helices into the amorphous lamellae (increasing the
thickness of the crystalline lamellae). The second mechanism suggests the penetration of amylose oriented transverse to the lamellar
stack into the amorphous lamellae, introducing disorder [45]. For
legume starches higher acid hydrolytic resistance was observed in
starches containing higher short:long chain ratios [46]. Vamadevan et al. [47] postulated that the length of internal chain segments
along the amorphous backbone of amylopectin inuences the organization of double helices in the crystalline lamellae, implying that
the organization of the amorphous components impacts the quality and level of crystallinity. A lower susceptibility of high-amylose
starches to acid hydrolysis has been attributed to a greater extent
of inter-chain associations between starch molecules, leading to
a more compactly organized amorphous region [48] and limited
swelling [49]. It thus appears that a number of factors in combination can inuence the structure of the amorphous lamella as
deduced by its susceptibility to hydrolysis. As shown in this investigation, a lamellar organization does not exist in AOS (Fig. 4). Indeed,
the lamellae in normal starch granules are formed by the amylopectin component, which is absent in AOS. It is likely that the
arrangement and packing of the amylose chains in the amorphous
parts of the AOS granules is such that their hydrolysis becomes ratelimiting at the concentration of HCl and incubation temperatures
used in this study.
Along with WAXS and SAXS data, three complementary imaging techniques conrmed that the morphology and ultrastructure
of AOS starch granules was signicantly different from other types
of starch granules. This observation was supported by (i) the virtual lack of birefringence in native granules viewed under polarized
light and absence of so-called growth rings (Fig. 3), (ii) extreme
variability in the granule shape and surface texture (Fig. 2), and
(iii) the presence of highly textured aggregates of nanocrystals
in the lintners (Figs. 7 and 8). Further, as deduced from a recent
second harmonic generation (SHG) microscopy study [50], AOS
granules are reduced in long-range hydrogen and hydroxyl bond
networks and crystalline order typical for hydrated amylopectin. It
is thereby clear that the silencing of genes coding for SBEs had a
large inuence on the biosynthesis of the AOS granules resulting
in a unique morphology and structure. Notably, while the hydrolysis of AOS reached a plateau at 45 wt% acid hydrolysis the NBS
and WBS lintners were degraded substantially further than this
level (Fig. 1). The structure of barley starch granules subjected to
more extensive acid hydrolysis in 15% (v/v) H2 SO4 [36] revealed
that the size distribution proles of WBS (87% hydrolyzed) and
NBS (67% hydrolyzed) lintners displayed peaks at DP 12, 25, and
higher DP up to 40, corresponding to linear, single branched, and
double branched molecules, respectively, prior to debranching. Our
study conrmed these results using lower levels of acid hydrolysis
generating somewhat larger dextrins (Fig. 10). Upon debranching,
WBS and NBS displayed a major population of linear chains with a
peak at DP 12 (Fig. 12), which has been described for other starch
lintners as well [37,43]. However, our study also revealed an additional population of very short chains at DP 26 (Fig. 6), which
was not found in the previous study [36]. This could be due to
the lower extent of lintnerization in this work, as the quantity of
short chains (DP 28) produced after debranching diminishes with
hydrolysis, as the short chains may represent defective structures
of crystallites, which are selectively removed at higher levels of
hydrolysis [43,20]. However, Song and Jane [37] debranched the
lintners with isoamylase alone, whereas in this work isoamylase
was used together with pullulanase and it has been shown that the
debranching reaction becomes more complete when both enzymes
are used (albeit resistant branches always remains in the dextrin
mixture) [20]. The lintners obtained in both our study and by Song

and Jane [37] contained small amounts of chains with DP up to

about 50 and greater in very minute amounts (Fig. 12), which was
similar to the reported chains in barley lintners by Song and Jane
[37] despite the large differences in the extent of hydrolysis. If these
dextrins really are single chains or branched dextrins that escapes
the attack by the debranching enzymes is difcult to know.
The inability of -amylase to completely digest acid-hydrolyzed
starches into maltose and maltotriose [51] indicates that branch
points were present in structured regions and resistant to acid
hydrolysis. These effects were observed in all three lintner samples
in our study, however at different extents. The exo-activity of amylase prevents its ability to bypass branch points and hence the
location of branch points inuences the extent of hydrolysis with
this enzyme. Particularly, the presence of very few resistant branch
points giving rise to the relatively larger AOS lintners (Fig. 11)
demonstrates that the branches were generally positioned close by
the non-reducing ends. This suggests that AOS is devoid of clustered
branches conned to the amorphous lamellae which is supported
by the lack of a lamellar organization in AOS, as shown by the
absence of a peak in the SAXS prole and the textured aggregates
of its nanoparticles.
The presence of resistant branch points in -limit dextrins of
lintners from waxy starch [15,20,51] was conrmed for barley
starch in our study (Fig. 11) and the proles were similar to previous
data [20]. Surprisingly, we detected the general features of the lintners already at 45 wt% acid hydrolysis in NBS and WBS, whereas
most structural analyses are reported at higher levels of acid
hydrolysis [15,20,37,38]. As the extent of hydrolysis increases, the
-amylolysis limit increases and proportion of remaining dextrins
decreases, demonstrating that more branches are removed as acid
hydrolysis progresses [20]. This observation supports the decrease
in production of short chains produced following debranching in
lintners at extensive levels of acid hydrolysis reported by Jane et al.
[43] and Bertoft [20] as discussed above.
Long chains (DP < 120) are often detected in cereal starch
lintners, which can be attributed to double-helical structures of
hydrolyzed and re-crystalized amylose, or segments of amyloselipid complexes [2] formed in the early stages of lintnerization
[32]. NBS at 45% acid hydrolysis contained dextrins of this size
and after debranching the longest chains were of DP about 60
(Fig. 13). Since WBS gave the same result these chains were likely
derived from amylopectin. However, the AOS lintners, derived
from amylose also contained long-chains, but the average DP of
these were two times smaller than for NBS and WBS (Table 3)
and the size-distribution prole demonstrates a different architecture of the AOS acid-resistant regions. Distinct repeating peak
distances of 56 glucose residues largely remained after debranching (Figs. 12 and 13), which approximately corresponds to the six
glucose residues required for one turn of a helix [5] and may suggest
that the lintners were composed of trimmed amylose helical structures of repeating periodicity. Interestingly, similar patterns have
also been reported previously for isoamylase-debranched Ngeli
amylodextrins of maize amylose-extender mutant starch granules [38]. Furthermore, the same repeating periodicity has been
observed in enzyme-resistant material from retrograded wheat and
amylose-extender maize starches [52]. This implies that the structure of AOS lintners is reminiscent to that of retrograded amylose.
The signicantly higher chain length in the AOS lintners and lower
number of chains (NC nearly 1 chain per dextrin, Table 3) suggests
that the acid-resistant fraction of AOS starch granules contains
double-helical structures with very little defects.
Although the structure of the acid-resistant AOS components
and the dynamics of acid hydrolysis of its amorphous components
were determined in this study, a complete understanding of the

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

organization of its crystalline and amorphous components needs

further investigations with complementary analytical tools.
5. Conclusion
The structure of lintners, derived from starch granules containing zero to 99% amylose, as determined by combined
chromatographic, spectrometric, and microscopic techniques,
demonstrates unique features of the AOS as compared to NBS
and WBS lintners. The differences can be translated to crystalline
structures of the native starch granules. AOS granules possessed
semicrystalline structures despite the lack of the amylopectin
component and nano-lamellar structure. The chain lengths of the
lintners correlated with the content of amylose and AOS lintners
had a distinct repeat motif corresponding to 56 glucose residues.
The data conrms that starch granules can be mostly composed
of amylose, albeit with very distorted structures, and that amylopectin is required to form the micro- and nano-structures typical
for normal starch granules.
Acknowledgments
The authors would like to thank Christine Lancelon-Pin for the
SEM observation, the Electron Microscopy Platform of NanoBioICMG (Grenoble, France) for granting access to the microscopy
facility, the French CRG at European Synchrotron Radiation Facility (Grenoble, France) for beam time allocation and Yu Ogawa
(CERMAV, Grenoble) for collecting the SAXS data. We thank Lantmnnen CoOp, Sweden, for kindly providing grains of the waxy
barley starch.
References
[1] S. Prez, E. Bertoft, The molecular structures of starch components and their
contribution to the architecture of starch granules: a comprehensive review,
Starch/Strke 62 (2010) 389420.
[2] S. Wang, L. Copeland, Effect of acid hydrolysis on starch structure and
functionality: a review, Crit. Rev. Food Sci. Nutr. 55 (2015) 10811097.
[3] A. Hermansson, K. Svegmark, Developments in the understanding of starch
functionality, Trends Food Sci. Technol. 7 (1996) 345353.
[4] S. Ball, H.P. Guan, M. James, A. Myers, P. Keeling, G. Mouille, A.P. Colonna, J.
Preiss, From glycogen to amylopectin: a model for the biogenesis of the plant
starch granule Bulon, Cell 86 (1996) 349352.
[5] A. Imberty, A. Bulon, V. Tran, S. Prez, Recent advances in knowledge of
starch structure, Starch/Strke 43 (1991) 375384.
[6] M. Carcio, A. Blennow, S.L. Jensen, S.S. Shaik, A. Henriksen, A. Bulon, P.B.
Holm, K.H. Hebelstrup, Concerted suppression of all starch branching enzyme
genes in barley produces amylose-only starch granules, BMC Plant Biol. 12
(2012) 223238.
[7] Y. Takeda, S. Hizukuri, C. Takeda, A. Suzuki, Structures of branched molecules
of amyloses of various origins: and molar fractions of branched and
unbranched molecules, Carbohydr. Res. 165 (1987) 139145.
[8] S.S. Shaik, T. Obata, K.H. Hebelstrup, K. Schwahn, A.R. Fernie, R.V. Mateiu, A.
Blennow, Starch granule re-structuring by starch branching enzyme and
glucan water dikinase modulation affects caryopsis physiology and
metabolism, PLoS One (2015), accepted.
[9] L. Li, H. Jiang, M. Campbell, M. Blanco, M.J. Jane, Characterization of maize
amylose-extender (ae) mutant starches. Part I: relationship between resistant
starch contents and molecular structures, Carbohydr. Polym. 74 (2008)
396404.
[10] S.S. Shaik, M. Carcio, H.J. Martens, K.H. Hebelstrup, A. Blennow, Starch
bioengineering affects cereal grain germination and seedling establishment, J.
Exp. Bot. 65 (2014) 22572270.
[11] K. Kainuma, D. French, Naegeli amylodextrin and its relationship to starch
granule structure. I. preparation and properties of amylodextrins from
various starch types, Biopolymers 10 (1971) 16731680.
[12] M. Yamaguchi, K. Kainuma, D. French, Electron microscopic observations of
waxy maize starch, J. Ultrastruct. Res. 69 (1979) 249261.
[13] S. Srichuwong, N. Isono, T. Mishima, M. Hisamatsu, Structure of lintnerized
starch is related to X-ray diffraction pattern and susceptibility to acid and
enzyme hydrolysis of starch granules, Int. J. Biol. Macromol. 37 (2005)
115121.
[14] H. Jacobs, R.C. Eerlingen, N. Rouseu, P. Colonna, J.A. Delcour, Acid hydrolysis of
native and annealed wheat, potato and pea starchesDSC melting features
and chain length distributions of lintnerised starches, Carbohydr. Res. 308
(1998) 359371.

317

[15] J. Wikman, A. Blennow, E. Bertoft, Effect of amylose deposition on potato

tuber starch granule architecture and dynamics as studied by lintnerization,
Biopolymers 99 (2013) 7383.
[16] M. Carcio, S.S. Shaik, S.L. Jensen, A. Blennow, J.T. Svensson, . Vincze, K.H.
Hebelstrup, Hyperphosphorylation of cereal starch, J. Cereal Sci. 54 (2011)
339346.
[17] M. Dubois, K.A. Gilles, J.K. Hamilton, P. Rebers, F. Smith, Colorimetric method
for determination of sugars and related substances, Anal. Chem. 28 (1956)
350356.
[18] J. Sargeant, Determination of amylose:amylopectin ratios of starches,
Starch/Strke 34 (1982) 8992.
[19] J.-P. Robin, Lintnerized starches. Gel ltration and enzymatic studies of
insoluble residues from prolonged acid treatment of potato starch, Cereal
Chem. 51 (1974) 389406.
[20] E. Bertoft, Lintnerization of two amylose-free starches of A- and B-crystalline
types, respectively, Starch/Strke 56 (2004) 167180.
[21] E. Bertoft, Composition of building blocks in clusters from potato
amylopectin, Carbohydr. Polym. 70 (2007) 123136.
[22] K. Koch, R. Andersson, P. man, Quantitative analysis of amylopectin unit
chains by means of high-performance anion-exchange chromatography with
pulsed amperometric detection, J. Chromatogr. A 800 (1998) 199206.
[23] E. Bertoft, On the nature of categories of chains in amylopectin and their
connection to the super helix model, Carbohydr. Polym. 57 (2004) 211224.
[24] J.-L. Putaux, S. Molina-Boisseau, T. Momaur, A. Dufresne, Platelet nanocrystals
resulting from the disruption of waxy maize starch granules by acid
hydrolysis, Biomacromolecules 4 (2003) 11981202.
[25] A. Regina, J. Blazek, E. Gilbert, B.M. Flanagan, M.J. Gidley, C. Cavanagh, J.P. Ral,
O. Larroque, A.R. Bird, Z. Li, M.K. Morell, Differential effects of genetically
distinct mechanisms of elevating amylose on barley starch characteristics,
Carbohydr. Polym. 89 (2012) 979991.
[26] E. Bertoft, A. Kllman, K. Koch, R. Andersson, P. man, The cluster structure of
barley amylopectins of different genetic backgrounds, Int. J. Biol. Macromol.
49 (2011) 441453.
[27] S. Wang, J. Blazek, E. Gilbert, L. Copeland, New insights on the mechanism of
acid degradation of pea starch, Carbohydr. Polym. 87 (2012) 19411949.
[28] T. Vasanthan, R. Hoover, Barley starch: production, properties, modication
and use, in: J. BeMiller, R. Whistler (Eds.), Starch: Chemistry and Technology,
third edition, Elsevier, Burlington, 2009, pp. 601628.
[29] J. Stark, X. Yin, The effect of physical damage on large and small barley starch
granules, Starch/Strke 38 (1986) 369374.
[30] H. Tang, H. Ando, K. Watanabe, Y. Takeda, T. Mitsunaga, Fine structures of
amylose and amylopectin from large, medium, and small waxy barley starch
granules, Cereal Chem. 78 (2001) 111115.
[31] J.P.J. Jenkins, R.E. Cameron, A.M. Donald, A universal feature in the structure of
starch granules from different botanical sources, Starch/Strke 45 (1993)
417420.
[32] W.R. Morrison, R.F. Tester, M.J. Gidley, J. Karkalas, Resistance to acid
hydrolysis of lipid-complexed amylose and lipid-free amylose in lintnerised
waxy and non-waxy barley starches, Carbohydr. Res. 245 (1993)
289302.
[33] J. Wikman, A. Blennow, A. Bulon, J.-L. Putaux, S. Prez, K. Seetharaman, E.
Bertoft, Inuence of amylopectin structure and degree of phosphorylation on
the molecular composition of potato starch lintners, Biopolymers 101 (2014)
257271.
[34] H. Angellier-Coussy, J.-L. Putaux, S. Molina-Boisseau, A. Dufresne, E. Bertoft, S.
Prez, The molecular structure of waxy maize starch nanocrystals, Carbohydr.
Res. 344 (2009) 15581566.
[35] P.I. Hansen, J.-L. Putaux, A. Blennow, S.B. Engelsen, Nanocrystal platelets of
A-and B-type starches: a preliminary study, in: V.P. Yuryev, P. Tomasik (Eds.),
Starch Science and Technology, Nova Science Publishers, 2008, pp. 4149.
[36] M. Abdullah, D. French, Reversible action of pullulanase, Nature 210 (1966)
200.
[37] Y. Song, J. Jane, Characterization of barley starches of waxy, normal, and high
amylose varieties, Carbohydr. Polym. 41 (2000) 365377.
[38] H. Jiang, S. Srichuwong, M. Campbell, J. Jane, Characterization of maize
amylose-extender (ae) mutant starches. Part III: Structures and properties of
the naegeli dextrins, Carbohydr. Polym. 81 (2010) 885891.
[39] R.N. Waduge, R. Hoover, T. Vasanthan, J. Gao, J. Li, Effect of annealing on the
structure and physicochemical properties of barley starches of varying
amylose content, Food Res. Int. 39 (2006) 5977.
[40] C. Grard, P. Colonna, A. Bulon, V. Planchot, Order in maize mutant starches
revealed by mild acid hydrolysis, Carbohydr. Polym. 48 (2002) 131141.
[41] L. Jayakody, R. Hoover, The effect of lintnerization on cereal starch granules,
Food Res. Int. 35 (2002) 665680.
[42] E. Bertoft, K. Piyachomkwan, P. Chatakanonda, K. Sriroth, Internal unit chain
composition in amylopectins, Carbohydr. Polym. 74 (2008) 527543.
[43] J. Jane, K. Wong, A.E. McPherson, Branch-structure difference in starches of Aand B-type X-ray patterns revealed by their Naegeli dextrins, Carbohydr. Res.
300 (1997) 219227.
[44] S.S. Kozlov, A. Blennow, A.V. Krivandin, V.P. Yuryev, Structural and
thermodynamic properties of starches extracted from GBSS- and
GWD-suppressed potato lines, Int. J. Biol. Macromol. 40 (2007) 449460.
[45] P. Jenkins, A. Donald, The inuence of amylose on starch granule structure,
Int. J. Biol. Macromol. 17 (1995) 315321.
[46] C. Biliaderis, D. Grant, J. Vose, Structural characterization of legume starches.
II. Studies on acid-treated starches, Cereal Chem. 58 (1981) 502507.

318

A. Goldstein et al. / International Journal of Biological Macromolecules 89 (2016) 305318

[47] V. Vamadevan, E. Bertoft, K. Seetharaman, On the importance of organization

of glucan chains on thermal properties of starch, Carbohydr. Polym. 92 (2013)
16531659.
[48] R. Hoover, H. Manuel, The effect of heatmoisture treatment on the structure
and physicochemical properties of normal maize waxy maize, dull waxy
maize and amylomaize V starches, J. Cereal Sci. 23 (1996) 153162.
[49] Y. Nakazawa, Y. Wang, Acid hydrolysis of native and annealed starches and
branch-structure of their naegeli dextrins, Carbohydr. Res. 338 (2003)
28712882.

[50] R. Cisek, D. Tokarz, M. Steup, I.J. Tetlow, M.J. Emes, K.H. Hebelstrup, A.
Blennow, V. Barzda, Second harmonic generation microscopy investigation of
the crystalline ultrastructure of three barley starch lines affected by
hydration, Biomed. Opt. Express 6 (2015) 36943700.
[51] T. Watanabe, D. French, Structural features of naegeli amylodextrin as
indicated by enzymic degradation, Carbohydr. Res. 84 (1980) 115123.
[52] M.J. Gidley, D. Cooke, A.H. Darke, R.A. Hoffmann, A.L. Russell, P. Greenwell,
Molecular order and structure in enzyme-resistant retrograded starch,
Carbohydr. Polym. 28 (1995) 2331.