Accepted Manuscript

Title: Quinoa starch: Structures, properties, and applications

Authors: Guantian Li, Fan Zhu

PII: S0144-8617(17)31354-1
DOI: https://doi.org/10.1016/j.carbpol.2017.11.067
Reference: CARP 13018

To appear in:

Received date: 17-9-2017

Revised date: 2-11-2017
Accepted date: 19-11-2017

Please cite this article as: Li, Guantian., & Zhu, Fan., Quinoa
starch: Structures, properties, and applications.Carbohydrate Polymers
https://doi.org/10.1016/j.carbpol.2017.11.067

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Quinoa starch: Structures, properties, and applications

Running title: Quinoa starch review

Guantian Li, Fan Zhu *

T
IP
School of Chemical Sciences, University of Auckland, Private Bag 92019, Auckland

R
1142, New Zealand

SC
* Correspondence, email: fzhu5@yahoo.com

U
N
Highlights
A

Isolation, composition, properties, and uses of quinoa starch summarized

M

Quinoa starch granules are very small (1 to 3 m) with low amylose contents
ED

Quinoa amylopectin has a relatively high content of super-long chains

Quinoa starch modification has great potential for Pickering emulsion application
PT

Comparative analysis between quinoa starch and other starch sources conducted
E
CC

Abstract

Quinoa (Chenopodium quinoa Willd.) has gained popularity worldwide due to the
A

attractive nutritional profile. It also has much potential for food security due to the

great genetic diversity. Starch is the main component of quinoa grain and makes up to

70% of the dry matter. The starch plays a crucial role in the functional properties of

quinoa and related food products. The starch granules are rather small (~13 m) with
1
relatively low amylose contents as compared with most of the other starches. Quinoa

amylopectin has a significant amount of short chains and super-long chains. These

unique features have generated research interest in using the starch for food and other

applications such as Pickering emulsions. This review summarizes the present

knowledge of the isolation, composition, granular and molecular structures,

T
physicochemical properties, modifications, and applications of quinoa starch. It

IP
becomes obvious that this starch has great potential for food and nonfood

R
applications.

SC
U
Keywords: Chenopodium quinoa; small granule; pseudocereal; starch modification;
N
Pickering emulsion; starch digestion
A
M

1. Introduction
Quinoa (Chenopodium quinoa Willd.) of the Chenopodiceae family is native to the
ED

Andes, and has been cultivated in the Andean Region for several thousand years
PT

(Taylor & Parker, 2002; Fleming & Galwey, 1995). Before the Spanish colonization

of South America, quinoa had been widely grown as a staple grain crop for its edible
E
CC

seeds (Valencia-Chamorro, 2003). In the late 1970s, quinoa production started to

experience a renaissance within South America, not only for domestic consumption
A

but also for export (Fleming & Galwey, 1995). Quinoa cultivation is in the process of

rapid expansion outside its traditional cultivated areas with good yields (FAOSTAT,

2017; Wang & Zhu, 2016). World production of quinoa has kept increasing in the

period of 19922014, exceeding 192 thousand metric tons in 2014 (FAOSTAT, 2017).
2
Quinoa plays a significant role in food security for its broad genetic diversity and an

extraordinary adaptability to a wide range of agro-ecological conditions (Alan, 2011).

It can grow from sea level to 4000 meters above sea level, at humidity ranging from

40% to 88%, and at temperatures from 4 to 38 C (Alan, 2011). It has a high

tolerance to adverse environmental conditions such as drought and saline

T
environments with low input costs (Jacobsen, 2003). The above-mentioned

IP
characteristics make quinoa a strategic crop for providing nutrition and food security

R
in the face of climate change (Ruiz et al., 2013). The genome of quinoa has been

SC
recently sequenced, providing the genetic basis for the improvements in agricultural

U
traits and food processing properties of this crop (Jarvis et al., 2017). Quinoa seeds
N
come in a variety of colors ranging from white to red and black (Vega-Galvez et al.,
A

2010). An example of the seed cross-section is presented in Supplementary Fig. 1.

M

In recent years, there has been renewed interest in quinoa due to its attractive
ED

nutritional features. The seed is a source of starch, protein, dietary fiber, fat, minerals,

polyphenols, and vitamins (Repo-Carrasco, Espinoza, & Jacobsen, 2003; Ruales,

PT

Valencia, & Nair, 1993; Vega-Galvez et al., 2010). It contains no gluten and can be a
E

gluten-free alternative for persons with celiac disease (Alvarez-Jubete, Arendt, &
CC

Gallagher, 2010).

The major component of quinoa seed is starch, which varies from ~30 to 70% of the
A

dry matter (Supplementary Table 1). The quality of quinoa food products can be much

affected by the properties of the starch (Wang, Opassathavorn, & Zhu, 2015b). Recent

research has shown that quinoa starch can be an ingredient for food and non-food

3
applications (Wang et al., 2015b; Wang & Zhu, 2016). Compared with other starches

from maize, potato, and wheat, there is a lack of systematic knowledge of quinoa

starch. This limits the further development of this crop and the utilization of the

starch. This review summarizes the present knowledge of the isolation, composition,

granular and molecular structures, physicochemical properties, modifications, and

T
uses of quinoa starch. Suggestions for the direction to improve the understanding and

IP
utilization of this starch are provided.

R
SC
2. Isolation of quinoa starch
Various milling and soaking approaches have been applied to remove the non-starch
U
N
components. The seeds were washed and steeped in water or alkaline solutions before
A
homogenizing in a blender (Araujo-Farro, Podadera, Sobral, & Menegalli, 2010;
M

Wright, Huber, Fairbanks, & Huber, 2002). Alternatively, the seeds were dry-milled

into flour before soaking in solutions (Li, Wang, & Zhu, 2016; Mundigler, 1998;
ED

Watanabe, Peng, Tang, & Mitsunaga, 2007). It should be noted that too harsh dry-
PT

milling conditions may induce damages to starch granules (Li & Zhu, 2017c; Qian &

Kuhn, 1999). The solution used to soak the seeds or flour could be deionized water or
E
CC

solution containing sodium hydroxide, sodium bisulfite, sodium dodecyl sulfate

(SDS), or sodium acetate to remove the other components such as protein, lipid,
A

saponins, and fibers (Atwell, Patrick, Johnson, & Glass, 1983; Li et al., 2016;

Srichuwong et al., 2017; Araujo-Farro et al., 2010). The enzymatic treatment is

effective in starch purification but the cost is high, which may not be suitable for large

sample preparation. The soaking time varied from 5 min to 1 week among different
4
studies (Atwell et al., 1983; Steffolani, Len, & Prez, 2013). Long soaking may give

rise to microbial issues, whereas the altered pH may induce damage to starch granules

(Lim, Lee, Shin, & Lim, 1999).

After soaking, the suspension was filtered and the starch in filtrate was recovered by

centrifuge. High centrifugation speed (> 2000 g) was applied to increase the

T
recovery of starch due to the small granule size (Li et al., 2016; Lindeboom, Chang,

IP
Falk, & Tyler, 2005b; Steffolani et al., 2013).

R
The starch cake was re-suspended in water to further remove the non-starch

SC
components and chemical reagents added. It should be noted that residue of the

U
reagents, if not washed off probably, may affect the physicochemical properties of
N
starch such as enzyme susceptibility (Zhang & Hamaker, 1999).
A

The starch yield was in range of 3053.3% (Jan, Panesar, Rana, & Singh, 2017;
M

Wright et al., 2002). The purity was from 93% to 99% (Jan et al., 2017; Mundigler,
ED

1998; Srichuwong et al., 2017). The quinoa starch isolations were mainly carried out

at laboratory-scale and there appears to be no reports on industrial isolations of this

PT

starch.
E
CC

3. Chemical composition of quinoa starch

Starch is mainly composed of two kinds of biopolymers: the linear amylose and the
A

branched amylopectin (Prez & Bertoft, 2010). The amylose contents of quinoa starch

have been measured by a range of methods based on iodine binding-

spectrophotometry/potentiometry, concanavalin A (Con A) precipitation, and size

5
exclusion chromatography (SEC) (Table 1). Amylose content estimated by iodine

binding-based methods ranged from 0.3% to 27.7% (Table 1). It is notable that the

amylose content calculated by subtracting the influence from amylopectin was

significantly lower (7.1%) than that calculated from the whole starch (25.4%) (Tang,

Watanabe, & Mitsunaga, 2002). Such a difference could be due to the long-chain

T
fraction of quinoa amylopectin also complex with iodine, causing an overestimation

IP
of amylose content (Tang et al., 2002; Vilaplana, Hasjim, & Gilbert, 2012). The lipids

R
of starch granules may affect the amylose content measured by iodine-binding based

SC
method by forming amylose-lipid inclusion complexes (Srichuwong & Jane, 2007).

U
Thus, the iodine binding method should be with other method for the estimation of
N
amylose content in quinoa starch.
A

The amylose contents estimated from SEC of whole starch were 410.9% (Table 1).
M

The values seem to be lower than those estimated by SEC of debranched whole starch
ED

which ranged from 3.5 to 27.0% (Table 1). The presence of long unit chains of

amylopectin may affect the amylose content estimated from debranched whole starch
PT

(Li & Zhu, 2017b).

E

The Con A binding based method has been also used for quantifying the amylose
CC

content of quinoa starch (Gibson, Solah, & McCleary, 1997). Only the branched

polysaccharides could form precipitates with Con A (Goldstein, Hollerman, &

A

Merrick, 1965). Apart from amylopectin, branched amylose may also get precipitated

in this process. Although the amylose content estimated by this method has been

reported as high as 19.7%, the majority of the studies reported a value of less than

6
10% (Table 1). The lower amylose content estimated by this method may due to the

over-estimation in the value from the iodine binding and SEC of debranched whole

starch based methods or underestimation of Con A method caused by amylose

precipitation.

The isolated starch contains minor components such as protein, lipid, ash, and fiber

T
(Table 1). The majorities of the studies reported the values of the minor components

IP
being less than 0.5% (Table 1). High contents of minor components suggest an

R
insufficient purification of starch. It should be borne in mind that these minor

SC
components, though small in quantity, may have effects on the functional properties

of starch (Srichuwong & Jane, 2007).

U
N
A
M
ED
E PT
CC
A

7
 Table 1. Chemical composition of quinoa starch
Amylose content Ash Lipid Fiber
No. Method A
Protein (%) Reference
(%) (%) (%) (%)
18 Iodine-SP 14.3327.66 de Briceo & Briceo, 1980
1 Iodine-SP 11 0.11 0.04 Atwell et al., 1983
Iodine-SP &
1 20 B, 4 C Praznik et al., 1999
SEC
19.820.6 D,
2 Iodine-SP Wright et al., 2002
20.621.0 E

T
1 Iodine-SP 25.4 F, 7.1 G Tang et al., 2002
Ando et al., 2002; Tari et al.,

IP
1H Iodine-SP 11.123.9
2003; Araujo-Farro et al., 2010a
8 Iodine-SP 0.312.1 Lindeboom et al., 2005b

R
26 Iodine-SP 7.725.7 Li et al., 2016

SC
2 Iodine-SP 9.4612.1 0.890.95 0.180.22 0.320.40 0.100.13 Jan et al., 2017
1 Iodine-PT 9.28 0.91 0.11 Lorenz, 1990
Qian & Kuhn, 1999; Linsberger-
1 H
Con A 8.2512.2

U Martin Lukasch, & Berghofer,

2012; Srichuwong et al., 2017
N
3 Con A 8.229.30 1.094.13 1.461.64 1.942.56 0.190.24 Steffolani et al., 2013
1 Con A 19.7 Nascimento et al., 2014
A
9 Con A 6.068.44 Li & Zhu, 2017b
4 SEC, debranch 24.727.0 Inouchi et al., 1999
M

8 SEC, debranch 3.519.6 0.141.23 Lindeboom et al., 2005b

Lindeboom, Chang, Tyler, &
16 SEC, debranch 3.519.5
Chibbar, 2005a
ED

9 SEC, debranch 12.623.7 Li & Zhu, 2017b

9 SEC 7.4910.88 Li & Zhu, 2017b
5 5.210.9 Watanabe et al., 2007
PT

Wolf, Macmasters & Rist, 1950;

H
1 0.241.3 0.20.34 0.20.02 0.143 Mundigler, 1998; Pagno et al.,
2015
E
CC

No.: number of samples studied; Iodine-SP: colorimetric-iodine-based method; Iodine-PT:

potentiometric titration-iodine- based method; SEC: whole starch analyzed on size exclusion
chromatography; SEC, debranch: debranched starch analyzed on size exclusion chromatography; Con
A

A: concanavalin A precipitation based method; A: amylose content (%) is based on starch content; B:
amylose content determined calorimetrically by whole starch solution; C: amylose content determined
by SEC method; D: un-defatted starch sample; E: defatted starch sample; F: amylose content calculated
by: 100%BV(whole starch)/BV(amylose); G: amylose content calculated by: 100% [BV(starch)
BV(amylopectin)]/[BV(amylose)BV(amylopectin)]; H: Number of samples applied for each literature

8
 4. Structure of quinoa starch

4.1.Morphology

Various techniques such as light microscope (LM), scanning electron microscope

(SEM), transmission electron microscopy (TEM), Coulter Counter (CC), and laser

light diffraction (LLD) have been employed to study the morphology of quinoa starch

T
granules (Supplementary Table 2). The individual starch granules could be released

IP
during the isolation process (Atwell et al., 1983; Qian & Kuhn, 1999). The size of

R
quinoa granule was mostly in the range of 0.42.0 m, which was smaller than that of

SC
most starches from other botanical origins (Supplementary Table 2). The shape of

U
quinoa starch was polygonal, angular, and irregular. The diversities in both the shape
N
and size of single quinoa starch granule is relatively small (Lindeboom, Chang, &
A

Tyler, 2004; Li & Zhu, 2017a). Light microscopy, limited in the resolution capacity,
M

should not be used to study the details of quinoa starch granules. TEM analysis
ED

showed that quinoa starch granule had a homogeneous outer layer with high density

and a hilum with low density (Supplementary Fig. 2) (Tang et al., 2002).
PT

Quinoa starch may present as aggregations (Supplementary Table 2). These spherical
E

or oblong shaped aggregates were between 10 to 30 m in size with 14,00020,000

CC

single starch granules (Fig. 1) (Ando et al., 2002; Lorenz, 1990; Ruales & Nair, 1994;

Srichuwong et al., 2017). It should be noted that these aggregates may give rise to the
A

artefacts of granule size distribution data. The formation of these aggregates may be

largely due to the presence of protein as adding pepsin facilitated their dissolution

(Atwell et al., 1983; Ruales & Nair, 1994).

9
 A B

T
R IP
SC
Figure 1. SEM photos showing aggregates (A) and starch granules (B) in quinoa

U
perisperms (A: Magnification: 5,000, sample from Peru; B: Magnification:
N
30,000, sample from South America) (Srichuwong et al., 2017; Li & Zhu, 2017a)
A
M
ED
E PT
CC
A

10
 4.2. Crystallinity

Quinoa starch has an A-type polymorph (Supplementary Table 3). The degree of

crystallinity of quinoa starch ranged between 21.543.0% (Supplementary Table 3).

The value from peak fitting method (the area ratio between crystalline peak and total

peak) appeared to be higher than those calculated from the ratio of crystalline area

T
(separated by a smooth line in spectrum) and total areas. The degree of crystallinity of

IP
quinoa starch has been reported to be lower than amaranth, garden orache, and normal

R
maize starches and higher than barley, adzuki, and kaiwa starches (Qian & Kuhn,

SC
1999; Steffolani et al., 2013; Tang et al., 2002; Wright et al., 2002). Such differences

U
may be due to the differences in the chemical structure and composition of starches
N
(Prez & Bertoft, 2010). The quinoa starch has been reported to have a significant
A

amount of Afp-chains which could contribute to the defects in crystalline lamella and
M

contribute to the low crystallinity in quinoa starch. The peak around 0.44 nm (d-
ED

spacing) in XRD spectrum is characteristic of amylose-lipid inclusion complexes.

This peak is not significant for quinoa starch, indicating quinoa starch had a low level
PT

of amylose-lipid complexation (Tang et al., 2002).

E

The nature of quinoa starch crystallinity can be probed by techniques other than XRD,
CC

such as solid-state nuclear magnetic resonance (ssNMR) and small-angle X-ray

scattering (SAXS), which are yet to be applied on quinoa starch (Lopez-Rubio,

A

Flanagan, Gilbert, & Gidley, 2008).

11
 R I
SC
U
Table 2. Chemical structure of quinoa starch, amylose, and amylopectin
Sample type No. max (nm) BV DP CL NC Molecular mass Analytical method Reference

N
Whole starch 1 27 Debranched whole starch on GPC (Bio-Gel P-10) Atwell et al., 1983
AM: 595604,

A
Whole starch 1 Whole starch on GPC (Sepharose CL-2B) Ruales & Nair, 1994
AP:575

M
Whole starch 4 587596 0.320.41 Iodine absorption spectrum Inouchi et al., 1999
4,600
Iodine absorption spectrum; whole starch on SEC
Whole starch 1 161,000, Mw: 1.13107 Praznik et al., 1999
ED system
DPw: 70,000
0.287 Tang et al., 2002; Ando
Whole starch 1A 602609 Iodine absorption spectrum
0.305 et al., 2002
PT
AM:601621, Whole starch on GPC (Sepharose CL-2B);
Whole starch 9 Li & Zhu, 2017b
AP:581608 debranched whole starch on GPC (Sepharose CL-6B)
Amylopectin 4 Debranched AP analyzed by HPAEC-PAD Inouchi et al., 1999
E

Iodine absorption spectrum;

Amylopectin 1 595 DPn: 6,675 21 317 Mn: 1.1106 Tang et al., 2002
debranched AP analysed by SEC
CC

0.190 to DPn :3,403 172 Mn: 5.5105 Iodine absorption spectrum;

Amylopectin 5 589590 1820 Watanabe et al., 2007
0.210 4,752 227 7.7105 debranched AP analysed by SEC
16.0
A

Amylopectin 9 Debranched amylopectin analyzed by HPAEC-PAD Li & Zhu, 2017b

17.0
Amylopectin 1 Isoamylase-treated starch analyzed on FACE Srichuwong et al., 2017
Iodine absorption spectrum;
Amylose 1 663 1.014 DPn: 921 73 11.6 Mn: 1.5105 Tang et al., 2002
isolated amylose debranched and analyzed by SEC
0.998 DPn: 822 Mn: 1.3105 Iodine absorption spectrum;
Amylose 5 648650 98119 69 5
Watanabe et al., 2007
1.101 1,054 1.710 isolated amylose debranched and analyzed by SEC

12
 R I
SC
No.: number of samples studied; max: the wavelength with the maximum absorbance; BV: blue value; DP: degree of polymerization; DPw: weight-average degree of

U
polymerization; DPn: number-average degree of polymerization; CL: average chain length; NC: number of chains per molecule; Mw: weight-average molecular mass; Mn:
number-average molecular mass; AM: amylose fraction; AP: amylopectin fraction; GPC: gel-permeation chromatography; SEC: size-exclusion chromatography; HPAEC-

N
PAD: high-performance anion-exchange chromatography with pulsed amperometric detection; FACE: fluorophore-assisted capillary electrophoresis; A: Number of samples
applied for each literature.

A
M
ED
E PT
CC
A

13
 4.3.Chemical structure

4.3.1. Starch

The iodine-binding behaviors of quinoa starch have been studied by several

researchers (Table 2). Blue value (BV) is defined as the absorption value of a light

with the wavelength of 680 nm when passing through a cell containing a mixture of

T
starch (Bourne, Haworth, Macey, & Peat, 1948). Both BV and max of a starch

IP
solution increase with increasing degree of polymerization (DP) of the linear glucan

R
molecules (Bailey & Whelan, 1961). Among different reports, max and BV of quinoa

SC
starch have been reported in the range of 578609 nm and 0.2870.41, respectively

U
(Table 2). However, the approach appeared to be spurious as the starch is a mixture of
N
amylose and amylopectin.
A
M

4.3.2. Amylose
ED

Iodine absorption spectrum (IBS) and SEC have been used to study the structure of

quinoa amylose (Table 2). BV and max of isolated quinoa amylose ranged from
PT

0.9981.101 and 648663 nm, respectively (Tang et al., 2002; Watanabe et al., 2007).
E

max has been also estimated in the amylose fraction from the SEC of quinoa starch,
CC

which ranged from 595621 nm (Li & Zhu, 2017b; Ruales & Nair, 1994). A

comparative study showed that BV of quinoa amylose was significantly lower than
A

that of barley and adzuki starch, which could be due to their differences in the

amylose content and structure (Tang et al., 2002).

14
Quinoa amylose had lower number-based degree of polymerization (DPnAM) and

average chain length (CLAM) than barley and adzuki amyloses and more number of

chains per amylose molecule (NCAM) than barley and adzuki amyloses (Tang et al.,

2002; Watanabe et al., 2007). A SEC study revealed a lower ratio of long-chain

amylose to short-chain amylose of quinoa starch than normal maize starch (Li & Zhu;

T
2017b). Quinoa amylose appeared to be smaller with more branches, which explained

IP
its lower BV value (Tang et al., 2002; Watanabe et al., 2007).

R
SC
U
N
A
M
ED
E PT
CC
A

15
 R I
SC
U
A B

N
10 6
Weight-based carbohydrate content

A
Molar-based carbohydrate content
5 Weight-based carbohydrate content
8 Molar-based carbohydrate content

M
4
Carbohydrate content (%)

Carbohydrate content (%)

6
3

4
ED 2

2 1
PT

0
0
E

-1
0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 90
CC

Degree of polymerization Degree of polymerization

A

Figure 2. Weight- and molar-based unit chain length distributions of typical quinoa amylopectin (A) and the ,-LDs (B) (sample S21). The

carbohydrate content (%) is calculated based on the total weight of carbohydrate (weight-based carbohydrate content, empty dots) or total molar

amount (molar-based carbohydrate content, solid dots) (Li & Zhu, 2017b).

16
 4.3.3. Amylopectin

4.3.3.1. General structural features

Both the max and BV of quinoa amylopectin were significantly higher than those of

barley and adzuki amylopectins (Tang et al., 2002). DPnAP, CLAP, and NCAP of quinoa

T
amylopectin appeared to be between those of barley and adzuki amylopectins (Tang et

IP
al., 2002).

R
Super-long unit chains have been observed in quinoa amylopectin through SEC of the

SC
debranched amylopectin with contents between 13 and 19% (Tang et al., 2002;

U
Watanabe et al., 2007). The high amount of these super-long chains may explain the
N
high max and BV values of quinoa amylopectin. It may also explain the great
A

differences in the amylose content estimated by iodine binding or SEC of debranched

M

starch based methods and those by Con A or SEC of whole starch based method as
ED

discussed above. The structure and function of these large-amount super-long chains

remain to be better studied.

E PT

4.3.3.2.Unit and internal chain composition

CC

Unit chain length distribution of amylopectin can be obtained by analysing

debranched amylopectin using techniques such as high-performance anion-exchange

A

chromatography with pulsed amperometric detection (HPAEC-PAD) and fluorophore-

assisted capillary electrophoresis (FACE) (Fig. 2) (Li & Zhu, 2017b; Srichuwong et

al., 2017; Zhu, Corke, & Bertoft, 2011). A large amount of short unit chains with DP <

17
12 and a shoulder of the HPAEC chromatogram at around DP 1820 were recorded

(Li & Zhu, 2017b; Srichuwong et al., 2017). Quinoa amylopectin had higher amounts

of short unit chains with DP < 12 and lower amounts of chains with DP of 1335 than

waxy maize starch (Inouchi et al., 1999). The ratio between short and long chains of

quinoa amylopectin tended to be higher than those of amylopectins from other sources

T
(Bertoft, Piyachomkwan, Chatakanonda, & Sriroth, 2008; Li & Zhu, 2017b).

IP
The average chain length of quinoa amylopectins (CLAP) (9 varieties) estimated by

R
HPAEC-PAD ranged from 1617 glucosyl residues, which were shorter than those

SC
quantified by using SEC (1821 glucosyl residues) (Li & Zhu, 2017b; Tang et al.,

U
2002; Watanabe et al., 2007). Such a difference could be readily attributed to the
N
different analytical methods used in different studies. Quinoa CLAP was shorter than
A

that of most of the amylopectins from other sources except for oat amylopectin as
M

estimated by the same analytical method (Bertoft et al., 2008; Li & Zhu, 2017b).
ED

External chains of amylopectin can be enzymatically removed by enzymes (-amylase

and/or phosphorylase a) to produce -limit dextrins or ,-limit dextrins (-LDs and

PT

,-LDs) (Bertoft, 2004). The amount of carbohydrate removed during the enzymatic
E

hydrolysis is termed -limit value or ,-limit value (Bertoft, 2004). The -limit value
CC

of quinoa amylopectin (1 variety) has been reported to be 59% (Ando et al., 2002).

Another study showed that ,-limit values for quinoa amylopectins were between
A

52.7 to 56.4% (Li & Zhu, 2017b). The varieties could coming from different

experimental conditions as well as method of calculation. The internal unit chain

18
length profile of quinoa amylopectin had a shoulder at DP 57 and two peaks at DP

910 and DP 33 (Fig. 2) (Li & Zhu, 2017b).

External chain length (ECL) and internal chain length (ICL) of quinoa amylopectins

were shorter than those of amylopectins from several of A-type starches (e.g., waxy

maize starch) and most of B- and C-type starches (e.g., potato starch) (Bertoft et al.,

T
2008; Li & Zhu, 2017b). Bertoft et al. (2008) categoried amylopectins into 4 groups

IP
based on the internal unit chain profiles. Group 1 amylopectins had the highest

R
amount of short internal unit chains and the lowest amount of long internal unit

SC
chains. Group 4 amylopectins had just the opposite compositional pattern. Group 2

U
and 3 amylopectins were intermediate between group 1 and 4 samples (Bertoft et al.,
N
2008). Quinoa amylopectin shared a lot of similarities with both the group 1 and
A

group 2 amylopectins, though it did not strictly fit into any group (Bertoft et al.,
M

2008).
ED

The unit chains in amylopectin could be classified as A-chains which do not carry

other chains, B-chains which carry other chains, and C-chain which has the sole
PT

reducing end (Peat, Whelan, & Thomas, 1952). The amount of A-chains can be
E

accurately evaluated from the ,-LDs where all the A-chains are maltose stubs
CC

(Bertoft, 2004). Molar amounts of A-chains in quinoa amylopectins (9 varieties)

ranged between 46.4 to 50.1% (Li & Zhu, 2017b). A portion of A-chains (DP 68) is
A

characteristics of the botanical source, which is termed fingerprint-A chains (Afp)

(Bertoft, 2004). The molar percentages of Afp in quinoa amylopectins has been found

to be higher than that of most of other amylopectins estimated by Bertoft et al. (2008).

19
Therefore, the unit chain profile of quinoa amylopectin appeared to be rather unique

among various types of starches.

The number of chains in amylopectin clusters may be approximately calculated from

the molar ratio between the short and long unit chains, which ranged between 10 to

16.6 (Li & Zhu, 2017b; Tang et al., 2002; Watanabe et al., 2007). However, the

T
accurate cluster structure could only be estimated from the isolated clusters which

IP
were well illustrated in many occasions by Bertoft and co-workers (Bertoft, 2013).

R
The future work would be to apply this analytical approach to include quinoa

SC
amylopectin.

5. Physicochemical properties of quinoa starch

U
N
A range of analytical techniques were used to study the physical changes of quinoa
A

starch suspension during heating and cooling as summarized below.

M
ED
E PT
CC
A

20
 R I
SC
Table 3. Gelatinization properties of quinoa starch

U
Starch Scanning rate Scanning
No. Instrument To (oC) Tp (oC) Tc (oC) H (J/g) A Reference
concentration (oC/min) range (oC)

N
Atwell et al., 1983; Tang et
10.316.7
1 DSC 3240% (w/w) 5 50.654.5 60.662.6 66.071.3 al., 2002; Srichuwong et al.,

A
J/g
2017

M
1 DSC 21% (w/w) 10 30120 59.9 64.5 71 1.66 Qian & Kuhn, 1999
4 DSC 33% (w/w) 52.257.4 54.261.9 67.668.5 7.310.5 Inouchi et al., 1999
2 DSC
ED
21% (w/w) 10 20120 51.451.8 55.756.1 64.364.5 12.6 Wright et al., 2002

1 DSC 20% (w/w) 54 62.2 71 11 Ando et al., 2002

8 DSC 50% (w/v) 10 40140 44.653.7 50.561.7 12.815.0 Lindeboom et al., 2005b
PT

3 DSC 25% (w/w) 10 25120 54.2555.72 61.6663.01 8.459.00 Steffolani et al., 2013

26 DSC 25% (w/w) 10 2590 50.158.3 56.265.0 65.874.9 10.814.4 Li et al., 2016
E

2 DSC 30% (w/w) 10 10120 64.3266.61 69.3671.56 75.7876.98 4.345.10 Jan et al., 2017
CC

1 M 48 62 Wolf et al., 1950

1 PM 1% 1 57 64 Atwell et al., 1983

1 KHSM 61 64.5 68 Lorenz, 1990

A

1 RH 5% (w/w) 2 55 Praznik et al., 1999

No.: number of samples studied; DSC: differential scanning calorimetry; M: microscope observation; PM: polarized microscope; KHSM: polarized microscope equipped
with a Kofler hot stage; RH: rheometer equipped with cone/plate; A: the unit cal/g is converted to J/g by a factor of 4.184; B: starch equivalent based on starch and moisture
contents; To: onset temperature; Tp: peak temperature; Tc: conclusion temperature; H: enthalpy change

21
 A
CC
EPT
ED
M
A
N
U
SC
R I
22
 5.1.Gelatinization measured by DSC and microscopy

Differential scanning calorimetry (DSC) and polarized microscope (may include a

Kofler hot-stage setting) were used to study the gelatinization properties of quinoa

starch (Table 3). Differences in estimation methods and experimental conditions of

DSC (e.g., starch concentration and heating rate) make the results among different

T
studies difficult for comparison (Zhu, 2015). Variations in gelatinization properties

IP
were observed within the same study (Inouchi et al., 1999; Jan et al., 2017; Li et al.,

R
2016; Lindeboom et al., 2005b). For example, Li et al. (2016) revealed that Tp and H

SC
of 26 quinoa starch samples ranged from 56.265.0 oC and 10.814.4 J/g,

respectively.
U
N
Comparative studies showed that the gelatinization temperatures of quinoa starch
A

were lower than those of amaranth, maize, garden orach, adzuki, kaiwa, sorghum,
M

millet, and wheat starches, and were similar to those of barley starch. H of quinoa
ED

starch was lower than that of amaranth, maize, and adzuki starches, were similar to

that of garden orach, kaiwa, and sorghum starches, and were higher than that of
PT

barley and wheat starches (Inouchi et al., 1999; Qian & Kuhn, 1999; Srichuwong et
E

al., 2017; Steffolani et al., 2013; Tang et al., 2002; Wright et al., 2002). Such
CC

differences in gelatinization properties among different starches could be attributed to

the differences in starch composition and structure (Srichuwong & Jane, 2007;
A

Vamadevan & Bertoft, 2015).

Correlation analysis shows that the gelatinization temperatures and H were closely

correlated with the fine structure of amylopectin (Li & Zhu, 2017a; Srichuwong et al.,

23
2017). For example, the high amount of short unit chains, especially the Afp-chains,

and low amounts of long unit chains in quinoa amylopectin may provide an

explanation for the low gelatinization temperatures of quinoa starch (Li & Zhu,

2017a).

The presence of amylose-lipid inclusion complexes can be detected by DSC with an

T
endothermic process occurring around 100 oC. Tang et al. (2002) showed that the H

IP
for amylose-lipid inclusion complexes in quinoa starch was lower than that in barley

R
and adzuki starches. Wright et al. (2002) found no endothermic process for the

SC
amylose-lipid complexes of quinoa starch. These reports suggested that the level of

U
amylose-lipid complex in quinoa starch was quite low, which agreed with the XRD
N
results (Section 4.2).
A
M

5.2.Swelling and solubility

ED

During starch gelatinization process, the extents of granule swelling and the release of

soluble material are usually monitored by the swelling and solubility tests. The
PT

starches are heated at various temperatures. The solubilized and undissolved materials
E

are separated and weighted, from which the swelling power (SP), water solubility
CC

index (WSI), and swelling factor (SF) could be calculated (Supplementary Table 4).

The increasing SP toward high temperature suggested that quinoa starch granules
A

could better maintain their integrity under heating than some other starches such as

potato starch. The experimental settings such as starch concentration, centrifuge

24
condition, stirring, and separation method could impact on the results, which makes

results from different studies difficult for comparison.

The SP and WSI of quinoa starch were negatively correlated with the amylose

content, suggesting amylose could restrict the swelling of starch granules (Li et al.,

2016; Lindeboom et al., 2005b). The SP at low temperatures (i.e., 55 oC) was

T
correlated with the amylopectin structural parameters such as Afp and the ratio of

IP
short-to-long chains, whereas the SP and WSI at higher temperatures (e.g., 95 oC) and

R
AML was more related to the internal structure of quinoa amylopectins such as the

SC
molar amounts of B-chain categories (Li & Zhu, 2017a). These results suggested the

U
roles of diverse chain categories in the swelling properties of starch as reviewed
N
previously (Srichuwong & Jane, 2007; Vamadevan & Bertoft, 2015).
A
M

5.3.Rheological properties
ED

5.3.1. Pasting

Pasting properties of quinoa starches were estimated by using Rapid Visco-analyzer

PT

(RVA), Brabender viscograph (BV), and rheometer equipped with a starch cell
E

(Supplementary Table 5). Within the same study, great variations in pasting properties
CC

were observed among different quinoa samples (Supplementary Table 5). This could

due to the high diversities in composition as well as the molecular and granular
A

structure of quinoa starch. In comparative studies, the PV of quinoa starch was higher

than that of most other starches such as waxy maize, wheat, millet, buckwheat, garden

orach, kaiwa, white sorghum, amaranth, and red sorghum starch (Praznik et al.,

25
1999; Qian & Kuhn, 1999; Srichuwong et al., 2017; Steffolani et al., 2013; Wright et

al., 2002). The high PV of quinoa starch corresponded well with the high SP as

described in Section 5.2. PV has been found positively correlated with the amounts of

amylose and long chain fraction of amylopectin. This may be due to the long glucan

chains restricting the swelling and improving the ability for granules to maintain their

T
structure (Li et al., 2016; Li & Zhu, 2017b). In contrast, opposite correlation pattern

IP
between PV and amylose was also reported (Lindeboom et al., 2005b). Another study

R
revealed a parabolic relationship between PV and amylose content (Srichuwong et al.,

SC
2017). These conflicting results indicated the relationship between PV and amylose

U
content is complicated. Other factors such as granule size, super-long chains in
N
amylopectin, and the presence of ghost structure may also contribute to the high PV
A

of quinoa starch (Debet & Gidley, 2007; Srichuwong et al., 2017).

M

5.3.2. Other rheological properties

ED

Small deformation oscillatory and shear-stability tests were applied on quinoa starch

(Jan et al., 2017; Lindeboom et al., 2005b; Praznik et al., 1999). The shear-stability of
PT

quinoa starch was higher than that of amaranth, waxy maize, millet, and buckwheat
E

starches, and was similar to that of wheat starch (Praznik et al., 1999). The high shear-
CC

stability of quinoa starch gel agreed with previous discussions on SP and pasting

behaviors. The shear-stability of quinoa starch was positively correlated with amylose
A

content (Lindeboom et al., 2005b). The relationships between rheological behaviors

and structures of amylose and amylopectin are still to be better studied.

26
 5.4. Retrogradation

The retrogradation of quinoa starch has been monitored by several approaches (DSC,

texture analysis, syneresis, freeze-thaw stability, and turbidity) (Supplementary Tables

6 & 7).

T
5.4.1. Amylopectin re-crystallization measured by DSC

IP
The amylopectin re-crystallization was monitored by DSC (Supplementary Table 6).

R
The starch retrogradation is significantly affected by the water content and storage

SC
conditions. The comparison of results of different studies may be meaningless (Wang,

U
Li, Copeland, Niu, & Wang, 2015a). The retrogradation percentage (R%, the
N
percentage ratio between the enthalpy change of retrogradation and gelatinization) of
A

quinoa starch was lower than that of kaiwa, sorghum, millet, maize, and wheat
M

starches, and was higher than that of amaranth starch (Srichuwong et al., 2017;
ED

Steffolani et al., 2013). The retrogradation temperatures have been found to be

negatively correlated with short amylopectin chains of DP 612 and positively

PT

correlated with long unit chains of DP 1324 (Srichuwong et al., 2017). The possible
E

reason for the low R% in quinoa starch could be due to the short ECL of quinoa
CC

amylopectin retarding the formation of double helical structure (Li & Zhu, 2017b).
A

5.4.2. Gel texture

Texture profile analysis (TPA) has been used to study the retrogradation of quinoa

starch (Supplementary Table 7). Hardness (HD) of quinoa starch was much lower than

27
that of kaiwa starch under the same conditions (Steffolani et al., 2013). Positive

correlations between amylose content and HD of quinoa starch gel have been

observed (Li et al., 2016; Steffolani et al., 2013). These confirm the central role of

amylose in starch gelation. Quinoa amylose is short in average chain length and

highly branched as discussed above, which could have impacts on the gel texture. The

T
texture properties of quinoa starch may also be affected by the amylopectin internal

IP
structure. For example, the ratio between short and long B-chains of ,-LDs of

R
quinoa amylopectins (9 samples) was negatively correlated with HD and positively

SC
correlated with adhesiveness (AD) (Li & Zhu, 2017a). As discussed above, the quinoa

U
amylopectin has a significant amount of super-long chains. These super-long chains
N
may also contribute to the starch gel texture (Tang et al., 2002; Watanabe et al., 2007).
A
M

5.4.3. Turbidity
ED

The changes in turbidity or clarity of starch gel could be used to reflect the

retrogradation process. The quinoa starch gel with less amylose showed lower
PT

turbidity (Jan et al., 2017). Like the other aspects of retrogradation, the turbidity can
E

be affected by a range of factors such as swelling and remnants of granules, leaching

CC

of amylose and amylopectin, the chain length distributions of amylose and

amylopectin, and minor components which could mask the inherent turbidity of starch
A

(Wang et al., 2015a).

5.4.4. Syneresis and freezethaw stability

28
As retrogradation progresses, water is expelled from starch gel due to recrystallization

and re-association of starch molecules. The syneresis is estimated by the amount of

water separated from storing the gel at low temperature (e.g., 4 oC) or from the freeze-

thaw process. The gel concentration, thermal history, and centrifugation settings could

significantly impact on the syneresis of starch gel (Zhu, 2015). Lindeboom et al.

T
(2005b) found diversities in freeze-thaw stability of 8 quinoa starch with the syneresis

IP
after the first cycle ranging from 035%. Therefore, good freeze-thaw stability of

R
quinoa starch gel can be achieved by selecting suitable quinoa variety. No correlations

SC
have been found between the freeze-thaw stability and amylose content of quinoa

U
starch, while the correlations between the syneresis or freezethaw stability and
N
amylopectin structure are yet to be studied (Ahamed, Singhal, Kulkarni, & Pal,
A

1996b; Lindeboom et al., 2005b).

M
ED

5.5.Enzyme susceptibility

Different studies used various approaches (source and category of enzyme, buffer
PT

type, pH, incubation temperature and time) to estimate enzyme susceptibility of

E

quinoa starch (Supplementary Table 8). Quinoa starch appeared to have higher
CC

enzyme susceptibility than most of the other starches such as barley, adzuki, maize,

millet, wheat, and sorghum starches (Inouchi et al., 1999; Srichuwong et al., 2017;
A

Tang et al., 2002).

The high specific surface area of quinoa starch has been suggested as the main reason

for the high enzyme susceptibility (Inouchi et al., 1999; Li et al., 2016; Srichuwong et

29
al., 2017; Tang et al., 2002). Lorenz (1990) suggested that the enzyme susceptibility

of quinoa starch was negatively correlated with the amylose content. Thus the

amylose may reduce the enzyme susceptibility at an earlier stage of hydrolysis. The

enzyme hydrolysis extent after 1 h of incubation was positively correlated to the

molar amount of Afp-chains and chains of DP 612 of amylopectin, while being

T
negatively correlated with CLAP (Li & Zhu, 2017a). This is due to the short unit

IP
chains of amylopectin such as Afp-chains are too short to form double helices and

R
cause defects in the crystalline structure, making it more susceptible to the enzyme

SC
attack (Vamadevan & Bertoft, 2015). The hydrolysis extent of quinoa starch after 24 h

U
was positively correlated with the molar amount of A-chains which are assumed to
N
forming the crystalline lamellae (Acrystal) (Li & Zhu, 2017a). This might suggest that the
A

enzyme hydrolysis rate is significantly affected by the amount of short-chains, while

M

the digestible starch is more related to the amount of chains forming crystalline
ED

region.
PT

6. Applications of quinoa starch

E

In recent years, there has been increased interest in quinoa starch for various food
CC

applications (Supplementary Table 9). Both native and modified starches were used

for various products including films and in encapsulations of food ingredients as

A

described below.

6.1. Native starch

30
6.1.1. Film

The native quinoa starch was blended with low-density polyethylene (LDPE) as

biodegradable fillers (Ahamed, Singhal, Kulkarni, Kale, & Pal, 1996a). Quinoa starch

based films were prepared with glycerolat various pH and drying conditions (Araujo-

Farro et al., 2010). The optimized conditions for film preparation were obtained as

T
21.2% of glycerol, pH 10.7, and drying at 36 oC for 14 h (Araujo-Farro et al., 2010).

IP
The film was colorless and had better mechanical properties and barrier properties

R
than the films made from maize starch, yam starch, and amaranth flour (Araujo-Farro

SC
et al., 2010).

U
Quinoa starch was also used to manufacture active biofilms with gold nanoparticles
N
(AuNPs) (Pagno et al., 2015). The addition of AuNPs increased the tensile strength
A

and UV absorption while decreasing the solubility of the resulting films, which is
M

desired for food packaging and coating. Moreover, the biofilms have been found
ED

having a good thermal stability and antimicrobial activity with great potential for

maintaining the shelf life of food products (Pagno et al., 2015).

PT

Starch based films and coatings have attracted much attentions in food industry for
E

the environmental compatibility (Jimnez, Fabra, Talens, & Chiralt, 2012). The
CC

above-mentioned studies suggested that quinoa starch can be a good thermoplastic

material for food packing applications.

A

6.1.2. Starch aggregates for flavor entrapment

31
Food ingredients can be encapsulated by various starch based systems at micro- or

nano-scales for controlled delivery (Zhu, 2017). The native quinoa starch granules

have been used as wall material to entrap vanillin (Tari, Annapure, Singhal, &

Kulkarni, 2003; Tari & Singhal, 2002). It seemed that the amylose content was

positively correlated with the encapsulation capacity of the starches, though the exact

T
reasons remain to the analysed (Tari & Singhal, 2002; Tari et al., 2003).

IP
6.2. Chemically-modified starch

R
Quinoa starch has been modified by octenyl succinic anhydride (OSA) to various

SC
degrees of substitution for stabilizing Pickering emulsions (Supplementary Table 9).

U
N
6.2.1. Preparation of OSA-modified starch
A

The preparation and functional properties of OSA-modified starches have been well
M

reviewed previously (Sweedman Tizzotti, Schfer, & Gilbert, 2013). Briefly, native
ED

quinoa starch was suspended in water with the addition of certain amounts of OSA.

The pH of solution was maintained at 7.47.9 and the reaction was completed when
PT

the pH reached a constant value (Rayner, Sj, Timgren, & Dejmek, 2012a; Timgren,
E

Rayner, Dejmek, Marku, & Sj, 2013). The OSA-modified quinoa starch was
CC

recovered by centrifugation and the degree of substitution was estimated by a

titration-based method (Rayner et al., 2012a; Timgren et al., 2013).

A

6.2.2. Preparation of Pickering emulsions containing OSA-starch

32
The OSA-modified quinoa starch has increased hydrophobicity, which is suitable for

the preparation of Pickering emulsions (O/W). OSA-modified quinoa starch was

mixed thoroughly with water phase and oil phase. The mixture was emulsified in a

high-shear mixer (Matos, Timgren, Sj, Dejmek, & Rayner, 2013; Timgren et al.,

2013).

T
Matos et al. (2013) prepared a Pickering double emulsion (W1/O/W2) stabilized by

IP
OSA-modified quinoa starch granules. A water-in-oil emulsion (W1/O) was prepared

R
first and then emulsified with W2 phase in the presence of OSA-modified quinoa

SC
starch (Matos et al., 2013).

U
N
6.2.3. Characteristics of Pickering emulsions stablished by OSA-modified starch
A

The particle size of Pickering emulsions decreased with increasing amounts of added
M

starch (Rayner, Timgren, Sj, & Dejmek, 2012b). Rheological analysis showed that
ED

the linear viscoelastic region of emulsion was in a short range of 0.0001 to 0.002,

which suggests low shear resistance (Rayner et al., 2012b). The degree of OSA
PT

substitution of 3% was optimal for stabilizing the Pickering-emulsion (Rayner et al.,

E

2012a). High ionic strength of buffer could narrow the size distribution of emulsions
CC

(Rayner et al., 2012a; Timgren et al., (2013). The Pickering emulsion formed by

OSA-modified quinoa starch remained stable even up to a storage period of 2 years

A

(Timgren et al., 2013).

For the Pickering double emulsions, the average size of inner particles was 3.1 m,

whereas the size of the droplets stabilized by starch granules showed a mean diameter

33
of 33 m (Matos et al., 2013). The majority of the encapsulated oils (90%) remained

after the storage over a period of 21 days (Matos et al., 2013).

Both the Pickering emulsion and Pickering double emulsion systems were employed

to produce powders with high oil contents (> 70%) through freeze-drying (Marefati,

Rayner, Timgren, Dejmek, & Sj, 2013; Marefati, Sj, Timgren, Dejmek, &

T
Rayner, 2015).

IP
The oil properties, especially the melting temperature, may impact on the stability of

R
emulsions during freeze-drying (Marefati et al., 2015). The partially-gelatinized OSA-

SC
modified quinoa starch gave the emulsion higher stability against the freeze-thaw

U
process than the non-heat treated starch (Marefati et al., 2013; Marefati et al., 2015).
N
The dual modification of H2SO4 and OSA increased the emulsion capacity of quinoa
A

starch (Saari, Heravifar, Rayner, Wahlgren, and Sj, 2016). Matos, Marefati,
M

Gutirrez, Wahlgren, & Rayner (2016) showed that the emulsion formed by OSA-
ED

modified quinoa starch in non-granular form was less stable than that by OSA-

modified quinoa starch granules.

E PT
CC
A

34
 T
R IP
SC
U
N
A
M
ED
E PT
CC

Figure 3. Pickering emulsions formed by OSA-modified quinoa starch granules

without (a, c, e) and with heat treatment (b, d, f) as observed by light microscopy
A

(ad) and SEM (e, f) (Marefati, Bertrand, Sj, Dejmek, & Rayner, 2017)

35
OSA-modified starches have been commonly used for the encapsulation of functional

food ingredients (Zhu, 2017). Curcumin has been encapsulated by a heated emulsion

stabilized by OSA-modified quinoa starch. The emulsion greatly preserved the

curcumin in a simulated intestinal digestion model system, which suggested a

successful delivery of this bioactive to the large intestine (Fig. 3) (Marefati et al.,

T
2017). Overall, the Pickering emulsions stabilized by OSA-modified quinoa starch

IP
granules appeared to have potential in novel formulations of food products,

R
pharmaceutical drugs, and cosmetics (Rayner 2012a; Marefati et al., 2013).

SC
6.3. Physically-modified starch
U
N
Dry-heating was used to modify the hydrophobicity of quinoa starch (Rayner et al.,
A

2012a; Timgren et al., 2013) (Supplementary Table 9). However, the emulsion
M

capacity of heat-treated starch was less efficient than OSA-modified quinoa starch
ED

(Rayner et al., 2012a; Timgren et al., 2013).

Quinoa starch was modified by high hydrostatic pressure (HHP) up to 600 MPa at
PT

different conditions (Li & Zhu, 2018; Linsberger-Martin et al., 2012). The quinoa
E

starch has been found to be more sensitive to HHP treatment than maize starch (Li &
CC

Zhu, 2018). Resistant starch (RS) content of HHP-treated quinoa starch was increased

with increasing pressure and temperature (Linsberger-Martin et al., 2012). Therefore,

A

choosing correct processing techniques and conditions may improve the quality of

quinoa starch products.

36
 7. Conclusions

Great variations in chemical composition, granular and molecular structures, and

physicochemical properties of quinoa starch were observed among many studies.

Quinoa starch is small in granule size (~12 m) with A-type polymorph. The

granules may present as aggregates. The amylose contains significant amounts of

T
short chains and branches. Apart from the genetics, the amylose content of quinoa

IP
starch can be affected by the presence of super-long chains in amylopectin. Quinoa

R
amylopectin is abundant in short unit chains (e.g., Afp-chains) and lack of long chains

SC
as compared with other kinds of amylopectins. The amylopectin structure gives

U
quinoa starch several special physicochemical properties such as low gelatinization
N
temperatures and slow retrogradation. The quinoa starch showed high SP and PV
A

during pasting event with a degree of resistance to shearing. There has been
M

increasing interest in using quinoa starch for food and nonfood applications. Native
ED

quinoa starch has been used in biofilm production and in microencapsulation of food

ingredients (e.g., oils) for its small granule size. The OSA-modified quinoa starch was
PT

much used in the stabilization of Pickering emulsions. These Pickering emulsions

E

showed several promising properties such as high emulsifying efficiency and stability.
CC

Physical modifications such as HHP has been applied on quinoa starch to obtain

several unique properties.

A

Compared with the starches isolated from major crops such as maize and potato,

research on both structure and applications of quinoa starch is still limited. Quinoa has

excellent adaptability to different environmental conditions due to the great genetic

37
diversity. The diversity in starch properties remains to be studied by assessing the

impacts of genetics and environments. Several important structural features such as

the cluster structure of quinoa amylopectin are still unknown. Although many studies

reported the quinoa starch properties, the different experimental settings used by

different researchers make the inter-study comparison difficult. The standardization of

T
the methods can improve this common issue that existed in our starch research

IP
community. The modifications and uses of quinoa starch remain much less

R
investigated. Several commonly used modifications such as ultrasound, annealing,

SC
cross-linking, and dual modifications could be applied on quinoa starch in the future.

U
Native and modified quinoa starches with promising properties could have application
N
potential such as fat replacement, novel food additives, pharmaceuticals, cosmetics,
A

papermaking, and textile, which still need to be studied.

M
ED

Acknowledgments

The Faculty Research Development Fund of the University of Auckland (3705617)

PT

supported this research. The figures are reprinted with permissions from the
E

publishers.
CC

References
A

Ahamed, N. T., Singhal, R. S., Kulkarni, P. R., Kale, D. D., & Pal, M. (1996a).

Studies on Chenopodium quinoa and Amaranthus paniculatas starch as

biodegradable fillers in LDPE films. Carbohydrate Polymers, 31, 157160.

38
Ahamed, N. T., Singhal, R. S., Kulkarni, P. R., & Pal, R. (1996b). Physicochemical

and functional properties of Chenopodium quinoa starch. Carbohydrate

Polymers, 31, 99103.

Alan, B. (2011). Quinoa, an ancient crop to contribute to world food security. In 37th

FAO conference.

T
Alvarez-Jubete, L., Arendt, E. K., & Gallagher, E. (2010). Nutritive value of

IP
pseudocereals and their increasing use as functional gluten-free ingredients.

R
Trends in Food Science & Technology, 21, 106113.

SC
Ando, H., Chen, Y. C., Tang, H. J., Shimizu, M., Watanabe, K., & Mitsunaga, T.

U
(2002). Food components in fractions of quinoa seed. Food Science and
N
Technology Research, 8, 8084.
A

Araujo-Farro, P. C., Podadera, G., Sobral, P. J., & Menegalli, F. C. (2010).

M

Development of films based on quinoa (Chenopodium quinoa, Willdenow)

ED

starch. Carbohydrate Polymers, 81, 839848.

Atwell, W. A., Patrick, B. M., Johnson, L. A., & Glass, R. W. (1983). Characterization
PT

of quinoa starch. Cereal Chemistry, 60, 911.

E

Bailey, J., & Whelan, W. (1961). Physical properties of starch I. Relationship between
CC

iodine stain and chain length. Journal of Biological Chemistry, 236, 969973.

Bertoft, E. (2004). On the nature of categories of chains in amylopectin and their

A

connection to the super helix model. Carbohydrate Polymers, 57, 211224.

Bertoft, E., Piyachomkwan, K., Chatakanonda, P., & Sriroth, K. (2008). Internal unit

chain composition in amylopectins. Carbohydrate Polymers, 74, 527543.

39
Bertoft, E. (2013). On the building block and backbone concepts of amylopectin

structure. Cereal Chemistry, 90, 294311.

Bourne, E., Haworth, N., Macey, A., & Peat, S. (1948). The amylolytic degradation of

starch. A revision of the hypothesis of sensitisation. Journal of the Chemical

Society, 924930.

T
de Briceo, Z. S., & Briceo, O. (1980). Evaluacion de la composicin quimica

IP
(composicin nutricional) de algunas entradas de quinua (Chenopodium

R
quinoa Wild) del banco de germoplasma de la Universidad Nacional Tcnica

SC
del Altiplano (Per). Anales Cientficos, 18, 125134.

U
Debet, M. R., & Gidley, M. J. (2007). Why do gelatinized starch granules not dissolve
N
completely? Roles for amylose, protein, and lipid in granule "Ghost" integrity.
A

Journal of Agricultural and Food Chemistry, 55, 47524760.

M

FAOSTAT (Statistics Division of Food and Agriculture Organization of the United

ED

Nations). Available from : http://www.fao.org/faostat/en/#search/quinoa 15

May 2017. Accessed on May 21, 2017.

PT

Fleming, J.E., & Galwey, N.W. (1995). Quinoa (Chenopodium quinoa). In: Williams,
E

J.T. (Ed.), Cereals and Pseudocereals. Chapman and Hall, London.

CC

Gibson, T. S., Solah, V. A., & McCleary, B. V. (1997). A procedure to measure

amylose in cereal starches and flours with concanavalin A. Journal of Cereal

A

Science, 25, 111119.

Goldstein, I., Hollerman, C., & Merrick, J. M. (1965). Protein-carbohydrate

interaction I. The interaction of polysaccharides with concanavalin A.

40
 Biochimica et Biophysica Acta (BBA)-General Subjects, 97, 6876.

Inouchi, N., Nishi, K., Tanaka, S., Asai, M., Kawase, Y., Hata, Y., et al. (1999).

Characterization of amaranth and quinoa starches. Journal of Applied

Glycoscience, 46, 233240.

Jacobsen, S. E. (2003). The worldwide potential for quinoa (Chenopodium quinoa

T
Willd.). Food Reviews International, 19, 167177.

IP
Jan, K. N., Panesar, P., Rana, J., & Singh, S. (2017). Structural, thermal and

R
rheological properties of starches isolated from Indian quinoa varieties.

SC
International Journal of Biological Macromolecules, 102, 315322.

U
Jarvis, D. E., Ho, Y. S., Lightfoot, D. J. Schmckel, S. M., Li, B., Borm, T. J., et al.
N
(2017). The genome of Chenopodium quinoa. Nature, 542, 307312.
A

Jimnez, A., Fabra, M. J., Talens, P., & Chiralt, A. (2012). Edible and biodegradable
M

starch films: a review. Food and Bioprocess Technology, 5, 20582076.

ED

Li, G., Wang, S., & Zhu, F. (2016). Physicochemical properties of quinoa starch.

Carbohydrate Polymers, 137, 328338.

PT

Li, G., & Zhu, F. (2017a). Amylopectin molecular structure in relation to

E

physicochemical properties of quinoa starch. Carbohydrate Polymers, 164,

CC

396402.

Li, G., & Zhu, F. (2017b). Molecular structure of quinoa starch. Carbohydrate
A

Polymers, 158, 124132.

Li, G., & Zhu, F. (2017c). Physicochemical properties of quinoa flour as affected by

starch interactions. Food Chemistry, 221, 15601568.

41
Li, G., & Zhu, F. (2018). Effect of high pressure on rheological and thermal properties

of quinoa and maize starches. Food Chemistry, 241, 380-386.

Lim, S. T., Lee, J. H., Shin, D. H., & Lim, H. S. (1999). Comparison of protein

extraction solutions for rice starch isolation and effects of residual protein

content on starch pasting properties. Starch/Strke, 51, 120125.

T
Lindeboom, N., Chang, L. R., Tyler, R. T., & Chibbar, R. N. (2005a). Granule-bound

IP
starch synthase I (GBSSI) in quinoa (Chenopodium quinoa Willd.) and its

R
relationship to amylose content. Cereal Chemistry, 82, 246250.

SC
Lindeboom, N., Chang, P. R., Falk, K. C., & Tyler, R. T. (2005b). Characteristics of

U
starch from eight quinoa lines. Cereal Chemistry, 82, 216222.
N
Lindeboom, N., Chang, P. R., & Tyler, R. T. (2004). Analytical, biochemical and
A

physicochemical aspects of starch granule size, with emphasis on small

M

granule starches: a review. Starch/Strke, 56, 8999.

ED

Linsberger-Martin, G., Lukasch, B., & Berghofer, E. (2012). Effects of high

hydrostatic pressure on the RS content of amaranth, quinoa and wheat starch.

PT

Starch/Strke, 64, 157165.

E

Lopez-Rubio, A., Flanagan, B. M., Gilbert, E. P., & Gidley, M. J. (2008). A novel
CC

approach for calculating starch crystallinity and its correlation with double

helix content: A combined XRD and NMR study. Biopolymers, 89, 761768.
A

Lorenz, K. (1990). Quinoa (Chenopodium quinoa) Starchphysico-chemical

properties and functional characteristics. Starch/Strke, 42, 8186.

Marefati, A., Bertrand, M., Sj, M., Dejmek, P., & Rayner, M. (2017). Storage and

42
 digestion stability of encapsulated curcumin in emulsions based on starch

granule Pickering stabilization. Food Hydrocolloids, 63, 309320.

Marefati, A., Rayner, M., Timgren, A., Dejmek, P., & Sj, M. (2013). Freezing and

freeze-drying of Pickering emulsions stabilized by starch granules. Colloids

and Surfaces A: Physicochemical and Engineering Aspects, 436, 512520.

T
Marefati, A., Sj, M., Timgren, A., Dejmek, P., & Rayner, M. (2015). Fabrication of

IP
encapsulated oil powders from starch granule stabilized W/O/W Pickering

R
emulsions by freeze-drying. Food Hydrocolloids, 51, 261271.

SC
Matos, M., Marefati, A., Gutirrez, G., Wahlgren, M., & Rayner, M. (2016).

U
Comparative emulsifying properties of octenyl succinic anhydride (OSA)-
N
modified starch: granular form vs dissolved state. PloS One, 11, e0160140.
A

Matos, M., Timgren, A., Sj, M., Dejmek, P., & Rayner, M. (2013). Preparation and
M

encapsulation properties of double Pickering emulsions stabilized by quinoa

ED

starch granules. Colloids and Surfaces A: Physicochemical and Engineering

Aspects, 423, 147153.

PT

Mundigler, N. (1998). Isolation and determination of starch from amaranth

E

(Amaranthus cruentus) and quinoa (Chenopodium quinoa). Starch/Strke, 50,

CC

6769.

Nascimento, A. C., Mota, C., Coelho, I., Gueifo, S., Santos, M., Matos, A. S., et al.
A

(2014). Characterisation of nutrient profile of quinoa (Chenopodium quinoa),

amaranth (Amaranthus caudatus), and purple corn (Zea mays L.) consumed in

the North of Argentina: Proximates, minerals and trace elements. Food

43
 Chemistry, 148, 420426.

Prez, S., & Bertoft, E. (2010). The molecular structures of starch components and

their contribution to the architecture of starch granules: A comprehensive

review. Starch/Strke, 62, 389420.

Pagno, C. H., Costa, T. M., de Menezes, E. W., Benvenutti, E. V., Hertz, P. F., Matte,

T
C. R., et al. (2015). Development of active biofilms of quinoa (Chenopodium

IP
quinoa W.) starch containing gold nanoparticles and evaluation of

R
antimicrobial activity. Food Chemistry, 173, 755762.

SC
Peat, S., Whelan, W., & Thomas, G. J. (1952). Evidence of multiple branching in

U
waxy maize starch. Journal of the Chemical Society, 45364538.
N
Praznik, W., Mundigler, N., Kogler, A., Pelzl, B., Huber, A., & Wollendorfer, M.
A

(1999). Molecular background of technological properties of selected starches.

M

Starch/Strke, 51, 197211.

ED

Qian, J., & Kuhn, M. (1999). Characterization of Amaranthus cruentus and

Chenopodium quinoa starch. Starch/Strke, 51, 116120.

PT

Rayner, M., Sj, M., Timgren, A., & Dejmek, P. (2012a). Quinoa starch granules as
E

stabilizing particles for production of Pickering emulsions. Faraday

CC

Discussions, 158, 139155.

Rayner, M., Timgren, A., Sj, M., & Dejmek, P. (2012b). Quinoa starch granules: a
A

candidate for stabilising food-grade Pickering emulsions. Journal of the

Science of Food and Agriculture, 92, 18411847.

Repo-Carrasco, R., Espinoza, C., & Jacobsen, S. E. (2003). Nutritional value and use

44
 of the Andean crops quinoa (Chenopodium quinoa) and kaiwa (Chenopodium

pallidicaule). Food Reviews International, 19 179189.

Ruales, J., & Nair, B. M. (1994). Properties of starch and dietary fibre in raw and

processed quinoa (Chenopodium quinoa, Willd) seeds. Plant Foods for

Human Nutrition, 45, 223246.

T
Ruales, J., Valencia, S., & Nair, B. (1993). Effect of processing on the physico-

IP
chemical characteristics of quinoa flour (Chenopodium quinoa, Willd).

R
Starch/Strke, 45, 1319.

SC
Ruiz, K. B., Biondi, S., Oses, R., Acua-Rodrguez, I. S., Antognoni, F., Martinez-

U
Mosqueira, E. A., et al. (2014). Quinoa biodiversity and sustainability for food
N
security under climate change. A review. Agronomy for Sustainable
A

Development, 34, 349359.

M

Saari, H., Heravifar, K., Rayner, M., Wahlgren, M., & Sj, M. (2016). Preparation
ED

and characterization of starch particles for use in pickering emulsions. Cereal

Chemistry, 93, 116124.

PT

Srichuwong, S., Curti, D., Austin, S., King, R., Lamothe, L., & Gloria-Hernandez, H.
E

(2017). Physicochemical properties and starch digestibility of whole grain

CC

sorghums, millet, quinoa and amaranth flours, as affected by starch and non-

starch constituents. Food Chemistry, 233, 110.

A

Srichuwong, S., & Jane, J. (2007). Physicochemical properties of starch affected by

molecular composition and structures: A review. Food Science and

Biotechnology, 16, 663674.

45
Steffolani, M. E., Len, A. E., & Prez, G. T. (2013). Study of the physicochemical

and functional characterization of quinoa and kaiwa starches. Starch/Strke,

65, 976983.

Sweedman, M. C., Tizzotti, M. J., Schfer, C., & Gilbert, R. G. (2013). Structure and

physicochemical properties of octenyl succinic anhydride modified starches: A

T
review. Carbohydrate Polymers, 92, 905920.

IP
Tang, H., Watanabe, K., & Mitsunaga, T. (2002). Characterization of storage starches

R
from quinoa, barley and adzuki seeds. Carbohydrate Polymers, 49, 1322.

SC
Tari, T. A., Annapure, U. S., Singhal, R. S., & Kulkarni, P. R. (2003). Starch-based

U
spherical aggregates: screening of small granule sized starches for entrapment
N
of a model flavouring compound, vanillin. Carbohydrate Polymers, 53, 4551.
A

Tari, T. A., & Singhal, R. S. (2002). Starch-based spherical aggregates: stability of a

M

model flavouring compound, vanillin entrapped therein. Carbohydrate

ED

Polymers, 50, 417421.

Taylor, J. R. & Parker, M. L. (2002). Quinoa. In Pseudocereals and less common

PT

cereals (pp. 93122). Springer Berlin Heidelberg.

E

Timgren, A., Rayner, M., Dejmek, P., Marku, D., & Sj, M. (2013). Emulsion
CC

stabilizing capacity of intact starch granules modified by heat treatment or

octenyl succinic anhydride. Food Science & Nutrition, 1, 157171.

A

Valencia-Chamorro, S. A. (2003). Quinoa. In B. Caballero, L. Trugo, & P. Finglas,

(Eds.), Encyclopedia of Food Sciences and Nutrition (2nd ed., Vol. IV, pp.

48954902). Oxford: Academic Press.

46
Vamadevan, V., & Bertoft, E. (2015). Structure-function relationships of starch

components. Starch/Strke, 67, 5568.

Vega-Galvez, A., Miranda, M., Vergara, J., Uribe, E., Puente, L., & Martinez, E. A.

(2010). Nutrition facts and functional potential of quinoa (Chenopodium

quinoa willd.), an ancient Andean grain: A review. Journal of the Science of

T
Food and Agriculture, 90, 25412547.

IP
Vilaplana, F., Hasjim, J., & Gilbert, R. G. (2012). Amylose content in starches:

R
Toward optimal definition and validating experimental methods.

SC
Carbohydrate Polymers, 88, 103111.

U
Wang, S., Li, C., Copeland, L., Niu, Q., & Wang, S. (2015a). Starch retrogradation: a
N
comprehensive review. Comprehensive Reviews in Food Science and Food
A

Safety, 14, 568585.

M

Wang, S., Opassathavorn, A., & Zhu, F. (2015b). Influence of quinoa flour on quality
ED

characteristics of cookie, bread, and Chinese steamed bread. Journal of

Texture Studies, 46, 281292.

PT

Wang, S., & Zhu, F. (2016). Formulation and quality attributes of quinoa food
E

products. Food and Bioprocess Technology, 9, 4968.

CC

Watanabe, K., Peng, N. L., Tang, H. J., & Mitsunaga, T. (2007). Molecular structural

characteristics of quinoa starch. Food Science and Technology Research, 13,

A

7376.

Wolf, M. J., Macmasters, M. M., & Rist, C. E. (1950). Some characteristics of the

starches of three South American seeds used for food. Cereal Chemistry 27,

47
 219222.

Wright, K. H., Huber, K. C., Fairbanks, D. J., & Huber, C. S. (2002). Isolation and

characterization of Atriplex hortensis and sweet Chenopodium quinoa

starches. Cereal Chemistry, 79, 715719.

Zhang, G., & Hamaker, B. R. (1999). SDS-sulfite increases enzymatic hydrolysis of

T
native sorghum starches. Starch/Strke, 51, 2125.

IP
Zhu, F. (2015). Isolation, composition, structure, properties, modifications, and uses

R
of yam starch. Comprehensive Reviews in Food Science and Food Safety, 14,

SC
357386.

U
Zhu, F., Corke, H., & Bertoft, E. (2011). Amylopectin internal molecular structure in
N
relation to physical properties of sweetpotato starch. Carbohydrate Polymers,
A

84, 907918.
M

Zhu, F. (2017). Encapsulation and delivery of food ingredients using starch based
ED

systems. Food Chemistry, 229, 542552.

E PT
CC
A

48