J. Food. Sci. Agric 1999 (Submitted)
Effects of Protein and -Glucans on Pasting and Thermal Properties of Oats
1 1 2 1
Meixue Zhou , Kevin Robards , Malcolm Glennie-Holmes , and Stuart Helliwell
School of Science and Technology, Charles Sturt University, PO Box 588, Wagga Wagga, NSW 2678, Australia
NSW Agriculture, Agricultural Research Institute, Wagga Wagga, NSW 2650, Australia
Specific enzymes were used to decompose protein and -glucans in wholemeal oats prior to measuring pasting and thermal
properties of the meal, using rapid viscoanalysis and differential scanning calorimetry. -glucans were responsible for a large
proportion of these properties of the meals while the contribution of protein was much less. An extra transition process was
observed using DSC tests on wholemeal oats compared to isolated starch. This transition was not associated with -glucan or
protein. Pasting and thermal properties were correlated with each other and with compositional data. Processing effects on
thermal properties were also noted.
Key words: DSC; oat; glucan; protein; pasting.
because of poor colour and poor flavour, and was
Introduction described by trained sensory panellists as having stronger
starchy, weaker watery flavours as well as an unpleasant
Pasting and thermal properties of starch are the key factors
adhesive mouthfeel (Zhou et al. 1999c).
responsible for the texture of cereal-based processed
products. Most previous studies of pasting properties of
Table 1 Composition of oat samples.
cereals have concentrated on isolated starch because of the
problems of separating the effects due to other
components. However, unlike that from most cereals, oat Component Content (g kg-1)
starch cannot easily be separated from other components Cultivar
of the grain and the process of isolation may change the Mortlock Yarran
properties of the starch. More importantly, from the Moisture 1381 1431
consumer's point of view the pasting and thermal Protein 1886 1116
properties of the whole oat product are critical to its Starch 5077 5761
acceptance. Thus, for studies on oats, it is more important -glucan 471 461
to determine the pasting and thermal properties of the Lipid 8 01 1011
whole meal. Of the non-starch components, lipid, -glucan
and protein will most influence these properties of the Series II: Eight varieties of oats (Bimbil, Carrolup, Cooba,
whole meal. The effect of lipid on pasting properties of Echidna, Euro, Mortlock, Pallinup & Yarran) were
oats has been examined (Zhou et al. 1999c). -glucans grown in Condobolin. Futher samples of Mortlock and
form viscous gums (Autio et al. 1987) and contribute Yarran were grown in Cowra and Parkes in NSW. These
significantly to water retention and processing behaviour. samples were used to examine the effect of oat variety (and
The high content of gum, especially -glucans, in wet- to a lesser extent, the effect of growing site) on the
milled oat bran had a marked effect on the viscosity of relationship between RVA and DSC data and the effect of
heat- and -amylase-treated bran slurries (Jaskari et al processing on DSC.
1995). Slurry viscosity was correlated with -glucan
concentration in oat flours and treatment of slurries with The oat samples were dehulled with a laboratory scale oat
-glucanase lowered viscosity significantly (Zhang and dehulling machine. Dehulled samples were ground with a
Moore 1997). There has been little work on the effect of Falling Number Grinder fitted with a 0.8 mm screen.
protein on pasting and thermal properties of oat meal. Moisture contents were determined by Near Infrared
Reflectance Spectroscopy (NIR) using a Technicon 450R
This study examined the effect of -glucan and protein on instrument based on calibrations provided by the Drought
the pasting and thermal properties of oat meal as measured Evaluation Unit of NSW Agriculture. Starch was
by Rapid Viscoanalysis and Differential Scanning determined by a total starch kit and -glucans were
Calorimetry (DSC). The effect of processing on the determined by a total -glucan kit both supplied by Mega-
thermal properties of oat meal was also examined. Zyme (Deltagen, 31 Wadhurst Drive, Boronia 3155, Vic.
Australia) and used according to the manufacturer's
Materials and methods instructions. Protein contents of the two commercial
samples were determined by Dumas combustion using a
Lecor CNS 2000 instrument (improved AOAC method
Groat samples 7.016 with modern technology). Lipid contents of the
Series I: Commercially grown samples of two Australian commercial samples were determined by extraction using
cultivars, Yarran and Mortlock, were used in this study. petroleum ether (bp 40-60, PE) for 16 h in Goldfisch
Table 1 characterises these two samples. These (plain-body) apparatus (AOCS official method Ai 3-75 -
commercial samples came from the same bulk lots as were modified). Protein and lipid contents of the other samples
used in previous studies (Zhou et al. 1999a, 1999b, were determined by NIR.
1999c). For food use purposes, Mortlock is preferred by
consumers, while Yarran is not acceptable to consumers
Preparation of Rolled Oats Results and discussion
Samples of each variety (10 kg) were dehulled and then
stored at 8C for the two days which had to elapse before RVA
processing could proceed. Samples (3.5 kg) were Rapid viscograms are normally carried out using freshly
processed following normal commercial practice using prepared slurries in distilled water. The use of buffer to
small scale steaming, kilning and rolling machines in replace distilled water had no effect on the viscograms.
Uncle Tobys Research & Development Centre However, to ensure the complete degradation of the protein
(Wahgunyah, Victoria, Australia). The conditions were: or -glucan, it was necessary to allow 1 hr of reaction time
steaming for 9 min, kilning at 100C for 45 min and 65C prior to the test. The viscograms were substantially
for 15 min, cooling to room temperature, resteaming for 5 changed by this pretreatment (Figure 1). The reduction in
min and rolling to a thickness of 55 m. viscosity following soaking can be attributed either to
physical changes due to hydration or the action of intrinsic
Enzymes enzymes. Consistent with this latter interpretation,
viscograms run in the presence of silver nitrate (0.1
Proteinase K (EC 220.127.116.11) was supplied by Sigma Chemical mMoles/g meal, a concentration sufficient to inactivate
Co. (St. Louis, MO USA). The -glucanase ((1,3)-, (1,4)--D- amylases, proteases and glucanases, Glennie-Holmes,
glucanase, EC 18.104.22.168) was taken from a total -glucan kit 1995) were indistinguishable from those of freshly
manufactured by Mega-Zyme (Deltagen, Boronia, Vic. prepared slurries.
Following soaking in buffer for 1 hr, most RVA
measurements were affected by variety. The peak
RVA measurements viscosities of Morlock and Yarran were significantly
Ground groat meal (4.0g, at 15% moisture content) was decreased (35% and 11%, respectively). There was a slight
slurried in phospate buffer (20 mM, pH 6.5, 25 mL). The increase in the trough of Mortlock (6%) while that of
pasting properties of the slurry were determined with a Yarran was decreased significantly (14%). Final viscosity
Rapid Visco-analyser (RVA, Newport Scientific, and setback of Mortlock were increased 42% and 64%,
Warriewood, NSW, Australia) using a previously respectively, whereas those of Yarran decreased 29% and
developed profile (Zhou et al 1999a) with a stirring speed 40%. The two characteristics not varietally affected were
of 960 rpm for 10 sec and 115 rpm for the remainder of time to peak viscosity (15% longer) and pasting
the test and with the temperature programmed to rise from temperature (9% and 4% higher for Morlock and Yarran,
40C to 90C in 3 min, to hold for 6.5 min, to cool to 40C respectively). These results indicate that Morlock
in 4.5 min and to hold for 5 mins. All measurements were contained a lesser amount of intrinsic enzymes in the
as previously described (Zhou et al 1998). resting grain than Yarran. The detection of enzymatic
activity agrees with reports of -glucanase in
To study the effect of proteinase and -glucanase on the ungerminated oat seeds (Wood et al. 1978). Doehlert et al.
viscogram, "blank" determinations were carried out by (1997) also suggested that the loss of apparent viscosity in
mixing phosphate buffer with meal in an RVA can and the raw and roasted oat meal slurries after an extended
can immersed in a 40C water bath for 1 hour before incubation period was partly due to the action of -
running the RVA. Similarly, tests with either proteinase or glucanase.
-glucanase in buffer were placed in the 40C water bath
for 1 hr before viscoanalysis. The times and quantities of Effect of added proteinase and -glucanase on
enzyme were chosen to ensure degradation of all the viscograms
protein or -glucans in the meal. All measurements were Viscograms of both meals were changed significantly
performed in duplicate. (Figure 2 and Table 2) by the addition of -glucanase.
Although the -glucan content of these two varieties
DSC measurements differed only slightly (Table 1), the addition of -
DSC was carried out by means of a Mettler Toledo DSC glucanase caused differential responses. Added -
820 Differential Thermal Analyser. Meals (4.0 mg) were glucanase had a greater effect on Mortlock with peak
suspended in an aqueous solution (15 µL; phosphate viscosity, trough and final viscosity being decreased by
buffer, proteinase in buffer, or -glucanase in buffer) in an 36%, 14% and 34%, respectively, whereas the reduction in
aluminium pan (medium pressure) or a stainless steel pan these measurements for Yarran was 12%, 11% and 6%,
(high pressure). To ensure complete enzyme activity, the respectively. The addition of -glucanase decreased the
meal slurries were put in a 40C oven for one hour and setback of Mortlock meals significantly (43%) while there
then moved to ambient temperature until testing was no significant effect on the setback of Yarran (2%
commenced. The temperature of the DSC was started at higher). -glucanase also caused significant increases in
25C, raised to 115C (the maximum temperature before time to peak viscosity and pasting temperature of Mortlock
the sealing of the aluminium pans fails) at the rate of (4% and 23%, respectively). In contrast, time to peak
10C/min and then cooled to 25C at the rate of 50C/min. viscosity and pasting temperature of Yarran were
unaffected by additions of -glucanase.
Additions of proteinase caused significant but minor
All analyses were performed at least in duplicate and changes to some of the RVA parameters of both varieties
results recorded as the mean and standard deviation. The and the effects were relatively small compared to -
data were analysed by single factor or multifactor analysis glucanase. Peak viscosity and trough of both varieties
of variance following Snedecor and Cochran (1967). were decreased by about 15% with added proteinase.
While final viscosity of both meals was little affected by
the addition of proteinase, the setback of both varieties
was increased significantly (9% and 30%, respectively for prepared. The statistically significant correlations are now
Morlock and Yarran). The time to peak viscosity of the discussed. Groat protein content was negatively correlated
Yarran meal was decreased significantly by adding with trough (-0.82**). -glucan content was significantly
proteinase, whereas that of Morlock was unchanged. The correlated with breakdown (0.82**), final viscosity (-
proteinase had no significant effect on the pasting 0.69*) and time to peak viscosity (-0.60*). The Tp1 of the
temperature of either meal. meals was positively correlated with protein (0.68*) and -
glucan (0.78**) content and negatively correlated with
-glucans contribute most of the viscosity of soluble lipid content (-0.73**). -glucan had a close correlation
extracts of oats (Autio et al 1987, Bhatty 1992), thus it is with both H2 (-0.92**) and Tp2 (-0.62*).
not surprising that decomposition of -glucans had a much
greater effect on meal pasting properties than The pasting properties showed close correlation with
decomposition of protein. As the two varieties contained thermal properties. Peak viscosity was positively
the same amount of -glucan as substrate but different correlated with H1 (r = 0.71**); Trough was negatively
levels of intrinsic glucanases, the pretreatment stage correlated with Tp1 (r = -0.74**); Breakdown was
permitted greater reduction in viscosity in the Yarran thus positively correlated with Tp1 (r = 0.80**) and negatively
reducing the opportunity for added enzymes to express full with H2 (r = -0.69*) and Tp2 (r = -0.70*); Final viscosity
activity. When the results of the standard (freshly mixed, was negatively correlated with Tp1 (r = -0.67*) and
without soaking) procedure and the -glucanase treated positively with H2 (r = 0.73**) and Tp2 (r = 0.66*);
test are compared, both varieties showed an identical Setback was correlated with Tp2 (r = 0.68*); Time to peak
(43%) reduction in viscosity. This overall reduction viscosity was negatively correlated with Tp 1 (r = -0.67*)
indicated that the -glucan had been completely destroyed. and pasting temperature was positively correlated to H2 (r
= 0.61*) and Tp2 (r = 0.78**). These results suggest that as
DSC the second transition has been identified as the result of the
The data obtained by DSC using wholemeals were the breakdown of the amylose:lipid complex (Hartunian Sowa
enthalpy and temperature corresponding to endothermic and White 1992), the pasting temperature measured by
transitions occurring during heating (Figure 3, Table 3). RVA was similarly asigned to this process.
Three transition processes were detected in Morlock and
Yarran, the first corresponding to the loss of starch Effect of processing on DSC
crystallinity (Paton, 1987), the second to the transition of Our previous study showed that processing of oats
the amylose-lipid complex (Hartunian Sowa and White changed the pasting properties of the meals significantly
1992; Zhou et al 1999b) and the third to interactions (Zhou et al 1999c), as also shown by Doehlert et al.
between other components of the meal and starch (Zhou et (1997), and Zhang et al. (1997). Processing also had a
al 1999b). The loss of starch crystallinity has been defined significant effect on thermal properties of the meals (Table
as the point of gelatinisation and thus H1 (Table 3) is 4).
termed the enthalpy of gelatinisation and Tp1 the
temperature of gelatinisation. Most DSC studies on oats The H1 of the processed meals was reduced in every
have been based on isolated starch and values reported for cultivar, on average by 1.4 J g-1. Tp1 was altered on
the gelatinisation and amylose:lipid complex transition average by –1.0C (Table 4) with a range between the
enthalpies were about 9 J g-1 and 3 J g-1 respectively varieties of +0.1 to -2.5C. Therefore, processing would
(Hartunian-Sowa and White 1992, Paton 1987). The use of reduce the energy input required to gelatinise the product,
wholemeals in present study acounts for the third trasition but individual varieties had a characteristic response to
and also for the reduced enthalpies (see Table 3). processing which was also affected by growing site. For
example, as shown in Table 4, Morlock grown at Cowra
There were no significant differences of enthalpy between had H1 reduced by 2.8 J g-1 whereas the H1 of the same
varieties for H1 and H2, whereas H3 differed in cultivar grown at Condobolin was reduced by only 0.4 J g-
varieties with Mortlock having higher enthalpy. Mortlock 1
. The non-zero value of the enthalpy of gelatinisation of
had significantly higher Tp1 and Tp3. There was no processed oats, H1, which occurred in all cultivars
difference in Tp2 between varieites. indicates that complete gelatinisation had not occurred
during processing despite the fact that grains are steamed
Effect of added proteinase and -glucanase on DSC twice before rolling. This was supported by the moisture
The two varieties responded similarly in peak temperature content of the grain, which did not change significantly
to the addition of -glucanase (Figure 3, Table 3). When during processing (Zhou et al 1999b). Nevertheless, there
the -glucans were broken down, Tp1 was increased by was a significant decrease in the transition enthalpy, H1,
8C, Tp2 by 5 - 6C and Tp3 by 3 - 5C. These data are following processing, indicating some disorganisation of
consistent with those obtained with the RVA (Table 2). the crystalline structure (Table 4).
With the exception of H3 of Mortlock and H2 of Yarran,
the enthalpies of the two varieties were little affected by The H2 was also reduced by processing in every cultivar
the addition of -glucanase. The changes following the (Table 4) with an average reduction of 0.4 J g-1. Of
addition of proteinase were smaller that those following particular interests was the similartity in H2 values of
the addition of -glucanase. Morlock and Yarran both before and after processing.
Since this enthalpy change is associated with amylose:lipid
complex, it suggests that the complex is similar in both
Correlations between RVA and DSC data
varieties despite the large difference in oil content (Table
1). Therefore, it is the proportion of amylose capable of
A correlation matrix (17 x 17, not presented) relating forming a complex with lipid which is limiting in both
compositional (protein, -glucan & lipid), RVA and DSC cultivars. Tp2 was increased, on average, by 2.0C after
measurements on the 12 oat samples (series II) was processing, and therefore the amylose:lipid complex was
stabilised by processing. Iowa State University Press. Ames, Iowa, USA.
Wood P J, Siddiqui I R, Paton D 1978 Extraction of high
References viscosity gums from oats. Cereal Chem 55 1038-
Autio K, Myllymäki O, Mälkki Y 1987 Flow properties of
Zhang D, Doehlert D C, Moore W R 1997 Factors
solutions of oat -glucans. J Food Sci 52 1364-
affecting viscosity of slurries of oat groat flours.
Cereal Chem 74 722-726.
Bhatty R S 1992 Total and extractable -glucan contents
Zhou M X, Glennie-Holmes M, Robards K, Helliwell S
of oats and their relationship to viscosity. J Cereal
1999a RVA protocol for determining the pasting
Sci 15 185-192. properties of oats. J Cereal Sci (Submitted).
Doehlert D C, Zhang D, Moore W R 1997 Influence of Zhou M X, Glennie-Holmes M, Robards K, Helliwell S
heat pretreatments of oat grain on the viscosity of 1999b. Effects of processing and short-term storage
flour slurries. J Sci Food Agric 74 125-131. on the pasting characteristics of slurries made from
Glennie-Holmes M 1995 Studies on barley and malt with raw and rolled oats. Food Aust (In press).
the RVA. (1) The effect of variation in physical and Zhou M X, Roberts G L, Robards K, Glennie-Holmes, M,
chemical parameters. J Inst Brew 101 11-18. Helliwell S 1998 Effects of sowing date, nitrogen
Hartunian-Sowa M, White P J 1992 Characterisation of application, and sowing rate on oat quality. Aust J
starch isolated from oat groats with different amount Agric Res 49 845-51.
of lipid. Cereal Chem 69 521-527. Zhou M X, Robards K, Glennie-Holmes M, Helliwell S
Jaskari J, Henriksson K, Nieminen A, Suortti T, Salovaara 1999c Effects of oat lipids on groat flour pasting
H, Poutanen K 1995 Effect of hydrothermal and properties. J Sci Food Agric 79:(In press).
enzymic treatments on the viscous behaviour of dry-
and wet-milled oat brans. Cereal Chem 72 625-631.
Paton D 1987 Differential scanning calorimetry of oat
starch pastes. Cereal Chem 64 394-399.
Snedecor G W, Cochran W G 1967 Statistical method. The
Table 2 Changes of RVA measurements with enzyme treatments
Variety Treatment PV T BD FV SB TTPV PT
(RVU) (RVU) (RVU) (RVU) (RVU) (min) (C)
Buffer + 0 hr 461a 205b 256a 532b 327c 6.48c 66.9c
Mortlock Buffer + 1 hr 411b 219a 192b 756a 537b 7.43b 73.0b
-glucanase + 1 hr 261c 189c 72c 496c 307c 7.72a 90.1a
Proteinase + 1 hr 375b 182c 193b 765a 583a 7.33b 72.7b
Buffer + 0 hr 410a 282a 128a 669a 387a 7.39b 86.5b
Yarran Buffer + 1 hr 265b 243b 22c 474c 232c 8.57a 89.7a
-glucanase + 1 hr 232c 217c 15d 445d 228c 8.31a 90.0a
Proteinase + 1 hr 239c 186d 53b 488b 302b 7.58b 90.0a
PV: Peak viscosity; T: Trough; BD: Breakdown; FV: Final viscosity; SB: Setback; TTPV: Time to peak viscosity; PT: Pasting temperature.
Table 3 Changes of DSC measurements with enzyme treatments
Variety Treatment H1 Tp1 H2 Tp2 H3 Tp3
(J g-1) (C) (J g-1) (C) (J g-1) (C)
Buffer + Ambient 5.7a 50.3a 1.2a 81.0a 1.3b 104.1a
Mortlock Buffer + 40C 6.2a 51.5b 1.2a 81.8a 1.2b 104.5a
-glucanase + 40C 5.9a 59.5c 1.3a 87.5b 0.7a 109.2b
Proteinase + 40C 5.8a 52.1b 1.3a 81.7a 1.4b 105.6a
Buffer + Ambient 5.1a 46.4a 1.2a 81.3a 0.5a 104.7a
Yarran Buffer + 40C 5.8b 48.6b 1.2a 82.8b 0.7a 107.1b
-glucanase + 40C 6.1b 56.7c 2.0c 87.1c 0.4a 109.8c
Proteinase + 40C 5.7b 49.2b 1.6b 82.3ab 0.5a 105.5a
H1, H2 and H3 are the enthalpies of first, second and third transition process, respectively; Tp1, Tp2 and Tp3 are the peak temperatures
of first, second and third transition process, respectively.
Table 4 Thermal properties of twelve samples before and after processing
Before Processing After Processing
H1 Tp1 H2 Tp2 H3 Tp3 H1 Tp1 H2 Tp2 H3 Tp3
(J g-1) (C) (J g-1) (C) (J g-1) (C) (J g-1) (C) (J g-1) (C) (J g-1) (C)
Bimbil 5.0 50.5 1.0 81.0 0.4 104.1 3.8 49.0 0.8 83.1 1.1 105.7
Carrolup 4.9 50.4 1.1 81.7 1.2 104.5 4.1 50.3 0.4 83.8 0.8 103.7
Cooba 4.9 50.5 1.2 81.9 0.8 103.9 3.1 49.4 0.8 83.0 0.4 106.4
Echidna 5.3 50.9 1.0 81.9 1.0 103.7 3.2 49.8 0.6 82.4 1.1 102.9
Euro 5.2 51.2 1.1 81.6 2.0 104.0 4.6 48.7 0.6 84.4 / /
Mortlock 5.0 51.7 1.2 81.6 1.6 101.8 4.6 51.2 0.6 84.5 1.1 105.4
Pallinup 4.7 52.4 0.9 81.8 0.9 101.9 3.1 50.5 0.8 84.5 1.1 104.1
Yarran 4.7 49.0 1.2 82.8 1.6 103.6 3.4 48.2 0.0 84.0 / /
Cowra-Mortlock 6.2 49.1 1.1 81.2 0.8 104.5 3.4 49.2 0.7 82.0 0.8 104.3
Cowra-Yarran 5.8 46.8 1.5 83.2 0.6 104.3 3.6 45.9 0.6 84.5 0.8 103.1
Parkes-Mortlock 5.9 48.8 1.3 81.4 0.7 103.1 4.7 47.8 0.8 84.8 0.9 104.2
Parkes-Yarran 4.1 45.7 1.2 82.0 1.0 102.9 2.7 45.7 0.8 85.0 1.0 104.9
Average 5.1 49.8 1.1 81.8 1.0 103.5 3.7 48.8 0.7 83.8 0.9 104.5
Captions for Figures
Figure 1 RVA viscograms of Morlock and Yarran before ( ) and after ( ) soaking in buffer for 1 hr.
Figure 2 Changes in RVA viscograms of Morlock and Yarran before ( ) and after the addition of either -glucanase ( ) or proteinase (
Figure 3 Effect of enzymatic treatments on thermal properties.