The effect of thermal processing and storage on the physicochemical properties and in vitro digestibility of potatoes

Abstract

Shepody potatoes were cooked using three common cooking methods – microwaving, boiling and pressure cooking. Microwaving for 2.5 min retained the highest amounts of slowly digestible starch (SDS, 19.6%) and resistant starch (RS, 48.8%) as compared to the other cooking treatments. Similarly, enthalpy and FTIR results (ratio of 1047/1022 cm⁻¹) were also higher for microwaved samples, again indicating incomplete gelatinisation. Potatoes were also boiled for 15, 30 or 45 min, followed by 1, 3 or 7 days retrogradation at 23 or 4 °C. Retrogradation enthalpy increased significantly (P ≤ 0.05) with increased storage time and decreased storage temperature; FTIR results displayed temperature dependency; at 4 °C, ordered structure increased with increasing storage time, whereas the opposite trend was seen at 23 °C. Lastly, formation of SDS and RS was favoured for longer boiling times (30 – 45 min), extended storage times (3–7 days) and 4 °C.

Graphical Abstract

Food matrix, potatoes, processing effects, resistant starch, retrogradation, starch digestibility

Introduction

Potatoes are an important source of carbohydrate energy for much of the world due to their availability and versatility in cooking. In their uncooked state, potatoes are highly resistant to digestion; however, upon cooking the granular structure of the potato starch is destroyed, creating a large amount of highly digestible starches. In a cooked state, potatoes have a high glycemic index – often in the range of 80 to 150 (white bread used as reference) (Foster-Powell et al., 2002), meaning that they are rapidly digested resulting in a large amount of glucose entering into the blood stream in a relatively short amount of time. This in turn creates an insulin spike, which if chronically experienced can be a risk factor for various types of metabolic syndromes, such as diabetes or obesity. For this reason, potatoes are largely regarded as being an unhealthy food, especially given their predominance in the production of deep fried snacks, such as potato chips or French fries.

In an effort to quantify the digestibility of starch-based foods, several assays have been created which categorise starch into three distinct and generally accepted fractions. Based on the framework described by Englyst et al. (1992), starch digested within 20 min in vitro is termed rapidly digestible starch (RDS), which represents the starch digested in the stomach's portion of digestion. This fraction is rapidly hydrolysed by the body and expeditiously absorbed into the bloodstream, giving rise to the aforementioned increased in blood glucose levels. The next fraction, slowly digestible starch (SDS), is starch released between 20 and 120 min post-ingestion, or in vivo, the portion digested approximately in the small intestine. SDS gives a more sustained supply of energy and leads to a gradual increased in blood glucose levels, evading the large spike in insulin levels seen with RDS. Lastly, resistant starch (RS), starch which entirely eludes digestion in the small intestine and passes to the colon. As such, RS acts as a novel pre-biotic fibre (Fuentes-Zaragoza et al., 2011), and contributes to increased satiety which may help curb obesity rates.

Some potato varieties possess up to 87% RS, however, cooking causes irreversible granular swelling due to the starches gelatinising, and the RS content can drop to as low as 1.5% (García-Alonso & Goñi, 2000). To increase the digestive resistance of potatoes, a prolonged post-cook cooling step can be introduced to facilitate the process of retrogradation. During gelatinisation, which is a key pre-step to retrogradation, the starch loses much of its ordered structures, and the starch polymers are freed into solution. When retrogradation begins, free amylose chains coalesce relatively rapidly into hydrogen bond stabilised double helices, creating digestion-resistant crystallites which possess high thermal stability (Haralampu, 2000). Due to amylopectin's bulk, it does not contribute to rapid retrogradation, but instead forms crystallites over longer time periods in excess of days or weeks that contribute more significantly to retrogradation crystallite stability (Russell, 1987). Together these two polymers form a more crystalline structure which contributes to a lower digestion as this complex is now less accessible to enzymatic hydrolysis. Previously, it has been shown that a cooling treatment being performed on cooked potatoes results in an increased RS content (Mishra et al., 2008). However, retrogradation in the context of a food matrix is more complicated as there may be interference from other components, such as endogenous protein, lipid and other non-starch carbohydrates.

The objective of this work was to examine the effect of different processing methods on the thermal properties, amount of crystalline order and the digestible starch fractions of potatoes. Potato tubers were cooked using three cooking methods (pressure cooking, microwaving and boiling) for three different cooking times, resulting in three different qualities of cooked potatoes, that is undercooked, optimally cooked and overcooked, respectively, in order to better understand the physicochemical properties of cooked potatoes and their in vitro starch digestibility. This work also characterised the effect of starch retrogradation at different temperatures and times following different boiling treatments, with the same physicochemical properties as previously mentioned determined.

Materials and methods

Materials

Shepody variety potatoes (Solanum tuberosum) were obtained from Fredericton, New Brunswick in 2013 where they were grown under standard agronomic practices. All chemicals used were of reagent grade.

Cooking treatments

For all cooking treatments, three potato tubers were selected with an approximate size of 7.5 × 5.0 × 5.0 cm to ensure consistency. Freeze-dried material from the three tuber samples were pooled following grinding. All analyses were completed using this pooled dry matter.

Boiling

Washed potato tubers were placed in a glass vessel (VWR 190 × 100 mm crystallising dish) with 1 L of boiling water (95–100 °C), covered with aluminium foil and cooked on a hotplate (IKA Werke RCT Basic with and ETS-D5 temperature controller) for 15, 30 or 60 min.

Microwave cooking

Washed potato tubers were placed in the microwave (1100 W, Panasonic NN-S759, Panasonic Corp., Osaka, Japan) and cooked for 2.5, 5 or 10 min at 50% power, and an operating frequency of 2450 MHz.

Pressure cooking

Washed potato tubers were placed inside the pressure cooker (Model IP-LUX-60, Instant Pot Co., Ontario, Canada) on a steaming rack above 250 mL of distilled water. The potatoes were cooked using the “steam” setting (115–118 °C, 80 kPa) for 15, 30 or 60 min.

Immediately following cooking, the tubers were sliced into 1 cm thick slabs and frozen at −20 °C, followed by freeze drying (−40 °C, 0.1 mBar; FreeZone 12, Labconco, MO, USA) and manually ground to pass through a 250 μm sieve by mortar and pestle. Samples were identified as B15, B30 and B60 for those from potatoes boiled for 15 min, 30 min and 60 min respectively; M2.5, M5 and M10 for those microwaved 2.5 min, 5 min and 10 min respectively; P15, P30 and P60 for those pressure cooked 15 min, 30 min and 60 min respectively. All samples were stored in air-tight plastic bags until further use. Freeze-dried native potato tuber without any cooking treatment served as control after being ground to form native potato flour (NPF).

Storage treatments

Washed potato tubers were boiled as described above for 15, 30 or 45 min. Retrogradation was allowed to take place under either ambient (23 °C ± 1 °C) or refrigerator (4 °C ± 1 °C) conditions. In both cases, the cooked potato tubers were placed in a plastic bag and retrogradation was allowed to continue for 1, 3 or 7 days, followed by lyophilisation and grinding to pass through a 250 μm sieve. Samples were stored in air-tight plastic bags until further use.

Characterisation

Thermal properties

Thermal analyses for starch gelatinisation and retrogradation in potato flours were performed using a differential scanning calorimeter (DSC Q20; TA Instruments, New Castle, DE, USA) equipped with an RCS90 refrigeration system. Samples (12 mg, dwb) were weighed into high-volume pans and distilled water was added to make suspensions with 70% moisture content. Pans were hermetically sealed and equilibrated overnight at room temperature, and then scanned from 5 to 180 °C at a heating rate of 10 °C min⁻¹. The instrument was calibrated using indium and an empty pan as reference. The transition temperatures and enthalpy of each curve were determined using Universal Analysis 2000 v4.5A (TA Instruments, New Castle, DE, USA). All samples were subject to two determinations.

Short-range order

The short-range order of the samples was determined using Fourier-transform infrared spectroscopy (FTIR). Infrared spectra of samples were recorded on a Digilab FTS 7000 spectrometer (Digilab USA, Randolph, MA, USA) according to the method used previously (Chung et al., 2009a). The moisture content of starch was equilibrated to 14% in a desiccator containing a saturated solution of ammonium nitrate for fourteen days before FTIR analysis. The amplitudes of absorbance for each spectrum at 1047 and 1022 cm⁻¹ were measured, representing the respective amounts of crystalline and amorphous character, respectively, and their ratio taken. All samples were subject to three determinations.

In vitro starch digestibility

Rapidly digestible starch, SDS and RS were determined using the method of Englyst et al. (1992), with minor modifications (Chung et al., 2009b). Porcine pancreatic α-amylase (0.45 g, P-7545, Sigma, St. Louis, MO, USA) was suspended in 4 mL of distilled water and stirred for 10 min, after which it was centrifuged at 1500 g for 10 min. The supernatant (2.7 mL) was mixed with 0.32 mL amyloglucosidase (3300 U mL⁻¹), diluted to 0.40 mL and 0.20 mL of 10 mg mL⁻¹ invertase solution (I4505, Sigma). The enzyme solution was freshly prepared before each digestion. Potato flour samples (100 mg) were weighed into glass test tubes, along with fifteen glass beads (4 mm in diameter). Pepsin (10 mg, P7125, Sigma, St. Louis, MO, USA) and 2 mL of 0.05 m HCl solution containing 5 mg mL⁻¹ guar gum were added to each tube, followed by incubation in a longitudinally shaking water bath (37 °C, 200 strokes min⁻¹) for 30 min. Immediately after, 4 mL of 0.5 m sodium acetate buffer (pH 5.2, 20 mm CaCl₂) and 1 mL of the enzyme solution were added to each tube. Duplicate 100 μL aliquots were removed from the test tubes at 20 (G₂₀) and 120 min (G₁₂₀) and the glucose content determined spectrophotometrically at 510 nm by incubating with 3 mL of glucose-oxidase-peroxidase-reagent (GOPOD) for 20 min at 50 °C.

Total starch (TS) was determined using Megazyme Total Starch kit (K-TSTA, Megazyme, Bray, UK). Samples (100 mg) were placed into glass test tubes and dispersed in 0.2 mL of 80% ethanol, followed by 2 mL of 90% dimethyl sulphoxide to dissolve RS. Thermostable α-amylase (100 μL, 3000 U mL⁻¹; Megazyme) and 3 mL MOPS buffer (50 mm, pH 7.0) with 5 mm CaCl₂ were added to the tube and the tube was then incubated in a boiling water bath for 12 min. The tubes were then placed in a 50 °C water bath and, once cooled, 4 mL of sodium acetate buffer (200 mm, pH 4.5) and 100 μL of amyloglucosidase (3300 U mL⁻¹; Megazyme) were added and the sample incubated at 50 °C for 20 min. The contents of the tubes were then emptied into a 100 mL volumetric flask and brought to volume. An aliquot (1 mL) was removed and centrifuged at 1800 g for 10 min, after which a 100 μL aliquot was mixed with 3 mL of GOPOD and incubated at 50 °C for 30 min, followed by measuring the absorbance spectrophotometrically at 510 nm to determine the glucose content. All samples were subject to two determinations.

Starch nutritional fractions were classified as follows:

$$\%\text{RDS}=G_{20}\times 0.9$$

$$\%\text{SDS}=(G_{120}-G_{20})\times 0.9$$

$$\%\text{RS}=\%\text{TS}-(G_{120}\times 0.9)$$

Statistical analysis

All samples were subject to at least two determinations. Statistics (t-test and one-way analysis of variance) were conducted with SAS (version 9.1 for Windows, SAS Institute, Cary, NC, USA). When appropriate, the difference among means was determined with Tukey's multiple comparisons. Pearson correlation coefficients were determined to evaluate relationships between variables. Statistical significance was set at the 5% level of probability.

Results and discussion

Effect of cooking treatments

Thermal properties of flour obtained from cooked potatoes

Thermograms and endothermic enthalpy of freeze-dried native potato tuber and cooked potato samples are displayed in Fig. 1a and b respectively. NPF gave the largest enthalpic transition peak, followed by sample M2.5, which displayed a larger enthalpic transition peak and shifted to a higher peak temperature than the control (Fig. 1a). The latter result was likely due to the fact that the microwaving treatment was not long enough to fully gelatinise the starch in the tuber. As can be seen on Fig. 1a, subsequent samples microwaved for longer times (5 and 10 min) display a significantly reduced peak which was more consistent with other samples that were boiled or pressure-cooked (P ≤ 0.05). The remaining samples all displayed similar curves with a broad peak which shifted to a lower temperature (Fig. 1a), possibly caused by rapid retrogradation during cooling of the cooked tubers from cooking temperature to −20 °C (Singh et al., 2003). With the exception of NPF, M2.5 and M5, no significant differences were observed in T_o or T_p among cooked samples (P > 0.05).

Figure 1

Thermograms (a) and endothermic enthalpy (b) of freeze-dried potato samples cooked using various methods for varying amounts of time. Differing letters above bars at each individual cooking method denote significant differences determined by Tukey's HSD test following Anova (P ≤ 0.05) (n = 2); no significant difference was found between enthalpy of boiled and pressure-cooked samples at a given cooking time as determined by t-test; the values from NPF were used as reference.

As shown in Fig. 1b, it can be seen that the enthalpy for sample M2.5 was significantly higher than other samples, and is consistent with the enthalpy of a native flour being gelatinised for the first time (Liu et al., 2007), providing further evidence that this sample was simply not fully cooked compared to other samples. As the severity of microwaving increased, the enthalpy correspondingly decreased. This trend was also present in the pressure-cooked group and boiled group as well, with 15 min of cooking having a larger enthalpy than 30 or 60 min of cooking. No significant differences were seen in ΔH between pressure-cooked and boiled samples at a given cooking time, indicating that pressure cooking had no advantages over boiling for this size of tuber sample.

Short-range order in starch

FTIR was used to measure the ratio of absorbance at 1047 and 1022 cm⁻¹, which has been shown to be sensitive to the amount of crystalline to amorphous character of the starches respectively (van Soest et al., 1995). A representative FTIR spectrum of cooked potato sample is shown in Fig. 2a, and the short-range ordered structure in starch determined by the band ratio of 1047/1022 cm⁻¹ in FTIR spectra of samples from different cooking treatments is shown in Fig. 2b. The data presented in Fig. 2b displayed a nearly identical trend to that of the DSC enthalpy in Fig. 1b (r = 0.94). In samples subject to microwaving for longer amounts of time, the ratio of crystalline to amorphous character decreased as the amount of gelatinised starch increases, with the absorbance decreasing from 0.91 (M2.5) to 0.70 (M10) (P ≤ 0.05). Samples in the boiling and pressure-cooked groups were also significantly different from each other in their respective groups. Under-cooked samples showed higher amounts of ordered starch structure whereas over-cooked samples had lower amounts (P ≤ 0.05). Cooking, or gelatinising, of starch is well known to increase the amount of disorder of a system, as seen by an increase in absorbance at 1022 cm⁻¹ and a decrease seen at 1047 cm⁻¹ in all starches (Sevenou et al., 2002). No significant differences were seen between boiled and pressure-cooked methods at a given cooking time as determined by a t-test (P > 0.05), indicating that the respective boiling and pressure-cooking treatments essentially cooked the potatoes to the same degree, which was also evident in the enthalpic data from DSC (Fig. 1b).

Figure 2

A representative FTIR spectrum of cooked potato sample (a), and effect of cooking type and length on the ratio of 1047/1022 cm⁻¹ (b). Numbers above cooking methods indicate cooking time. Differing letters above bars at each individual cooking method denote significant differences determined by Anova following Tukey's HSD test (P ≤ 0.05) (n = 2); no significant difference was found between the ratio in boiled and pressure-cooked samples at a given cooking time as determined by t-test; the values from NPF were used as reference.

Digestible starch fractions in cooked potatoes

The effect of cooking treatment on in vitro starch digestibility of potato samples is shown in Table 1. Native Shepody potatoes had a RS content of 57.2%, an SDS content of 25.4%, and an abnormally high RDS content of 17.4%. Previous work with Shepody potatoes indicated a much lower RDS content of approximately 5% RDS (Lu et al., 2012), however this could be attributed to the potatoes being stored for an extended period of time prior to analysis, or different environmental conditions for growth despite the use of the same variety. Potatoes stored at low refrigerator temperatures (4 °C) have been shown to have a higher content of reducing sugars after storage for 3 month, as compared to potatoes stored at other, elevated temperatures (9 and 11 °C) (Coffin et al., 1987). This increase in reducing sugar content can skew assay results, artificially increasing the RDS content of the potatoes being tested.

Table 1

Digestible starch fractions of potato tubers subjected to various processing treatments

Cooking conditions	RDS (%)	SDS (%)	RS (%)
Control NPF	17.4 ± 0.6	25.4 ± 2.6	57.2 ± 2.1
Microwaving
2.5 min	31.6 ± 1.4c	19.6 ± 0.1a	48.8 ± 1.4a
5 min	76.6 ± 0.2b	10.6 ± 1.4b	12.8 ± 1.5b
10 min	88.4 ± 1.2a	6.2 ± 1.0c	5.4 ± 0.5c
Pressure cooking
15 min	83.4 ± 0.7b	8.9 ± 0.5a	7.7 ± 0.6a
30 min	88.7 ± 1.1a	4.8 ± 1.2b	6.5 ± 0.1a
60 min	86.2 ± 0.6a	6.4 ± 0.2b	7.4 ± 0.8a
Boiling
15 min	81.5 ± 1.0c	9.9 ± 0.6a	8.6 ± 0.4a
30 min	83.8 ± 0.0b**	6.6 ± 0.1b*	7.9 ± 0.1b*
60 min	86.4 ± 0.4a	6.4 ± 0.3b	7.2 ± 0.7b

Data are mean ± SD; n = 4 (two replicates, two determinations/replicate). Values in a same column for a given cooking method followed by different letters are significantly different as determined using Tukey's HSD test following Anova (P ≤ 0.05); * and ** indicate significant differences at P ≤ 0.05 and 0.01, respectively, between boiling method and pressure-cooking method at a given cooking time as determined by t-test; The values from NPF were used as reference.

Regardless of cooking methods, RDS amounts in cooked potatoes significantly increased whereas SDS and RS significantly decreased with extended cooking time (P ≤ 0.05, Table 1). With the exception of microwaving, there were no significant differences in RDS, SDS or RS between optimally cooked and over-cooked potatoes, indicating that starch gelatinisation reached its peak under optimal cooking conditions. Sample of M2.5 had the highest RS and SDS contents at 48.8% and 19.6% respectively (P ≤ 0.05). With increasing microwaving time, RS and SDS content significantly dropped, with RDS conversely increasing. The higher RS and SDS in M2.5 and, to a lesser extent, M5, is likely due to the cooking process not fully gelatinising all of the starch in the potato thus leaving more granules intact and in their native form. Microwaving potatoes has also been seen to create a more dense structure with smaller pores than boiled potatoes, which could limit amylolysis of the tubers (Mulinacci et al., 2008). Both boiling treatments of 30 or 60 min essentially completely cooked the potatoes to the same degree, resulting in RS contents of 7.9% and 7.2%, respectively, with the RDS content of B60 increasing to reflect the decrease in RS. This is consistent with previous data for boiled potatoes, which regardless of potato type (floury, waxy or general purpose) had an RS content of 5.8% to 10.5% (Monro et al., 2009). For pressure cooking, SDS was maximised with 15 min of cooking to a level of 8.9%. It is likely that the pressure-cooking treatment also gelatinised the starches to a similar degree, as all times used for pressure cooking had RS contents ranging from 6.5% to 7.7%, with no significant differences among P30 and P60 samples. These results were to be expected, given that it has been shown that the degree of gelatinisation, as evident from Fig. 1b, is related to the in vitro starch digestibility of potato starch (Parada & Aguilera, 2009). Furthermore, significant differences were seen on RDS (P ≤ 0.01), SDS (P ≤ 0.05) and RS (P ≤ 0.05) amounts between B30 and P30 samples (Table 1), showing that pressure cooking produced significant more amount of rapid digestible starch than boiling method, even though no significant differences were seen on DSC and FTIR results (Figs 1b and 2b).

Effect of storage conditions

Thermal properties of cooked and stored potato samples

Potato tubers were boiled for 15, 30 or 45 min and then separated into two groups: one to be retrograded at 4 °C and the other at 23 °C (Table 2). In regard to the thermal parameters, T_o and T_p were largely similar between samples of both differing storage and boiling times, with few, if any, significant differences present. Some significant differences were, however, seen in T_c for samples boiled for 30 and 45 min, and subsequently stored for longer periods and between different storage temperatures (23 °C vs. 4 °C) (P ≤ 0.05). More differences were seen when evaluating the effect of retrogradation temperature. The samples stored at 4 °C developed narrower retrogradation peaks than those at 23 °C, as indicated by lower T_c values at 4 °C in most cases. Although few differences were seen in gelatinisation temperatures, ΔH proved to be more interesting, as most of the samples retrograded at 4 °C were significantly higher than their 23 °C counterparts (P ≤ 0.05), and enthalpy generally increased with the extent of boiling in samples stored either at 23 °C or at 4 °C (P ≤ 0.05). Notably, samples boiled for 45 min and stored at 4 °C, the enthalpy increased significantly with the storage time, with the highest ΔH value seen at 7 days of storage at 4 °C (P ≤ 0.05). This result was well elaborated in a previous study (Kim et al., 1997a).

Table 2

Thermal properties of potato samples subjected to retrogradation for 1, 3 or 7 days at room temperature (23 °C) or refrigerator conditions (4 °C) after boiling for 15, 30 or 45 min

Storage conditions	23 °C				4 °C
	To (°C)	Tp (°C)	Tc (°C)	ΔH (J g−1)	To (°C)	Tp (°C)	Tc (°C)	ΔH (J g−1)
15 min
1	48.1 ± 0.1aA	61.4 ± 0.1aB	83.8 ± 0.4aA	6.7 ± 0.1aB	48.0 ± 0.2aA	63.1 ± 0.6aA	83.4 ± 1.9aA	6.9 ± 0.1aA
3	47.6 ± 0.4aA	62.1 ± 0.5aA	86.1 ± 2.1aA	6.7 ± 0.1aB	48.1 ± 0.1aA	62.9 ± 0.7aA	86.5 ± 0.1aA	7.0 ± 0.1aB*
7	47.8 ± 0.3aA	62.2 ± 0.4aA	86.7 ± 0.2aA	6.6 ± 0.1aB	48.2 ± 0.2aA	63.0 ± 0.9aA	83.3 ± 0.5aB*	7.1 ± 0.1aC*
30 min
1	47.9 ± 0.1aA	61.2 ± 0.0aB	82.7 ± 0.1cA	7.0 ± 0.1aA	48.4 ± 0.3aA	63.7 ± 0.4aA**	85.6 ± 0.0aA***	7.3 ± 0.0aA*
3	48.3 ± 0.8aA	62.1 ± 1.1aA	87.8 ± 0.3aA	7.0 ± 0.1aA	47.2 ± 0.4aA	63.1 ± 0.4aA	83.7 ± 0.6bB*	7.5 ± 0.2aAB
7	47.7 ± 0.6aA	62.8 ± 0.9aA	85.7 ± 0.6bA	6.8 ± 0.1aAB	48.4 ± 0.1aA	64.5 ± 0.9aA	85.3 ± 0.2aA	7.4 ± 0.1aB*
45 min
1	47.7 ± 0.4aA	61.9 ± 0.0aA	84.1 ± 0.7bA	7.1 ± 0.0aA	47.9 ± 0.5aA	63.8 ± 0.7aA	85.9 ± 0.6aA	7.2 ± 0.2bA
3	47.7 ± 0.0aA	62.0 ± 0.2aA	87.1 ± 0.0aA	7.1 ± 0.0aA	47.1 ± 0.2aA*	63.9 ± 0.4aA*	84.1 ± 0.0bB	8.1 ± 0.2aA*
7	48.5 ± 0.7aA	63.3 ± 1.4aA	87.3 ± 0.2aA	7.2 ± 0.2aA	47.7 ± 0.2aA	63.6 ± 0.1aA	84.4 ± 0.0bAB**	8.2 ± 0.0aA*

Data are mean ± SD, n = 2. Values in a same column for a given boiling time followed by different lowercase letters differed significantly; values in a same column at a given storage period followed by different uppercase letters differed significantly, both determined by Tukey's HSD test following Anova (P ≤ 0.05). Values of samples stored at 4 °C followed by *, **, and *** indicate significant differences at P ≤ 0.05, 0.01 and 0.001, respectively, from its 23 °C counterpart at a given boiling time and storage period as determined by t-test.

Effect of storage on short-range order in starch

Figure 3 shows the short-range ordered structure in starch of potato samples determined by the ratio of absorbance at 1047 cm⁻¹ and 1022 cm⁻¹ in FTIR spectra. Samples boiled for 30 or 45 min had their ratio increased with storage time (P ≤ 0.05), but only at 4 °C. This trend was present in the enthalpy of the same samples as well (Table 2) and is consistent with previous literature which demonstrated that lower temperatures (i.e. 4 or 10 °C) are more conducive to nucleation and crystallite propagation in potato starches (Kim et al., 1997b). This would explain why no significant differences were seen in samples retrograded at 23 °C. Additionally, short-range order increased in samples retrograded for the same amount of time at 4 °C that were boiled for longer amounts of time, with B45 7D having a ratio of 0.82, whereas B15 7D had a ratio of 0.78; no significant differences were observed in the 2 °C samples again. The increased treatment may have helped to disrupt the majority of the native structures, such as starch granules or protein networks enclosing them (Choi et al., 2008), and therefore, allowing the amylose and amylopectin more freedom of movement.

Figure 3

Effect of storage length and temperature on the ratio of 1047/1022 cm⁻¹ in starch for potato tubers boiled for 15, 30 or 45 min. Differing lowercase letters above bars at a given boiling time denote significant differences; different uppercase letters above bars at a given storage temperature and days differed significantly, both determined by Tukey's HSD test following Anova (P ≤ 0.05) (n = 2). * above a bar stored at 4 °C indicates significant difference at P ≤ 0.05 from its 23 °C counterpart at a given boiling time and storage period as determined by t-test.

Effect of storage on in vitro starch digestibility

Table 3 displays the results of the Englyst assay for samples boiled and stored for 1, 3 or 7 days at 4 or 23 °C. Overall, samples stored at 4 °C and extended stored period tended to have lower RDS amounts than their 23 °C counterparts, particularly after 3 days retrogradation, but overall RDS levels decreased as retrogradation time increased, regardless of temperature (P ≤ 0.05). This is likely due to the starch fractions more easily coalescing into a more crystalline state at 4 °C over longer retrogradation periods, leading to lower RDS. When examining the SDS and RS content of these samples, only few significant differences were observed. There was no clear trend that temperature had either a positive or negative effect on SDS or RS content. The increased length of time for retrogradation to occur could also allow for more interaction between different components in the potatoes, such as proteins, other carbohydrates, etc., which could aid or hinder in the creation of structures that may alter the in vitro starch digestibility (Escarpa et al., 1997). Few significant differences were seen between samples when either boiling time or retrogradation time was considered and may in part be due to the fact that when initially cooked, only minor, albeit significant, differences were seen in the enthalpy (Fig. 1b), indicating that all the samples had achieved a similar level of granular disruption following the treatment. However, it is clear that boiling for 45 min followed by 7 days retrogradation produced more SDS and RS, as well as less RDS, than those boiled or stored for a shorter period of time. This result likely stems from the extensive starch gelatinisation accompanying prolonged boiling, which in turn led to more extensive retrogradation during storage, as is evident in the enthalpy of samples B45 7D for both retrogradation temperatures (Table 2).

Table 3

Digestible starch fractions of potato samples boiled (15, 30 45 min) and stored at either room temperature (23 °C) or at refrigerator temperatures (4 °C) for 1, 3 or 7 days

Storage conditions	23 °C			4 °C
	RDS (%)	SDS (%)	RS (%)	RDS (%)	SDS (%)	RS (%)
15 min
1	82.4 ± 0.4aB	9.3 ± 0.3aA	8.3 ± 0.1aA	85.5 ± 0.3aA*	7.5 ± 0.2aA*	7.0 ± 0.5bA
3	82.7 ± 0.0aA	7.1 ± 1.4aA	10.2 ± 1.4aA	80.3 ± 0.6bA*	10.6 ± 1.4aA	9.0 ± 0.8abA
7	79.6 ± 0.4bB	10.0 ± 0.7aA	10.4 ± 0.3aA	81.5 ± 0.5bA	8.7 ± 0.6aA	9.8 ± 0.0aA
30 min
1	83.9 ± 0.7aAB	7.2 ± 1.5aA	8.9 ± 0.8aA	83.5 ± 0.1aB	8.7 ± 0.6bA	7.8 ± 0.7bA
3	83.0 ± 0.1aA	8.9 ± 0.9aA	8.2 ± 0.8aA	81.8 ± 0.4abA*	8.8 ± 0.5bA	9.5 ± 0.9aA
7	84.1 ± 0.5aA	6.6 ± 0.1aB	9.4 ± 0.4aA	79.7 ± 1.0bA*	10.8 ± 1.1aA*	9.5 ± 0.1aB
45 min
1	85.6 ± 0.3aA	7.0 ± 0.5bA	7.4 ± 0.2bA	84.5 ± 0.2aA*	8.7 ± 0.3bA*	6.7 ± 0.1cA*
3	82.5 ± 0.3bA	9.2 ± 0.4aA	8.3 ± 0.1abA	81.6 ± 0.3bA*	11.0 ± 0.5aA*	7.4 ± 0.2bA*
7	80.6 ± 0.3cB	10.1 ± 0.1aA	9.4 ± 0.4aA	79.5 ± 0.4cA	10.6 ± 0.4aA	9.9 ± 0.1aA*

Data are mean ± SD. n = 4 (tworeplicates, two determinations/replicate); Values in a same column at a given boiling time followed by different lowercase letters differed significantly; values in a same column at a given storage period followed by different uppercase letters differed significantly, both determined by Tukey's HSD test following Anova (P ≤ 0.05); values of samples stored at 4 °C followed by * indicate significant differences at P ≤ 0.05 from its 23 °C counterpart at a given boiling time and storage period as determined by t-test.

The effects of retrogradation were quite pronounced when comparing the RS content of retrograded and non-retrograded samples. Sample B30 had a RS content of 7.9%, and samples B30 1D, B30 3D and B30 7D had RS contents of 7.8%, 9.5% and 9.5%, respectively, for samples retrograded at 4 °C. SDS was also affected by storage conditions. The B30 control (Table 2) contained 6.6% SDS, and all retrograded samples stored at 4 °C had an SDS content ranging from 8.7% to 10.8%. Both RS and SDS significantly increased with increasing storage time (P ≤ 0.05), which is consistent with literature. Mishra et al. (2008) studied the effect of a cooling treatment (4 °C, 2 days) and found that both SDS and RS increased significantly, and with significant decreases seen in RDS, and similar results were seen by others as well (Kingman & Englyst, 1994; García-Alonso & Goñi, 2000; Monro et al., 2009). Although all of the aforementioned papers assayed the in vitro digestibility of potatoes, significant differences remain among the studies. Many of these differences, however, can be attributed to the length of the boiling treatment used, pre-cooling potato tuber preparation, as well as the variety of potato used.

In the case of Mishra et al. (2008), a waxy-type potato was used, whereas Monro et al. (2009) examined waxy, floury and general purpose potatoes. The extent of conversion of RDS to SDS and/or RS by the use of retrogradation was markedly different, with different varieties responding differently to the treatment and could be attributed to differences in the ratio between amylose and amylopectin between varieties. In addition, the differences in their respective branch chain sizes, which will heavily influence the quantities of RDS, SDS and RS produced could also be a contributor.

Conclusion

In all cases, an increase in cooking time caused an increase in the in vitro starch digestibility of microwaved, pressure-cooked or boiled samples, with SDS and RS generally decreasing. Both thermal and FTIR data indicated that as the length of cooking increased, the amount of ordered structure in the system decreased. Stored cooked potatoes at 4 °C for extended time (i.e. 7 days) significantly enhanced starch retrogradation and increased retrogradation enthalpy and the short-range ordered structure in starch, which well explained the significant increase of SDS and/or RS amount in cooked and stored potatoes. This study also indicated that boiling or pressure-cooking potatoes for equal amounts of time, namely 15 or 60 min, are essentially equal cooking treatments, as indicated by non-significant differences in their relative amounts of RDS, SDS and RS, enthalpy, and ratio 1047/1022 cm⁻¹. Further research should be conducted to explore the effects of boiling/refrigeration storage cycles on the in vitro starch digestibility of cooked potatoes.

Conflict of interest

The authors declare no conflict of interest.