Effects of acetylation, acid-thinning and oxidation on Chrysophyllum albidum (African star apple) kernel native starch

Starch is a biological macromolecule with myriad of industrial uses. The growing need for starch and native starch deficiencies have necessitated researches into under-exploited starch sources and modifications, respectively. Chrysophyllum albidum (African star apple, ASA) seed is a waste consisting of 63.94% carbohydrate in its kernel. Its starch content can be fully utilised, industrially, if it is in a suitable form and its properties understood. That necessitated the isolation of ASA kernel starch, its modification and properties assessment. Acetylation, acid-thinning and oxidation were performed on the native Chrysophyllum albidum kernel starch, NACA, to produce the acetylated Chrysophyllum albidum kernel starch, ACCA, acid-thinned Chrysophyllum albidum kernel starch, ATCA, and oxidised Chrysophyllum albidum kernel starch, OXCA, respectively. The physicochemical properties of the native and modified starches were investigated with established methods. The yield of starch was 43.76%. The acetylated and oxidised starches had low degrees of substitution. FTIR results confirmed the introduction of carbonyl functionality into the acetylated and oxidised starches. Starch morphology revealed smooth, small rounded and truncated ellipsoid granules with diameter ranges of 7–20 μm. Swelling power of modified starches improved, except for the acid-thinned, and increased with temperature. Acetylated and oxidised starches had significantly (p < 0.05) higher water and oil absorption capacities, respectively. Acid-thinning significantly improved starch pasting properties and reduced retrogradation tendency. Modification improved the gelatinisation transition temperature of native starch.


Introduction
Starch is a biopolymer next to cellulose and chitin in abundance (Tharanathan 2005). It is composed majorly of two macromolecules-amylose and amylopectin, occurring in 20-30 and 70-80% proportions, respectively. The two proportions vary depending on the plant botanical origin, cultivar, and plant level of maturity. Starch is found in plant fruits, grains, tubers, seeds, roots, and stems (Jyothi et al. 2010;Tharanathan 2005). Its application spans the food, pharmaceutical, cosmetic, textile, construction, and paper industries (Jyothi et al. 2010;Tharanathan 2005). In its native form however, starch is characterised by deficiencies, which limit its use optimally in the industry. Among these deficiencies are less freeze-thaw stability thus retrogradation, high gelatinisation temperatures and high gel turbidity, narrow peak viscosity range, lack of clarity, susceptibility to precipitation of cooled paste, poor process tolerance, weak cohesive and rubbery paste upon heating (Deka and Sit 2016;Shah et al. 2017). These deficiencies require starch quality improvement, which is achieved by modification. The most prominent starch modification methods are chemical derivatisation, which introduce alteration of functionalities on the starch chain. Oxidation, acetylation, acid-thinning and hydropropylation have improved starch qualities (Adebowale et al. 2002;Lawal 2009;Omojola et al. 2011). Oxidation leads to conversion of hydroxyl groups on starch glucose units to carbonyl and carboxylic functionalities and starch chain scission. The resultant effects are enhanced starch solubility as a result of depolymerisation, starch stability due to the carboxylic group and reduced setback viscosity (El Halal et al. 2015;Isah et al. 2016;Olayinka et al. 2013). Starch acetylation is commonly achieved using an 1 3 anhydride to make a starch ester. The introduction of bulky acetyl group, for instance, increases water permeation and thus swelling power and solubility (Siroha et al. 2019). Acidthinning of starch improves starch pasting properties and resistance to retrogradation (Oderinde et al. 2020).
Chrysophyllum albidum, Linn (African star apple), is a tropical evergreen tree locally referred to in South Western Nigeria as ''agbalumo'' and "udara" in the Eastern part. The fruit is a berry having five flattened glossy seeds (Ajayi and Ifedi 2015) or lesser in certain instances probably due to abortion (Oyelade et al. 2005). The seeds are used for local games and discarded (Emudainohwo et al. 2015) and usually litter (Akin-osanaiye et al. 2018). The pericarp of the seed shell has been reported to contain about 20.86% starch (Ibrahim et al. 2017), while the seed flour contains 63.94% carbohydrate (Ajayi and Ifedi 2015). C. albidum seed could be a veritable industrial raw material if its starch possesses exceptional physicochemical properties. This is more important in a developing country, like Nigeria, in which the conventional starch sources are major staple foods. Researches are therefore needed to be directed towards unearthing unconventional starch sources (Akinterinwa et al. 2014) to stem potential food insecurity in the nearest future. The aim of this work therefore was to extract African star apple kernel starch, derivatise the starch by oxidation, acetylation, and acid-thinning and determine the properties of the modified starches.

Materials
Ripe African star apple fruits were procured from Oja-Odan Market in Yewa North Local Government Area, Ogun State, Nigeria. The seeds were extracted from the berries, screened of immature ones and adhering pulp was rinsed off with tap water. The cleaned seeds were dried at 40 ± 2 °C for five days in forced-air oven (OV/125, Genlab Limited, Cheshire, England). The dried seeds were cracked open, the shells discarded, the kernels milled into flour with a grinder (Marlex Excella, KIL, Daman, India) and stored in polythene bag for later use.

Starch extraction
The extraction of starch from African star apple kernel flour was performed using the modified  method. African star apple kernel flour was defatted by the cold extraction process using n-Hexane for 72 h with intermittent shaking. A 25% w/v defatted kernel flour suspension was made in 4 L of distilled water. The suspension pH was adjusted to 8.0 with sodium hydroxide solution (0.05 M) stirring continuously for 4 h at ambient temperature. It was centrifuged for 30 min at 4500 rpm (Rotanta 460 R, Hettich GmbH & Co. KG, Tuttlingen, Germany)). The pellet was dispersed in distilled water (4000 mL), muslin cloth was used to screen the starch, and the suspension centrifuged at 4500 rpm for 30 min. The starch pellet was washed two times and dried at room temperature for 48 h to yield the native African star apple starch, NACA.

Starch modification
Starch oxidation, acetylation and acid-thinning were done by the method of Lawal (2004) using sodium hypochlorite, acetic anhydride and dilute hydrochloric acid, respectively. The oxidised, acetylated and acid-thinned starches were labelled OXCA, ACCA, and ATCA, respectively.

Starch oxidation
A 10% NACA suspension was prepared in distilled water and adjusted to pH 9.5 using 2 M sodium hydroxide. Sodium hypochlorite (10 g) was added to the suspension over a period of 30 min at a pH range of 9-9.5 with continuous stirring. The reaction was allowed to proceed for additional 10 min after adding the hypochlorite. The pH of the reaction medium was adjusted to 7 with 1 M hydrochloric acid solution. The starch suspension was centrifuged, washed four times with distilled water with centrifugation in-between washings. The oxidised starch pellet was dried for 2 days in the drying cabinet and kept in air-tight polythene bag for analyses.

Starch acetylation
A 100-g weight of NACA was dispersed in 500 mL of distilled water and stirred for 20 min, while the pH was adjusted to pH 8.0 using 1 M NaOH. 10 g of (CH 3 CO) 2 O was added to the mixture for 1 h while maintaining a pH range of 8.0-8.5. The reaction proceeded for 5 min after the addition of (CH 3 CO) 2 O. The pH of the slurry was adjusted to 4.5 using 0.5 M HCl. The mixture was centrifuged for 30 min at 4500 rpm. The starch pellet was washed four times with distilled water and air-dried at 30 ± 2 °C for 48 h and kept in air-tight polythene bag for analyses.

Starch acid-thinning
A 100-g weight of NACA was suspended in 500 mL of 0.15 M HCl and stirred for 8 h at 50 °C. The modified starch was centrifuged for 10 min at 4500 rpm. The starch pellet was washed four times with distilled water with centrifugation in-between washings. The acid-thinned starch was airdried for 2 days in the drying cabinet and kept in air-tight polythene bag for analyses.

Chemical composition
The method of Nwinuka et al. (2005) was used for the determination of chemical composition of the native and modified starches.

Acetyl content
The per cent acetyl content of ACCA and the degree of substitution were determined as described by . A 5 g of sample in 50-mL water with the addition of phenolphthalein indicator was titrated to a permanent pink end point with 0.1 M sodium hydroxide. Upon the completion of titration, 25 mL of 0.45 M NaOH was added to the flask, sealed and vigorously shaken for 30 min for saponification. The saponified mixture was titrated with 0.2 M hydrochloric acid until the phenolphthalein colour disappeared. The native starch was similarly treated as a blank. The acetyl content (%) and the degree of substitution, DS, are calculated using Eqns. (1) and (2), respectively.

Carboxyl and carbonyl contents
The carboxyl and carbonyl contents of OXCA were determined by the method of Lawal (2004). To determine the carboxyl content, 5 g of oxidised starch sample slurried in 25 mL of 0.1 M HCl was stirred and filtered after 40 min. (1) The residue was washed free of chloride, heated on a steam bath to gelatinise and titrated with 0.1 M sodium hydroxide to a phenolphthalein end point. The native starch was similarly treated as a blank. The carboxyl group content of OXCA is calculated using Eq. (3) with NACA serving as blank where CX = carboxyl content; OT = OXCA titre; CT = NACA titre; SM = Mass of sample The carbonyl content was determined using the hydroxylamine method. Oxidised starch suspension (2 g in 100 mL distilled water) was gelatinised by heating in a water bath at 90 ± 2 °C and then cooled to 40 °C. The pH was adjusted to 3.2. Hydroxylamine reagent (15 mL) was added. (The hydroxylamine reagent: 12.5 g of reagent grade hydroxylamine hydrochloride in water + 0.5 M sodium hydroxide (50 mL). The solution was made to 250 mL with distilled water.) The starch sample was covered with aluminium foil and placed in a water bath at 40 °C. After 4 h, the reaction mixture was rapidly titrated to a pH value of 3.2 with 0.1 M hydrochloric acid to determine the excess hydroxylamine. The carbonyl content of OXCA is calculated using Eq. (4) with NACA used as blank where CN = carbonyl content; OT = OXCA titre; CT = NACA titre; SM = Sample mass

Granule morphology
The scanning electron microscopy of the samples was performed on Phenom Pro Desktop SEM (Thermo Fisher Scientific) at an acceleration voltage of 25 kV. All of the samples were observed.

Swelling power
Swelling power of starch samples were done at a temperature range of 55-95 °C, as reported by Lawal (2004). Starch sample (0.1 g) was accurately weighed into a clean, dried, test tube of weight (M 1 ). Distilled water was added to the test tube, and agitated on an orbit shaker for 30 s. The suspension was heated for 30 min at 55 to 95 °C (a temperature at a time) on a thermostated water bath. The mixture was cooled to room temperature and centrifuged for 15 min at 4500 rpm. The test tube and wet starch pellet were weighed as M 2 . Starch swelling power (g/100 g) is calculated using Eq. (5) where M 1 and M 2 are as defined.

Water and oil absorption capacities
Water and oil absorption capacities of starches were determined by the method of Lawal (2005). Starch sample (1 g) (M) was weighed into a clean, dry test tube. 10 mL of distilled water (d = 1.0 g/mL −1 ) or oil (Power Oil, Nigeria; d = 0.98 gmL −1 ) was added. The suspension was mixed thoroughly for 30 s and allowed to stand for 30 min. The volume of water or oil supernatant was measured (V). The mass of water or oil absorbed was expressed as g/100 g starch on a dry weight basis using Eq. (6).

Pasting properties
Pasting properties of the starches were determined using a Rapid Visco Analyser (RVA-4 Newport Scientific, Warriewood, NSW, Australia) according to the method of Lawal (2011) with modifications. A 3.0-g sample of starch, corrected to 14% moisture, was dispersed in 25-mL distilled water. A programmed heating and cooling cycle was employed at constant shear rate, where the sample was held at 50 °C for 1 min, heated to 95 °C in 7.5 min and then held at 95 °C for 5 min. It was subsequently cooled to 50 °C within 3 min and then held at 50 °C for 2 min. The rotation speed is maintained at 960 rpm for 10 s and then at 160 rpm throughout the remainder of the process. The analysis was done in duplicate.

Gelatinisation properties
Differential scanning calorimetry measurements were performed in a DSC3 (Mettler Toledo Co., Switzerland) as described by Xu et al. (2017). Starch suspension (33.3% w/v) was prepared with distilled water. Accurately weighed sample (± 0.1 mg) was sealed in an aluminium sample pan. It was equilibrated for 24-h ambient temperature. The DSC curves were performed under dynamic nitrogen atmosphere (50 mL min −1 ) and heating rates of 10 °C min −1 . An empty sample holder was used as reference and the runs were performed by heating the samples from 30 up to 120 °C. The onset temperature (To), peak temperature (Tp), conclusion temperature (Tc) and gelatinisation enthalpy (ΔH) were obtained by data analysis software STARe Software 16.10.

Statistical analyses
The analyses were performed in triplicate and expressed as means ± standard deviation of measurements. The statistical evaluations of significant differences were performed with analysis of variance using SigmaPlot for Windows 14.0 (Systat Software, Inc.).  (Xiao et al. 2011). Recovered starches after modification of native African star apple kernel starch were not significantly different. Their differences could be attributed to the different effects of modification types and washing ) as performed during acetylation and oxidation. The moisture, crude protein and fat contents of starches ranged from 3.95 ± 0.01-5.45 ± 0.00%, 0.30 ± 0.17-0.56 ± 0.44, and 0.13 ± 0.12-0.30 ± 2.41%, respectively. The differences in per cent moisture, crude protein and crude fat of native and modified kernel starches were significant (p < 0.05), with the native starch having significantly higher values than the modified. The moisture contents of modified African star apple starches were not significantly different except for OXCA (3.95 ± 0.01%). ATCA, however, had the highest moisture content of 4.91 ± 0.02%. Differences in moisture contents can be attributed to differences in the nature of starches, the durations and the methods of drying of isolated starches. Variations in the crude protein contents of starches cut across both native and modified starch forms. Crude fat of native and modified African star apple kernel starches ranged from 0.13 ± 0.12 to 0.72 ± 0.08%. Significant differences exist among the different starches except for ACCA (0.20 ± 0.86%) and OXCA (0.18 ± 0.21%). The reductions in chemical constituents of starches following modifications are attributable to the abilities of the modifying reagents to degrade and erode, and loss of degraded molecules with wash water. The low protein and fat contents are measures of the starch purities.

Physicochemical properties
Acetylated African star apple kernel starch, ACCA, had an acetyl content of 1.61 ± 1.83% (Table 1), while the oxidised African star apple kernel starch, OXCA, had carbonyl, and carboxyl contents of 0.28 ± 0.48 and 0.34 ± 0.93%, respectively. The values obtained showed low degrees of substitution which were within recommended limits (FAO 2018). The acetyl contents were similar to those obtained by Akhter et al. for Indian horse chestnut (Akhter et al. 2015). The values obtained for carboxyl and carbonyl contents were similar to those obtained for oxidised jack bean  but higher than that of breadfruit ).

Surface morphology
The results of the scanning electron microscopy of African star apple kernel starches (Fig. 1) revealed starch granules having shapes ranging from small rounded to truncated ellipsoid granules. They were about 7-20 μm in diameter with smooth surfaces. Modifications imparted no significant differences in the surface morphologies of African star apple starches. Retained surface and structural morphologies can be attributed to less severe conditions of modification (El Halal et al. 2015) as confirmed by the degrees of substitutions of acetyl, carboxyl and carbonyl in the native starch. Moin et al. (2017) reported granule fissures, surface erosion and clumping together of starch granules following acid treatment of Irri and Basmati rice starches. Figure 2 presents results of the Fourier-transform infrared spectroscopy, FTIR, of African star apple kernel starches. Prominent absorption bands characterising the native and modified starches were observed around 3285-3378, 2922-2933, 1728-1732, 1634-1648, 1148-1153, 1066-1077, 923-997 and 849-853 cm −1 regions. Similarities exist in the regions of the appearance of some prominent bands. In the 3700-3000 cm −1 regions, broad bands due to hydrogen-bonded hydroxyl (-OH) stretching were NACA ACCA OXCA ATCA

Effect of temperature on swelling power
The results of the effect of temperature on swelling power of native and modified African star apple kernel starches are represented by Fig. 3. Starch swelling measures the extent of water uptake during gelatinisation of starch. Figure 3 shows that temperature variations affected the swelling power of African star apple kernel starches. The swelling power of native and modified starches had positive correlations with temperature variations, increasing gradually with temperature increase from 55-95 °C. The increase in swelling power with increasing temperature is attributable to hydrogen bond cleavage with resultant-free hydroxyl groups and gradual water absorption. Among African star apple kernel starches, on the average, OXCA had a significantly (p < 0.05) higher swelling power (1245.97 g/100 g), followed by ACCA (1077.09 g/100 g) and NACA (842.57 g/100 g), in that order. The least swelling power was observed in ATCA (263.61 g/100 g). Starch acetylation introduced bulky acetyl groups into ACCA chain which disrupted the regularity in structure thus facilitating entrance of water into the starch granules (Afolabi et al. 2012), while oxidation of African star apple starch introduced carboxyl and carbonyl functionalities which increased weakening of hydrogen bonding and thus percolation of water molecules (Pietrzyk et al. 2014). The observed lower swelling power of ATCA can  Figure 4 represents the results of the water and oil absorption capacities of African star apple kernel starches. Water absorption capacity, WAC, was significantly higher (p < 0.05) in ACCA (200.00 ± 0.00 g/100 g) than the native and other modified starches. OXCA had the highest oil absorption capacity (OAC) followed by ACCA and ATCA, respectively. Increased WAC by ACCA could be attributed to the introduction of acetyl group into the starch, which due to steric effects and disruption of hydrogen bonding allowed water percolation. Acid-thinning has been reported to decrease WAC by Reddy et al. (2015). Low WAC by ATCA can be attributed to amylose content erosion as a result of acid hydrolysis, which resulted in reduced hydrophilic groups on starch. Increased OAC following acetylation could be due to the introduction of hydrophobic methyl group of the acetyl functionality. OAC of ATCA was significantly higher than that observed for NACA. Generally, water absorption is desirable in controlling consistency, gelling and functional attributes of starches during preparations. High OAC improves palatability, mouthfeel and flavour retention in food preparations (Tharanathan 2005).

Pasting properties
Starch composition and molecular structure affect pasting properties. Modification significantly (p < 0.05) affected the pasting properties of native African star apple kernel starch (Table 2). Peak viscosity, PeV, is a measure of starch ability to freely swell in water before physical breakdown and indicates water absorption capacity of starch (El Halal et al. 2015). All the starch derivatives, except ACCA, had lower peak viscosities than their native counterpart, NACA. The most significant reduction in peak viscosity was observed in ATCA. Reddy et al. (2015) reported similar observations attributable to slow water molecules percolating into the resulting stiff starch granules left after the amorphous amylose layer has been eroded by thinning which prevented granules swelling and α-1,4-glycosidic linkages breakage in the amylopectin chains. Breakdown viscosity (peak viscosity minus trough viscosity), BrV, measures the extent of swollen  starch granule rupture and amylose leaching. All modified African star apple kernel starches showed increase in BrV, except ATCA with the least BrV. The results of the BrV showed that all modifications did not improve the mechanical shearing and thermal stabilities of African star apple kernel starch, except acid-thinning (Ikegwu et al. 2009). Starch modification caused a significant reduction in the final viscosity, FnV, with ATCA having the least. Reduced viscosity following acetylation can be attributed to weakened and disintegrated starch integrity. That may be due to the introduced acetyl groups and starch depolymerisation which led to reduced starch molecular weight (Sun et al. 2016), while decreased FnV, after oxidation, is attributable to partial breaking of the glycosidic bonds due to oxidation thus decreased molecular weight of starch molecules and loss of resistance to shear and integrity of starch granule and so, lower viscosity (Vanier et al. 2012). The setback viscosity, SeV, is a measure of the degree of retrogradation of starch. A high setback value indicates a high tendency of starch to retrograde (Chan et al. 2009). The SeV of the modified starches were all lower than that of the native starch. While the acetylated starch had the highest retrogradation tendency among the modified African star apple kernel starches, the acid-thinned had the least retrogradation tendency. Reduction in pasting temperature, PaT, is desired for economy as lesser fuel is consumed in industrial processes requiring starch gelatinisation. The results of the PaT showed that the temperatures required to cook all modified African star apple kernel starches were lower than that of native African star apple kernel starch. Similar observations of reduced PaT following acetylation (Ibikunle et al. 2019), acid-thinning ) and oxidation (Pietrzyk et al. 2012) have been reported.

Gelatinisation properties
The ranges of onset temperature (T o ) (72.2-75.8 °C), peak temperature (T p ) (79.9-80.9 °C), conclusion temperature (T c ) (82.1-86.1 °C) and enthalpy (ΔH) (11.7-15.6 J/g) of gelatinisation of native and modified African star apple kernel starches are presented in Table 3. African star apple kernel starch exhibited reduced gelatinisation transition temperatures following derivatisation. This can be attributed to weakened starch structures following the introduction of -CO/-COOH and CH 3 CO-functionalities into the starch chain in respect of the oxidised and acetylated (Bello-Perez et al. 2010) starches, respectively, thus enhanced water absorption and leaching of amylose Similar reports of decreased gelatinisation temperatures have been reported for acetylated maize and barley (Nunez-Santiago et al. 2010) and oxidised bean (Vanier et al. 2012) starches. On the contrary, acid hydrolysis (Martins et al. 2018), acetylation and oxidation (Sanchez-Rivera et al. 2005) have been reported to increase gelatinisation temperatures of some modified compared to native starches by other workers.

Conclusions
African star apple kernel starch was isolated and modified to produce the acetylated, oxidised and acid-thinned derivatives. Starch morphology remained unchanged following modifications. The presence of carbonyl/carboxyl and acetyl functionalities as confirmed by FTIR showed that the native starch was modified. Oxidation improved the swelling and gelation properties of African star apple kernel starch making it suitable as filler and gelling agent. Acetylated and oxidised African star apple kernel starches can serve as thickener and flavour retention agent due to their improved water and oil absorption capacities, respectively. Improved pasting properties were observed in the modified starches with the acid-thinned starch having the most significant improvement. Acid-thinned African star apple kernel starch is a potential raw material for lowviscosity quick-cook preparations and frozen products due to its significantly improved pasting properties.