Partial characterization of starches from major banana (matooke) cultivars grown in Uganda

Abstract The SEM of starch from five EAHB cultivars showed a mixture of irregular granule shapes with smooth surfaces. The starches' average diameter size ranged between 16.31 and 21.98 μm. The moisture, protein, and ash content of the starches were: 11.12%–11.84%, 0.1% and 0.23%–0.47% respectively. The amylose content of these starches was between 11% and 13%. The starches peak viscosity ranged between 488.42 and 558.66 RVU. The EAHB starches exhibited relatively low pasting temperatures (<75°C), a high peak viscosity (488.42–558.71 RVU), high level of viscosity breakdown (235.00–311.92 RVU) and low set‐back values (61.21–104.33 RVU). In general, the EAHB starches' WHCs and SPs increased substantially at 80°C with maximum SP (12.43–14.27 g water/g starch) and solubility (12.52%–14.19%) values obtained at 90°C. The starch clarity ranged from 1.7% to 2.3% and followed the same pattern. The starches exhibited poor freeze‐thaw stability as they had high syneresis (68% to 72%) after 10 freeze‐thaw cycles.


| INTRODUCTION
Starch is a polymer of glucose, which serves as the main energy reserve in higher plants. The main botanical sources of starch are wheat, maize, potato, and cassava with only minor quantities from rice and other starches produced (Waterschoot, Gomand, Fierens, & Delcour, 2015).
Starch is an inexpensive, abundant, biodegradable and renewable material that is available from a variety of botanical sources. Since the end of the 20th century, an assortment of applications for starch has been discovered with respect to food and non-food industries (Bello-Pérez & Paredes-López, 2009). Due to broadening of the spectrum of starch applications as well as the desire by development agencies to promote regional economies and sustainable agricultural activities, use of nontraditional starch botanical sources, such as roots and tubers (Hoover, 2001), tropical fruits (Zhang, Whistler, BeMiller, & Hamaker, 2005), among others has been sought in recent decades. Kayisu, Hood, and Vansoest (1981) having isolated and characterized starch from green bananas, subsequent studies on banana starch raised interest with regard to carrying out more investigations in its properties as well as its suitable applications as results revealed that banana starch had peculiar physicochemical, functional, and digestibility features (Zhang & Hamaker, 2012) in comparison with the traditional starch sources (BeMiller, 2007). Bananas are mainly grown in the tropical and sub-tropical regions of the world with an estimated annual world production of about 105 million MT in 2012 according to FAO.
Banana production for export is considered a different economic and technological activity to banana production as a staple.
Production for export relies on only a few varieties, which were selected for their high yields, durability in long distance transport, consistent quality and unblemished appearance. Likewise, EAHB production for industrial raw material as a starch source requires relying on a few selected cultivars based on high yields especially in terms of starch content and consistent quality in terms of physicochemical starch properties.
Studies have revealed that the physicochemical and functional properties of banana starch are dependent on variety, regional climatic conditions, and harvesting periods (Bello-Perez, Agama-Acevedo, Sanchez-Hernandez, & Paredes-Lopez, 1999). An assortment of different banana varieties and cultivars are produced at regional levels despite the Criollo variety being the most commercialized banana cultivar worldwide. For instance, the East African highland bananas (EAHB)-AAA cultivars are commonly cultivated and are a staple food for the Great Lakes region of East Africa (Gold, Bagabe, & Ssendege, 1999) and in particular Uganda where they are a key component in both food security and agricultural sustainability (Tenywa, Isabirye, Lal, Lufafa, & Achan, 1999).
An array of cultivars of the EAHB is extensively produced and consumed in many parts of Uganda, where some cultivars are more preferred than others mainly due to the cooking and tasting characteristics.
The EAHB is mainly consumed when green during which stage starch is the main component (≥70% Dry Weight Basis (DWB) (Muranga, 1998).
Due to the bananas' perishability leading to high postharvest losses and low value to bulk ratio, production of banana flour from the EAHB has been sought in Uganda and neighboring countries.
Although, banana starch characterization has been carried out extensively in the last two decades (Bello-Perez et al., 1999;Eggleston, Swennen, & Akoni, 1992;Muranga, 1998;Nwokocha & Williams, 2009;Waliszewski, Aparicio, Bello, & Monroy, 2003;Zhang & Hamaker, 2012), some characterization of starch from a few EAHB cultivars has been done by only one researcher (Muranga, 1998) yet the botanical source of starch is determinant of the chemical composition, granule morphology, amylose:amylopectin ratio and arrangement of the macro molecules, which in turn determine the physicochemical and functional properties of the starch. Since the EAHB are set to transform from being a regional food staple to a crop with numerous industrial applications, the characterization of starch from EAHB cultivars was considered.
Consequently, studies on the properties of starches from regional banana cultivars have increasingly become important due to the possibility of diversifying their applications, which would transform these bananas into a commercially viable commodity (Zhang et al., 2005).
The examination of the rheological and morphological properties of starches from bananas is of particular importance in the characterization and understanding of their functional properties.
Thus, the aim of this study was to characterize the chemical composition, average particle size, morphology, swelling, solubility, pasting properties, paste clarity, and freeze thaw stability of starch isolated from the Enyeru, Mbwazirume, Nandigobe, Mpologoma, and Bukumu East African highland banana cultivars from Western Uganda which represent the three sub groups for the EAHB available and also are among the most popular cultivars. The results of this study are expected to provide useful technical information about the physicochemical and functional properties of these starches, which will provide a foundation for enhancing their application in the food industry in substitution of commercially available starches.

| Starch Isolation
Starch was prepared by a modified method of Muranga (1998). In brief, the bananas were peeled, sliced, and crushed in a blender after holding in NaOH solution for 10 min. The slurry was diluted with water, strained over 500 μm sieves and the filtrate held at 4°C. After 6 hr, the starch was washed repeatedly and left to dry at room temperature for 72 hr. The dried starch was milled, passed through a 200 μm sieve and stored in air tight glass containers till further analyses.

| Chemical composition
The moisture, ash, fat, and crude protein were determined by standard procedures (AOAC, 1995). Purity of isolated starch was determined polarimetrically using an Automatic polarimeter (Atago, AP-300). Amylose content was determined according to Williams et al.'s Williams, Kuzina, and Hlynka (1970) method. Ten mL of 0.5 N KOH were added to 20 mg of starch samples, mixed thoroughly and topped up in 100 ml volumetric flask with distilled water. Ten ml aliquots were pipetted in 50 ml volumetric flasks to which 5 ml of 0.1 N HCl and 0.5 ml of iodine reagent were added sequentially. The flasks were diluted to the mark and absorbance was measured at 625 nm.
An amylose and amylopectin blend standard curve was used to measure the amylose content.

| Starch granule morphology
The granule micrographs were obtained using a JSM 6060 model SEM (JEOL Co., Ltd, Japan). The EAHB starch powders were placed on the SEM stubs using a two-sided adhesive tape (Nisshin EM Co., Ltd). The specimen stub was subsequently coated with Pt-Pd using an MSP-1S magnetron sputter coater (Vacuum Device Inc.). The coated samples were then analyzed using the SEM operated at 10.0 KV.

| Particle size analyses
Particle size was analyzed with a Laser scattering particle size analyzer (SALD-7100, Shimadzu Corporation) installed with a batch sample cell using methyl propanol as the solvent. The percentages of the number of granules corresponding to the different sizes were recorded and average granule diameter was obtained directly using the equipment's software.

| Pasting properties
The pasting properties of the EAHB starch samples were measured by a Rapid ViscoAnalyser (RVA-4, 1998, Newport Scientific Pty. Ltd, Australia). The General pasting method STD1 profile (ICC, 1995) was used for all the samples with a modification of holding temperature to 90°C instead of 95°C due to the high altitude of our laboratory location. Data were directly calculated from the pasting curve, using Thermocline for Windows v3.0 (TCW3) software for the RVA.

| Swelling power and solubility
These were determined as described by Waliszewski et al. (2003).
The supernatant was decanted and the volume determined. The sediment was dried in an oven for 2 hr at 110°C and the swelling power (SP) determined by difference.

| Paste clarity
Paste clarity was determined according to the method by Waliszewski et al. (2003). In brief, 4% starch suspension in a screw cap tube was heated in boiling water with vigourous shaking at 5 min intervals.
After 30 min, the samples were cooled to room temperature and refrigerated to 6°C for 72 hr. The percentage transmittance at 650 nm was determined every 24 hr against a water blank in a Cary 100 UV-Vis spectrophotometer, (Agilent Technologies).

| Freeze-thaw stability
This was evaluated using Bello-Perez et al.'s (1999) method. Briefly, 5 ml of starches suspensions (5% w/v) were rapidly heated to 75°C in a water bath subjected to constant agitation. After a 30 min hold time, the gels were cooled and stored at 20°C, and after 18 hr, they were thawed to 28°C for 6 hr. The water exuded by the gels was determined gravimetrically by vortexing the thawed gels for 15 s followed by centrifugation at 1620g for 10 min. The amount of water exuded after each freeze-thaw cycle was measured and expressed as a percentage of water separated. Ten freeze-thaw cycles were performed in total.

| Chemical composition
The chemical composition of the starches from the different EAHB cultivars is presented in Table 1.
The moisture content (MC) of the starch from the five EAHB cultivars ranged from 11.1% to 11.8% and protein content was 0.1% (Table 1), which were close to the results reported by Kayisu et al. (1981) of 10.8% MC and 0.2% protein content. Muranga (1998) reported the MC for starch from EAHB to range from 9.7% to 14.4% and protein content to be 0.1%. Lipids were not detected by the usual methodologies used to analyze proximate composition of foods (AOAC, 1995). This unusual low content of oil in the starch from the EAHB cultivars presents potential technological use of the starch in many industrial applications as challenges arising from rancidity of the oil in the starch are not expected to arise in the EAHB starch. The ash content (0.23%-0.47%) was higher than that reported by Kayisu et al. (1981) of 0.02% and Muranga, 1998) of 0.1%-0.2% but was in close range with that reported by Eggleston et al. (1992) for plantains (0.27%-0.34%), plantain hybrids (0.28%-0.32%) and cooking bananas (0.35%-0.41%). The high ash content may be attributed to the high mineral content of the banana starch especially potassium rather than impurities as the starch obtained was over 99% pure. Bello-Pérez, de Léon, Agama-Acevedo, and Paredes-López (1998) reported higher results of protein (2.03%), fat (2.46%) and ash (0.54%) for banana.
The amylose content of the starches from the five EAHB cultivars was in the range of 11.96-12.83% (Table 1). Though amylose is a smaller fraction in most starches, it has a large influence on the starch properties due to their structural contribution to the amorphous component of the starch granules. Different researchers have reported varying amylose contents of banana starch: 16% (Kayisu et al., 1981); ~17% for Cavendish (Garcia & Lajolo, 1988); 40.7% for Valery (Waliszewski et al., 2003); 10%-11% for plantains (Eggleston et al., 1992); 38.6%-43.8% for six Kluai cultivars (Vatanasuchart, Niyomwit, & Wongkrajang, 2012). Muranga (1998)   The EAHB starches showed a mixture of irregular granule shapes of elongated, spheroid and oval granule shapes with smooth surfaces devoid of dents (Figure 1). These results were in agreement with those of other researchers who reported that banana starch was composed of smooth irregularly shaped granules with different shapes (Kayisu et al., 1981;Muranga, 1998

| Granule size
The average diameter size for the starches from the five EAHB cultivars was 16.31-21.98 μm (

| Swelling Power
The SP results of the five EAHB cultivar starches are given in Table 3.
At <70°C, SP values were minimal (<1 g water/g starch) and there was no significant difference in the SP values registered for all the five EAHB cultivar starches (

| Banana starch pasting profile
The pasting of the EAHB starches exhibited gradual viscosity increase with increasing temperature during the heating stage (50°C-90°C) until peak viscosity was attained ( The increase in viscosity during the heating cycle is influenced by the extent of amylose leaching, granular swelling and the extent of friction between swollen granules. The higher peak viscosity displayed by Nandigobe could be attributed to its greater SP (Table 3) and relatively large granule size (Table 1). Srichuwong, Sunarti, Mishima, Isono, and Hisamatsu (2005) -Pérez et al., 2006), which is an interesting feature for products that require high viscosity at lower processing temperature, and also an advantage for the development of new products with thermolabile ingredients.

| Paste clarity
The starches for all the banana cultivars followed the same trend.
The starch clarity (i.e. transmittance value) of the starches from the EAHB cultivars ranged from 1.7% to 2.3% ( Figure 2). Results showed these starches to be far less translucent than white and yellow plantain starch at the same concentration (~6% and 7.5% respectively) (Nwokocha & Williams, 2009). Hoover, Sailaja, and Sosulski (1996)  Values are means (n = 3). Values with the same superscript letter are not statistically significant at the 5% level.

| Freeze-thaw stability
Freeze-thaw stability is one of the quality characteristic of starch gels.
When a starch gel is subjected to repeated freezing and thawing cycles, it releases water a condition termed as syneresis. The extent of syneresis is a measure for its freeze-thaw stability (Goff, 2004).The starch from the cultivars of the EAHB exhibited poor freeze-thaw stability ( Figure 3). All the EAHB cultivar starches had high syneresis (>55%) after the first freeze-thaw cycle, which indicates poor freeze-thaw stability. However, the same starches did not undergo much syneresis during the subsequent freeze-thaw cycles and after 10 freeze-thaw cycles, the total syneresis was between 68% and 72% for the EAHB starches.
Mpologoma starch, which had the smallest mean granule diameter and lowest SP, exhibited the lowest freeze-thaw stability F I G U R E 2 Paste clarity for starch from five EAHB cultivars (Bukumu, Nandigobe, Mpologoma, Enyeru and Mbawzirume) F I G U R E 3 Freeze thaw stability for starch from five EAHB cultivars (Bukumu, Nandigobe, Mpologoma, Enyeru and Mbawzirume) while Mbwazirume and Nandigobe starch starches, which had the biggest mean granule diameter and highest SP respectively, exhibited the highest freeze-thaw stability. Singhal and Kulkarni (1990) attributed smaller starch granule size to lower freeze-thaw stability. The exudation of water from the starch gels is probably due to separation of the phases during ice crystal formation resulting in concentration of amylose and amylopectin in the unfrozen matrix and when heat is applied, expulsion of water from inter-and intra-molecular associations occurs (Soni, Sharma, Srivasta, & Gharia, 1990).
Earlier findings have demonstrated low freeze-thaw stability for plantain banana starch (Bello-Perez et al., 1999) and white plantain starch (Nwokocha & Williams, 2009). The EAHB starch freezingthawing results imply that these starches may not be appropriate for use in their native state in food systems involving refrigeration or freezing processes.