Application of heat-moisture-treated banana flour as composite material in hard biscuit

ABSTRACT

Composite biscuits made of heat-moisture-treated banana flour and wheat flour with ratio ranging from 25 to 75% were characterized for its postprandial glucose response in mice, diffraction pattern by X-ray diffraction, organoleptic and texture properties. Biscuit made of 100% wheat flour was used as a control. Composite biscuits resulted in lower blood glucose response, lower glycemic index and load due to its resistant starch and slowly digestible starch content as well as lipid-starch complex. Composite biscuit made of 75% banana flour is potential to use as a source for long lasting energy product as it displayed stable blood glucose response for 4 hours. Among sensory attributes, color and taste of biscuits were the most sensitive to the addition of the banana flour. The difference in the hardness of biscuits using instrumental analysis did not significantly affect the acceptability of texture perceived by consumers except at 75% of the banana flour.

RESUMEN

Las galletas compuestas elaboradas con harina de trigo y harina de plátano tratada con calor y humedad (HMT) en una proporción que oscila entre 25% y 75% fueron sometidas a pruebas para determinar su respuesta de glucosa postprandial en ratones, su patrón de difracción por rayos X y sus propiedades organolépticas y de textura. A manera de control se utilizó una galleta elaborada con 100% de harina de trigo. Las galletas compuestas produjeron menor respuesta de glucosa en sangre, menor índice y carga glucémicos debido a su contenido de almidón resistente y de almidón de lenta digestión, así como al complejo lípido-almidón. La galleta compuesta elaborada con 75% de harina de plátano puede utilizarse como fuente de productos energéticos de larga duración, ya que mostró una respuesta estable de glucosa en sangre durante cuatro horas. Respecto a sus atributos sensoriales, el color y el sabor de las galletas fueron los que mostraron mayor sensibilidad a la adición de harina de plátano. La diferencia en la dureza de las galletas, comprobada mediante análisis instrumental, no afectó significativamente la aceptabilidad de la textura por parte de los consumidores, excepto en la galleta elaborada con 75% de harina de plátano.

KEYWORDS:

• Biscuit
• banana flour
• heat moisture treatment
• postprandial glucose
• slowly digestible starch
• in vivo

PALABRAS CLAVE:

• galleta
• harina de plátano
• tratamiento con calor y humedad
• glucosa postprandial
• almidón de lenta digestión
• in vivo

1. Introduction

Biscuit is a baked product, usually made from wheat flour, sugar, and fat with low moisture content (Mamat & Hill, 2018). For several reasons, research on wheat flour substitution in food product has been increasingly studied. For instance, wheat gluten has been known to trigger celiac disease, non-celiac gluten-sensitivity and wheat allergy (Sharma et al., 2020). For certain countries wheat flour import volume is so high that the country’s food security is undermined. Indonesia for instance, according to USDA (2020) has imported wheat in 2018–2019 with a total volume of 11.8 million tons. Wheat flour substitution with other sources of flour is thus important to tackle wheat-related disorders as well as food security problem.

Among other sources of flour or starch such as breadfruit (Marta, Cahyana, Arifin et al., 2019), arrowroot (Handarini et al., 2020), banana flour is a potential subtitute abundantly available in tropical countries (Cahyana et al., 2018; Utrilla-Coello et al., 2014). Although native banana flour has been reported to contain high resistant starch (Bi et al., 2017; Cahyana et al., 2019; Zhang et al., 2005) which means that it exhibits low glycemic index, its modificiation using heat moisture treatment displays another promising result particularly in terms of starch hydrolisis rate (Cahyana et al., 2019). Heat-moisture-treated (HMT) banana flour exhibits a higher content of slowly digestible starch compared to its native starch according to the result of in vitro assay (Cahyana et al., 2019). Therefore, applying the modified banana flour with high content of slowly digestible starch in biscuit is interesting to carry out.

Studies on banana starch application in food product have been reported such as in paste (Alimi et al., 2017), cakes (Segundo et al., 2017) and ice cream (Yangilar, 2015). In terms of HMT-banana flour, previous study demonstrated that HMT-banana flour displayed low swelling volume (Cahyana et al., 2019). Therefore the product such as biscuit which does not need high swelling volume is a good choice to apply besides the reason of its commonly consumed product. Preliminary study showed that biscuit made of 100% of HMT-banana flour was not acceptable organoleptically, therefore it is important to produce composite biscuit made of HMT-banana flour and wheat flour. The aim of the study was to examine the effect of HMT-banana flour in composite biscuit on blood glucose response, texture, and organoleptic properties.

2. Materials and methods

2.1. Materials

Pisang Kapas (Musa acuminata colla) was used as source of banana flour at ripening stage 1 (entirely green). The banana was obtained from local market Gede Bage (Bandung, Indonesia). Wheat flour and other ingredients were purchased from a local supermarket.

2.2. Banana flour preparation

Immediate flour preparation was carried out once the fresh samples were obtained. Banana flour preparation was carried out according to a method explained in other studies (Cahyana et al., 2018; Marta, Cahyana, Djali et al., 2019). First of all, the unpeeled banana fruit was washed in water. It was then cut into approximately 2 mm thick. The slices were then dipped in water for 10 min, followed by draining. It was then dried in an oven at 50°C for 24 hours. The pulp slices were then milled by a grinder and passed through 80 mesh screens. The flour was then stored for later analysis at 25°C packed in polyethylene plastic bags.

2.3. Thermal modifications

Heat moisture treatment was performed as reported in other studies (Cahyana et al., 2019; Marta, Cahyana, Arifin et al., 2019) by adjusting banana flour moisture content to 30% followed by heating the flour at 100°C for 8 hours. It was then stored for 24 hours in a refrigerator at 4–5°C. The modified flour was dried in an oven at 50°C for 24 hours followed by grinding and sieving through 100 mesh.

2.4. Hard biscuit preparation

Ingredients used for hard biscuit production (Table 1) were as follows: HMT-banana flour and wheat flour (total 200 g), fine granulated sugar (70 g), margarine (80 g), egg-yolk (60 g), skimmed milk powder (100 g), water (5 g), and sodium bicarbonate (1 g). HMT-banana flour was used in composite biscuit to replace 25, 50, and 75% of wheat flour. Composite biscuits containing 25, 50, and 75% of HMT-banana flour are denoted as BFB25, BFB50, and BFB75 respectively. Biscuit made of 100% wheat flour (WFB) was used as a control.

Table 1. The difference in biscuit composition.

Tabla 1. Diferencia en la composición de las galletas.

Ingredient (gram)	Sample
	WFB	BFB25	BFB50	BFB75
HMT-banana flour	0	50	100	150
Wheat flour	200	150	100	50
Margarine	80
sugar	70
Egg yolk	60
Skimmed milk	100
Baking powder	1
water	5

WFB (wheat flour biscuit) refers to biscuit with 100% wheat flour. BFB25, BFB50, and BFB75 are respectively composite biscuits made of 25, 50, and 75% of HMT-banana flour.

WFB (galleta de harina de trigo) se refiere a la galleta hecha con 100% de harina de trigo. BFB25, BFB50 y BFB75 son galletas compuestas elaboradas con 25%, 50% y 75% de harina de plátano HMT, respectivamente.

Margarine and sugar were creamed for 10 minutes in a mixer. Egg yolk and skimmed milk powder were then added and mixed for 5 minutes with intermediate scraping. At the end of the mixing process the mixture of flour (depending on the ratio of HMT banana flour and wheat flour), sodium bicarbonate and water were gradually added to dough by kneading until producing smooth dough. After that, the dough pieces were sheeted to a final thickness of 0.4 cm using wood roller. Hard biscuit dough was molded with a circular mold (5 cm diameter) and baked in a baking oven for 15 minutes at 160–165°C. The hard biscuit were let to cool down for 30 minutes and packaged in sealed polyethylene bags for storage at 25°C until further analysis. Biscuit production was carried out in three different batches for each type of biscuit.

2.5. Vivo test

All procedures involving mice were carried out in compliance with the Animal Welfare Act, guidelines established by the Institutional Animal Care and Use Committee under the active protocol number 1193 at Universitas Padjadjaran. Male Wistars weighing approximately 200 g and of 8–12 weeks age were used in this study. The animals were housed in polypropylene cages in a legalized laboratory animal facility with controlled temperature (∼24°C) and lighting (12 h day/night cycle) for a 7-day adaptation period. They were given the feed and drinking water ad libitum. After adaptation period, the rats were given the same portion of biscuit approximately 1.5 gram biscuit. WFB contained 53.4% available carbohydrate while BFB75, BFB50, and BFB25 contained respectively 58.4, 56.7, and 55.07%. The study was conducted after a 12 hours fasting. Blood samples were taken in the fasting state and 30, 60, 90, 120, 150, 180, and 210 minutes after the meal. Blood samples were taken on test scripts and analyzed using glucometer. The area under curve (AUC) was calculated following the procedure explained in another study (Zhang & Hamaker, 2009).

2.6. Color evaluation

Color scale for L*, a* and b* of biscuit was measured using a Spectrophotometer CM-5 (Konica Minolta Co., Osaka, Japan). The color parameters L*, a*, and b* are indices reflecting lightness, redness/greenness and yellowness/blueness, respectively. Hue and ΔE (difference in coloration) were calculated according to equations:

(1) $\mathrm{H}\mathrm{U}\mathrm{E}=ta{n}^{-1}\left(\frac{b\ast }{a\ast }\right)$(1)

(2) $\mathrm{\Delta }E={\left(\mathrm{\Delta }{L}^2+\mathrm{\Delta }{a}^2+\mathrm{\Delta }{b}^2\right)}^{1/2}$(2)

where Δa, ΔL, and Δb represent the deviations of the individual values from the respective values of control biscuit of WFB.

2.7. X-ray Diffraction (XRD)

Powder X-ray diffraction analysis of wheat flour and HMT banana flour from biscuit was carried out on powder using X-ray diffractometer (PANalytical X’Pert PRO seri PW3040/x0, Malvern, UK) with Cu Kα value of 1.54 radiation with a 2θ range of 3º–50º using a voltage of 40 kV and filament current 30 mA.

2.8. Hardness (TPA)

The hardness of the biscuit was determined using TA-XT (Stable Micro Systems, UK) equipped with a 3-point bend rig (HDP/3 PB) using a 5 kg load cell and heavy duty platform. The experiment was carried out at a test speed of 2.0 mm/sec and trigger force was automatic at 50 g. The maximum force at which the biscuit was broken into two major pieces was recorded as hardness.

2.9. Hedonic sensory test

Hedonic sensory test of the hard biscuit was conducted with 15 trained consumer panels with ages from 18 to 25, male and female. Samples were analyzed one day after baking. Hard biscuits were presented as whole pieces, placed on white plastic dishes coded with three-digit random numbers. The biscuit samples were served to the panels in a random fashion. Hard biscuit were evaluated based on the acceptability of their color, texture, taste and aroma on a 6-point hedonic scale. The scale ranged from “extremely like” to “extremely dislike,” corresponding to the highest and lowest scores of “6” and “1” respectively.

2.10. Statistical analysis

Data were collected in triplicate and analyzed using ANOVA followed by posthoc test of Duncan at a significance level of 5% (p < 0.05). SPPS software was used in this analysis.

3. Results and discussion

3.1. Postprandial glucose and glycemic index

Response of postprandial glucose of mice fed with composite banana biscuits is presented in Figure 1. Glucose response was followed for 4 hours. Biscuit made of 100% wheat flour (WFB) was used as a control biscuit. The result showed that the change in plasma glucose following consumption of control biscuit reached as high as approximately 100 mg/dL. Compared to the control biscuit, composite banana biscuit exhibited lower change in peak of plasma glucose reaching as much as 40–50 mg/dL. Bearing in mind that in this experiment mice were fed with the same weight portion of biscuit but different available carbohydrate content. Therefore AUC (area under curve) of the glucose response directly reflects rather glycemic load than glycemic index. The AUC of composite biscuits within 2 hours ranged from 39.15 to 53.43% of the AUC of WFB (Table 2), suggesting that glycemic load of the composite biscuits was much lower than that of the control. This experiment thus demonstrated that producing biscuit by replacing wheat flour with 25–75% HMT-banana flour would be beneficial in reducing at least approximately half of glycemic load of control biscuit for the same weight portion. Conducting the experiment in the same weight portion of biscuit instead of the same available carbohydrate gives a practical information on how blood glucose response would be when food industries replace wheat flour content with HMT-banana flour in the same weight portion of biscuit.

Table 2. Area under curve (AUC), hardness and color properties of control and composite biscuits.

Tabla 2. Área bajo la curva (AUC), dureza y propiedades de color de las galletas de control y las galletas compuestas.

Sample	AUC relative to WFB (%)	Hardness (g)	L*	a*	b*	ΔE	^oHue
WFB	100 ± 0^a	4434.72 ± 94.55^a	71.66 ± 0.50^a	9.80 ± 0.34^a	42.22 ± 1.28^a	-	76.93 ± 0.05^a
BFB25	53.43 ± 3.68^b	3992.80 ± 341.36^b	62.91 ± 1.70^b	10.13 ± 0.59^a	33.66 ± 0.45^b	12.3 ± 0.27 ^c	73.25 ± 0.71^b
BFB50	40.87 ± 5.43 ^c	3460.98 ± 285.57 ^c	55.92 ± 0.74 ^c	10.32 ± 0.21^a	30.76 ± 1.80 ^c	19.48 ± 0.50^b	71.45 ± 0.66^b
BFB75	39.15 ± 6.54 ^c	2965.29 ± 450.80^d	53.84 ± 0.19^d	10.58 ± 1.18^a	27.74 ± 1.45^d	22.99 ± 0.16^a	69.12 ± 1.14 ^c

WFB (wheat flour biscuit) refers to biscuit with 100% wheat flour. BFB25, BFB50, and BFB75 are respectively composite biscuits made of 25, 50, and 75% of HMT-banana flour. All values are means of triplicate determination ± S.D. Means within columns with different superscripts are significantly different (p ≤ 0.05).

WFB (galleta de harina de trigo) se refiere a la galleta hecha con 100% de harina de trigo. BFB25, BFB50 y BFB75 son galletas compuestas hechas con 25%, 50% y 75% de harina de plátano HMT, respectivamente. Todos los valores son medias determinadas por triplicado ± DE Las medias dentro de las columnas con distintos superíndices son significativamente diferentes (p ≤ 0.05).

Figure 1. Postprandial glucose response of biscuit from different ratio of HMT-flour. WFB refers to biscuit with 100% wheat flour. BFB25, BFB50, and BFB75 are respectively composite biscuits made of 25, 50, and 75% of HMT banana flour.

Figura 1. Respuesta postprandial de la glucosa de galletas elaboradas con diferentes proporciones de harina HMT. WFB se refiere a la galleta hecha con 100% de harina de trigo. BFB25, BFB50 y BFB75 son galletas compuestas elaboradas con 25%, 50% y 75% de harina de plátano HMT, respectivamente.

It is worth noting that although the composite biscuits fed to mice were at the same weight as the control one, the portion of composite biscuits contained slightly higher available carbohydrate. In other words although mice were fed with slightly higher content of available carbohydrate in composite biscuits, the AUCs of postprandial glucose response within 2 hours were lower than that of the control. It suggests that glycemic index of composite biscuits can be predicted whether or not it was lower than that of the control. Taking into account the higher available carbohydrate in composite banana biscuits but lower AUC than that of the control, it indicates that composite biscuits possessed lower glycemic index.

Furthermore the AUC within 2 hours decreased with the increase in available carbohydrate from banana flour. It indicates that HMT-banana flour plays a role in decreasing glycemic index and glycemic load of biscuit. The decrease was however small when the HMT-banana flour content was increased from 50% to 75% to replace wheat flour, suggesting that the content of 50% HMT-banana flour is sufficient to decrease glycemic index and load remarkably.

Raw banana flour was reported to contain high resistant starch (RS) of approximately 86–90% (Cahyana et al., 2019). Following modification by HMT, the flour was reported to contain RS and slowly digested starch (SDS) as much as approximately 50% and 34.5% respectively (Bi et al., 2017; Cahyana et al., 2019). Although a decrease in the RS content due to HMT modification was reported, the RS content along with high content of SDS in HMT-banana flour may be sufficient to decrease postprandial glucose response. It is obvious in the present study that high content of RS and SDS in biscuit made of HMT-banana flour resulted in a beneficial effect with lower glycemic index and load which is good for patients with diabetic syndrome.

The lower glycemic index of banana flour biscuit might also result from the complex formation between banana starch with lipid from either margarine or egg yolk mixed in the biscuit as shown in Figure 2. XRD result showed a sharp peak at about 25° (2θ) indicating a starch-lipid complex. Formation of V-type crystallites following a complex formation between amylose and lipid has been reported in another study (Lopez-Rubio et al., 2008). The presence of peak was clearly observed from biscuit containing 75% banana flour. Starch-lipid complex is resistant to digestion, hence higher RS content. Therefore the lower glycemic index of the composite biscuit may stem from high initial RS content per se as well as higher starch-lipid complex.

Figure 2. X-ray diffraction pattern of biscuits from different ratio of HMT-starch. HMT-BF is banana flour modified by HMT. WFB is biscuit made of 100% wheat flour. BFB25 and BFB75 are respectively biscuits made of 25 and 75% of HMT banana flour.

Figura 2. Patrón de difracción de rayos X de galletas hechas con diferentes proporciones de almidón HMT. HMT-BF es harina de plátano modificada por HMT. WFB es la galleta hecha con 100% de harina de trigo. BFB25 y BFB75 son galletas elaboradas con 25% y 75% de harina de plátano HMT, respectivamente.

Interestingly while the control biscuit exhibited a decrease in postprandial glucose with the time, composite banana biscuits maintained their postprandial glucose particularly BFB50 and BFB75. The composite biscuit of BFB75 even displayed a tendency of slight increase in plasma glucose after 3 hours of consumption. This maintained plasma glucose level may not be due to gluconeogenesis because the trend was not observed in the other biscuits. Therefore it can be related to the high content of SDS in HMT-banana flour as reported in another study (Cahyana et al., 2019). This finding demonstrates that the characteristic of biscuit with slow release of glucose is encouraging to create a product exhibiting long lasting energy properties. The slow increase in plasma glucose and the maintained glucose level within longer time are several benefits of HMT-banana flour biscuits for patients with diabetic syndrome to avoid hypoglycemia for longer time.

3.2. Texture and color properties

Control biscuit and composite of banana biscuits are presented in Figure 3. The texture and color of biscuits are tabulated in Table 2. Visually the control biscuit was brighter while the composite banana flour biscuits exhibited a darker appearance with the increase in the content of HMT-banana flour. This visual observation was confirmed by instrumental analysis showing that the control displayed a higher L* (71.66) compared to the composite banana flour biscuits ranging from 62.9 to 53.84. Inclusion of HMT-banana flour also resulted in higher a* values (reddish) and lower b* values (less yellowish tones).

Figure 3. Biscuits from different ratio of HMT-banana flour. WFB refers to biscuit with 100% wheat flour. BFB25, BFB50, and BFB75 are respectively composite biscuits made of 25, 50, and 75% of HMT banana flour.

Figura 3. Galletas que contienen diferente proporción de harina de plátano HMT. WFB se refiere a galletas hechas con 100% de harina de trigo. BFB25, BFB50 y BFB75 son galletas compuestas hechas con 25%, 50% y 75% de harina de plátano HMT, respectivamente.

The difference in the color among biscuits can be also observed from the value of ΔE. Compared to the control, ΔE values ranged from 12.3 to 23 suggesting that the color of composite biscuits were different from the color of control biscuit. The Hue value which ranged from 69 to 76.9 suggests that the addition of HMT-banana flour to biscuit resulted in a drawback of darker appearance. The darkening of food product with the inclusion of banana starch was also observed in cookie (Roman et al., 2019). The darker appearance may be related to the Maillard or caramelisation reaction taking place during baking. The difference in biscuit color can also be linked to the difference in color of wheat flour (whitish) and banana flour (cream color) (Cahyana et al., 2019; Marta, Cahyana, Djali et al., 2019). Furthermore HMT treatment on banana flour resulted in darker appearance than the native one (Cahyana et al., 2019). In terms of biscuit texture, characterization by Texture analyzer showed that the hardness of control biscuit which was 4434 g decreased to as low as 2965 g with the increase in the HMT-banana flour content. This may be related to the higher gluten content in the control biscuit than the composite biscuits. Moisture content might also play a role in hardness of biscuit. The relation between moisture content and hardness is worth further investigation.

3.3. Hedonic sensory properties

Four organoleptic parameters were measured including color, taste, aroma and texture (Figure 4). The difference between composite biscuits and control for taste and aroma was observed at 25% and 50% of HMT-banana flour respectively (p < 0.05), suggesting that taste attribute is more sensitive to the HMT-banana flour addition compared to aroma. The inclusion of HMT-banana flour as much as 50% did not result in different taste and aroma (p > .05) among the composite biscuits, but the substitution with 75% of HMT-banana flour led to significant difference compared to the rest of composite biscuits (p < .05).

Figure 4. Organoleptic parameters of biscuits from different ratio of HMT-flour. WFB refers to biscuit with 100% wheat flour. BFB25, BFB50 and BFB75 are respectively composite biscuits made of 25, 50, and 75% of HMT-banana flour.

Figura 4. Parámetros organolépticos de las galletas hechas con diferente proporción de harina HMT. WFB se refiere a galletas hechas con 100% de harina de trigo. BFB25, BFB50 y BFB75 son galletas compuestas hechas con 25%, 50% y 75% de harina de plátano HMT, respectivamente.

The obvious difference in the panelist preference was also observed in the color attribute. For this parameter, addition of 25% HMT-banana flour led to a significant difference from that of the control (p < 0.05). The statistical difference was also observed among composite biscuits made of different ratio of HMT-banana flour, suggesting that color attribute is also sensitive to the addition of HMT-banana flour. The panelist preferred control biscuit to composite ones. This may be related to the value of L* as shown in instrumental analysis which was higher in control biscuit. Together with taste attribute, color attribute of biscuit can be regarded as the most sensitive sensory attributes to the addition of HMT-banana flour. The trend in lower acceptability of taste and color of biscuits with the increase in banana flour inclusion was also observed in another study (Norhidayah et al., 2014).

In terms of texture, although the biscuit hardness measured by instrumental analysis exhibited a statistical difference among the samples (p < 0.05), the addition of as much as 50% HMT-banana flour did not significantly reduce the acceptability of biscuit texture compared to the control when characterized by organoleptic test (p > 0.05). It suggests that the difference in the biscuit hardness detectable by instrumental analysis did not sufficiently affect the consumer’s acceptability of biscuit texture as observed from hedonic test, particularly until the addition of 50% of HMT-banana flour.

4. Conclusion

By and large HMT-banana flour is potential to use as a raw material for long lasting energy product particularly biscuit. Composite biscuits made of HMT-banana flour exhibited lower glycemic index and glycemic load than biscuit made of 100% wheat flour probably due to high content of RS and SDS in HMT-banana flour as well as lipid-starch complex formation. Moreover composite biscuit made of 75% HMT banana flour displayed a characteristic of long lasting energy product as it maintained postprandial blood glucose response to be relatively stable within 4 hours, a feature characteristic of slowly digested starch present in HMT-banana flour. The composite biscuit thus is a good product for diabetic patients not only due to its lower glycemic index and load but also its long lasting energy properties hence avoiding hypoglycemia for longer time. The addition of HMT-banana flour decreased the lightness, °Hue and resulted in different color compared to the control. The reduction in hardness was also noticed with the increase of HMT-banana flour content, but this reduction did not affect the consumer acceptability of the biscuit texture except at 75% HMT-banana flour. Organoleptic test indicates that among sensory attributes, color, and taste are the most sensitive to the addition of HMT-banana flour. The drawback however is not a big issue to handle. Addition of certain food additives can be an alternative to solve the problem. Further investigation on human intervention using the composite biscuits made of banana flour is warranted particularly to diabetic patients.

Acknowledgments

The authors would like to express a sincere gratitude to Kemenristek-DIKTI, Indonesia for the support to carry out this research

Disclosure statement

The authors declare no conflict of interest