Characterization of corn, cassava, and commercial flours: Use of amylase‐rich flours of germinated corn and sweet potato in the reduction of the consistency of the gruels made from these flours—Influence on the nutritional and energy value

Abstract Malnutrition appears in weaning age and is usually due to weaning food which is of low nutritional value. This problem led us to investigate the study of the physicochemical and functional properties of cassava flours and corn flours, and the fluidification of the gruels made from these flours by germinated yellow corn and sweet white potato flours. To do this, the approximate chemical composition, physical and functional properties, and ability of amylase‐rich flours to digest the starch in order to reduce consistency were evaluated. From these analyses, it emerges that the chemical composition, and physical and functional properties are influenced by the nature and the treatment undergone by the flours. It appears that the amylase‐rich flours that we used at a concentration of 1%–3% during the preparation of the gruels significantly reduced their consistencies. Given their strong liquefying power, this reduction was more marked with germinated corn flour where 1% permits to obtain desired consistency with 21.50 g of DM of bitter cassava flour, thereby multiplying the energy density and nutritional value of this flour by 5.18. It also appears that the action of flours rich in amylases was depending on the concentration, the nature of the flour, its composition, and the treatment undergone. In view of all these results, we can therefore consider the formulation of a weaning food with the consistency, and energy and nutritional value necessary for the proper growth of children.

the formulation of infant cereals, most of the infant meals offered in Cameroon are imported products at almost inaccessible prices for most families. To overcome this inaccessibility, they themselves produce complementary foods for their babies. However, failure to control key steps in the food processing process can impact on the nutritional value of food (macronutrients and micronutrients) and incorporate a risk of malnutrition in growing children (Gobson & Hotz, 2007). This infant malnutrition occurs mainly between the ages of 6 and 24 months, normally corresponding to the end of exclusive breastfeeding and the introduction of solid foods (De Benoist, 1995). In Cameroon, 33% of children under five suffer from malnutrition, more than half (18.4%) of which is severe (INS, 2014). This malnutrition is due to the fact that the infant gruels prepared daily from cereal flours (corn, e.g.), roots (cassava), and tubers (potato, sweet potato) by the parents are of low energy and nutritive density because they use low concentrations in dry matter of flours (5-10 g of MS). This low concentration of dry matter used during cooking of the spray liquid is due to the starch which swells during cooking and increases the consistency of the spray liquid. To give the gruels a sufficiently fluid consistency acceptable to the child's digestive system (30 ml/Kg body weight), mothers limit the proportion of flour to water and prepare gruels, which affects the energy and nutritional density of these (WHO, 1998). However, according to WHO and UNICEF, weaning gruels must have flow rates between 100 and 160 mm/30 s for a dry matter concentration of at least 30% (Zannou-Tchoko, Ahui-bitty, Kouame, Kouame, & Dally, 2011). Therefore, the most effective solution to increase the energy intake of children seems to be the implementation of treatments that reduce the consistency of gruels with high concentration in dry matter. One of these treatments is the depolymerization of starch which can be obtained by enzymatic hydrolysis. This hydrolysis is possible thanks to the use of flours or amylase extracts. Yadang, Mbome Lape, and Ndjouenkeu (2013) have demonstrated the increased capacity of unfermented sweet potato flour to reduce the viscosity of porridges made from fermented potato flour. In the same vein, Trèche (1999) demonstrates that the preparation of gruels of improved energy density can be done by using exogenous sources of amylases such as germinated cereal flours or amylase-rich flours (ARF) which are the resultant of the germination of cereals such as rice (Singhavanich, Jittinandana, Kriengsinyos, & Dhanamitta, 1999), sorghum (Usha, Lakshmi, & Ranjani, 2010), barley (Elenga, Massamba, & Silou, 2012), and corn (Sodipo & Fashakin, 2011). In addition, this treatment will reduce the consistency and increase the nutritional value and energy density of weaning gruels. This work therefore aims to improve the fluidity, nutritional value, and energy density of gruels made from corn (Atp), cassava (bitter cultivar), and commercial flour by adding flour rich in amylases (germinated yellow corn and white sweet potato flours).

| Plant material
The plant material consists of seeds of Atp variety corn (yellow), cassava roots (bitter cultivar), tubers of local sweet potato variety (white) and commercial flour. The choice of these foods is justified by their common uses in the formulation of infant flours.

| Methods
Supply of food raw material: Dry corn seeds, cassava roots, and potato tubers were purchased from the Dschang and Santchou IRAD stations, respectively. They were used to produce the various flours.

| Flour production
Corn seeds obtained at IRAD were separated into three batches. The first lot was used for the production of sprouted corn flour and the other two batches for the production of corn flour. Indeed, for the production of germinated corn flours, the previously sorted corn seeds were washed and soaked for 48 hr. They were then removed from the water, spread out, and mixed with ashes and then for 96 hr, they were left in the shade on a cloth that we watered every day (02 watered/ day) until the germination process was well underway and the roots appeared. Afterward, we proceeded to drying in a "Venticell" oven at a temperature of 50 ± 0.5°C for 45 hr. The seeds were then degerminated, ground, and sieved (Ø = 400 μm). The second batch was sorted, ground, and subsequently sieved (Ø = 400 μm). The third batch of each variety was also sorted, skinned, ground, and sieved (Ø = 400 μm). The flours obtained were then packed in plastic bags and stored in a desiccator. The cassava roots obtained were transformed into flour according to the method described by Bindzi (2012). This cassava flour was separated into two batches, one of which was roasted (120°C, 25 min).
Local sweet potato tubers were washed and peeled. They were cut into slices (5-7 mm thick) using a stainless steel knife and dried for 24 hr in a "Venticell" oven set at 45 ± 0.5°C. The dried samples were then ground using a "Moulinex," and the flour obtained was sieved (400 μm). The various corn flours, cassava flours, and commercial flour purchased on the market were subsequently characterized. Amylase-rich flours were incorporated at 60°C during the preparation of the gruels.

| Methods of analysis
The water content of each flour was obtained after drying in the oven at 105°C until constant weight and the dry matter determined after subtraction of the water content. Leach and Schoch (1959) method was used to determine water retention capacity. The mass density was determined using the method described by Okaka, Okorie, and Ozo (1991). pH, fibers, protein, ash, mineral (Ca, Cu, Fe, Mg, Na, P, and K), and fat and carbohydrate contents were determined according to the method described by the AOAC (1990), starch content according to the method described by Jarvis and Walker (1993). The method described by Fischer and Stein (1961) was used to measure reducing sugars. The titratable acidity was performed according to the method described by AFNOR (1982). The method described by Reddy, Manju, and Love (1999) was used to measure phytates in different flours. The cyanogenic glycoside content of the different flours was evaluated according to the method described by Makkar, Siddhuraju, and Becker (2007). The energy density was determined by a precise method by combining all the ingredients providing energy and using the coefficients of Atwater and Benedict (1899). DE = (G × 4) + (P × 4) + (L × 9). After the characterization of the flours, these were used for the preparation of the gruels with the introduction of amylase-rich flours after cooking (Figure 1). Flours in concentrations ranging from 15 to 25 g MS were used for amylase-rich flour concentrations ranging from 1% to 3%. Flow velocities of the different gruels were subsequently measured using a Bostwick consistometer (Bookwalter, Peplinski, & Pfeifer, 1968).

| Statistical analysis
The results of the analyses carried out were expressed as averages plus or minus deviations. The means were analyzed by the ANOVA test at the 5% probability threshold, and the Duncan test was used to compare the means. The graphs were drawn using Excel 2013 software. A correlation matrix and principal component analysis (PCA) between the physicochemical, functional, and consistency properties of the gruels were also performed using SPSS software version 20.0 and XLSTAT 2016. The analyses carried out showed the existence in very small quantities of phytates in corn and commercial flours and an absence in cassava samples. These results could be due to crop conditions and the degradation of these compounds by phytases during plant development as reported by Medoua, Mbome Lape, Agbor-Egbe, and Mbofung (2005), who observed a reduction in the phytates content of white yam tubers as maturation time evolved. The content of cyanogenetic compounds in cassava flours is higher than that found by Bindzi (2012), which was 0.37 mg HCN equivalent/100 g DM in cassava flour of a variety of the cultivated in the Central Cameroon region. This could be due to the drying method. Indeed, Bindzi (2012) demonstrated that drying cassava slice in the oven did not allow a better evaporation of cyanide from cassava compared to drying in the sun. Moreover, these results are not in agreement with those of Onimawo and Akpovwo (2006), who showed that during roasting at 100°C of pigeon pea, there was a significant decrease in cyanide contents (from 68 to 23 mg/100 g). Overall, the levels of cyanoge-

| Ratio between the different minerals
Minerals represent the nutritive part of the flours. According to Kotowska and Wybieralski (1999), the nutritive value of edible vegetable parts is largely determined by the following ratios: K:Mg, Ca:Mg, Ca:P, and K:(Mg + Ca) (Table 3). Nutrient ratios present in Table 3 show that the mineral ratios were influenced by the nature of the flours and the treatment. According to Majkowska-Gadomska and Wierzbicka (2008), the optimal Ca:Mg ratio should approximate 3, and the Ca:P ratio should be within the 1.2-2.2 range. The difference between this and our results can be explained by the nature of the flours, the treatment, and the culture conditions. A higher ratio is indicative of magnesium or phosphorus deficiency. All analyzed flours showed a higher potassium-to-magnesium ratio and the same for potassium-to-total magnesium and calcium ion ratio. According to Radkowski, Grygierzec, and Soek-Podwika (2005), the optimal ratios K:Mg are 6 and K:(Mg + Ca) are 1.6-2.2. The difference observed can be due to fertilized and cultural conditions. Highest K:Mg ratio was noted in roasted bitter corn flour and the lowest in yellow corn flour.  (Tester & Karkalas, 1996). In addition, there is a better retention of undehulled corn flour, which would be the consequence of the elimination of fibers contained in the films and which would be a factor limiting water retention by the starch.

| Functional properties of the different flours
The dehulling eliminates the phytates, which makes the phosphorus complex and thus limiting their capacities to bind water. For cassava flours, the high water retention is between 60 and 70°C with a peak at 80°C followed by a water retention fall between 80 and 90°C.
These results are in agreement with those of Nuwamanya, Baguma, Emmambux, Taylor, and Rubaihayo (2010), who demonstrated that the gelatinization temperature of cassava starch is between 60 and 70°C and that after 80°C of heating of the latter, there was a burst.
The burst at 80°C is due to bursting of starch granules.
As far as the swelling rate is concerned, we are witnessing the same phenomena as with water retention capacity. The cassava flours have higher swelling power due to lower content in protein.
Protein reduces the establishment of bound between starch and water ( Figure 3).

| Physical properties (MD and pH), WAC, titratable acidity, and SI of different flours
The physical properties of flours (mass density and pH) are shown in Table 4. The mass density varies from 0.47 g/ml for bitter cassava flour to 0.91 g/ml for decoupled yellow corn flour. There is a significant difference (p < 0.05) between cassava and corn flours for this parameter due to the particle size of the samples (Adebowale, Sanni, & Oladapo, 2008) and the nutrient composition. These values are higher than those for corn flour obtained by Klang (2015), which was 0.55 g/ml. This difference is due to the difference of variety, treatment applied to it, and drying method used. Low densities (<0.5 g/ml) are recommended for the preparation of infant flours.
The pH has an influence on the water retention capacity of flours and their rate of demotion. It is also an important parameter in the production of high viscosity starch gels. In our study, its value varies from 6.44 for dehulled yellow corn flour to 5.55 for roasted bitter cassava flour. There is a significant difference (p < 0.05) in pH values between corn and cassava samples. This would be explained by the fact that the cassava samples were fermented and during this process there was synthesis of organic acids leading to a reduction in pH. The pH range obtained is the same with that of Fana et al. (2015). The titratable acidity which indicates the free acidity of the flours varies from 1.92 ml eq NaOH/100 g of DM for commercial flour to 3.92 ml eq NaOH/100 g of DM for roasted bitter cassava flour. As the results show, it is little affected by the treatment applied and the nature of the sample. The water absorption capacity as presented in Table 3  commercial flour. This would be explained by the composition of the commercial flour which would be rich in soluble compound.
The value obtained is greater than that of Abioye, Ade-Omowaye, Babarinde, and Adesigbin (2011), obtained in plantain flour. The difference observed is due to the nature of the sample, the chemical composition, the drying mode, and the treatment applied. be also attributable to low starch content of corn (Table 1). This table also shows that the incorporation of amylase-rich flours for the preparation of the various gruels makes it possible to multi- an increase in dry matter. It also appears that the viscous character of corn, cassava, and commercial gruels disappears in the presence of germinated corn and sweet potato flours. This is due to the hydrolytic action of amylases which degrade large starch molecules into smaller molecules (maltodextrins, maltose...) whose reduced swelling capacity (Klang, 2015). The germinated corn flours, as well as those of sweet potato, thus make it possible to predigest the starch in order to make the gruels more digestible and easy to consume.

| Evolution of the flow velocities of different gruels as a function of the concentrations of different amylase-rich flours
The evolution of the flow velocities of gruels in which the various amylase flours were added after cooking is given in Figure 5. This

| Effect of the use of amylase-rich flours on the nutritional value of different gruels with flow velocities between 100 and 160 mm/30 s
This Table 7 shows that the introduction of amylase sources during the preparation of gruels based on different flours implies an increase in nutritional value compared to gruels not supplemented.
As a result, the multiplication of the nutritional value of the sup-

| Effect of the use of amylase-rich flours (ARF) on the energy density of different gruels with a flow velocity between 100 and 160 mm/30 s
The influence of the use of the two amylase flours on the energy density of the gruels (   also obtained by Elenga, Massamba, Kobawila, Makosso, and Silou (2009), who demonstrated that the energy density of traditional gruels is around an average of 30-60 Kcal/100 ml of gruel and that such gruels are not able to effectively cover the nutritional needs of infants and young children in addition to breastmilk especially that the frequency of daily consumption of gruels is two times/day (Trêche, de Benoist, Benbourzid, & Delpeuch, 1995).    In the case of our analysis, three components F1, F2, and F3 explain 100% of the variations with respective contributions of 63.32%, 27.89%, and 8.79% (Figure 6).   The PCA diagram of variables, also called variable correlation circles, and the variable dendogram allow us to visualize the grouping between these factors. This figure confirms once again the correlations but also the proximity between the different variables studied. These figures also show that the formation of these axes is dependent on the variables; this is how carbohydrates, lipids, proteins, cyanide, water retention capacity, swelling power, K, Na, Cu, Fe, pH, amylose, amylopectin, phytates, mass density, fibers, and flow velocities due to germinated yellow corn flour allow the formation of the F1 axis. The grouping of solubility index, reducing sugar, starch, P, Ca, and total phenols allows the formation of the F2 axis. Mg and flow velocities due to white sweet potato flour allow the formation of F3 axis. From these analyses, it emerges that the three axes make it possible to group the variables into four blocks according to PCA and dendogram (Figures 7 and 8).
We also subjected the observations to a hierarchical bottom-up classification with a view to classifying the different classes and this allowed us to group them into three classes as shown in Figure 9.
Concerning the variables, the first grouping shows that germinated yellow corn flour and K are significantly and positively correlated (r = 0.827). This is explained by the fact that the amylases present in these flours require the presence of cofactors (ions) for their activity. The K would play this role at some concentration. These amylases are therefore metalloenzymes. These results corroborate those of Noman, Hoque, Sen, and Karim (2006), who demonstrated that the ions present in flours rich in amylases bind TA B L E 11 Pearson correlation coefficient (r) matrix between physicochemical and functional properties and flow velocities very strongly with these amylases and they are therefore necessary for activity and stability.
The second grouping is that between Ca, Mg, reducing sugars, proteins, pH, mass density, cyanide content, titratable acidity, solubility index, fibers, phytates, ashes, lipids, Fe, copper, phenols, and water retention capacity. This grouping shows that there is a strong negative correlation between phytates and the follow-  and lipids are found on the surface of starch granules and limit the binding of water molecules by the granule (Debet & Gidley, 2007). Cyanide is present in plants as cyanogenetic glycosides.
The hydrolysis of this molecule during different treatments would release in addition to HCN a carbohydrate molecule capable of retaining water and thus increase this parameter. Indeed, proteins mask the starch molecule and more particularly these constituents thus reducing access to water or starch-water bonds. Zhang and Hamaker (1998) showed that the proteins present in sorghum flour reduced starch-water binding. It has also been found that the removal of gluten from wheat flour increases the water retention rate of its starch (Jenkins et al., 1987). Lipids are present in starch in the form bound to amylose (amylose-lipids). They therefore reduce the degree of solubility of this molecule and the ability of these −OH groups to establish bonds with water (prevents amylose-water bonds). Reducing sugars are polysaccharide monomers, especially starch. They have a low swelling power and a low water retention capacity compared to these polysaccharides. Their presence in large proportion in a matrix would therefore lead to a reduction in this parameter. Finally, there is a strong correlation between the solubility index and proteins (r = 0.684), Ca (r = 0.945), phenols (r = 0.756), and reducing sugars (r = 0.987).
Depending on the type of fatty acids and amino acids constituting the lipid and protein chain, respectively, this will influence their solubilities. Indeed, the richer the lipids are in phospholipids, the more soluble they are; the richer the proteins are in hydrophilic amino acids, the more soluble they are. This would lead us to conclude that these flours are rich in this type of compound. Calcium and reducing sugars (glucose, fructose, maltose) are low molecular weight molecules naturally soluble in water. This would mean that their high proportion in a matrix would make it more soluble.
Another grouping is phosphorus. Although not grouped to any other molecule, it is very important in the water intake of flours, the swelling power, and the ability of amylases to hydrolyze starch.
The fourth grouping is the starch, WAC, Na, SwP, AMYP, AM, carbohydrates, and WSPF. From this grouping, it appears that SwP, which is similar to water retention capacity, is strongly correlated to carbohydrates (r = 0.803), AMYP (r = 0.987), AM (r = −0.987), and water absorption capacity (r = −0.833). Carbohydrates are the main molecules responsible for water retention in flours, thanks in particular to the existence of their −OH groups. Indeed, the carbohydrate content of a food would mean less fat and protein and therefore fewer compounds preventing water retention by the starch flour and their swelling. Absorption capacity is the ability of flours to retain water and trap it. This imprisonment would therefore lead to a water stress condition in the environment, thus reducing the availability of water necessary for swelling. Amylose and amylopectin are the two components of starch. The swelling power of flours is mainly linked to starch and more particularly to its two constituents. The instability of the flour gel formed is due to the high amylose content of the flour. Indeed, amylose molecules are more soluble than amylopectin molecules due to the size of its chain and the lack of steric clutter on this molecule. It will therefore tend to dissolve in water, causing the bonds formed with water to break (Herrera, Aguirre, & Castaño, 2017). This rupture leads to a demotion of the flours, the fall of the swelling power, and the viscosity of the flours. Therefore, a flour rich in amylose will swell less. There is also a correlation between water absorption capacity and AMYP and AM content (r = −0.802 and 0.802, respectively). Starch and particularly amylose and amylopectin are the main molecules responsible for the rheological and functional properties of flours. In particular, it is responsible for the water retention and absorption capacity of the flours. This explains their relationship with water absorption capacity. The negative correlation between amylopectin and water absorption capacity is due to the fact that this molecule is sterically hindered and these intramolecular −OH groups have difficulty establishing bonds with molecules. Moreover, this steric clutter makes this molecule compact and difficult to detach so difficult to absorb water. The variables mentioned above, although not significantly correlated to the action of white sweet potato flour, could still influence its activity.
Despite this, Na was found to be positively correlated with white sweet potato action. Indeed, just like Ca, K, and Fe, this ion plays the role of enzymatic cofactor and is therefore essential to the activity of these enzymes or enzymes in white sweet potato flour.
The analysis of the classification of flours by hierarchical ascending classification made it possible to define three classes (Figure 8).
The first is made from dehulled corn and commercial flours, the second from corn, and the third from cassava and roasted cassava.

| CON CLUS ION
At the end of this work, the aim was to chemically and functionally characterize the flours and to evaluate the ability of germinated yellow corn flour and white sweet potato flour to reduce the consistency of cassava flour-based porridges, maize, and commercial, and it appears that the approximate chemical composition and physical and functional properties of the flours vary significantly from one plant to another, from one variety to another, and from one treatment to another. The gruels made from corn, commercial, and cassava flour without flour rich in amylases are low concentrations of dry matter, and the increase in its concentration leads to an increase in consistency. The addition of flours rich in amylases during cooking resulted in a fluidification, an increase in the nutritional value, and energy density of the gruels. This reduction proved to be better with germinated maize flour where with 1%, flow rates of between 100 and 160 mm/30 s were obtained for 21.50 g DM of bitter cassava flour, thereby multiplying the energy density and nutritional value of this infant flour by 5.18.
It also emerges from these analyses that the action of these amylase-

CO N FLI C T O F I NTE R E S T
The authors do not have any conflicting interests.

E TH I C A L R E V I E W
This study does not involve any human or animal testing.

I N FO R M E D CO N S E NT
None.