Technological Properties of Wheat/Trifoliate Yam (Dioscorea dumetorum) Hardened Tubers Composite Flours

The ability of trifoliate hardened-yam flours to partially substitute wheat flour in food formulations was assessed.Three varieties of hardened-yam flour were incorporated in wheat flour in proportions of 0, 10, 20, 30, 40, and 50% (w/w). Samples were evaluated for protein content, Zeleny sedimentation index, Hagberg falling number, functional properties (WAC, WSI, and OAC), and some rheological properties including dough rupture pressure (P), extensibility (L), stability (P/L), and deformation energy (W). Results showed that trifoliate hardened-yam flours do not have acceptable baking properties as pictured by the low Zeleny sedimentation index and the low Hagberg falling number. Protein quality (Zeleny index, 31) of wheat flour helped to compensate gluten deficit of yam flours, but the amylasic activity determined by the Hagberg falling number could not be adjusted, which resulted in a loss of extensibility (L) of the paste at 10% substitution. Multivariate analysis of experimental data regrouped wheat flour and all wheat/hardened-yam treated with kanwa composite flours in one homogeneous cluster. Although wheat/hardened-yam treated with kanwa composite flours had physicochemical and functional properties similar to wheat, the inadequate diastasic activity makes them inappropriate for bread making, marking the strongest influence of that parameter.


Introduction
Trifoliate yam (Dioscorea dumetorum) tubers are good source of nutrients and energy [1].Their storage under tropical ambient conditions makes them susceptible to a hardening phenomenon characterised by loss of the ability to soften during cooking [2,3].In an attempt to add more value to D. dumetorum as an important source of food and energy, Medoua et al. [4,5] developed schemes for processing of its hardened tubers into flours and suggested that these flours can be used in bakery.
Composite flour can be defined as a mixture of flours, starches, and other ingredients intended to replace wheat flour totally or partially in bakery and pastry products [6].Composite flour has several advantages for developing countries including the reduction of wheat flour importation and promotion of high-yielding, native plant species [7].In this regard, several developing countries have initiated programmes to evaluate the feasibility of alternative locally available flours as a substitute for wheat flour [8] and several authors successfully developed composite flours where wheat flour is partially replaced by cassava [9], taro [10], banana [11], yam [12], sorghum [8], soybean [13], cocoyam [14], chickpea [15], lupin [16], sweet potato [17], or maize [18].
To the best of our knowledge, no study has been performed so far to assess the aptitude of trifoliate hardenedyam to partially substitute wheat flour in food formulation.The present study was therefore designed to evaluate the aptitude to develop trifoliate hardened-yam-wheat composite flours.Generally, composite flours are developed to be used in bakery and in pastry-making and are suitable to produce bread, biscuits, noodles, and various cakes.Evaluating technological value of composite flour can be done either by direct appreciation, corresponding to the implementation of a production test at a reduced scale, or by indirect appreciation, corresponding to chemical, functional, and rheological analysis.This last alternative was chosen in the present study.

Materials.
Trifoliate yam, D. dumetorum cv., yellow tubers were harvested at maturity in a farm at Esse in the Centre Region of Cameroon.They were immediately transported to the laboratory in Yaoundé and stored under prevailing tropical ambient conditions (19-28 ∘ C, 60-85% RH) for 14 days.Hardening of stored tubers was characterised by a rough and fluffy surface of peeled tubers as opposed to the smooth and moist surface of freshly harvested ones.After storage, the samples of hardened tubers were thoroughly washed with water, peeled, chopped into chips of 1 cm thickness, and divided in three lots.Samples of the first lot were not pretreated, samples of the second lot were pretreated by fermenting the yam slices during 7 days in a closed system as described by Medoua et al. [5], and samples of the third lot were pretreated by soaking yam slices in a solution of 1% kanwa alkaline salt for 16 h as described by Medoua et al. [4].The samples of each lot were, respectively, dried at 40 ∘ C in a ventilated oven for 24 h and the dried samples were ground into flour to pass through a 400 m sieve.
2.2.1.Functional Properties.Water absorption capacity and water solubility index were determined as described by Bello et al. [23] and Anderson et al. [24].Oil absorption capacity was estimated by centrifuging a known quantity of flour saturated with oil as described by Sosulski [25].The least gelatinisation concentration was estimated according to the method of Coffman and Garcia [26].The bulk density of the paste obtained after mixing 100 g of flour with a given quantity of water to obtain a water content of 60% was determined according to the method described by Okezie and Bello [27].Emulsion activity and stability were determined according to the method described by Neto et al. [28].Flour sample (1 g) was mixed with 3 mL of distilled water and 3 mL of cotton oil in a graduated tube.Then, the mixture was shaken for 10 min in a KS10 agitator and centrifuged at 2500 rpm for 5 min.Emulsion activity was calculated by dividing the volume of the emulsified layer by the volume of total content in the tube.The emulsion stability was determined by heating the emulsion for 30 at 80 ∘ C and then cooled and centrifuged again at 2500 rpm for 5 min.The emulsion stability was expressed as the % of emulsifying activity remaining after heating.The colour derivative functions of  * [chroma = ( * 2 +  * 2 ) 1/2 ] and  * [hue angle = tan −1 ( * / * )] were also calculated.

Rheological Properties. Alveograph characteristics
(tenacity , extensibility , stability /, and deformation energy ) were determined under conditions of constant dough water content (40%) and mixing time (28 min), according to the AACC International Approved Method 54-30.02[29] using the Alveo-Consistograph (Chopin, Villeneuve-La-Garenne, France) with a built-in diaphragm pump to supply air for inflating the tested dough piece.

Statistical Analysis.
All measurements were carried out in triplicate.Statistical analyses of data were performed using SPSS 10.1 (SPSS Inc., Chicago, Illinois, USA) and STATIS-TICA 6 (Data Analysis Software System, StatSoft, Inc., USA).Analysis of variance (ANOVA), Duncan multiple range test, principal component analysis (PCA), and hierarchical cluster analysis (HCA) were performed.Statistical significance was defined at  ≤ 0.05.

Results and Discussion
Table 1 displays some physicochemical properties of wheat and yam flours.It was noted that protein content of yam flours (9.9-8.4 g/100 g) is relatively low compared to wheat flour.Zeleny sedimentation index which defines the capacity of proteins to form aggregate in acid medium (related to its content in glutelin) was low for various yam flours; in all the cases, it was below the minimum acceptable value set at 15 mL (Norm ISO 5529), marking the absence of glutelin in yam flours.The diastasic activity of yam flour was very strong as illustrated by the small values of Hagberg index.This strong diastasic activity led to a maximum starch liquefaction (specially for hardened-yam flour and fermented hardened-yam flour) and a low water retention capacity, which explained why, compare with wheat flour, hardenedyam flours presented a lower viscosity during cooking as pictured by the higher values of least gelatinisation concentration.In general, these results showed that hardened-yam flours do not have acceptable baking properties.
The protein content and the gluten quality (Zeleny index, 31) of wheat flour used in this study could help compensating protein deficit of different yam flours, but the amylasic activity determined by the Hagberg falling number was just at the standard level and could hardly adjust the enzymatic level in composite flours.
Table 2 displays some physico-chemical properties of wheat/hardened-yam composite flours.After incorporation of wheat flour, the baking properties were significantly improved.Zeleny index remained in acceptable level until a substitution rate of 50% for hardened-yam and fermented hardened-yam flours, and 30% for hardened-yam treated with kanwa flour.
In general, water absorption capacity (WAC), oil absorption capacity (OAC), water solubility index (WSI), and the least gelification concentration (LGC) increased with the substitution level.This evolution was similar to the one noted by Njintang et al. [30] during substitution of wheat flour by taro flour.Increasing of WAC with substitution of hardened-yam flours suggested that wheat/hardened-yam composite flours could be used in food systems such as bread making that require lots of water to improve mechanical characteristics of the dough.Since variation of OAC was associated with presence of nonpolar chains, results of the present study suggested that substitution of wheat flour by hardened-yam flours significantly increased oil fixation sites in the composites flours.Thus, wheat/hardened-yam composite flours could be potentially useful in the structural interaction of food, especially for flavour retention and the taste improvement of products.
Emulsion activity (EA) and emulsion stability (ES) of wheat/hardened-yam composite flours significantly decreased with the increase of hardened-yam flour substitution.This result agreed with the study reported by Yatsumatsu et al. [31] who correlated the decrease of EA with the increase of fibre content, as it was reported that D. dumetorum hardened tubers are rich in fibres [32].
In this study, a decrease of bulk density was noted with the increase of substitution rate of various hardenedyam flours.This is valuable if one considers the works reported by Nelson-Quartey et al. [33] and Akubor and Badifu [34] which, respectively, showed that a decrease of bulk density promotes proper formulation of infant foods, and that decrease of the density during incorporation of Treculia africana flour in wheat flour is good for production of cookies.
To have a better understanding of relations existing between basic parameters of flours based on their physicochemical and functional properties, experimental results were analysed using HCA and PCA method.
Original data introduced for analysis, [ = (16, 10)], comprised 10 variables measured in 16 samples, where each entry matrix was the mean value of three replicates.The data were autoevaluated before analysis.
Figure 1 displays results of HCA.In this dendogram, there are four main groups for the similarity 0.6 (similarity index = 100 − ( link / max ) * 100).Wheat flour forms a similar group with wheat/hardened-yam treated with kanwa composite flours (K10, K20, K30, K40, and K50).This suggested that, until 50% substitution, wheat/hardened-yam treated with kanwa composite flours preserves the physicochemical and functional properties of wheat flour.Wheat/hardened-yam composite flours (H) and wheat/fermented hardened-yam composite flours (F) are spatially away from the group formed by wheat and form a single group at the index of similarity of 0.4 characterized by a small Hagberg falling number (Table 2).
Analysis of varimax-rotated principal components gave similar trends and characteristics to those observed in HCA, but with the advantage that correlations between variables and samples become clearer.Varimax-rotated loading factors representing correlations between principal components   (PC) and original data are presented in Table 3 with their variances and Eigen values.PC1 included 56.2% of the variance in the introduced data and loadings, suggesting that it has a significant contribution to the functional properties.PC2 describes 32.45% of the total variance and has as main contributor physicochemical properties and emulsion activity.PC3 and PC4 only describe 5.35 and 3.84% of the total variance, respectively.Thus, PC1 and PC2 representing 88.65% total variance in the data can be enough to restitute the maximum of information contained in the original data.Results of HCA for variables are presented in Figure 2. In the dendogram obtained, three groups were observed for  a similarity index of 0.6.The diastasic activity of flours represented by the Hagberg falling number is spatially separated from the main group formed by proteins, density, Zeleny index, LGC, WSI, EA, and ES and forms a single group at a similarity index of about 0.1 with WAC and OAC.This suggested that technological properties of wheat/hardenedyam composite flours at substitution levels used in this study are mainly influenced by the diastasic activity and correlation that may exist between diastasic activity, WAC, and OAC. Figure 3 presents the scores and the varimax-rotated loadings of the first two principal components.In this figure, it is remarkable to note that wheat (W) and the various wheat/hardened-yam treated with kanwa composite flours (K10, K20, K30, K40, and K50) form a group that is distinguished by a high Hagberg falling number, a low OAC, and a low WSI.The second group formed by wheat/hardenedyam composite flours (H10, H20, H30, H40, and H50) and wheat/fermented hardened-yam composite flours (F10, F20, F30, F40, and F50) is mainly characterised by a low Hagberg falling number, a high OAC, and a high WSI.At this level of analysis, it could be suggested that the technological properties of wheat/hardened-yam treated with kanwa composite flours are similar to those of wheat.However, a more detailed observation of this group can allow subdividing it into three subgroups consisting, respectively, of W, K10-K20-K30, and K40-K50, which are differentiated by variation in WAC, protein content, Zeleny index, and EA.
Rheological properties of composite flours assessed by alveograph analysis are presented in Table 4.The effect of a partial substitution of wheat flour by various hardenedyam flours induced changes in the alveograph characteristics.For all composite flours, a drastic decrease of all parameters measured was noted at 10% substitution rate, followed by a progressive increase with increasing substitution rate.The alveograph characteristics in all cases were below the minimum threshold for the use of flour in bakery.This trend is different from the one noted by Yatsumatsu et al. [31] for wheat/taro composite flours and by Balla et al. [35] for  wheat/sorghum composite flours describing a linear increase of alveograph parameters with increasing substitution rate.
Results of this study suggested that wheat/hardened-yam composite flours are not suitable for bread making at all substitution levels and treatments type used in the study.
Considering the fact that the Zeleny index, indicating the capacity of proteins to form a viscoelastic mass, is in most cases at an acceptable level (Zeleny index ≥ 15, Table 2), and that extensibility () of the paste is mainly due to this capacity of gluten, it was surprising to note a loss of extensibility at 10% of substitution even after using a hard wheat with a protein content of 16 g/100 g DM.It was suggested that the phenomenon was due to the strong diastasic activity of composite flours (Table 2), which could make the paste soft and sticking.Hence, an inactivation of enzymes in hardenedyam flours before the mixture with wheat flour could help improving the alveograph characteristics of wheat/hardenedyam composite flours and make these flours more suitable for bread production.
The colour of flours can be an important parameter that can directly influence the acceptability of a product.CIE  *  *  * parameters of wheat/hardened-yam flours are presented in Table 5.In general, substitution level changes the CIE  *  *  * parameters of composite flours. * parameter (luminance) increases with substitution of hardened-yam flour and is not significantly affected by the increase of substitution of hardened-yam treated with kanwa and hardenedyam fermented flours. * parameter (red) decreases with the increase of the substitution level of various hardened-yam flours, while an increase was noted for  * parameter (yellow).In general, a partial substitution of wheat flour by different hardened-yam flours in study led to a significant increase of  * parameter (chroma) that measures colour saturation, and hue angle ( * ), which shows the brightness and yellow colour of composite flours obtained.

Conclusion
Under standard condition of bread making, flours made from D. dumetorum hardened tubers did not present any baking property.A partial substitution of wheat flour by hardenedyam flours straightened the baking properties of the flours (except for the Hagberg falling number which remained low) and significantly influenced the functional properties of the flours.Multivariate analysis (HCA and PCA) of experimental data brought together the different flours studied in two groups defined in two principal components: the first group was composed by wheat and various wheat/hardened-yam treated with kanwa composite flours (K10, K20, K30, K40, and K50) and was characterised by a high Hagberg falling number and a low WAC and WSI, while the second group was composed of wheat/hardened-yam composite flours (H10, H20, H30, H40, and H50) and wheat/fermented hardened-yam composite flours (F10, F20, F30, F40, and F50) and was characterized by a low Hagberg falling number and a high WAC and WSI.In all the cases the inadequate diastasic activity was the limiting factor preventing the wheat/hardened-yam composite flours to be appropriate for bread making.

Figure 2 :
Figure 2: Dendogram (HCA) for the physicochemical and functional properties of flours.

Figure 3 :
Figure 3: Scores and varimax-rotated loadings for the first two principal components.

Table 1 :
Some physicochemical properties of experimental flours.

Table 2 :
Physicochemical and functional properties of wheat/hardened-yam composites flours.

Table 3 :
Loadings for the first four varimax rotated principal components.