Physicochemical and Technofunctional Properties of Irvingia wombolu Kernels Subjected to Various Cooking Treatments
Abstract
This study was conducted to assess the impact of various cooking treatments on the physicochemical and technofunctional characteristics of Irvingia wombolu kernels, which are bitter tasting. They were harvested in the town of Bertoua, East Cameroon, and then underwent drying (D), roasting (R), drying + roasting (D + R), drying + boiling (D + B), roasting + boiling (R + B), and drying + roasting + boiling (D + R + B). The findings showed all the treatments increased the amount of lipids, but only D and R increased the amount of protein. The minerals that were more prevalent and concentrated in roasted kernels were Ca, P, Mg, and K. Antinutrients were considerably (p < 0.05) decreased by all treatments, particularly roasted and boiled kernels. The R + B treatment exhibited high technofunctional properties in terms of swelling, water, and oil retention capacities. The secondary metabolite content (phenols and flavonoids) was reduced by about 40% in all treatments (p < 0.05). However, there was a notable (p < 0.05) increase in antioxidant power, as demonstrated by the ferric reducing antioxidant power (FRAP) assay, the 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging assay, and the hydroxyl radical (ROH) inhibitory capacity with roasted and boiled kernels, respectively. Amino acids were significantly destroyed by D and D + R; however, they were better conserved by R. Therefore, R is recommended as a suitable cooking method for these kernels to benefit from their properties.
1. Introduction
According to world food statistics, nearly 200 million people are malnourished in Africa, with at least one-quarter dying each year. In the world in general and the countries of the Congo Basin, agroforestry appears to be a fast-growing farming system and contributes to improving human living conditions, notably by enabling the discovery of new food varieties following the example of kernels with interesting nutritional properties [1].
These new species include Irvingia wombolu, a local fruit tree in the Irvingiaceae family that produces nontimber forest products (NTFPs) commonly used by local populations [2]. It grows wild in the humid zone of the Gulf of Guinea and is found in the central and southwest regions of Cameroon under the name “Ndo′o” or “bush.” The fruit is made up of various parts, including the seeds or kernels, which can be sweet or bitter, and a nonedible pulp [2]. The kernels obtained after extraction of the I. wombolu fruit have a high socioeconomic potential and are used in traditional pharmacopeia to treat certain diseases, notably dysentery and yellow fever [3]. The use of each part differs significantly. For example, dried, smoked, or powdered kernels are used extensively to thicken stews [4]. One study showed that the thickening capacity of this species was better than that of Irvingia gabonensis [5]. Onyeke et al. [6] report on the importance of kernels in their role as an emergency food capable of supplying carbohydrates and proteins once consumed. In addition to these technofunctional properties, Ejiofor et al. [7] revealed that kernels are a major source of lipids, which can make up 70% of their dry matter. These fats are used in many industries, notably the food industry, in the production of margarine. Its richness in functional polysaccharides makes its flour one of the most prized in the formulation of high-consistency products [8]. Furthermore, Ikhatua and Abiodun [9] reported that I. wombolu kernels contain 2.50% water, 21.20% carbohydrates, 6.50% protein, and 67.50% lipids. In the same vein, Onyeke et al. recorded a high mineral content, notably Ca (119.50 mg), Mg (151.75 mg), and P (195.31 mg) [6].
In order to facilitate preservation, limit bitterness, and use kernels over a longer period, people apply various processing methods to kernels that modify their technofunctional and physicochemical properties. These include roasting (R), drying (D), fermentation, smoking, and boiling [10, 11]. Joseph [12] reported that smoking reduces bitterness and prolongs shelf life by reducing the water content of I. wombolu kernels. In addition, Awono and N’doye [13] found that R reduced kernel swelling, promoted flavor development, and reduced bitterness. Despite the multitude of studies cited, the culinary treatment of kernels to determine their nutritional value has not yet been fully addressed. There are virtually no studies on the effects of culinary treatments on the nutritional and antioxidant properties of I. wombolu kernels. This lack of information prompted us to measure the effect of processing on the nutritional, technofunctional, and antioxidant properties of I. wombolu kernels harvested in Bertoua.
2. Materials and Methods
2.1. Nature and Scope of the Study
The present study is purely quantitative and experimental. It consisted of determining the physical, chemical, and antioxidant properties of I. wombolu seeds after various treatments. Kernel samples were collected in the East Cameroon region, more precisely in the town of Bertoua (4°57 ′ N, 13°66 ′ E), 340 km from Yaoundé. This region is bordered to the north by Adamaoua, to the east by the Democratic Republic of Congo, and to the west by the provinces of Sud and Centre. The area occupies part of the southern plateau of Cameroon. The climate is subtropical with four seasons. Temperatures are high throughout the year, peaking at 30°C, and rainfall is relatively abundant (1500–2000 mm/year). Soils are hydromorphic and not very ferritic; vegetation is a contact zone between savannah and forest.
2.2. Sample Used
The fruit, harvested from all four corners of the city of Bertoua, was sorted, cleaned with soap and water, disinfected with bleach, and rinsed to remove any impurities before being cut into pieces by hand. The I. wombolu kernels (Figure 1) collected after shelling underwent various treatments.
2.3. Methods
2.3.1. Processing I. wombolu Kernels
Once the kernels had been collected, they were transported to the Biochemistry Laboratory of the Faculty of Science at the University of Dschang, where they underwent various transformation processes on 200 g portions. The kernels, separated into six batches, underwent D at 50°C for 35 h, R at 120°C for 7 min, boiling of dry kernels at 90°C for 7 min, R of dry seeds under the same conditions, R followed by boiling under the same conditions, and finally D followed by R and boiling. Once the treatments were complete, the kernels were cooled in a desiccator before mechanical grinding using an Anton Moulinex (power: 800 W for 5 cycles of 1 min each). The resulting powders were packed in polyethylene bags before being stored in a desiccator.
2.3.2. Determination of the Physicochemical, Technofunctional, and Antioxidant Properties of Irvingia Kernels
2.3.2.1. Determining Physical Properties
2.3.2.2. Physicochemical Analysis
Once the ash had been obtained, it was digested with HNO3/HCl, and the filtrates or digestates obtained were read by an atomic absorption spectrophotometer (BIOBASE BK-D590, spectrophotometer) at wavelengths characteristic of Ca, Mg, Na, K, P, Fe, and Zn [15]. The ratios (Ca:P, Ca:Mg, and Na:K) used to assess the availability of certain minerals were also determined.
2.3.2.3. Determination of Antinutrient Content and Antinutrient/Mineral Ratios
After extraction in 1% MeOH/HCl, condensed and hydrolyzable tannins were, respectively, determined after mixing with vanillin and FeCl3·6H2O prepared in HCl [18]. Optical densities were read at 500 nm for condensed tannins and 660 nm for hydrolyzable ones. Contents expressed as milligram catechin equivalent per gram extract (mg CE/g) were determined from the catechin calibration line. The physical method of Koziol [19], consisting of measuring the height of the foam generated by stirring 1 g of powder in 5 mL of distilled water, was used to determine the saponin content (milligram/100 g). The bioavailability of minerals in the body was assessed by determining the molar ratios between phytates or oxalates and divalent cations.
2.3.2.4. Determination of pH, Functional Properties, and Color of Irvingia Kernels
2.3.2.5. Influence of Different Culinary Treatments on the Rheological Properties of Irvingia Kernels
These parameters were evaluated using the protocol of Sanchez et al. [22], which involved mixing 3 g of the various almond powders with 20 mL of distilled water (the concentration being 15%) and introducing the whole into a Rapid Visco Analyzer (MCR-52, Anton Paar) at a starting temperature of 50°C. The first values were obtained after rotation at 960 rpm for 10 s, before falling back to 160 rpm for the rest of the process. The process steps were holding for 1 min at 50°C and heating between 50°C and 90°C at a velocity of 6°C min−1, cooling at 50°C at a velocity of 12°C/min, and another holding at 50°C for 2 min. The RHEOPLUS/32 Service V3.62 software 21000071-33086 was used to analyze the various results and to highlight the following parameters: pasting temperature (PT), pasting time (Ptim), peak viscosity (PV), holding viscosity (HV), final viscosity (FV), breakdown (BD = PV − HS), setback (SB = FV − HS), stability ratio (STR = HV/PV), and finally setback ratio (SBR = FV/HV).
2.3.2.6. Influence of Different Cooking Treatments on Thermal Properties and PRAL Index of Irvingia Kernels
2.3.2.7. Influence of Different Culinary Treatments on the Amino Acid Profile of Irvingia Kernels
Amino acid composition expressed in millimoles/liter was determined according to the protocols of John et al. [24] and Savych et al. [25]. The first step was total hydrolysis of the proteins in 6 M HCl at 100°C for 17 h, followed by concentration of the hydrolysate in rotavapor and suspension in citrate buffer pH 2.2. The suspension was filtered through a Millipore membrane (0.45 μm) directly into vials. Amino acid profiling and quantification were performed after the injection of 1 μL of each extract into a Pursuit XRs C18 Analytical column, 150 × 4.6 × 3 μm, and into a second Zorbax Eclipse Plus C18, 4.6 × 12.5 × 5 mm Ultrahigh-Performance Liquid Chromatography column (Agilent 1290 Infinity II) fitted with a Diode Array Detector (1260 Infinity II). The starting temperature was 40°C, and the elution speed was 1.5 mL/min. Peaks were detected at 230 nm emission and 360 nm excitation. OpenLab CDS software was used to display values in mmol/L.
2.3.2.8. Influence of Different Culinary Treatments on the Antioxidant Potential of Irvingia Kernels
2.3.2.8.1. Extraction of Secondary Metabolites
A 1∗6 factorial device using EtOH at 96° as a solvent was used to extract secondary metabolites. To achieve this, 100 g of the various powders was introduced into a 500-mL beaker and mixed with 250 mL of the solvent. The mixture was then macerated manually at 8 h intervals at a temperature of 28°C ± 0.5°C for 48 h according to [26]. At the end of maceration, the mixture was filtered through muslin and then through Whatman No. 4 paper before the extracts were dried at 40°C for 30 h. The dry extracts obtained were stored at −10°C.
2.3.2.8.2. Determination of Phenol and Flavonoid Content
Secondary metabolites were quantified on extracts diluted to 2 mg/mL. Phenols were quantified according to the protocol of Gao et al. [27]. To achieve this, 10 μL of extracts was mixed with 200 μL of Folin–Ciocalteu 1/10; then, 1.39 mL of distilled water was added. After standing for 3 min, 400 μL Na2CO3 (20%) was added, and the mixture was heated in a water bath (Model TT 9052, Techmel & Techmel, United States) at 40°C for 20 min. The optical densities of the solutions were read at 760 nm using a UV-Vis spectrophotometer, Sican 2301, against a blank prepared under the same conditions. Phenol content, expressed as mg GAE (gallic acid equivalent)/gram of extract, was calculated using the calibration line for gallic acid (0.2 g/L).
Flavonoids were determined according to the protocol of Bahorun et al. [28]. This involved mixing 100 μL of extract with 1.4 mL of distilled water and 30 μL of NaNO2 (5%), followed by a 5-min rest. Subsequently, 200 μL AlCl3 (10%), 200 μL NaOH (10%), and 240 μL distilled water were added before the mixture was vigorously stirred. The optical densities of the solutions were read at 510 nm using a UV-Vis spectrophotometer, Sican 2301, against a blank prepared under the same conditions. Flavonoid content, expressed as mg CE/g of extract, was calculated using the catechin calibration line (0.4 g/L).
2.3.2.8.3. Determination of the Antioxidant Potential of Ethanolic Extracts From Different Treatments of I. wombolu Kernels
The protocol of Oyaizu [30] using TPTZ (2,4,6-Tris(2-pyridine)-s-triazine) was used to assess the reducing capacity of Fe3+ to Fe2+. It involved mixing 30 μL of extracts at concentrations of 200, 100, 50, 25, and 12.50 μg/mL with 900 μL of TPTZ solution, followed by incubation at 50°C in the dark for 30 min. Wavelengths were read at 595 nm against a blank consisting of 100 μL of distilled water. Results were compared with vitamin C and BHT prepared under the same conditions. The Trolox calibration curve was used to quantify the reducing power, expressed as micromole Trolox/gram extract.
The protocol described by Nagulendran et al. [31] was used to assess the inhibitory capacity of the hydroxyl radical (ROH). Indeed, 60 μL of FeSO4 (1 mM), 90 μL of 1,10-phenanthroline (1 mM), 2.4 mL of phosphate buffer (0.2 M, pH 7.8), 150 μL of hydrogen peroxide (H2O2) 0.17 M, and 1.5 mL of extracts prepared at concentrations of 200, 100, 50, 25, and 12.5 μg/mL were successively mixed in test tubes. The results were compared with vitamin C and BHT prepared under the same conditions. The mixture was incubated at room temperature for 5 min, and the absorbance was read at 560 nm against the blank prepared under the same conditions with distilled water.
2.4. Statistical Analysis
Results expressed as mean ± standard deviation (first analysis and two replicates for each analysis or n = 3) were subjected to an ANOVA to highlight differences between samples. Where significant differences were noted, a Fisher post hoc test at the 5% probability threshold was used to compare results using Minitab 18.1 software. A principal component analysis (PCA) was performed using XLSTAT 2014 software.
3. Results and Discussion
3.1. Physical Characteristics of I. wombolu Fruit and Kernels
The physical parameters of I. wombolu fruits and kernels collected in Bertoua are presented in Table 1. These parameters provide information on the plant’s state of maturity. The average mass of fruit was 74.30 g, while the percentage of extracted kernels was 6.24 g (8.40% of the fruit). These results are lower than those of Van Dijk [32], who found an average fruit mass of 97.80 g and a kernel content of 14.72%. Indeed, for good plant growth and almond yield, the plant should be located in an environment with little human activity, such as forests. However, the fruit in this study was harvested right in the center of Bertoua, where there is a high level of human activity. Similarly, the harvesting season for the fruit in this study was dry, which would have resulted in water stress or loss of water through diffusion and evaporation, leading to a reduction in mass. These observations are confirmed by data from Kang et al. [33], who report that high rainfall favors the development of I. wombolu. In the dry season, however, rainfall is almost nonexistent. Fruit volume was 0.56 cm3 while kernel volume was 0.09 cm3. Ejiofor et al. [7] found 0.90 and 0.10 cm3 to be the volumes of the fruit and kernels, respectively, thus confirming the hypothesis that plant growth zones affect their volumes. Mass density, which represents mass per unit volume, is important in the choice and efficiency of equipment to be used in fruit processing, as well as in the final quality of the product [34, 35]. This parameter ranged from 69.33 g·cm−3 (almond) to 132.67 g·cm−3 (fruit). These values were higher for fruit and lower for kernel than those of [7], which were 107.6 g·cm−3 for fruit and 103.8 g·cm−3 for kernel, respectively. According to Ameen et al. [36], the fruit’s physical properties are affected by storage conditions and climatic growing conditions.
| Parameters | Irvingia wombolu fruit | Irvingia wombolu almond |
|---|---|---|
| Mass (g) | 74.30 ± 0.90a | 6.24 ± 0.37b |
| Volume (cm3) | 0.56 ± 0.04a | 0.09 ± 0.01b |
| Density (g/cm3) | 132.67 ± 0.40a | 69.33 ± 0.37b |
- Note: Mean ± standard deviations (n = 3) bearing the letters a, b, c… on the same line differ significantly at 5% of probability.
3.2. Influence of Cooking Treatments on the Nutritional Composition of I. wombolu Kernels
Table 2 shows the nutrient composition of the various I. wombolu kernels. The nutritional evaluation of foods tells us a great deal about their content of important macro- and micronutrients. Nutrition specialists need to know this information in order to select appropriate foods and create nutritional tables. Water content predicts shelf life and alterations that can affect foods and their textural–sensory properties. The recommended threshold value is less than 14% in food matrices stored in humid areas [37].
| Treatments | Dry matter (%) | Ash (%) | Fiber (%) | Glucids (%) | Proteins (%) | Fat (%) | Energy (kcal/100 g) |
|---|---|---|---|---|---|---|---|
| D | 9.36 ± 0.02a | 2.90 ± 0.00c | 4.60 ± 0.09b | 11.22 ± 0.65b | 13.60 ± 0.46b | 58.78 ± 0.46c | 628.26 ± 2.27c |
| R | 3.65 ± 0.87b | 2.45 ± 0.00d | 3.70 ± 0.07c | 19.14 ± 0.93a | 9.90 ± 0.18d | 60.69 ± 0.18bc | 662.34 ± 2.20ab |
| D + R | 4.35 ± 0.21b | 2.85 ± 0.00c | 3.21 ± 0.10d | 8.04 ± 0.64c | 19.59 ± 0.20a | 64.93 ± 0.23abc | 658.68 ± 1.61b |
| D + B | 3.22 ± 0.32b | 3.90 ± 0.00b | 3.37 ± 0.10d | 9.20 ± 2.15bc | 11.85 ± 0.20c | 68.01 ± 0.18a | 696.26 ± 5.10a |
| R + B | 4.46 ± 0.81b | 2.20 ± 0.00d | 4.88 ± 0.10a | 13.08 ± 2.98ab | 11.00 ± 0.50c | 64.37 ± 0.46abc | 675.67 ± 4.60ab |
| D + R + B | 4.13 ± 0.19b | 4.35 ± 0.00a | 2.30 ± 0.18e | 15.11 ± 1.51ab | 9.71 ± 0.46d | 65.84 ± 0.46ab | 691.88 ± 6.49ab |
- Note: Means ± standard deviations (n = 3) bearing the letters a, b, c… in the same column differ significantly at 5% of probability.
- Abbreviations: D, drying; D + B, drying and boiling; D + R, drying and roasting; D + R + B, drying then roasting and boiling; R, roasting; R + B, roasting and boiling.
Table 2 shows that the treatments significantly (p < 0.05) reduce the water content from 9.36 (raw) to 3.22 (drying + boiling [D + B]). Heat treatments promote the embrittlement of inter- and intramolecular bonds involving water, facilitating its evaporation [10]. Tambo et al. [38, 39] noted that, during R, the high dry heat was responsible for breaking bond energies, facilitating the elimination of bound and free water. The contents obtained are lower than those of Kouame et al. [40], which varied between 16% and 20%, demonstrating the value of culinary treatments in preserving kernels. After the removal of water and organic matter by calcination, we are left with only ash, which gives us information on the quantity of minerals present in the food [41]. This parameter varied between 4.35 (triple treatment) and 2.20 (roasting + boiling [R + B]). R significantly reduced this parameter, and this would be the consequence of a low decomplexation of antinutrient ash in these seeds [10].
These observations are contrary to those of [42], who noted a significant drop in ash and mineral content with treatments requiring long hydric treatment, such as soaking, blanching, and boiling. The values obtained are all below 5%, which would make the seeds suitable for the formulation of supplemental foods [43]. Onojah et al. [44] obtained a lower content (1.71%) than those of this study in Irvingia gabonensis kernels. This confirms the value of seed treatments in reducing hidden hunger. Fiber facilitates the intestinal transit of food, helping to prevent constipation and certain diseases such as colon cancer, diabetes, obesity, cardiovascular disease, and coronary heart disease [45]. The fiber content of the various samples ranged from 2.30% to 4.88%. Dried kernels (D = 4.60%) and those roasted then boiled (R + B = 4.88%) showed significantly (p < 0.05) higher contents than the other treatments. These results are contrary to those of Dongmo et al. [10], who noted a significant drop in this parameter with boiling and R. Tchimmoe et al. [46] obtained a lower content (2%) than those recorded in this study, and this may be due to the difference in treatments, climatic conditions, fruit ripeness, and the quality of the growing soil. Carbohydrate content was also determined, as carbohydrates are the macronutrient of choice in human nutrition, with an intake of between 40% and 50% [47]. This parameter ranged from 8.40% to 19.14%. It was significantly lower in kernels dried and then roasted. This may be due to the intensity of caramelization and the Maillard reaction during this double treatment, as recorded by [11]. Both reactions reduce the availability and concentration of sugars. Nevertheless, Dongmo et al. [10] and Klang et al. [48] reported a significant drop in carbohydrate content with boiling. Likewise, the contents obtained are lower than those recommended by Codex Stan [43] for food formulation, suggesting the addition of cereals or tubers in the case of the formulation. These results are lower than the 71% obtained by Amalia et al. [49] in canistel nuts and could be explained by biological and environmental factors such as plant age, tree load, fruit maturity stage, physiological state of the fruit at the time of analysis, soil, cultivation technique, and sample cooking (baking). Proteins are vital to our bodies, performing numerous functions including enzymatic, energetic, transport, pH regulation, and structuring [50]. Proteins of animal origin have a higher biological potential than those of plant origin. Potentiation of this biological capacity is possible through cooking treatments. The protein content of the various samples ranged from 9.71% to 19.59%. Roasted kernels (9.90%) and dried, roasted, and boiled kernels (9.71%) showed significantly (p < 0.05) lower protein contents than the other treatments. The reduction in protein availability during the Maillard reaction occurring during these treatments is thought to be the cause. Furthermore, leaching during boiling in the triple treatment (drying + roasting + boiling [D + R + B]) would explain its very low content. These results are higher than those of Onimawo et al. [51], who reported 13.27% in I. wombolu kernels collected in Nigeria. These variations in protein content can be explained by various factors such as the genetic makeup of the plant, climatic conditions, fruit maturity stage, soil quality, and sample preparation. Given these results, the consumption of 100 g of a portion of these seeds would contribute between 50% and 110% to the coverage of this nutrient in children aged 6–59 months [52]. These observations demonstrate the importance of treatment and the search for new sources of nutrients. Lipids play a fundamental role in our body’s energy supply (9 kcal/g), thermal protection and diffusion, the formation of biological membranes, and the transport and transfer of nerve impulses [45]. The results of this study show that lipids are the most abundant nutrients, corroborating the data of Onimawo et al. [51]. Lipid content ranged from 57.78% to 68.01%. Dried and boiled kernels (D + B) and dried, roasted, and boiled kernels (D + R + B) showed significantly (p < 0.05) higher lipid contents than the other treatments. Furthermore, the improvement of this treatment with all treatments could be explained by the low water content having led to an increase in the concentration of nutrients such as lipids and proteins. The Maillard reaction that takes place during the R process leads to a concurrent reduction in proteins and carbohydrates, unlike lipids. In addition, it was found that the treatments that had undergone boiling had a higher lipid content, which could be explained by the higher temperature of the R process, which resulted in less oxidation [11]. Dangang et al. [11] also reported that culinary treatments such as blanching and R were responsible for the disruption of cells and cell membranes, as well as the complexes formed between lipids and other chemical elements, leading to an increase in their content. Because of this high content, it would be advisable to include very low proportions of these kernels in formulations to avoid creating obesity-type nutritional disorders. Ogunsina et al. [53] obtained values similar to those in this study. Lipids are triglycerides made up of fatty acids; some are essential (omega 3, 6, and 9), and others can contribute to disease if they are too saturated. It is therefore advisable to reduce the cooking time and temperature of these kernels to preserve their lipid profile. The contribution of these kernels to energy coverage was assessed and found to be very high with D + B = 696.26 kcal and low with the control or dried sample (D = 628.26 kcal). This variation may be due to differences in macronutrient proportions (mainly lipids) in different samples. There was a positive correlation between lipid content and energy value. The dried and boiled (D + B) seeds with the highest lipid content were also the most energy-rich. These results are higher than those of Stadlmayr et al. [54] and Ciqual [55], which were 47 kcal in orange and 155 kcal in avocado, respectively. Given their high energy value, these kernels could provide energy security for the Cameroonian population in general and those of Bertoua in particular, especially those with medium or high levels of activity and children who spend a lot of time during the day.
3.3. Influence of Cooking Treatments on the Composition and Availability Ratio of Minerals in I. wombolu Kernels
Analysis of the micronutrient composition of these kernels reveals their value in managing and combating hidden hunger. Micronutrients such as potassium help maintain acid–base balance, osmotic pressure, nerve impulse transmission, and muscle contraction [56]. Calcium is involved in the fortification, maintenance, and formation of teeth and bones; blood coagulation; the transmission of nerve impulses; biological activity as a cofactor in metabolic reactions; and also in improving the functional properties of flour by promoting interactions with water [57]. Magnesium and zinc act as cofactors for several enzymes but are also involved in the proper functioning of the brain, heart, and skeletal muscles [58]. Iron is involved in the formation of hemoglobin, muscle myoglobin, and enzymes essential to cellular metabolism, helping to combat osteoporosis and iron deficiency anemia. Phosphorus plays an important role in maintaining functional properties by stabilizing gelatinized granules [38].
Table 3 shows that the zinc content of I. wombolu kernels varies between 2.01 and 20.34 mg/100 g and is lowered by all other treatments (R, D + R, R + B, and D + B) except for the triple treatment. As for Fe, there was no significant drop (p > 0.05) in any of the treatments (R, D + R, and R + B). Its content varied between 0.02 and 0.60 mg/100 g, and D + R + B yielded the highest content, in contrast to R. Calcium is the most abundant mineral in these kernels, and its content was significantly (p < 0.05) lowered by kernels that received drying and then roasting (D + R) and improved by R. Dangang et al. [11] also identified calcium as the major mineral in jackfruit seeds. Assessment of calcium availability by determining Ca/P and Ca/Mg ratios shows that its high content saturates the passage sites in the intestinal lumen, thus limiting its absorption. Indeed, FAO/WHO [59] reports that for calcium absorption, Ca/P and Ca/Mg ratios must be 0.5 and 1, respectively, while in this study, these ratios ranged from 4.96 to 15.34 and from 6.61 to 32.10, respectively. Phosphate and magnesium content improved with R + B, with values of 198.17 and 155.16 mg/100 g, respectively. R alone and D + B, on the other hand, significantly (p < 0.05) lowered both minerals. Sodium content was significantly (p < 0.05) improved by D + R while R contributed to the opposite effect. This content dropped from 22.18 to 11.57 mg/100 g. Potassium content was reduced by all treatments, from 215.99 mg/100 g in the dried sample to 162.03 mg/100 g in the dried and then boiled sample. What is more, the Na/K ratio is well below 1, making these kernels suitable for hypertensive people.
| Parameters | D | R | D + R | D + B | R + B | D + R + B |
|---|---|---|---|---|---|---|
| Zn | 2.01 ± 0.42b | 0.00 ± 0.00c | 0.00 ± 0.00c | 0.00 ± 0.00c | 0.00 ± 0.00c | 20.34 ± 0.90a |
| Fe | 0.30 ± 0.01c | 0.02 ± 0.00e | 0.13 ± 0.01d | 0.36 ± 0.07b | 0.03 ± 0.00e | 0.60 ± 0.01a |
| Ca | 1119.50 ± 0.70e | 2550.00 ± 14.10b | 961.00 ± 1.40f | 2964.00 ± 1.60a | 1285.00 ± 7.07d | 1720.00 ± 56.60c |
| P | 195.32 ± 1.20ab | 181.91 ± 1.70c | 193.69 ± 2.30ab | 193.29 ± 2.20b | 198.17 ± 1.70a | 181.50 ± 1.20c |
| Mg | 151.75 ± 1.20b | 136.14 ± 0.70d | 145.36 ± 2.30c | 92.33 ± 2.90f | 155.16 ± 1.70a | 102.83 ± 1.20e |
| Na | 19.10 ± 1.50b | 11.57 ± 0.10d | 22.18 ± 0.60a | 20.20 ± 0.14b | 16.74 ± 0.50c | 12.03 ± 1.10d |
| K | 215.99 ± 4.13a | 200.22 ± 1.99b | 201.64 ± 3.99b | 162.09 ± 4.49d | 175.70 ± 5.60c | 165.33 ± 9.07cd |
| Na/K | 0.09 ± 0.00c | 0.11 ± 0.00a | 0.10 ± 0.00b | 0.07 ± 0.00d | 0.10 ± 0.00bc | 0.07 ± 0.01d |
| Ca/Mg | 7.38 ± 0.08e | 6.61 ± 0.04f | 32.10 ± 0.07a | 16.73 ± 0.73c | 8.28 ± 0.07d | 18.73 ± 0.12b |
| Ca/P | 5.73 ± 0.04e | 4.96 ± 0.05f | 15.34 ± 0.20a | 9.48 ± 0.25c | 6.48 ± 0.09d | 14.02 ± 0.06b |
- Note: Means ± standard deviations (n = 3) bearing the letters a, b, c… in the same column differ significantly at 5% of probability.
- Abbreviations: D, drying; D + B, drying and boiling; D + R, drying and roasting; D + R + B, drying then roasting and boiling; R, roasting; R + B, roasting and boiling.
These data also show that except for untreated kernels and those dried and then roasted, all other treatments produced calcium levels sufficient to cover the daily requirement of 1200 mg/day [60]. As far as Mg is concerned, only the dried + boiled sample could not cover the daily requirement of 100 mg after eating 100 g of these kernels [61]. Nevertheless, consuming 100 g of these kernels would by no means provide the proportions (2.1–28 mg) required to cover Fe needs [59]. On the other hand, they would cover potassium and phosphorus requirements [59, 62].
The significant loss of calcium noted in the various treatments is thought to be a consequence of their solubilization during soaking [49]. Notwithstanding this, its increase in triple-treated kernels (D + R + B) raises a question as to the effectiveness of the quantification methodology. Tambo et al. [63] also noted such variability in the content of this parameter in different samples. The content obtained in dried kernels is close to that of [46], which was 3.44 mg/100 g. The reduction in Fe content with R would be the consequence of a low denaturation of the antinutrient–Fe and macronutrient–Fe complexes, thus reducing its availability. Tambo et al. [39] observed the opposite effect after R cassava flour. Compared with the work of [10], the Fe data in this study are 5 times lower, necessitating supplementation with a Fe source to avoid iron deficiency anemia in people consuming this. As for Ca, D does not facilitate its dissociation from antinutrients such as phytates and oxalates, hence its low availability in untreated kernels. Nevertheless, Dongmo et al. [10] and Klang et al. [64] have reported that boiling or blanching facilitates the solubilization loss of minerals such as Ca, unlike R, which improves this parameter through the degradation of complexes formed with antinutrients. The contents obtained are 7 to 20 times higher than those of [46], which were 127.08 and 125.80 mg/100 g in boiled and boiled-roasted Irvingia gabonensis kernels harvested in southern Cameroon. Dangang et al. [11] explain this variation by the geographical difference in harvesting areas, the plant species, and the different conditions of culinary treatments. The high P and Mg contents in untreated and R + B samples are consistent with the results of [48, 64] and [38, 39]. These authors report that the aforementioned treatments improve the gelatinization of starch and amylose, facilitating the release of phosphorus and magnesium ions. The high temperatures involved in this process probably promote the destruction of organic parts and the release of previously bound elements. Ndomou et al. [46, 65] obtained lower levels of phosphorus and magnesium in rice flour and mandes from Irvingia gabonensis. As for Na, the values obtained by [46] are lower than those obtained in this study, demonstrating the value of increasing the conditions of the various culinary treatments to reduce sodium content and facilitate food intake by hypertensive individuals. As for potassium, the results obtained were higher than those of [44], which were 84.50 mg/100 g. Overall, it can be seen that each method generates selective changes in mineral composition depending on the physicochemical phenomena involved. Optimal processing should take into account the nutritional requirements being sought. Additional analyses would also enable us to characterize the actual bioavailability of elements after ingestion.
3.4. Influence of Culinary Treatments on Antinutrient Composition and Mineral Availability Ratio of I. wombolu Kernels
Phytates, oxalates, and tannins are negatively charged, enabling them to easily complex proteins and divalent cations, thus reducing their bioavailability [66]. Tannins are also responsible for the onset of certain cancers, notably esophageal cancer, when consumed more than 500 mg/day [67]. Table 4 shows that phytate content, which varied between 54.92 and 36.62 mg/100 g, was significantly (p < 0.05) lowered by all treatments except the triple treatment (D + R + B). Oxalate content, which varied between 24.75 and 15.78 mg/100 g, was significantly increased by R. These results are similar for phytates to those of [11], who reported a slight reduction in phytate content with boiling and R. For oxalates, however, they are contrary to those of [38, 39], who instead observed a strong reduction in oxalate content with R. There was also a sharp drop in the content of these two antinutrients with D + B and R + B samples. Phytates are thermolabile and highly soluble molecules, hence their reduction during these treatments. Phytate levels in this study are significantly lower than those of [68], who reported 391.53 mg/100 g in kernels of an Irvingia variety. As for oxalates, they are significantly lower than those of Fotso et al. [69] in soybeans. This demonstrates the value of applying culinary treatments to improve nutrient digestibility by reducing antinutrients. Moreover, the levels of these two antinutrients are below the threshold values (180 mg for phytates and 250 mg for oxalates) recommended by [52] for the formulation of supplementary feeds. Assessment of the availability of divalent cations (Fe, Ca, and Mg) by calculating the oxalate or phytate/ion ratio confirms the availability of Ca and Mg in all these samples, as the molar ratios obtained with these two samples are all below 0.25. This availability was better in the boiled kernels, confirming the high elimination of antinutrients by solubilization. Unlike these two minerals, the Fe present in these samples is unavailable, as the ratios are greater than 1. Moreover, this Fe availability was very low in the boiled samples. This may reflect a loss of Fe during this treatment or the stronger interactions between these antinutrients and Fe. Ca and Mg availability ratios were lower than those of [10], while Fe availability ratios were higher. Supplementation with Fe-rich and available feeds when formulating supplement feeds, including these kernels, would therefore be necessary. Saponin levels were not significantly affected by the treatments, although R followed by boiling eliminated these molecules from the kernels. In fact, R weakens the kernel wall, creating porosities that facilitate the entry of water during boiling and the hydrolysis of the complexes formed by saponins, followed by their elimination by solubilization. Similarly, their apolar structures make them less compatible with water, hence their high elimination rates. The levels obtained are similar to those of [68].
| Parameters | D | R | D + R | D + B | R + B | D + R + B |
|---|---|---|---|---|---|---|
| Phytates (mg/100 g) | 54.92 ± 8.63a | 39.67 ± 4.32b | 39.67 ± 0.00b | 39.67 ± 4.32b | 36.62 ± 0.00b | 54.92 ± 4.32a |
| Oxalates (mg/100 g) | 22.50 ± 0.00a | 24.75 ± 3.18a | 9.00 ± 0.00d | 11.25 ± 3.18cd | 15.78 ± 3.18bc | 20.25 ± 3.18ab |
| Saponins (mg/100 g) | 0.10 ± 0.00a | 0.10 ± 0.00a | 0.10 ± 0.00a | 0.10 ± 0.00a | 0.00 ± 0.00a | 0.10 ± 0.00a |
| Condensed tannins (mg CE/g of extract) | 3.40 ± 0.16ab | 4.17 ± 0.18ab | 4.27 ± 0.10a | 2.71 ± 0.01b | 3.61 ± 0.17a | 3.79 ± 0.32c |
| Hydrolyzable tannins (mg CE/g of extract) | 4.74 ± 0.34a | 3.94 ± 0.10b | 3.48 ± 0.42b | 1.42 ± 0.20d | 2.07 ± 0.05c | 1.75 ± 0.07cd |
| Phytates/Fe | 15.29 ± 0.35b | 24.63 ± 1.27b | 8.73 ± 0.17b | 5.62 ± 0.07b | 149.98 ± 33.29a | 178.09 ± 62.53a |
| Phytates/Ca | 0.003 ± 0.00a | 0.003 ± 0.00a | 0.003 ± 0.00a | 0.001 ± 0.00b | 0.003 ± 0.00a | 0.001 ± 0.00b |
| Phytates/Mg | 0.01 ± 0.00a | 0.01 ± 0.00a | 0.01 ± 0.00a | 0.01 ± 0.00a | 0.01 ± 0.00a | 0.01 ± 0.00a |
| Oxalates/Fe | 20.67 ± 0.48c | 25.61 ± 1.32c | 13.76 ± 0.27c | 4.68 ± 0.06c | 225.29 ± 50.01b | 407.35 ± 143.05a |
| Oxalates/Ca | 0.004 ± 0.00a | 0.003 ± 0.00b | 0.001 ± 0.00d | 0.001 ± 0.00d | 0.004 ± 0.00a | 0.002 ± 0.00c |
- Note: Means ± standard deviations (n = 3) bearing the letters a, b, c… in the same column differ significantly at 5% probability.
- Abbreviations: CE, catechin equivalent; D, drying; D + B, drying and boiling; D + R, drying and roasting; D + R + B, drying then roasting and boiling; R, roasting; R + B, roasting and boiling.
They are also below the threshold value (2000 mg/day), which suggests that consumption of these kernels does not promote diabetes, obesity, gastrointestinal disorders, or liver toxicity and makes the fruit more acceptable organoleptically by reducing the astringency they cause [70]. The condensed and hydrolysable tannin contents ranged, respectively, from 2.17 to 4.17 mg EC/g and from 1.42 to 4.74 mg EC/g. Given these results, D + B resulted in a significant reduction in these factors, while R had the opposite effect. This shows that these compounds are more soluble than thermolabile. Cooking with water favors hydrolysis of the complexes formed between tannins and fibers, particularly in the cell wall, thus promoting their loss. Dangang et al. [11] noted a significant reduction in these factors during R. These results suggest that the nature of tannins and the inter- and intramolecular forces that bind them vary from one species to another. Similarly, hydrolysable tannins decrease in all treatments, whereas condensed tannins do not. The strength of the molecular bonds holding each of these molecules together would explain these variations. Data from [44] for condensed tannins and [65] for hydrolysable tannins are lower than those recorded in this study. A difference between the food matrices as well as an improvement in the availability of these antinutrients through thermal hydrolysis of the complexes during the various culinary treatments would explain these results. In addition, the results obtained are well below the daily dose that could lead to cancer, demonstrating the low toxicity of these kernels. Tannins are known to be negative for human health. However, recent research has revealed that the phenolic structure of tannins can be exploited in the food, nutraceutical, and pharmaceutical industries [71].
3.5. Influence of Culinary Treatments on pH, Functional Properties, and Color of I. wombolu Kernels
Functional properties have an impact on product application. The pH provides information on a product’s taste acceptability, water retention capacity, and swelling rate [38]. Table 5 shows that this parameter varies between 5.76 and 5.51. It was improved by treatments including boiling but lowered by R. During boiling, the organic acids present in the kernels are easily removed by solubilization. Klang et al. [48] also noted a rise in pH during potato blanching. The values obtained are lower than the 7 reported by [39] as being that of a good food intake. This suggests an increase in processing time to reduce as many organic acids as possible. Similarly, we note that samples with low pH possess low water retention. Dongmo et al. [10] reported a positive correlation between pH and water retention, demonstrating that acidic pHs reduce interactions between almond molecules (proteins, fibers, carbohydrates, etc.) and water. Ndamitso et al. [72] obtained higher values than those in this study in Irvingia kernels from Niger. The aridity of the Niger almond growing and harvesting area would have limited the development of organic acids, resulting in a higher pH. The apparent viscosity of products that occurs during processing as a result of water retention can be assessed by water retention capacity and swelling rate [64]. Water retention capacity and swelling rate increased significantly (p < 0.05) with all treatments, particularly those including boiling. In addition, the high P and Ca content of these processed kernels, particularly those boiled, improves their water retention through electrostatic interactions and, above all, the stability of the complexes to shear [49]. A comparison between roasted and boiled samples for these two parameters shows that the Maillard reaction that occurs during R leads to a reduction in the availability of polar amino acids (Lys and Arg), thus lowering water retention and swelling rate. The results obtained show that these treated kernels can be used in the formulation of many products as thickeners. Organoleptic properties, as well as the ability to limit O2 diffusion, enhancing product preservation, can be characterized by its oil retention capacity [63]. This parameter ranged from 58.49% to 70.39% and was significantly (p < 0.05) improved by the culinary treatments except for the triple treatment (D + R + B). The treatments improve the concentration of hydrophobic amino acids such as Ala, Pro, and Gly, promoting hydrophobic interactions with fats [73]. Similarly, the increase in lipid content following the various treatments would also promote better hydrophobic interaction between fats, facilitating the retention of exogenous oil. Given these results, which are very high compared with those of [72], which were 26% in raw kernels, culinary treatments appear to be necessary to improve oil retention. These kernels could therefore be used in the formulation of cakes and doughnuts, as they require flours with good oil retention in order to maintain and improve their taste properties. The ability of these kernels to stabilize emulsions and create interfacial tensions that can reduce repulsions between liquids of different natures is assessed by the ORC/WRC ratio. The closer this ratio is to 1, the higher the emulsifying capacity of the matrix. These analyses show that the best emulsion-stabilizing properties are obtained with triple-treated kernels (D + R + B) and the lowest with dried kernels (D). The treatments resulted in the formation of globular proteins, which can integrate between the oil/water interfaces. Ndomou et al. [65] obtained higher ratios than those recorded in this study in infested and noninfested rice flours. This suggests that I. wombolu kernels are suitable for emulsion formulation and stabilization. This is observed at the local level through its use in the preparation of many sauces. Color is responsible for the choice and acceptability of a product. It is assessed by the parameters l∗ (whiteness), a∗ (redness), and b∗ (yellowness). The results show that whiteness (L∗) was low in roasted samples, in contrast to boiled samples. The products of caramelization, the Maillard reaction, lipid oxidation, and vitamins obtained during R would be responsible for the depreciation of whiteness [74]. A solubilizing elimination of these red-, green-, black-, and yellow-colored compounds with the treatments would explain the results with boiling. As for redness (a∗), it varied between 3.33 and 5.73 with a clear improvement during R, particularly R alone. The presence of adequate water content would be conducive to the good mobilization of substrates for the Maillard reaction and caramelization during this treatment, resulting in the formation of red-colored compounds. These results are confirmed by the high carbohydrate content. Klang et al. [48] reported a positive correlation between carbohydrate content and reduced whiteness of potato flour. Yellowness (b∗) and browning index moved in the same direction as redness (a∗). These parameters are inversely proportional to whiteness, as they result from the degradation of whiteness following various food alteration reactions (lipid oxidation, nonenzymatic and enzymatic browning). The whiteness and redness values obtained are better than those of Sobowale et al. [75] on germinated and fermented Bambara flours. The germination leads to a significant production of reducing sugars following starch hydrolysis, and this depreciates the white-to-red color that is characteristic of the new products formed [76]. The table shows that triple-treated samples (D + R + B) are ideal for the formulation of products requiring good whiteness (ice cream and baby food).
| Parameters | D | R | D + R | D + B | R + B | D + R + B |
|---|---|---|---|---|---|---|
| pH | 5.56 ± 0.02b | 5.51 ± 0.01b | 5.54 ± 0.04b | 5.76 ± 0.08a | 5.71 ± 0.05a | 5.76 ± 0.03a |
| WRC (%) | 32.00 ± 2.83b | 39.00 ± 4.24ab | 42.00 ± 5.66ab | 47.00 ± 7.07ab | 48.00 ± 0.00a | 45.00 ± 9.90ab |
| Swelling power (%) | 77.26 ± 1.96a | 79.48 ± 1.57a | 80.51 ± 6.44a | 82.70 ± 1.35a | 82.74 ± 0.26a | 81.78 ± 2.67a |
| ORC (%) | 58.49 ± 0.07b | 62.57 ± 1.77ab | 63.43 ± 0.97ab | 63.54 ± 2.53ab | 70.39 ± 3.41a | 59.26 ± 7.51b |
| WRC/ORC | 1.83 ± 0.16a | 1.62 ± 0.22ab | 1.53 ± 0.23ab | 1.43 ± 0.28ab | 1.47 ± 0.06ab | 1.33 ± 0.13b |
| l∗ | 67.71 ± 0.01c | 60.73 ± 0.01f | 64.66 ± 0.01e | 69.83 ± 0.04b | 65.71 ± 0.01d | 71.78 ± 0.01a |
| a∗ | 3.84 ± 0.02c | 5.73 ± 0.01a | 4.88 ± 0.01b | 3.38 ± 0.02e | 3.44 ± 0.01d | 3.33 ± 0.01f |
| b∗ | 22.17 ± 0.01d | 25.06 ± 0.02a | 23.25 ± 0.01c | 21.39 ± 0.01f | 24.75 ± 0.00b | 21.43 ± 0.01e |
| Browning index | 42.54 ± 0.04d | 58.23 ± 0.04a | 48.64 ± 0.03c | 38.93 ± 0.03e | 49.42 ± 0.01b | 37.71 ± 0.01f |
- Note: Means ± standard deviations (n = 3) bearing the letters a, b, c… in the same column differ significantly at 5% probability.
- Abbreviations: CE, catechin equivalent; D, drying; D + B, drying and boiling; D + R, drying and roasting; D + R + B, drying then roasting and boiling; ORC, oil retention capacity; R, roasting; R + B, roasting and boiling; WRC, water retention capacity.
3.6. Influence of Cooking Treatments on the Rheological Properties of I. wombolu Kernels
The rheological parameters of food matrices depend not only on the starch but also on the chemical complexity and association between the various chemical constituents of the food [48]. They enable us to mimic the behavior of a matrix during cooking and its stability in treatments [10]. To define the time and energy required to cook the kernels, the starching time and temperature were evaluated. Table 6 shows that, except for triple treatment, packing temperature decreased with cooking treatments, particularly R, while the opposite effect was noted with temperature. The significant hydrolysis of bond energies during triple processing (D + R + B), which transforms molecules from crystalline to amorphous forms, is responsible for the reduction in processing temperature [77]. Authors such as [49] have noted a reduction in time and starching temperature during the blanching of canistel kernels. It also follows that R reduced the starching time but increased the temperature, unlike boiling. The thermal conducting and diffusing effect of the water contained in boiled kernels is thought to be responsible for the reduction in gelatinization temperature [49]. PV is the viscosity obtained when the temperature is raised and is influenced by chemical composition, notably protein, lipid, and especially starch content, as well as swelling power [78]. It varied between 642.30 and 1720.50 cP and was higher in kernels that had undergone more than one treatment. Likewise, boiling gave the kernels the greatest opportunity to gain volume during heat treatment. Klang et al. [48] reported that blanching resulted in gelatinization without dissociation of the molecule and weakening of its covalent bonds, favoring good water retention and swelling during cooking. In addition, the richness in hydrophilic amino acids in triple-treated kernels (D + R + B) and in phosphorus in boiled kernels would enable them to maintain water bonds during cooking. It was also reported that the low pHs of roasted kernels would not facilitate intermolecular interactions with water [64]. The values obtained are far lower than those of Zhang and Hamaker [79], which were 2016 cP, confirming their agreement with the viscosity of supplement foods (0.5 to 2 cP). During cooking, pellets are subjected to shear stress, the ability of which is assessed by the HV [49]. A small variation in this parameter with the PV indicates low retrogradation and therefore high stability. It ranges from 99.82 to 632.30 cP. There is a strong variation between this parameter and PV in the treated samples, reflecting their low shear stability, as demonstrated by the setback and stability ratio values. The weakening of chemical bond energies during culinary treatments and the high Ca and K contents in treated kernels would reduce the ability of phosphorus present at the surface of molecules to maintain bonds with water. These positive ions will form electrostatic interactions with phosphors, thus reducing the stabilizing interactions of phosphate on proteins [63]. The reduction in stability with treatment (stability ratio from 0.56 to 0.12) and the increase in shear during cooking (setback ratio from 1.48 to 7.31) are in line with the reduction in protein content with treatment. In addition, the higher lipid content of treated samples reduces the thermal stability of amylose [10]. During cooking, the phenomenon of advanced gelatinization could lead to the bursting of granules, which, once cooled, would reassociate [80]. This phenomenon, known as breakdown, is inversely proportional to stability [80]. It ranged from 321.90 to 1107.50 cP and increased with treatment, confirming the reduced stability and low shear strength of treated kernels. As boiled kernels are generally the most affected by shearing, this suggests that their treatment conditions should be reduced in order to limit retrogradation. The values obtained are higher than those of Chung et al. [81], which were 161 cP. Combining kernels with carbohydrate sources would reduce the rate of retrogradation and thus facilitate food intake. After baking, the viscosity obtained is called the FV and depends on the thermal stability of the food’s chemical constituents [82]. It also enables us to select the right flour for the formulation of supplementary feeds, as its viscosity represents that of the feed intake [10]. Indeed, all the viscosities obtained are in the range 0.5–2000 cP for food intakes in children aged 6–59 months [38]. This parameter varied between 1182.50 and 663.40 cP and was lowered by all treatments, particularly R. Nevertheless, there was little variation between this parameter and PV in the dried-roasted samples, which would confirm their high stabilities. The macromolecules present in untreated samples are not denatured, which would increase their swelling power on cooling. Similarly, the dissociation of macromolecules during processing is responsible for the lower FV [77]. Untreated kernels could therefore be used to produce jellies and cakes. After gelatinization, the ability to form a gel is assessed by setback [78]. Gelatinization caused molecules to burst, reducing their viscosity and stability on cooling [49, 80]. In addition, Reference [49] noted that a loss of gelling capacity was associated with high calcium content in blanched canistel nuts, as is the case in kernels of this Irvingia variety. They are also similar to the results of [10], who noted a drop in setback when corn flour was processed. These results ranged from 752.95 to 301.70 cP but remained lower than the values (1127–1747 cP) obtained by [77] in canistel nut starch. This demonstrates the low gelling capacity of these kernels and consequently low retrogradation.
| Samples | D | R | D + R | D + B | R + B | D + R + B |
|---|---|---|---|---|---|---|
| Peak time (s) | 486.95 ± 1.48b | 465.70 ± 0.42e | 458.95 ± 0.07f | 473.50 ± 0.71c | 470.95 ± 1.48d | 504.50 ± 0.71a |
| Pasting temperature (cP) | 92.73 ± 0.39a | 66.12 ± 0.17c | 67.43 ± 0.04b | 61.75 ± 1.06d | 67.53 ± 0.03b | 58.57 ± 0.61e |
| PV (cP) | 767.90 ± 17.11e | 829.95 ± 4.17d | 642.30 ± 0.99f | 971.00 ± 5.66c | 1720.50 ± 21.92a | 1026.00 ± 4.24b |
| HV (cP) | 429.50 ± 6.36d | 99.82 ± 0.25f | 312.50 ± 0.71e | 529.50 ± 4.95b | 632.30 ± 3.82a | 484.40 ± 2.26c |
| Breakdown (cP) | 321.90 ± 0.14e | 734.65 ± 1.91b | 332.35 ± 3.75e | 672.50 ± 10.61c | 1107.50 ± 0.71a | 539.00 ± 1.41d |
| FV (cP) | 1182.50 ± 79.90a | 729.95 ± 1.48c | 663.40 ± 7.92c | 868.00 ± 29.70b | 934.00 ± 4.24b | 920.90 ± 0.14b |
| Setback (Cp) | 752.95 ± 86.34a | 630.10 ± 1.27b | 350.95 ± 7.14cd | 338.50 ± 24.75d | 301.70 ± 8.06d | 436.50 ± 2.40c |
| Stability ratio (HV/PV) | 0.56 ± 0.02a | 0.12 ± 0.00d | 0.49 ± 0.00b | 0.55 ± 0.00a | 0.37 ± 0.01c | 0.47 ± 0.00b |
| Setback ratio (FV/HV) | 2.75 ± 0.23b | 7.31 ± 0.00a | 2.12 ± 0.02c | 1.64 ± 0.04d | 1.48 ± 0.02d | 1.90 ± 0.01c |
- Note: Means ± standard deviations (n = 3) bearing the letters a, b, c… in the same column differ significantly at 5% probability.
- Abbreviations: CE, catechin equivalent; D, drying; D + B, drying and boiling; D + R, drying and roasting; D + R + B, drying then roasting and boiling; FV, final viscosity; HV, holding viscosity; PV, peak viscosity; R, roasting; R + B, roasting and boiling.
3.7. Influence of Cooking Treatments on Thermal Properties and PRAL Index of I. wombolu Kernels
Thermal properties provide information on cooking conditions and product homogeneity. They are influenced by physicochemical composition and the interactions between them. Table 7 shows that the specific heat capacity, which represents the energy required to increase or vary the temperature of the seeds in the smallest unit, varied between 176.74 and 161.92 kJ/kg/°K. It was significantly (p < 0.05) reduced by treatments, in particular R. The lowering of this energy can be explained by the treatments, which reduced the crystallinity of the molecules present in the kernels, facilitating their gelatinization [83]. The high water content in dried (untreated) samples requires more energy for cooking and therefore for raising the temperature. A comparison between treatments shows that R facilitates better dispersion of particles within the molecule, as well as weakening inter- and intramolecular bonding forces. Raji et al. [84] obtained values (177.31–196.45 kJ/kg/°K) similar to those in this study. These data also show that processed kernels are more energy-efficient during cooking. The ability to diffuse heat within a matrix is characterized by its thermal conductivity [85]. It ranged from 17.47 to 20.08 W/m·K. It was lowered by all treatments, notably those including D + torrefaction and D + B. These results can be explained by their low water content. Indeed, Reference [85] reported that water is a good heat carrier thanks to the hot steam it releases during heating. Raji et al. [84] obtained values close to these. These results show that heat is distributed evenly throughout the kernels during cooking. Heat exchange between the product and its environment is assessed by thermal diffusivity [23]. This capacity was very high with triple treatment (8.50∗10−6 m2/s) but low with D + R (7.60∗10−6 m2/s). The high protein content of kernels dried and then roasted could explain this result. The proteins are thought to form compact molecular networks, preventing the release or diffusion of heat to the surface. Similarly, boiled kernels show better thermal diffusivity, suggesting that these treatments were applied before the kernels were cooked. Nevertheless, R would be favorable insofar as it would limit the formation of vapors due to heat loss during product packaging. The same results (8.17–8.81∗10−6 m2/s) were obtained in the same range when formulating pasta [84]. The PRAL index refers to the ability of a formulation or food to raise metabolic acidity or basicity after consumption [86]. This index ranged from −8.02 to −27.45 and was negatively correlated with pH. Indeed, kernels with a high pH had a low PRAL index and vice versa. Moreover, this index was below 0 in all kernels, suggesting that the treatments applied would be conducive when cooking food for obese, diabetic, and renal failure sufferers [87]. This low index thus confirms that almond consumption would not lead to metabolic acidosis and consequently to possible cancers. According to [88], the low PRAL index would not affect the musculature of the elderly by creating muscle frigidity. Indeed, Kataya et al. [89] noted a reduction in muscle stiffness in Japanese women consuming foods with a high PRAL index. The high value of the PRAL index in roasted and dried kernels would be linked to proteins, which are known for their strong capacity to increase the acidity of products, notably due to their acidic amino acid compositions but also following their decomposition with the production of ketone bodies.
| Samples | D | R | D + R | D + B | R + B | D + R + B |
|---|---|---|---|---|---|---|
| Specific heat capacity (kJ/kg/°K) | 176.74 ± 0.39a | 161.92 ± 2.37d | 166.31 ± 0.99bc | 162.50 ± 0.41cd | 164.03 ± 2.59bcd | 166.57 ± 1.21b |
| Thermal conductivity (W/m·k) | 20.08 ± 0.03a | 18.53 ± 0.27b | 17.48 ± 0.53c | 17.47 ± 0.27c | 18.16 ± 0.06b | 18.61 ± 0.03b |
| Thermal diffusivity (m2/s∗10−6) | 8.27 ± 0.01b | 8.25 ± 0.07bc | 7.60 ± 0.12d | 8.11 ± 0.01c | 8.24 ± 0.04bc | 8.50 ± 0.03a |
| PRAL index | −8.63 ± 0.15a | −8.02 ± 0.13a | −26.98 ± 0.07d | −15.10 ± 0.63c | −11.12 ± 0.48b | −27.45 ± 0.15d |
- Note: Means ± standard deviations (n = 3) bearing the letters a, b, c… in the same column differ significantly at 5% probability.
- Abbreviations: D, drying; D + B, drying and boiling; D + R, drying and roasting; D + R + B, drying then roasting and boiling; R, roasting; R + B, roasting and boiling.
3.8. Influence of Cooking Treatments on the Amino Acid Profile and Concentration of I. wombolu Kernels
The use of protein-rich matrices in the formulation of supplementary feeds in particular takes into account the biological quality of their proteins. Table 8 shows the composition and concentration of amino acids in kernels that have undergone different treatments and shows that both the number and concentration are influenced by cooking treatments. The number of amino acids identified ranged from 6 to 10, demonstrating a low level of variability compared with animal proteins, necessitating supplementation with legumes such as soy during formulation. Similarly, this variability could be the consequence of amino acid destruction during extraction or incomplete protein hydrolysis [90]. Among the amino acids identified, only four (His, Lys, Phe, and Met) were essential, confirming the low biological value of I. wombolu kernel proteins. These four essential amino acids were identified in the roasted kernels, while three were identified in the dried and then boiled kernels and the first two (His and Lys) in the other samples. This can be explained by the ability of R + B to break down the constitutive proteins (mainly hydrophobic amino acids) located in the wall, thereby releasing the essential hydrophobic amino acids [90]. We also note that essential amino acid content increases with R and decreases with boiling. Fellows [91] reported that the solubilization loss of these amino acids was responsible for the decrease in their concentrations. Nevertheless, the use of Lys in the Maillard reaction that takes place during the R of kernels would explain the drop in its content [90]. Among the nonessential amino acids, Arg was identified in roasted kernels, and its concentration decreased during R and boiling. This confirms the negative effect of ebullition on nutrient retention. In the case of serine, the drop during boiling is explained by leaching. This small glucoforming amino acid has significant antioxidant activity and also plays a role in covalently modulating substrate binding at the active site of certain enzymes [92]. Gly, Glu, Pro, and Asp were identified in all the kernels, with a clear improvement with the different treatments, particularly the triple treatment. The combination of heat treatments would have facilitated the release of amino acids present in cell membranes and walls. In addition, the richness of acidic amino acids is thought to be responsible for lowering the pH of roasted kernels and their odor properties [93]. Cooking treatments, except for D + R, identified Ala at proportions ranging from 0.075 to 0.203 mmol/L. This once again raises questions about the sensitivity of the method and the protein hydrolysis conditions used.
| Samples | D | R | D + R | D + B | R + B | D + R + B |
|---|---|---|---|---|---|---|
| Essential amino acid (mmol/L) | ||||||
| His | 0.024 | 0.100 | 0.027 | 0.095 | 0.181 | 0.081 |
| Lys | 1.919 | 1.787 | 1.948 | 1.860 | 1.775 | 2.104 |
| Phe | 0.000 | 0.006 | 0.000 | 0.004 | 0.000 | 0.000 |
| Met | 0.000 | 0.006 | 0.000 | 0.000 | 0.000 | 0.000 |
| No essential amino acid (mmol/L) | ||||||
| Arg | 0.000 | 0.011 | 0.000 | 0.000 | 0.007 | 0.000 |
| Ser | 0.000 | 0.146 | 0.000 | 0.098 | 0.000 | 0.088 |
| Gly | 0.126 | 0.386 | 0.073 | 0.634 | 0.408 | 0.388 |
| Pro | 0.817 | 0.866 | 0.748 | 1.337 | 0.911 | 0.986 |
| Glu | 0.019 | 0.076 | 0.059 | 0.055 | 0.063 | 0.071 |
| Asp | 0.823 | 0.920 | 0.920 | 0.753 | 0.872 | 0.931 |
| Ala | 0.000 | 0.203 | 0.000 | 0.105 | 0.152 | 0.075 |
- Abbreviations: D, drying; D + B, drying and boiling; D + R, drying and roasting; D + R + B, drying then roasting and boiling; R, roasting; R + B, roasting and boiling.
3.9. Influence of Cooking Treatments on the Phytochemical Composition and Antioxidant Activity of I. wombolu Kernels
3.9.1. Phytochemical Content
Phytochemicals limit the occurrence of numerous pathologies such as cancer, diabetes, and obesity, thanks to their ability to limit the formation of intracellular free radicals [94]. These properties are due to the presence of ortho and para hydroxyl groups on their surfaces [95]. The evaluation of these properties enables us to confer a functional character to the food. Flavonoids are a class of phenols with proven anticancer and antiobesity properties [46]. Unlike phenols, they are not very polar due to the number of benzene rings [96]. Table 9 shows that flavonoid content in I. wombolu kernels evolved positively with treatment. In particular, R improved extraction and flavonoid concentration. These results concur with those of Jeong et al. [97], who noted an improvement in flavonoid content during R. Indeed, during R, the bond energies holding flavonoids to other molecules in the wall are broken [98]. As for boiling, leaching during processing is thought to be the main cause of a reduction in this parameter. Flavonoid levels were higher than those reported by Tchimmoe et al. [46] in boiled and roasted Irvingia gabonensis kernels. The collection area and the conditions of the different treatments are thought to account for these differences. Phenolic content ranged from 92.30 ± 9.00 to 33.33 ± 3.62 mg GAE/g and was significantly (p < 0.05) reduced by all treatments. These results are inconsistent with those of Oladele et al. [99], who reported the opposite effect during R. Ofeole et al. [100] report in their work the value of respecting treatment temperatures to reduce the destruction of phytochemical compounds in plants. These results suggest that kernels could be used to limit cell oxidation. However, due to a lack of material, no specific polyphenol or fatty acid chromatographic analyses were carried out.
| Parameters | Flavonoids (mg CE/g of extract) | Phenols (mg GAE/g of extract) |
|---|---|---|
| D | 12.36 ± 1.94c | 92.30 ± 9.00a |
| D + R + B | 27.74 ± 1.94ab | 33.33 ± 3.62c |
| R | 28.29 ± 0.39ab | 37.18 ± 0.00c |
| D + R | 31.03 ± 0.39a | 62.18 ± 0.90b |
| R + B | 27.74 ± 1.94ab | 33.33 ± 3.63c |
| D + B | 25.81 ± 0.00b | 39.11 ± 6.35c |
- Note: Means ± standard deviations (n = 3) with different letters in the same column are significantly different at the 5% probability threshold.
- Abbreviations: CE, catechin equivalent; D, drying; D + B, drying and boiling; D + R, drying and roasting; D + R + B, drying then roasting and boiling; GAE, gallic acid equivalent; R, roasting; R + B, roasting and boiling.
3.9.2. Evaluation of the Antioxidant Capacity of Different Kernels
3.9.2.1. Ability to Trap the DPPH Radical
Improving the antioxidant pool involves consuming functional foods. These can limit damage to DNA and other human cells [101]. In order to assess a plant’s functional potential, the DPPH test is one of the best known, as it mimics the oxidation phenomenon in vivo. For the reference antioxidants, it can be seen that the antifree radical activity of BHT is very low compared to vitamin C with all the concentrations used (Figure 2a). This can be explained by the configuration of OH functions around the benzene ring, which does not facilitate the departure of protons, and by the concentration of phenolic compounds [102]. However, concerning the variation of DPPH Radical inhibitory capacity at different concentrations of I. wombolu extracts, it was recorded that R recorded the high values of this parameter compared to other treatments (p < 0.05) at the concentrations of 200 and 100 μg/L, respectively (Figure 2b). This is in agreement with the work of Djikeng et al. [98], who noted better activity with roasted tiger nuts and a link between phenol content and DPPH trapping power. However, the values obtained with the other treatments are lower than those found by Kenfack et al. [103] in spiny green Sechium edule (IC50 less than 200 μg/mL). This might be explained by the treatment, which leads to the destruction of antioxidants. The classification by Souri et al. [104] showed that extracts from I. wombolu kernels are not good antioxidants, as the EC50s are all above 200 μg/mL. This suggests a combination with plants with high antioxidant capacities when used.
3.9.2.2. Reducing the Power of Ferric Ions in Different Extracts (FRAP)
Cell and molecule alteration reactions are initiated by transition metals such as Fe, and their reductions to a less electrophilic state would protect against the various alterations [105]. Their reductions require not only the presence of proton-donating compounds but, above all, electrons [106]. For the reference antioxidants, it was recorded that vitamin C showed very high values compared to BHT with all the concentrations used (Figure 3a). Concerning the effect of treatments on the reducing power of ferric ions, it was recorded that extracts obtained after the treatment R + B recorder (72.48 μmol of Trolox/g) the high value (p < 0.05) at the concentration of 200 μg/mL, while R showed the high value (50.26 μmol of Trolox/g) at the concentration of 100 μg/mL (Figure 3b). In general, the reducing potential of I. wombolu extracts on Fe was very low compared to the potential of vitamin C. The concentration of phenolic compounds and potential electron-donating tannins in redox reactions would explain this low variation [107, 108]. Similarly, the configuration of OH groups around the benzene rings of phenolic compounds limits the departure of the proton associated with that of electrons [106]. The observed activities are in agreement with those of [11], who noted that the ability of jackfruit kernels to stabilize and reduce cellular oxidation depended on culinary treatments. These results thus suggest combining kernels with vitamin C to potentiate antioxidant power and limit the harmful effect of synthetic antioxidant consumption on health (cancer, immune system deficiency, and kidney failure). The results obtained are less significant than those of Kenne et al. [105], which ranged from 19.61 to 301.72 μmol of Trolox/g in different turmeric extracts.
3.9.2.3. ROH Scavenging Capacity of Various Extracts
Evaluation of the ROH scavenging capacity of the various extracts is another parameter that confirms the antioxidant potential of I. wombolu kernel extracts. Figure 4a shows that the evolution of the ROH inhibition capacity of BHT decreases with concentrations, but it presents very high values of this parameter than vitamin C at different concentrations. For the effects of treatment on the ROH scavenging capacity of the various extracts of I. wombolu kernel, it has been noted that dried + roasted samples presented high values at the concentrations 200 and 100 μg/mL, respectively (Figure 4b). This observation is in line with the work of Kenfack et al. [102], who reported the highest inhibitory capacities with braised yams for the Sapa variety. Indeed, culinary treatments improve the availability of flavonoids and their capacities to liberate electrons, which are known to be powerful antioxidants [109]. Nevertheless, this potential remains very low compared to that of BHT, confirming once again the low antioxidant power of these kernels. The inhibitory potential observed with these kernels is lower than that of Kenfack et al. [102], and this can be explained by the duration of the various treatments, the flavonoid content, and the hardness of the wall limiting the extraction of antioxidants. Activities obtained by Kenfack et al. [103] in Sechium edule extract were greater than those of this work and can be explained by the destruction of some antioxidants like tannins.
3.10. PCA of Variables
The PCA diagram (Figure 5) of the variables, also known as the correlation circles of the variables, shows the grouping of these factors (nutritional value, phytochemical properties, antioxidant potential, mineral and antinutrient content). Not only can correlations be visualized, but proximities between the different treatments can also be seen directly. As for the observations (marked here by the different treatments carried out on I. wombolu kernels), they illustrate the contributions of each observation. It shows that the F1 component contributing 43.80% is associated with the D + R treatment and the F2 component contributing 19.69% is associated with the R and D treatments, respectively. The biplot associating the F1 and F2 axes shows that the treatments involving boiling contribute most to the conservation of a large number of minerals and that grilling alone eliminates the maximum number of antinutritional factors.
4. Conclusion
At the end of this study, the assessment of the effect of culinary treatments on the physicochemical and technofunctional properties of I. wombolu (bitter-tasting) kernels harvested in the town of Bertoua revealed that among the various treatments carried out, the one with the lowest water content but highest protein content was the sample that had undergone D + R. R preserved and improved mineral content, while all treatments, particularly D + B, considerably reduced antinutrient content, namely tannins, saponins, phytates, and oxalates. For the technofunctional properties, R + B improved water and oil retention, while color was best preserved with the triple treatment (D + R + B). As for viscosity, it increased with R + B, while phenol content decreased with culinary treatments, unlike flavonoid content. The antioxidant potential was very low in these samples, although the kernels roasted and then boiled were the most active among the samples. The treatments resulted in significant destruction of amino acids. These results suggest that taking into account the processing conditions of I. wombolu kernels can reduce nutrient loss and enhance the nutritional status of consumer populations.
4.1. Highlights of the Work
- ➢
I. wombolu kernels’ protein content was increased by the D and torrefaction process, while samples with high mineral concentrations were recorded with R treatment.
- ➢
The number of antinutrients in I. wombolu kernel samples decreased after R + B.
- ➢
The technofunctional properties, such as water-holding capacity and oil-holding capacity, were better with the R + B treatment of I. wombolu kernels.
- ➢
The antioxidant power has been improved with the R + B of I. wombolu kernel.
- ➢
The treatments of D and D + R resulted in significant destruction of amino acids, whereas R preserved the amino acids identified in I. wombolu kernel.
4.2. Limitations of the Studies
- •
Evaluation of the bioavailability of the nutrients using Wistar rat models.
- •
Evaluation of the influence of variety and treatments on the physicochemical properties of kernels.
- •
Evaluation of the in vivo and in vitro digestibility of the different nutrients.
4.3. Recommendation
- •
Raise public awareness of the nutritional benefits and antioxidant activity of certain culinary treatments of I. wombolu fruit kernels.
- •
Encourage the consumption of I. wombolu fruit kernels in Cameroon while promoting the treatments that best preserve their virtues.
- •
To popularize knowledge of the culinary treatments best suited to the development of the population.
Ethics Statement
The authors have nothing to report.
Disclosure
This work is completely original, according to the authors, as it has never been published before and is not being considered for publication in any other journal.
Conflicts of Interest
The authors declare no conflicts of interest.
Author Contributions
Ulrich Landry Kamdem Bemmo: conceptualization (lead), investigation (equal), methodology (equal), writing—original draft (equal), writing—review and editing (equal). Serge Cyrille Houketchang Ndomou: conceptualization (equal), data curation (equal), writing—original draft (equal), writing—review and editing (equal). Stephano Tene Tambo: methodology (equal), writing—review and editing (equal). Jean Marcel Bindzi: Formal analysis (equal); writing—original draft (equal); writing—review and editing (equal). Luc Martial Kack Bea: investigation (equal), methodology (equal), writing—original draft (equal), writing—review and editing (equal). Hilaire Macaire Womeni: supervision (lead), validation (equal), writing—original draft (equal), writing—review and editing (equal). François Ngoufack Zambou: supervision (lead), validation (equal), writing—original draft (equal), writing—review and editing (equal). All authors read and approved the final manuscript.
Funding
No funding was received for this manuscript.
Acknowledgments
We would like to express our gratitude to the participants of the survey and also to the Director of the National Institute of Food Technology and Management-Thanjavur (NIFTEM-T) and Professor Venkatachalapathy Natarajan, Head of the Food Engineering Department, for facilitating us in performing the viscoamylographic, color, and amino acid profile parameters.
General Statement
Additional Information. No additional information is available for this article.
Open Research
Data Availability Statement
The data that support the findings of this study are available from the corresponding authors upon reasonable request.
