The results showed that the soaking time followed by the decolorizing time exhibited the most significant (p < 0.05) effect on the extraction yield; while the soaking temperature had the least significant (p < 0.05) effect on the extraction yield (Table 1). As also shown in Figure 1, the extraction yield differed from 23.7% to 93.8% depending on the chemical extraction conditions. This was higher than the extraction yield reported for hydrocolloid from Prosopis flexuosa DC seeds (9%–13%) [13], acid extracted mesquite seed gum (17.7%) and acid extracted mesquite seed gum (24.9%) [10]. Singh and co-researchers [14] reported a relatively low extraction yield (26%) for the mucilage from Cassis javanica seed obtained by chemical extraction. It should be noted that the higher extraction yield did not reflect a desirable effect in most cases. For example, the chemical extraction resulting in higher yield led to a crude seed gum with lower solubility. This could be due to the presence of impurities and insoluble matter induced by chemical extraction at an elevated soaking temperature over a long time.

The extraction efficiency also depends on the purity and functional properties of the crude gum extract. Figure 1a,b showed that the extraction yield increased in most cases by prolonging the decolorizing time and soaking time. This could be due to the availability of solvent for a longer time which increased the driving force of mucilage out of the seeds, thereby increasing the extraction yield [15]. The increase in polysaccharide yield could be explained by the large effect of extraction time and temperature on the mass transfer rate of the water soluble polysaccharides in the cell wall [16]. On the other hand, it could be interpreted by the passage of impurities such as pigments or tannic substances from the brown tegument or the seed shell into the crude extract [17]. After soaking in solvent, the endosperm is swollen; therefore the presence of accumulated solvent around the endosperm leads to binding of the soluble components, thus increasing the extraction yield [17].

The chemical extraction at elevated soaking temperature for a longer time led to an increase in the extraction yield (Figure 1a,b). This observation was also reported by previous researchers [18]. This could be due to a reduction in the viscosity of seed mucilage at elevated temperature. Thus, the mucilage can be easily released due to less sticky seeds, thereby increasing the extraction yield. This observation could be also explained by the fact that the solvent molecules have more energy to break the chemical bonds between constituent particles in compounds at elevated temperature, thus increasing the extraction yield [15,16]. The net dissolution process becomes an endothermic reaction (requires energy), if the degree of heating for the dissolution process is less than the heating level required to rupture the solid. Under this condition, elevated temperature facilitates the dissolution process by giving more energy to break apart the bonds in the solid. As explained by Piñeiro and co-researchers [19], elevated temperature results in a higher extraction yield due to two different mechanisms. First, elevated temperature results in weakness of the bonds between the polysaccharide gum and its matrix, thereby increasing its solubility. Second, the high extraction temperature results in the degradation of compounds. Therefore, elevated temperature is a driving force facilitating the extraction of the degraded compounds. Although elevated temperature is favourable for the extraction of carbohydrates, the thermal degradation however, induces undesirable effects on the functional properties.

The results indicated that the viscosity of crude durian seed gum varied from 3.3 mPa.s to 24.3 mPa.s depending on the extraction conditions. The changes of viscosity could be related to the significant (p < 0.05) effect of the soaking process on the content and structure of the protein fraction present in the structure of the gum. The viscosity of a hydrocolloid is dominated by concentration, molecular weight and structure (linear, branched or slightly branched). In general, durian seed gum did not induce a very high viscosity after dispersing in water. This might be due to the fact that durian seed gum does not have a high molecular weight (Mw) structure as reported previously [20]. It has been proven that the gum with a larger Mw causes stronger viscosity than the gum with a lower Mw. Our previous study [20] revealed the considerable effect of extraction conditions on the molecular structure of durian seed gum. This might be responsible for significant changes in the apparent viscosity of the gum as a function of the extraction variables. In addition, the apparent viscosity also depends on the chemical composition and structure of the gum. As also reported earlier [20], the extraction conditions significantly affected the chemical structure (i.e., carbohydrate, protein and lipid fraction) of the crude durian seed gum. This could be also responsible for significant changes in the apparent viscosity of durian seed gum as a function of extraction variables. As also stated by Balaghi and co-researchers [21], several factors (such as neutral sugars, uronic acid, methoxyl group content, configuration and position of glycosidic linkages, molecular weight and conformational properties) play a crucial role in the rheological properties of the hydrocolloid.

In the current study, the range of viscosity was lower than the viscosity reported for Prosopis flexuosa DC seed mucilage (58 mPa.s–277 mPa.s at a shear rate of 64 1/s) [13], achi (Brachystegea eurycoma), and basil seed gum (581.37 mPa.s at a shear rate of 98.9 1/s) [22]. However, the rheological properties of hydrocolloid depend on shear rate, temperature, pressure and shearing time [23]. As stated by previous researchers [24], the rheological properties of a hydrocolloid are mainly influenced by the droplet size and distribution, droplet volume fraction, droplet concentration, deformability and the inter-droplet forces. The chemically-extracted crude durian seed gum showed a typical “pseudoplastic or shear-thinning behavior”. The viscosity of the sample decreased with increasing shear rate. As the shear rate increased, the droplet–droplet interaction became deformed and eventually disrupted, thus resulting in a size reduction of the flocks thereby decreasing the viscosity [25]. Previous researchers also reported a similar shear-thinning behavior in the following plant gums: basil seed gum [3], cress seed (Lepidium sativum) gum [26], and Alyssum homolocarpum seed gum [27].

The soaking temperature and decolorizing time exhibited the most and least significant (p < 0.05) effect on the viscosity of the crude durian seed gum. The present study revealed that the viscosity increased as the extraction was carried out for a short decolorizing and soaking time (Figure 2). In general, the viscosity of crude durian seed gum solution increased when the soaking temperature was increased (Figure 2). This might be explained by the more efficient extraction of polysaccharides at the elevated soaking temperature which increased the viscosity. On the other hand, the extraction at an elevated soaking temperature led to an increase in the total soluble solids, thus inducing a higher thickness. This is in agreement with that reported by Ozkanli [28]. Singthong and his co-researchers [11] also reported that the extraction at elevated temperature and extended time provided a high extraction yield, which was responsible for a high viscosity for Yanang gum.

The protein content is a key biological aspect of a gum. The presence of protein in polysaccharides can induce an inflammatory response in tissues, thereby inhibiting the pharmacological application of polysaccharide-based materials [29]. Our previous study [20] revealed that the protein fraction and amino acid composition of durian seed gum were significantly influenced by extraction variables which had a significant (p < 0.05) effect on the protein content of durian seed gum. As shown in our previous study [20], the most abundant amino acids of durian seed gum were leucine (30.9%–37.3%), lysine (6.04%–8.36%), aspartic acid (6.10%–7.19%), glycine (6.07%–7.42%), alanine (5.24%–6.14%), glutamic acid (5.57%–7.09%), valine (4.5%–5.50%), proline (3.87%–4.81%), serine (4.39%–5.18%), threonine (3.44%–6.50%), isoleucine (3.30%–4.07%), and phenylalanine (3.11%–9.04%). The results indicated that the soaking process led to a reduction in the protein content when the extraction was carried out at elevated temperature (>36 °C) and prolonged time (Figure 3).

The “protein fraction” can be responsible for the surface activity and this hydrophobic nature is an obligatory requirement for gums to adsorb onto interfaces. The present study revealed that the chemical extraction under the medium soaking temperature (<36 °C) resulted in low thermal denaturation of the protein fraction present in the gum structure. This observation was also reported by Abd El-Hady and Habiba [30]. They reported that the soaking process decreased the protein level of the crude gum. They concluded that the soaking process resulted in the leaching of some of the water-soluble proteins from the gum structure into the soaking medium. They revealed that the soaking treatment led to an improvement in the in vitro digestibility, enhanced the phosphorus availability (by reducing phytate and phytate phosphorus), and reduced tannins, polyphenols, and α-amylase inhibitors.

In the current study, the protein content of crude durian seed gum ranged from 3.8% to 9.3%, depending on the chemical extraction conditions. This value was almost similar to the protein content reported for Opuntia ficus indica (7.3%) [25], crude locust bean gum (4.3%–5.0%) [31], and malva nut gum (6.7%) [32]; while it was lower than that reported for Prosopis flexuosa seed gum (10%–20%) [13]. On the other hand, it was greater than the protein content reported for basil seed gum (2.17%) [3], Mesquite seed gum (2.5%) [10], cress seed gum (0.6%) [26], and Krueo Ma Noy pectin (3.29%) [33]. According to Glicksman [34], the protein content in commercial guar gum, locust bean gum and Tara gum was found to be 10%, 8% and 3.13%, respectively. The results indicated that the protein content decreased by prolonging the soaking time from 1.46 h to 6.90 h. Conversely, it increased when the chemical extraction was carried out for a longer soaking time (>7 h). However, the short soaking process (1.46) resulted in the highest protein content in the studied range of soaking times (Figure 3). On the other hand, the reverse trend was observed with the soaking temperature. As shown in Figure 3, the protein content increased with increasing temperature from 15.5 °C to 36 °C. As also reported by Koocheki and co-researchers [27], the protein content was increased due to the greater rate of mass transfer at the higher temperature. Conversely, the further increase in soaking temperature led to a reduction in the protein content. The protein reduction could be explained by thermal denaturation caused by the relatively high soaking temperature.

Adewusi and Osuntogun [35] illustrated that the soaking process for a short time (<30 min) would promote nutrient retention by maintaining a higher content of protein; while the soaking process at high temperature (90 °C) for a long time (>30 min) led to continuous reduction of the protein and total solids. Above 32 °C, protein retention was observed. Protein retention reached a maximum at a temperature of 60 °C which was associated with the denaturation of the protein system, heat-induced Millard response and protein–protein interaction [35]. This could also account for a reduction in the protein content of crude durian seed gum when the soaking process was carried out at an elevated temperature (>36 °C). The results indicated that the interaction between the chemical extraction variables did not show any significant (p > 0.05) effect on the protein content (Table 2). In the current study, the decolorizing time exhibited an insignificant (p > 0.05) effect on the protein content (Table 2). It should be noted that the protein content was more significantly (p < 0.05) influenced by the soaking time rather than the soaking temperature (Table 2). In fact, the soaking time had the most significant (p < 0.05) effect on the protein content (Table 1). Therefore, it should be considered to be the most critical factor affecting the protein content present in the molecular structure of the gum.

The present study revealed that the natural polymer from durian seed had a relatively low solubility ranging from 9.1% to 36.0% (Figure 4). This was less than the solubility reported for carob gum (~50%) [36], xanthan gum and guar gum (>40%) [37]. On the other hand, the crude durian seed gum showed an almost similar solubility to Lepidium Perfoliatum seed gum (18%–25%), and locust bean gum (30%) [37]. The results indicated that durian seed gums extracted under different experimental conditions showed different degrees of solubility. This might be due to the significant (p < 0.05) effect of the extraction conditions on the chemical and molecular structure of durian seed gum. In fact, the different degree of solubility might be explained by the significant effect of extraction conditions on the carbohydrate composition of durian seed gum. As shown in our previous study [20], galactose was the most abundant monosaccharide in the carbohydrate composition of durian seed gum. It should be noted that galactose content plays a significant role in the solubility of various plant gums. A plant gum containing more galactose has better solubility than a gum with lower galactose content. In addition, the carbohydrate analysis revealed the presence of a high content of glucose and a low percentage of arabinose and xylose in the chemical structure of durian seed gum. Our previous study [20] reported the significant effect of extraction variables on the monosaccharide composition (i.e., galactose, glucose, arabinose and xylose) of durian seed gum, thereby affecting its solubility.

On the other hand, the low water solubility could be due to the presence of impurities (i.e., husk, germ fractions), insoluble matter and large particles present in the crude gum [38]. Our previous study [20] also revealed the presence of protein and lipid fraction in the chemical structure of both crude and purified durian seed gum. In fact, different purification techniques did not exclusively remove the protein and lipid fraction from the structure of durian seed gum. This might be responsible for the low solubility of the natural polymer from durian seed. Westphal and Jann [39] stated that the protein fraction was more soluble in phenol and the polysaccharide was more soluble in water. Therefore, the polysaccharide fraction containing high protein content showed low solubility. It was hypothesized that the considerable effect of extraction conditions on the amino acid composition and fatty acid profile of durian seed gum might be another reason for the different degree of solubility. As reported earlier [20], the extraction conditions significantly influenced the amino acid composition of the protein fraction present in the chemical structure of durian seed gum. The different degree of solubility might be also due to the significant effect of extraction conditions on fatty acid composition of durian seed gum [20]. Our previous study [20] also reported the presence of saturated and unsaturated fatty acids (i.e., palmitic acid (C 16:0), palmitoleic acid (C 16:1), stearic acid (C 18:0), oleic acid (C 18:1), linoleic acid (C 18:2), and linolenic acid (C 18:2)) in the chemical structure of durian seed gum. Our preliminary study indicated that the crude gum containing a lower content of the hydrophobic lipid fraction, saturated fatty acid (i.e., palmitic acid (C 16:0) and stearic acid (C 18:0)) showed a higher solubility than the crude gum.

In addition to chemical composition, the molecular structure of durian seed gum was also influenced by extraction conditions, as reported in our previous study [20]. The crude gum with a larger molecular weight (Mw) and particle size has more difficulties in the water absorption and solubilization process. Our preliminary study indicated that the crude gum with lower solubility had a larger particle size than the crude gum with a smaller particle size. Pollard and co-workers [38] also reported that the solubility of carob galactomannan did not exceed 50%. They explained that this could be due to the presence of large particles in the crude carob gum. In the current study, both decolorizing and soaking variables significantly (p < 0.05) affected the solubility of crude durian seed gum. The solubility was significantly (p < 0.05) influenced by the main effects of all extraction variables and the interaction effect of soaking time and temperature (Table 2).

Hydrocolloids behave differently from common food ingredients due to their hydration process [40]. In ordinary food ingredients such as sucrose, when subjected to water, they dissolve starting with the outside layer. Under this condition, the size decreases and they are fully dissolved in water. When hydrocolloids are added to water, they absorb water and swell like a sponge [41]. The water solubility of a homogenous polysaccharide is different from a polysaccharide-protein structure. The natural polymer from durian seed is a heterogeneous polysaccharide plus protein fraction; therefore, its solubility depends on the interaction between the monosaccharide and the protein fractions in the aqueous system. According to Tolstoguzov [42], the protein-polysaccharide structure may lead to three different interaction patterns in the solution: (I) compatible or co-solubility, (II) association or complex coacervation and (III) incompatibility. In association, the phase separation of both biopolymers from the phase containing solvent takes place during the complex coacervation. This occurs because of the electrostatic interaction between differently charged biopolymers [43,44]. However, the most usual pattern corresponds to the thermodynamic incompatibility between protein and polysaccharide molecules, thus reducing the solubility of the protein-polysaccharide mixture in the aqueous solution. Thermodynamic incompatibility occurs because the interaction between similar biopolymers is energetically more favorable than the interaction between different biopolymers [45]. This incompatibility results in the formation of two separated phases [44,46].

The soaking temperature had the most significant (p < 0.05) effect on the solubility (Table 2). In fact, the neutral gums are less soluble than those containing uronic acids, which are often in the polyelectrolyte form (charged or ionic). When the hydrocolloids dissolve in water, they give different conformations (spherical, random coil or rod-like). During the soaking process, the seed absorbs moisture, swells up and some water- soluble nutrients leach into the soaked water by simple diffusion. This swelling behavior depends on the nature and ratio of the matrix and solvent, the soaking temperature and the length of soaking time [47]. A relatively wide range of solubility could be also due to the presence of different protein content. In fact, the water solubility of a homogenous polysaccharide is different from a polysaccharide-protein structure (or heterogeneous polysaccharide).

As also stated by Laaman [48], particle size is also a fundamental issue affecting the solubility. The larger particles correspond to a coarser mesh size which take longer to dissolve due to the longer time required for the water to penetrate into the matrix. Whereas a smaller particle takes a shorter time to penetrate the water and become fully soluble [48]. Koocheki and co-researchers [37] also reported the maximum solubility (25%) of Lepidium perfoliatum seed gum. They described that the low solubility of Lepidium perfoliatum seed gum might be related to the presence of impurities in the chemical structure of the crude gum. A natural gum may not be fully dissolved in water due to the presence of insoluble particles [48]. Therefore, further purification is required to reduce the insoluble matters and consequently enhance the solubility of the crude gum.

The elastic modulus (G′) and viscous modulus (G″) are the critical parametric data in the oscillation test. The elastic modulus (G′) is an indicator of the deformation energy accumulated in the sample during the shear process, thus representing the gel or solid behavior of the sample. The viscous modulus (G″) is an indicator of the deformation energy accumulated in the sample during the shear and lost to the sample, showing the liquid behavior of the sample [49]. If the elastic modulus (G′) is higher than the viscous modulus (G″), the sample behaves more like a solid with gel like behavior and weak elastic behavior. As stated by Simas-Tosin and co-workers [50], the elastic modulus (G′) is related to the solid response of the material and viscous modulus (G″) is related to the fluid response of the material. Therefore, a sample with higher elastic value (G′) has more rigidity than the sample with lower G′[51]. Conversely, if the elastic modulus (G′) is lower than the viscous modulus (G″), the energy used to deform the material is dissipated viscously, and the sample behaves like a liquid [49].

In the current study, the degree of elastic modulus (G′) varied from 1.48 Pa to 9.46 Pa; while the degree of viscous modulus (G″) ranged from 0.28 Pa to 4.51 Pa (Figure 5a,b). This observation indicated that the chemically-extracted durian seed gum showed more elastic behavior rather than viscous behavior at the low frequency (0.1 Hertz). In fact, chemically-extracted durian seed gum had a typical gel network. However, if the ratio of viscous modulus (G″) to elastic modulus (G′) (tan δ) is greater than 0.1, it means that the sample is not a true gel [51]. This sample has a structure between a concentrated biopolymer and a true gel, thus indicating weak gel like behavior. In this sample, chain entanglements and macromolecule interconnections are not permanent; therefore, they could be disrupted by inducing high shear rates [51]. This confirms the pseudoplasticity of crude durian seed gum in the steady state experiments. This observation was also reported by Mandala and co-researchers [51] for locust bean gum and xanthan gum.

The degree of elastic modulus (G′) of crude durian seed gum was similar to that reported by Wang and co-researchers [52] for 0.5% flaxseed gum (3.5 Pa–5.5 Pa); while it was less than 3% maize starch (MS) gel (9.8 Pa–27.9 Pa) [52], xanthan gum 0.15% and locust bean 0.09% (>100 Pa) [51] at low frequency. Figure 5a also showed that the low elastic behavior was achieved by reducing the decolorizing time and prolonging the soaking time. Table 2 demonstrated that soaking time and decolorizing time exhibited the most and least significant (p < 0.05) effect on the elastic modulus (G′). As shown in Figure 5b, the degree of viscous behavior (G″) increased when the soaking time and temperature decreased. The results indicated that the soaking process for a short time at low temperature led to an increase in the viscous behavior (G″) of crude durian seed gum.

The degree of viscous modulus (G″) of crude durian seed gum ranged from 0.28 Pa to 4.51 Pa. It was similar to G″ reported for 3% maize starch (MS) gel (0.8 Pa–6.2 Pa) and 0.5% flaxseed gum (0.4 Pa–3.1 Pa) [52]. In general, the chemically-extracted durian seed gum showed solid (gel like) behavior at low concentration (0.5%, w/w) and low frequency (0.1 Hertz). However, the degree of solid (or gel like) or liquid (or viscous) behavior depends on the gum concentration and frequency measurement. Wang and co-workers [52] found that the flaxseed gum showed gel-like behavior at low concentration (0.5%), since the elastic modulus (G′) was extremely greater than the viscous modulus (G″). They also reported that both the elastic modulus (G′) and viscous modulus (G″) of flaxseed gum increased with increasing gum concentration. The precipitation in alcohol and soaking in acetic acid yielded a less viscous solution. As shown in Table 2, the soaking temperature exhibited the most significant (p < 0.05) effect on the viscous modulus. In general, the viscous behavior of durian seed gum was more significantly influenced by the soaking process rather than the decolorizing process.

Durian fruits (D. zibethinus) were purchased from the local market (Selongor, Malaysia). They were selected based on size uniformity and free of visual defects. Initially, the ripened fruits were de-husked by cutting along the suture on the back of the lobules, and their seeds were removed, cleaned, and rinsed thoroughly with sterile distilled water. The washed fruit seeds were partially dried by air circulation due to prevent the germination and prolong the storage life. They were then packed in plastic bags and stored in a dry and cool place (10 °C ± 2 °C) until the extraction process. In addition, glacial acetic acid (≥99.7%), methanol, hexane, isopropanol, petroleum ether, ethanol (95%) and absolute ethanol (99.9%) were purchased from Fisher Scientific (Leicestershire, UK).

Chemical extraction of durian seed gum was carried out according to the method described by Singh and co-researchers [14] with a minor modification. Initially, durian seeds were chopped into small pieces and air dried before milling into the seed flour. Then, the seed flour was subjected to the defatting process to extract the lipid fraction from the seed flour. The defatting process was carried out successively using hexane and isopropanol (60:40) at room temperature (25 °C ± 1 °C). The solvent residue was eliminated through the centrifugation at 1400g for 15 min (Avanti J-25 Centrifuge, Krefeld, Germany). Then, defatted seed flour (1 kg) was exhaustively decolorized using ethanol at different decolouring times (60 min–180 min). Subsequently, the decolourized seed flour was filtered and then soaked in 1% aqueous acetic acid under different soaking conditions (times, 4 h–12 h; temperatures, 25 °C–55 °C). Then, the slurry was filtered with a Muslin cloth filter and the filtrate was precipitated with 95% ethanol [53]. The precipitate was washed successively with acetone/ether (1:1), and then kept in a fume hood at room temperature overnight. The crude gum was freeze dried at −50 °C for 48 h using a freeze drier (Labconco Freezone 18, Model 77550, Kansas City, MO, USA). It should be noted that the homogenized-gum solution (10%) was pre-frozen at −20 °C for 24 h prior to freeze-drying process. The chemical extraction was carried out in duplicate for each extraction run.

Response surface methodology was applied to investigate the effect of three extraction variables namely decolorizing time, soaking time and soaking temperature on physicochemical and functional properties of gum from durian seed. A central composite design (CCD) including six centre points, six axial points and eight factorial cube points was used to create twenty treatments (Table 3). Polynomial multiple regression (PMR) was applied to describe the variation of the functional properties of durian seed gum as a function of three extraction variables. Significant differences were evaluated by PMR at the 95% confidence level. The creation of experimental design and further data analysis was carried out by using Minitab V. 15 (Minitab Inc., Pine Hall Rd, State College, PA, USA).