In vitro digestibility of normal and waxy corn starch is modified by the addition of Tween 80

Abstract

Aqueous dispersions of normal and waxy corn starch (3% w/w) were mixed with Tween 80 (0, 7.5, 15, 22.5 and 30 g/100 g of starch), and gelatinized (90 oC, 20 min). Optical microscopy of the gelatinized starch dispersions (GSDx; x = Tween 80 concentration) revealed that the microstructure was characterized by a continuous phase of leached amylose and amylopectin entangled chains, and a dispersed phase of insoluble remnants, called ghosts, on whose surface small granules were observed, imputed to Tween 80. The apparent viscosity of the GSDx decreased as the concentration of Tween 80 increased (up to about 70-90%). FTIR analysis of dried GSDx indicated that Tween 80 addition decreased short-range ordering. The content of rapidly digestible starch (RDS) and resistant starch (RS) fractions tended to increase significantly, at the expense of a significant decrease of slowly digestible starch (SDS) fraction, an effect that may be attributed to the increase of amorphous structures and starch chain-surfactant complexes. The RDS and RS increase was more pronounced for normal than for waxy corn starch, and the significance of the increase was dependent on Tween 80 concentration. Overall, the results showed that surfactant can affect largely the digestibility of starch chains.

Keywords: Starch; Tween 80; in vitro digestibility.

1. Introduction

Synthetic surfactants are commonly used in food matrices to modify texture and stabilize microstructures (e.g., emulsions). Via electrostatic effects, synthetic surfactants are able to stabilize water-oil interfaces, interact with food molecules (e.g., lipids) and modify morphology [1]. For instance, polysorbates are added up to 0.5% (v/v) concentration to make ice creams smoother, easier to handle, and more resistant to melting [2]. The quality of bread products can be improved by surfactants, by increasing crumb softening and reducing undesirable staling (e.g., short-range re- crystallization) effects, after cooking and cooling [3, 4, 5, 6]. The effect of Tween 20 on edible films surface tension has been studied [7]. Synthetic surfactants are commonly incorporated in food products via single and multiple emulsions in sauces, dressings, dips, desserts, ice cream, beverages, soups, and delivery systems [8]. Surfactants may impact the food structure through numerous physical mechanisms. In particular, synthetic surfactants may interact with starch chains (amylose and amylopectin), which are major components of numerous food staples. The nature and mechanisms of synthetic surfactant and starch chains have been explored to some extent. Strong amylose-surfactant interactions were detected by X-ray diffraction patterns of the insoluble residues of wheat starch [9]. Viscosimetry and differential scanning calorimetry measurements showed that synthetic surfactants are able to complex with amylopectin molecules as well as with amylose [10]. Synthetic emulsifiers have the ability of retarding the retrogradation of amylopectin chains [11]. Small-angle X-ray scattering studies revealed that synthetic anionic surfactants interact with starch chains by forming phases consisting of starch-surfactant aggregates arranged in long- range structures similar to lyotropic mesophases formed by the pure surfactants alone [12]. It was also reported that synthetic surfactants, like Tween 80, increases the free volume between the starch chains as reflected by reduced viscosity [13].

The aforementioned studies indicate that synthetic surfactants have the ability of altering the properties of food matrices. In turn, such alterations may modify the susceptibility of food components to acidic and enzymatic hydrolysis. Despite the widespread usage of synthetic surfactants in the food industry, the issue of digestibility of food components has been scarcely explored. Very few studies have explored the digestibility features food components in the presence of synthetic surfactants. Although the effect of synthetic surfactants on the digestibility of lipids has attracted some attention [14-16], studies regarding starch digestibility are still lacking. Results in this line should be valuable for the proper design of food staples with proven textural properties and tailored digestibility features.

The aim of this work was to assess the effect of the the synthetic surfactant Tween 80 on the in vitro digestibility of normal corn starch. To the best of our knowledge, this issue has not been explored in the literature. To contrast the result and explore the role of amylose chains, the study was also conducted for waxy corn starch.

2. Materials and Methods
2.1. Materials

Normal corn starch (CAS number 9005-25-8, amylose content 25.3%, moisture content < 15%, pH 4.8; ash < 0.5%, protein < 0.1%), waxy corn starch (CAS number 9679, only trace amounts of amylose, moisture content < 15%, pH 4.8; ash < 0.5%, protein < 0.1%), pancreatin from porcine pancreas (300U/mL, P1750) and amyloglucosidase (95U/mL, A7095) were purchased from Sigma- Aldrich (St Louis, MO, USA). Total starch (TS) and resistant starch (RS) kits were purchased from Megazyme International Ireland Ltd. (Bray Business Park, Bray, Co., Wicklow, Ireland). Tween 80 (Polyoxyethylene (20) sorbitan monooleate, Canarcel TW80, XT80200, CAS number 9005-65-6) was obtained from Canamex Quimicos S.A. de C.V. (Mexico City, Mexico). Deionized water was used in all experiments.

2.2. Preparation of samples

Aqueous dispersions of normal and waxy corn starch (3% w/w) were added with Tween 80 (0, 7.5, 15, 22.5 and 30 g/100 g of starch), and mixed with an Ultra-Turrax (T-25, IKA Works, Inc., Wilmington, NC, USA) homogenizer at 10,000 rpm during 5 min. The homogenized dispersions were then completely gelatinized (90 oC, 20 min) in a water bath with constant stirring at 250 rpm. The gelatinized starch dispersions were coded as GSDx, where the x stands for the amount of Tween 80 with respect to starch contents in the dispersions. GSDx were allowed to cool down to room temperature (aprox. 20 oC), and left to rest for 1 h to allow equilibration between starch chains and surfactant molecules. The GSDx were used as such for optical microscopy observation and apparent viscosity determination. A portion of GSDx were placed in a convection oven (Riossa, Mexico City, Mexico) and air-dried at 35 ºC until constant weight was achieved ( 24 h). The dried GSDx powders were kept in sealed bags until required for ATR-FTIR spectroscopy and in vitro digestibility analysis .

2.3. Optical microscopy

The microstructure of the GSDx was assessed with an optical microscope (Olympus BX45, Tokyo, Japan) coupled to an image analyzer system (Axiocam ERc5s camera and Zen blue edition software, Carl Zeiss Microscopy GmbH, Jena, Germany). GSDx were stained with iodine to obtain enhanced visualization of the morphology of the starch-Tween 80 mixture. Selected micrographs at 40× were used for illustration purposes.

2.4. Apparent viscosity

The apparent viscosity of GSDx was determined with a Physica MCR300 rheometer (Physica Messtechnik, GmbH, Stuttgart, Germany) with a double gap cylinder measuring system (DG 26.7, 2 mm gap width). Measurements were conducted by using ~10 mL of the samples, and left to stabilize for 5 minutes for structure recovery. Temperature maintenance was settled with a Physica TEK 150P temperature control system. Flow curves were determined by varying the shear rate in the range from 0.01 to 1000 s-1. All measurements were made by triplicate.
.
2.5. ATR-FTIR

Dried GSDx samples were characterized by ATR-FTIR spectroscopy. A Perkin Elmer spectrophotometer (Spectrum 100, Perkin Elmer, Waltham, MA, USA) equipped with a crystal diamond universal ATR sampling accessory was used. A spectrum of the empty cell was measured to correct the background effect. The average of three scans from 400 to 4000 cm-1 at 1 cm-1 resolution was collected. All spectra were deconvoluted using Gaussian and Lorentzian functions.

2.6. In vitro starch digestibility

The in vitro digestion of normal and waxy corn starch (raw and gelatinized) was performed using the Englyst et al. method [17] with slight modifications [18]. Briefly, starch samples were incubated with porcine pancreatic α-amylase (A3176, Sigma Aldrich, St Louis, MO, USA) and amyloglucosidase (A7095, Sigma Aldrich, St Louis, MO, USA) enzymes at 37 °C for 120 min. Starch classification was based on the rate of hydrolysis, RDS (digested within 20 min), SDS (digested between 20 and 120 min) and RS (undigested after 120 min).

3.1. Results and discussion
3.1. Optical microscopy

Figure 1 illustrates the morphology (optical images at 40×) of the GSDx with 0.0 (figures 1.a and 1.b), 15.0 (figures 1.c and 1.d), 22.5 (figures 1.e and 1.f) and 30.0 g (figures 1.g and 1.h) of Tween 80/100 g of starch. The left and right panel micrographs correspond to normal and waxy corn starches, respectively. Gelatinized starch dispersions without Tween 80 (Figures 1.a and 1.b) were characterized by a starch chain-rich (mainly amylose) expelled from the starch granules during the heating process, and a disperse phase formed by insoluble remnants that were undisrupted by heating. Amylose and amylopectin chains in the continuous phase are strongly entangled, adding high viscosity to the dispersion. On the other hand, the insoluble remnants (ghosts), are irregular structures that confer elasticity to the system [19]. Arrows in figures 1.a and 1.b indicate the presence of the Maltese Cross (MC) remnants, corroborating that insoluble remnants can be seen as the shell of native starch granules after most soluble chains have been leached out. Debet and Gidley [20] postulated that proteins play an important role in the integrity of starch ghosts by forming complexes with surface polysaccharides. Cross-linking of polysaccharide chains involves double helices also contribute to the integrity of starch ghosts. Zhang et al. [21] used enzymatic digestion methods to conclude that the condensed polymeric surface structure of ghost particles is mainly composed of nonordered but entangled amylopectin (and some amylose) molecules, with limited reinforcement through partially ordered enzyme-resistant structures based on amylose (e.g., for corn starch) or amylopectin (e.g., for potato starch). Figures 1.c to 1.h show that ghosts from normal and waxy corn starch lingered dispersed in the medium, despite the addition of Tween 80 (denoted as T80). In Figures 1.c, 1.d, 1.g and 1.h, T80 can be appreciated as small granular structures on the surface of ghosts, which were more numerous the higher the concentration of T80. These structures can be related to the formation of giant or wormlike micelles formed by weak electrostatic interactions of surfactant molecules [12, 22]. Notice that these structures were not observed in the GSD0 samples shown Figures 1.a and 1.b. The Tween 80 molecule contains 18 carbon chain long hydrophobic moiety, with four extended hydrophilic moieties, one of which has a tagged alkyl chain. The molecule is a multi-headed structure, which tends to form micelles in solution of reduced size [23]. Tween 80 is known to form complexes with dispersed amylose chains [24], and to affect the size and structure of amylose nanoparticles [25]. It is apparent that residual hydrophobicity allowed such complexes to be adsorbed onto the surface of the insoluble remnants. In turn, the complex microstructures adsorbed on the ghost surface can act as barriers to the action of hydrolyzing agents, including digestion enzymes [14].

3.2. Apparent viscosity

The apparent viscosity-shear rate curves of the GSDx made from normal corn starch are given in Figure 2.a. GSD0 exhibited typical shear-thinning behavior of weakly flocculated dispersions, where the shear thinning behavior was induced by the deformation of the flocs aligned with the shear field [8]. The flow behavior of the GSDx incorporating Tween 80 present a more complex behavior. At very low shear rates ( 0.03 s-1) the apparent viscosity increased slightly, probably because of the formation of complexes between Tween 80, amylose/amylopectin soluble chains and ghosts, which upon shearing flocculated and stretched elastically [26]. Eventually, the flocs aligned themselves with the shear field, and the apparent viscosity began to drop moderately but continuously ( 0.1- 1.0 s-1). At about 1.0 s-1 the apparent viscosity suffered a sharp downward drop. It is likely that the shearing forces caused disruption of the flocs, with the apparent viscosity achieving a minimum value at about 4.0-5.0 s-1. The apparent viscosity increased sharply at about 7.0-10.0 s-1 probably due to reformation of bonds between the different structures contained in the GSDx. At shear rate values  than about 20.0 s-1 the apparent viscosity of the GSDx achieved a plateau pattern. The above results indicate that Tween 80 interacted with the gelatinized starch gel entities in a very complex fashion, and discerning the nature of this is beyond the scope of this study. Nevertheless, it may be inferred that Tween 80-starch chain interactions modified the gel microstructure. A similar pattern was exhibited by waxy corn GSD0 (Figure 2.b), although the apparent viscosity values were higher than for normal corn GSD0. The addition of Tween 80 decreased the apparent viscosity to achieve values of about 1.0-2.0 Pa.s for GSD30.0. Furthermore, GSD7.5 and GSD15 exhibited a similar pattern than GSD0, although the apparent viscosity values were lower. The difference in the viscosity-shear rate pattern exhibited by the waxy corn GSDx could be explained from the fact that waxy corn only contains traces of amylose. In such case, the GSDx viscosity is mainly due to amylopectin (the starch component with high molecular weight) leached out from the starch granule during heating. Besides, interaction of Tween 80 was likely to occur with both amylose and amylopectin chains [10].

3.3. FTIR analysis

The analysis of the FTIR signal focused only on the fingerprint region of polysaccharides (about 1100-950 cm-1). The signal was deconvoluted by means of three Gaussian signals to quantify the relative contributions of peaks at about 995, 1022 and 1045 cm-1. The region ~950-1100 cm-1 has been linked to C-O-C interactions and it has been proven that these interactions are sensitive to changes in molecular arrangement within the short-range order [27]. The band at about 1048 cm-1 is sensitive to ordering content, while the band at about 1022 cm-1 has been related to vibration of amorphous components. On the other hand, the band at about 995 cm-1 is sensitive to hydrated ordered domains [27]. Commonly, the ratio R1047/1022 is considered as an index of the short-range ordering. However, the quantification of absorbance bands is commonly hampered by overlapping effects, which can lead inaccurate estimates. To tackle this problem, deconvolution of the FTIR spectrum is commonly conducted via, e.g., Gaussian basis functions. The spectrum was baseline corrected by drawing a straight line, then deconvoluted with Gaussian basis functions with a half- width of 15 cm-1 and a resolution enhancement factor of 1.5. Figure 3.a illustrates the experimental FTIR signal and the corresponding deconvolution Gaussian components. In this case, the 1048 band appears shifted to about 1043, an effect that can be attributed to, e.g., hydration and presence of other compounds (surfactants). Figure 3.b presents the behavior of the ratio R1047/1022, with the component 1047 taken as the closest one to such wavenumber. For both normal and waxy corn starches, the ratio decreased monotonously as Tween 80 concentration increased. The ratio R1047/1022 reflects the contribution of short-range ordered structures, like double and triple helices, relative to amorphously distributed starch chains. In this way, the decrease of the ratio R1047/1022 indicates that the surfactant molecules disrupted the formation of short-range ordered structures via, e.g., formation of complexes. Although the behavior was observed for both normal and waxy corn starches, the effect was more pronounced for normal starch, pinpointing that the amylose chains are the ones that govern the interaction with the surfactant molecules. Figure 3.c presents the behavior of the ratio R995/1022 as the surfactant concentration increased. Also, the ratio R995/1022 showed a monotonous decrease, reflecting the relative reduction of hydrated crystalline structures dispersed in the continuous phase of the gelatinized starch dispersion. Overall, the results presented in Figure 3 showed that Tween 80 increased the relative amount of amorphous and poorly ordered microstructures.

3.4. In vitro digestibility

The results of the in vitro digestibility of the dried GSDx are exhibited in Figure 4.a for normal corn starch and Figure 4.b for waxy corn starch. After drying of GSD0, the dispersed starch chains form double- and triple-helices in compact arrangements, led to the production of a relative large amount of SDS fraction. However, a decrease in the SDS fraction (due to a reduction of the ordered structures) occurred, while the RDS and RS fractions tended to increase depending of the concentration of Tween 80 added. The significance of the changes in the starch fractions are indicated by lower case letters placed above the bar of each fraction in Figure 4. In fact, for normal corn starch the RS fraction increased from 7.43% for GSD0 to 22.58% for GSD30. A similar trend was observed for waxy corn starch, with the increase being of 6.92% for GSD0 to 22.36% for GSD30. Interestingly, the content of the RS fraction was about 4-fold for GSD30 respect that of GSD0. However, the concentration of Tween 80 had a saturation-like effect on the in vitro digestibility, as the RS fraction showed a significant increase up to about a concentration of 15 g Tween 80/100 g of normal corn starch, but non-significant increases as Tween 80 concentration increased to 22.5 and 30 g Tween 80/100 g of normal corn starch. The effect was more pronounced for normal than for waxy corn starch. The increase of the RS fraction can be attributed to the formation of starch-surfactant complexes, while the increase of the RDS fraction was due to the increased amorphous fraction as detected by FTIR analysis (Figure 3). In this way, Tween 80 plays the role of forming surfactant-starch chain complexes with the ability of resisting the action of hydrolysis agents [5], leading to increased RS fractions. Also, the aggregation of Tween 80 microstructures, on the surface of ghost particles (figures 1.c to 1.h) suggests the formation of a protective barrier that limits the access to hydrolyzing agents. This effect was observed for lipid droplets by Chang and McClements [14], who postulated that synthetic surfactants can adsorb to the surfaces of lipid droplets and form protective coatings that inhibit the ability of lipase to access the underlying lipid phase. On the other hand, the increase of the RDS fraction is not clear at all, although it is suggested that such effect could be linked to the formation of local domains and to blocking re-crystallization (e.g., formation of double helices) of dispersed starch chains. A result would be the observed increase of the RDS and RS fractions at the expense of the SDS fraction. The effect was less pronounced for waxy corn starch (Figure 4.b) since the SDS content for the sample without surfactant (GSD0) was smaller (about 62%) than for normal corn starch. Interestingly, the distribution of the digestible starch fractions for high surfactant contents in waxy corn starch is similar to that in normal corn starch. Amylose chains are hardly present in waxy samples, which suggests that the increase of the RS fraction can be also ascribed to the covering of surface of insoluble remnants or ghosts (Figure 1), an effect that retards the action of enzymes.

4. Conclusions

The effect of Tween 80 on the in vitro digestibility of corn starch was assessed. The incorporation of the synthetic surfactant increased both the RDS and RS fractions at the expense of the SDS fraction. The increase of the RDS fraction can be attributed to the obstruction of the short-range re- crystallization of dispersed starch chains and blocking of hydrolyzing agent access to insoluble remnants, while the increase of the RS fraction to the formation of starch chain-surfactant complexes. The effect was less pronounced for waxy corn starch, suggesting that amylose chains play an important role in the formation of slowly digestible complex structures. The results in this study point out to the notion that surfactants have the ability of modifying the in vitro digestibility features of gelatinized starches.