Effects of two contrasting dietary polysaccharides and tannic acid on the digestive and physicochemical properties of wheat starch

Abstract In this study, konjac glucomannan, κ‐carrageenan, and tannic acid were selected to study the effects of different combinations on the in vitro digestibility and physicochemical properties of wheat starch. Results showed that the addition of konjac glucomannan, κ‐carrageenan, and tannic acid could decrease the digestion of starch and increase the content of resistant starch. Besides, the two polysaccharides weakened the extent of tannic acid on starch digestion. Moreover, although the two polysaccharides had different effects on the in vitro digestion of starch, they had no significant increase in the content of resistant starch. DSC and XRD results demonstrated that the polysaccharides and tannic acid showed synergistic effects on the rebuilding of starch microstructure. FTIR results further manifested that κ‐carrageenan and konjac glucomannan could significantly increase the strength of hydrogen bonds in starch. At the same time, the addition of tannic acid would weaken the molecular interaction between polysaccharides and starch. SEM and CLSM results showed that tannic acid added to the polysaccharide–starch mixture not only interacted with starch but also influenced the structure of polysaccharide gel.

among dietary polysaccharides, polyphenols, and starch are mainly dependent on the molecular structure of polysaccharides (linear, branched, charged, or uncharged) and polyphenols (Gao et al., 2017;Zhu, 2015). However, it remains unclear how the interaction between polysaccharides and polyphenols with different molecular structures may affect the functional properties of starch.
Two contrasting dietary polysaccharides were selected in this study. Konjac glucomannan (KGM) is a nonionic polysaccharide, which formed by the main chain composed of D-mannose and Dglucose linked by β-1,4 glycosidic bonds (Shah et al., 2015). It is reported that konjac glucomannan can affect the structure and physical properties of starch, thus improving the characteristics of starchy food (Lafarge & Cayot, 2018). In our previous work, konjac glucomannan addition is found to reduce the digestion of wheat starch through forming a barrier around the gelatinized starch molecule . κ-carrageenan (KC) is an anionic, linear polysaccharides extracted from red seaweed (Necas & Bartosikova, 2013). Studies have shown that the strength of interactions between κ-carrageenan and starch in water depends on the types and quantities of negative charge of the polysaccharide (Lascombes et al., 2017).
Tannic acid is an important natural polyphenolic active substance widely present in plants. Due to abundant phenolic hydroxyl groups, tannic acid can interact with amino acid residues in the active site of digestive enzymes, thereby inhibiting its activity (Funke & Melzig, 2005;Li et al., 2019;Zhao et al., 2013). Also, the addition of tannic acid to wheat starch could spontaneously form complex, thus affecting the rheological properties and microstructure of starch (Wei et al., 2019). Our previous research has found that the gelatinization processing does not give negative effect but even promote the nutritional value of tannic acid in wheat starch such as phenolic content and antioxidant activity . Therefore, the main objectives in the present work were to determine how the in vitro digestion, gelatinization, retrogradation, and microstructure of wheat starch were affected by different combinations of konjac glucomannan, κ-carrageenan, and tannic acid.

| Sample processing
In the experimental groups, the addition ratio of konjac glucomannan (KGM) and κ-carrageenan (KC) was 3%, and the addition ratio of tannic acid (TA) was 0.2%. The concentrations of KGM, KC, and TA were chosen according to the preliminary experiments and the reference Zhang et al., 2020). The mixing percentages of added substances were based on the mass ratio of wheat starch (w/w). Tannic acid was prepared into a solution of 0.01 g/ml and stored at 4℃.
Wheat starch (2 g) was accurately weighed, added with 30 ml distilled water, and prepared in a centrifuge tube. After mixed well with the additions, the samples were shaken in the water bath at 95℃ for 20 min. The gelatinized mixtures were prefrozen for 24 h in a −20℃ freezer and dried for 48 h in a freeze dryer (science-12N; Ningbo Xingzhi Biotechnology Co., Ltd.) at −50℃ (Xie et al., 2018). After freeze-drying, the samples were crushed and screened with 100 mesh. The powder samples were stored in a vacuum drying centrifugal tube at 4℃.

| Measurement of in vitro digestibility
In vitro digestion of wheat starch was conducted following the measurement described by Se˛czyk et al. (2017) with minor modifications. The brief steps were described below: Firstly, wheat starch (200 mg) was dispersed in 2 ml distilled water, mix well, and shaken in the water bath at 95℃ for 20 min. After the samples cooling to room temperature, 10 ml pepsin solution (0.01 M, KCl-HCl buffer, PH = 1.2) was added and incubated at 37℃ for 1 h. Next, the volumes of the samples were replenished to 25 ml with sodium phosphate buffer (PH = 7), and each group was added with 5 ml alpha-amylase buffer (2 U/ml) and incubated at 37℃ for 2 h. At the 10 min, 20 min, 30 min, 40 min, 50 min, 60 min, 90 min, and 120 min of the reaction, 1 ml of the reaction solution was, respectively, taken into a new centrifugal tube to inactivate the enzyme in a metal bath at 100℃ for 5 min. After that, each tube was incubated for 1 h at 60℃ with 3 ml sodium acetate buffer (PH = 4.75) and 21 μl starch glycosidase solution (40 U/ml). At the end of incubation, 20 μl sample solution was taken, and the glucose content was determined by the DNS method (Saqib & Whitney, 2011). Finally, the amount of

| Differential scanning calorimetry (DSC)
Differential scanning calorimetry (DSC-7; Perkin Elmer, Norwalk, CT, USA) was used to assess the process of gelatinization. For each group, 100 mg wheat starch and 400 μl distilled water mixed well, and 10 μl of the mixture was taken out for testing. Before the thermal scanning, the sample was sealed in the gland and balanced at 4℃ for 12 h. Using an empty crucible as a reference, the prepared samples were heated from 20 to 120℃ at 10℃/min. The transition temperatures of the onset (T o ), peak (T p ), and endothermic enthalpy of gelatinization (ΔH) were recorded.

| X-ray diffraction (XRD)
X-ray diffraction (XRD) patterns of the powder samples were measured with Cu-Ka radiation using a Powder X-ray diffractometer (Rigaku Mini Hex-II, Japan). The region of scanning ranged from 10º to 70º, and the speed of goniometer was 4º/min operated at 40 mA and 40 kV (Sukhija et al., 2016). According to the X-ray diffraction pattern, software Jade6 was used to calculate the relative crystallinity of each group (Jing et al., 2019;Xie et al., 2018).

| Fourier transform infrared spectroscopy (FTIR)
The FTIR spectra were run using a TENSORⅡ spectrometer (Bruker Corporation). Two mg powder samples were dispersed in 200 mg KBr, which then was extracted by air for 1 min using a tablet pressing machine mold. The spectra were recorded in transmission mode from 4,000 to 400 cm −1 at room temperature.

| Confocal light scanning microscopy (CLSM)
Two ml distilled water and 200 mg starches were mix well, and the samples were shaken in a water bath at 95℃ for 20 min. Five

| Statistic analysis
Data were presented as means ± SEM of triplicate. Analysis of variance was used to contrast the significant difference by using one-way analysis of variance (ANOVA). The significant differences (p < .05) were evaluated and displayed by different letters.

| Effects of different combinations of konjac glucomannan, κ-carrageenan, and tannic acid on the in vitro digestibility of wheat starch
Compared with the control group, the digestibility rates of starch in the experimental groups were all reduced ( Figure 1a). Among them, the curve of group TA was at the lowest position, which indicated that tannic acid would have a more obvious inhibition on the digestion of starch. As shown in Figure 1b, the contents of RDS in the five experimental groups were significantly lower than that in the control group, while the contents of RS were markedly increased (p < .05).
However, the contents of SDS in each experimental group showed no striking difference, expect for group KC. In particular, due to the different ways to affect the starch digestion, group TA had the highest RS content in the experimental groups. This may be due to the interaction between tannic acid and starch (Funke & Melzig, 2005;Zhao et al., 2013). Although the difference in the content of RS between group KC and group KGM was not noteworthy, there was a remarkable difference in the contents of RDS and SDS. Specifically, the content of RDS in group KC was lower than group KGM, while the content of SDS was higher than group KGM. These may be due to the difference in the intermolecular interaction of the gel network between the two polysaccharides.
Among the three single experimental groups, the content of RDS in KGM group was lower than KC and TA group. In comparison, the content of SDS in KC group was higher than KGM and TA group, which showed that KC was more suitable for reducing postprandial blood glucose. Moreover, compared with polysaccharide groups, the combination of tannic acid and polysaccharide did not increase the variation of the three types of starch. Nevertheless, compared with the tannic acid group, the combination of tannic acid and polysaccharide markedly decreased the content of RS (p < .05). The same reduction in the inhibitory effect also occurred in the previous study (Sun et al., 2018). This change may be related to the polysaccharide gel hindering the direct contact of tannic acid with starch.

| Effects of different combinations of konjac glucomannan, κ-carrageenan, and tannic acid on the thermal properties of wheat starch
The DSC method is commonly applied to measure the energy changes in starch, due to crystallite melting or formation of ordered structures. Table 1 showed the onset temperature (T o ), the peak temperature (T P ), and the gelation enthalpy (ΔH gel) obtained from the DSC thermogram ( Figure 2  . As is known to all, the gelation enthalpy (ΔH) reflects the energy intake of gelation, which is an indicator of the degree of molecular order loss in the particles during gelation (Koteswara et al., 2014).
Therefore, the decrease in the gelation enthalpy (ΔH) in the experimental groups reflected the partial gelatinization of starch in the system. Compared with the single additive group, the gelation enthalpy (ΔH) of group KC + TA and group KGM + TA was further decreased, which reflected the synergistic effect of dietary polysaccharides and polyphenols in inhibiting the formation of ordered structures.

| Effects of different combinations of konjac glucomannan, κ-carrageenan, and tannic acid on the recrystallization characterization of wheat starch
X-ray diffractometry is applied to determine the crystalline structures of starch and its degree of crystallinity (Blazek & Gilbert, 2011).
Many crystalline structures were formed when the starch molecules rearranged in the retrogradation process. It could be seen that the control group had diffraction peaks around 15°, 18°, and 20° at 2θ (Figure 3a), which indicated that the freeze-dried wheat starch mainly had obvious A-type crystals. Compared with the control group, the remaining five groups only had diffraction peaks around 18 ° and 20 °. Besides, the angle of the diffraction peaks had a slight change (Figure 3a). This means that these five groups were mixtures of A-type crystals and V-type crystals (Maache-Rezzoug et al., 2011). As shown in Figure 3b, compared with control group (17.2%), the relative crystallinities of the five experimental groups were significantly reduced (p < .05), especially in the four groups with polysaccharides. Besides, the relative crystallinities of group KC + TA and group KGM + TA were lower than the starch with these three substances alone. This result was supported by the thermal analysis showed above. Additionally, the relative crystallinities of group KGM + TA were memorably lower than that of group Note: *n = 3, T o = onset temperature (°C); T p = peak temperature (°C);ΔHgel = enthalpy of gelatinization (Jg −1 ). KCκ-carrageenan; KGMkonjac glucomannan; TA-tannic acid. All data were means of triplicates.
Results are expressed as mean ± SEM. Means with the same letter are not significantly different (p > .05).
KC + TA, which indicated that the inhibiting effect of KGM on starch retrogradation was better than that of KC. This discrepancy might be attributed to the interaction between the two different polysaccharide and starch.

| Effects of different combinations of konjac glucomannan, κ-carrageenan, and tannic acid on the molecular interaction characterization of wheat starch
To have a better understanding of the molecular structure of wheat starch, FTIR is applied to analyze the functional group shifts of starch reacted with other compounds. Figure 4 represented the similar major peaks of the six samples with the variant amplitudes.
These results demonstrated that the addition of polysaccharides and polyphenols to starch did not produce new functional groups . However, the molecular interaction still ex- strengthened, which related to the CH 2 deformation on the carbon skeleton . In contrast, other experiment groups were weakened to varying degrees. The contrary effect was related to the simple linear structure of κ-carrageenan, that is, it was more likely to enter the interior of starch molecules and thus interacted with the carbon skeleton (Bui et al., 2019). This was consistent with the trends of the RDS and SDS, which could be used in explaining the influence of KC on the starch digestion.
Compared with the control group, the peak intensity of the experiment groups was decreased at 1639 cm −1 , which indicated that the absorption of free water was inhibited (

| Effects of different combinations of konjac glucomannan, κ-carrageenan, and tannic acid on the morphological characterization of wheat starch
The molecular morphology of freeze-dried starch in the six groups was significantly different, which suggested different interactions between the additives and starch. Compared with the group control, the starch of the four groups with dietary polysaccharides formed a denser and smooth structure with smaller holes ( Figure 5). When the starch was added with tannic acid, the molecular borders present more irregular crimps. As for group KC and group KGM, the molecular surface of freeze-dried wheat starch after adding tannic acid had smaller pore diameter. Besides, the holes in wheat structure formed by konjac glucomannan were smaller than that of κ-carrageenan.
Similar results also happened between group KC + TA and group KGM + TA.
In addition, the gelatinized starch molecules without being freezedried also showed significant differences( Figure 6). Compared with control group, the starch granules in group KC were slightly enlarged.
However, the starch granules in group KGM and group TA were reduced to different degrees, especially in group KGM. This indicated that these two polysaccharides had different effects on the starch structure. Compared with the two polysaccharides group, the group KC + TA and group KGM + TA significantly inhibited the expansion of starch molecules after the addition of tannic acid (Figure 6b, c, e, f). Especially, the size of the starch molecule reduced most obviously in group KC + TA.

| DISCUSS IONS
Our previous study showed that tannic acid reacted with starch granules and altered their physicochemical properties (e.g., gelatinization and retrogradation) and the microstructure properties, and demonstrated tannic acid reduced the starch digestion by enzyme inhibition and the changed structures . Konjac glucomannan also played an important role in the microstructure of wheat starch and its digestibility, which acted by forming a barrier around the granules . Similar result was also found by κ-carrageenan in this study. Besides, the two polysaccharides in the present study could weaken the inhibition of tannic acid on starch hydrolysis. This change may be related to the polysaccharide gel hindering the direct contact of tannic acid with starch.
Moreover, FTIR results showed that tannic acid could weaken the interaction between the polysaccharides and starch molecules.
These results suggested that tannic acid and polysaccharides might   (Lascombes et al., 2017). It can be concluded that the addition of tannic acid increased the negative charge density in the system, which might reduce the interaction between κ-carrageenan and starch molecules, so that more κ-carrageenan formed a gel around starch molecules to inhibit the expansion of starch.

| CON CLUS IONS
Overall, the two polysaccharides weakened the inhibition of tannic acid on starch hydrolysis. Besides, tannic acid reduced the interaction between the polysaccharides and starch molecules.
These results suggested that tannic acid and polysaccharides starch might be competitively bound to starch. Further research found that the two polysaccharides and tannic acid had a synergistic effect on the formation of ordered starch microstructure, which made the starch structure more compact and inhibited the expansion of starch. Additionally, tannic acid strengthened the microstructure of the starch with dietary polysaccharides. The understanding of the starch microstructure caused by the two polysaccharides, and tannic acid could be useful for explaining the starch digestion.

ACK N OWLED G M ENTS
The work was supported by the National Key Research and Development Program of China (2017YF0400300) and National Natural Science Foundation of China (31901700).

CO N FLI C T O F I NTE R E S T
There is no potential conflict between authors and others that bias our work.

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

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.