Food Chemistry 155 (2014) 311–318

Contents lists available at ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

Structural characteristics and crystalline properties of lotus seed resistant starch and its prebiotic effects Yi Zhang a,b, Hongliang Zeng a,b, Ying Wang a, Shaoxiao Zeng a,b, Baodong Zheng a,b,⇑ a b

College of Food Science, Fujian Agriculture and Forestry University, Fuzhou 350002, Fujian, China Institute of Food Science and Technology, Fujian Agriculture and Forestry University, Fuzhou 350002, China

a r t i c l e

i n f o

Article history: Received 24 August 2013 Received in revised form 10 January 2014 Accepted 14 January 2014 Available online 24 January 2014 Keywords: Lotus seed resistant starch Structural characteristics Crystalline properties Bifidobacteria proliferation

a b s t r a c t Lotus seed resistant starch (LRS) is a type of retrograded starch that is commonly known as resistant starch type 3 (RS3). The structural and crystalline properties of unpurified LRS (NP-LRS3), enzyme purified LRS after drying (GP-LRS3), and enzyme purified LRS (ZP-LRS3) were characterized. The result showed that the molecular weights of NP-LRS3, GP-LRS3, and ZP-LRS3 were 0.102  106, 0.014  106, and 0.025  106 Da, respectively. Compared with native starch and high amylose maize starch (HAMS), LRS lacked the polarization cross and the irregularly shaped LRS granules had a rougher surface, B-type crystal structure, and greater level of molecular order. The FT-IR measurements indicated no differences in the chemical groups. Analysis by 13C NMR indicated an increased propensity for double helix formation and higher crystallinity in LRS than in the two other types of starch. Moreover, LRS was more effective than either glucose or HAMS in promoting the proliferation of bifidobacteria. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Lotus seeds are obtained from members of the genus Nelumbo, especially Nelumbo nucifera, which is an important specialty crop that has been grown in China for thousands of years, and is currently cultivated mainly in Fujian, Hebei, Hubei, Jiangsu, Hunan, Zhejiang, Jiangxi, and Taiwan (Wu et al., 2007). The content of starch in lotus seeds is approximately 500 g/kg (dry basis) with a high amount of amylose (40%, w/w)(Yuan-yuan, Bao-dong, Shao-xiao, Fan, & Shu-zheng, 2011), which may contribute to the formation of resistant starch type 3 (RS3). Resistant starch (RS) consists of starch and products of starch digestion that are poorly absorbed by the small intestine (Escarpa, Gonzalez, Morales, & Saura-Calixto, 1997) and are completely or partially fermented in the colon (Wang, Jin, & Yuan, 2007). The RS3 variant is primarily formed by hydrogen bonding between retrograde amylose molecules. The destruction of the crystal structure and dissolution of amylose during the gelatinization and retro-gradation process causes migration of the starch molecular Abbreviations: POM, polarized optical microscopy; SEM, scanning electron microscopy; XRD, X-ray diffraction; FT-IR, Fourier transform infrared spectroscopy; 13 C NMR, solid state 13C nuclear magnetic resonance spectroscopy; HAMS, high amylose maize starch; RS, resistant starch; RS3, resistant starch type 3. ⇑ Corresponding author at: Institute of Food Science and Technology, Fujian Agriculture and Forestry University, Fuzhou 350002, China. Tel.: +86 591 83736738; fax: +86 591 83739118. E-mail address: [email protected] (B. Zheng). http://dx.doi.org/10.1016/j.foodchem.2014.01.036 0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.

chain and binding of its terminal region to form a double helical structure. Together with these changes, extension, curling, and folding of the original disordered free starch enables hydroxyl groups on the molecular chains to form hydrogen bonds easily. A crystalline structure is formed by ordering a number of different double helices over a particular region or by folding starch chains (Eerlingen & Delcour, 1995). Some potential physiological effects of resistant starch are the prevention of gastrointestinal diseases; reduction of blood glucose levels, the insulin response, and levels of serum cholesterol; prevention of cardiovascular disease; modification of colonic micro-flora; promotion of bacterial growth; and promotion of mineral absorption (Brouns, Kettlitz, & Arrigoni, 2002). The prebiotic properties of RS3 have been studied recently. The results have shown that RS3 increases the growth of bifidobacteria in the gastrointestinal tracts of rats (Rodríguez-Cabezas et al., 2010). In addition, in vitro studies have reported that high amylose maize starch (HAMS) enhances the survival of bifidobacteria in acidic buffer or bile acid solutions as a result of a bacterial adhesive effect (Barczynska, Slizewska, JoChym, Kapusniak, & Libudzisz, 2012). No studies have been conducted on the effect of LRS3 on bifidobacteria proliferation. The present study characterized the structural and crystalline properties of lotus seed RS3 by using size exclusion chromatography (SEC), refractive index (RI) analysis, multi-angle laser light scattering (MALLS), polarized optical microscopy (POM), scanning electron microscopy (SEM), X-ray diffraction (XRD), Fourier transform infrared (FT-IR) spectroscopy, and solid state 13C nuclear

312

Y. Zhang et al. / Food Chemistry 155 (2014) 311–318

magnetic resonance (NMR) spectroscopy. The particle characteristics of lotus seed RS3 were compared with native starch and high amylose maize starch (HAMS), which is a typical resistant type-2 starch. In addition, the effects of LRS on the proliferation of bifidobacteria were assessed and compared with those of glucose (GLU) and HAMS by assessing the changes in optical density (OD). This investigation has provided a theoretical foundation for understanding the structural changes that occur during the preparation of LRS, and provides preliminarily insight into the relationship between the structure and function of LRS. 2. Materials and methods 2.1. Preparation of lotus seed native starch Approximately 2.5 kg of fresh lotus seeds (Green acres (Fujian) food Co., Ltd, Fujian, China) were homogenized in a fruitpulper (MJ-60BM01A, Guangdong Midea Electric Manufacturing Co., Ltd, Guangdong, China) with twice mass as much water. The filtrate was passed through 149-lm mesh screen and left to settle for 6 h, and the supernatant was discarded. The precipitate was washed with distilled water, and then put aside for 4 h. The supernatant was then discarded and the washed precipitate was placed in an oven (DJG-9053A, Shanghai Yiheng Technology Co., Ltd, Shanghai, China) for 13 h at 45 °C to obtain the lotus seed native starch (Zeng, 2007). The moisture of obtained lotus seed native starch was 11.8% determined by using a halogen rapid moisture tester (SFY-6, Shenzhen Electronic Technology Co., Ltd, Shenzhen, China). The dried lotus seed native starch was then extracted using a grinder (FW135, Tianjing Taisite Instrument Co., Ltd, Tianjing, China), and passed through a 149-lm mesh screen.

suspensions were heated at 121 °C for 10 min in an autoclave, cooled to room temperature, and stored at 4 °C for 24 h. The gelatinous suspensions were subsequently stirred well and hydrolysed by incubation in the presence of thermostable a-amylase at 95 °C for 2 h in an orbital incubator-shaker. The processing method was the same as that used to prepare GP-LRS3. 2.3. Determination of Mw and Mw/Mn A 20-mg sample of starch was dispersed in 5 ml of 50 mM LiBr in 90% DMSO (HPLC grade, Sigma Chemical Co., St. Louis, MO) at 90 °C for 2 h on a stirrer-heater module, followed by stirring for 24 h using a magnetic stirrer at room temperature. Dispersed samples were centrifuged for 15 min at 13,500g. The supernatants were filtered through 0.45-lm nylon syringe filters, and then 1-ml samples were injected into the SEC-MALLS-RI system that comprised a pump (P2000, Spectra System, San Hose, CA), injector with a 1-ml loop, SEC column, MALLS (632.8 nm, DAWN DSP, Wyatt Technology, Santa Barbara, CA), and an RI detector (Optilab DSP, Wyatt Technology, Santa Barbara, CA) connected in series. The flow rate of the eluent (50 mM LiBr in DMSO) was 0.6 ml/min 2.4. Polarized optical microscopy A small amount of starch sample was placed on an object slide with a drop of mixed liquor (1:1 glycerol/water (v/v)). The morphologies of dispersed starch granules and changes in cross polarization were observed using a polarized optical microscope (BA300POL, Mike Audi Industrial Group Co., Ltd, Xiamen, China) equipped with a camera set (Deheng Image DH-M3100UC, Shanghai, China) (Li, Ward, & Gao, 2011).

2.2. Preparation of lotus seed resistant starch

2.5. Scanning electron microscopy

For NP-LRS3, 250 g (dry weight) of native lotus seed starch was suspended in 1 L of distilled water in a 2-L conical flask. The starch suspensions were heated at 121 °C for 10 min in an autoclave (SYQ-DSX-280B, SHENAN Medical Devices, Shanghai, China), cooled to room temperature, and stored at 4 °C for 24 h. The gelatinous suspensions were subsequently dried at 45 °C in a drying oven (DJG-9053A, YIHENG Instrument, Shanghai, China), ground, and passed through a 185-lm mesh screen. The resulting product was NP-LRS3 (Onyango, Bley, Jacob, Henle, & Rohm, 2006; Zhang & Jin, 2011). For GP-LRS3, 250 g (dry basis) of LRS3 was suspended in 500 ml of citric acid buffer (pH 6.0) in a 1-L conical flask. The suspensions were hydrolysed by the addition of thermostable a-amylase (10,000 U/ml, obtained from ANKOM, New York, USA) at 95 °C for 1 h in an orbital incubator-shaker (SHA-C, Guo Hua Electric Applicance, Changzhou, China) at 128 rpm. After adjusting the pH to 4.5 with a citric acid solution (4 mol/L), glucoamylase (300 U/ml, obtained from Sigma, St. Louis, USA) was added (5000 U/g of starch) and incubated at 60 °C for 1 h in an orbital incubatorshaker at 128 rpm. The suspensions were centrifuged (L-530, Xiang Yi Laboratory Instrument Development Co., Ltd, Hunan, China) at 2850g for 10 min. The resulting precipitates or residues were washed 3 times with distilled water and ethanol solutions of different concentrations (75%, 85%, and 95%; all determined on a v/v basis). The residues were dried at 45 °C, ground, and passed through a 185-lm mesh screen. The resulting product was GP-LRS3. The GP-LRS3 content in LRS3 was determined as the ratio of the weight of GP-LRS3 (dry basis) relative to the weight of LRS3 (dry basis). For ZP-LRS3, 250 g (dry weight) of native lotus seed starch was suspended in 1 L of distilled water in a 2-L conical flask. The starch

The dried starch materials were deposited on copper stubs using double-adhesive tape and coated with gold. The particle morphology of starch samples were visualized using SEM (PHILIPS-XL30 ESEM, Philips-FEI, Netherlands) at an acceleration of 20 keV. 2.6. X-ray diffraction The starch powder was scanned through the 2h of 5°–45° using X-ray diffractograms (X’Pert Pro MPD, Philips, Netherlands). Traces were obtained using a Cu-Ka radiation detector with a nickel filter and scintillation counter operating under the following conditions: 40 kV, 30 mA, scattering slit 0.25 nm, K-Alpha1 wavelength 1.78901 Å, K-Alpha1 wavelength 1.7929 Å, Ratio K-Alpha2/KAlpha1 0.5, and scanning rate of 0.02°/min. The degree of crystallinity of samples was quantitatively estimated and analysed with Peakfit v4.12 (SeaSolve Software Inc., Framingham, USA) (Ciolacu, Kovac, & Kokol, 2010; Tester, Karkalas, & Qi, 2004):

C CL ¼

SC  100 ð%Þ ST

SC CL ¼

SSC  100 ð%Þ ST

C L ¼ C CL þ SC CL

ð1Þ

ð2Þ ð3Þ

where CCL is the proportion of crystalline region, SC is the crystallization area, ST is the total area, SCCL is the proportion of sub-crystalline region, SSC is the sub-crystallization area, and CL is the degree of crystallinity.

313

Y. Zhang et al. / Food Chemistry 155 (2014) 311–318

2.7. Fourier transform infrared spectroscopy

3. Results and discussion

Two mg of starch samples were ground and dried under an infrared lamp with 150 mg potassium bromide in an agate mortar. This removed both the free and crystalline water in the samples, so that it could eliminate the interference of water molecules with the absorption peaks. The mashed powder was put into a vacuum compression and pressed into a sheet. Starch samples were scanned using Fourier Transform Infrared Spectrometer (AVATAR360, Nicolet Company, USA). The spectra were recorded in transmission mode from 400 to 4000 cm1 at a resolution of 4 cm1 at room temperature.

3.1. Measurement of Mw and Mw/Mn values of starches

2.8.

13

C Nuclear magnetic resonance spectroscopy

Approximately 200–300 mg of starch samples were scanned using solid-state 13C CP/MAS NMR (AVANCE III 500, Bruker Co., Ltd, Switzerland) at a 13C frequency of 125.7 MHz. The spectral width was 40 kHz (Tan, Flanagan, Halley, Whittaker, & Gidley, 2007). The relative degree of crystallinity of samples was estimated quantitatively, and was analysed with Peakfit v4.12 (Atichokudomchai, Varavinit, & Chinachoti, 2004; Lin, Wang, & Chang, 2008):

RC L ¼

SF  100 ð%Þ ST

ð4Þ

where RCL is the relative degree of crystallinity, SF is fitting peak area of C1, and ST is the total peak area of C1. 2.9. Proliferation rate of bifidobacteria All fermentation experiments were conducted in a fermentation medium (Regmi, Metzler-Zebeli, Ganzle, van Kempen, & Zijlstra, 2011; Siew-Wai, Zi-Ni, Karim, Hani, & Rosma, 2010). The fermentation medium was adjusted to pH 7.0 either with 0.5 M acetic acid or 0.5 M NaOH (Chen et al., 2012). The four types of carbohydrates used were glucose (GLU; Sinopharm Chemical Reagent Co., Ltd, Shanghai, China), LRS3, GP-LRS3, and HAMS (60% amylose, Beststarch Creatmaterial Co., Ltd, Fujian, China). The maximum OD value of bifidobacteria was achieved at 600 nm. The number of bifidobacteria was assessed by determining the OD600 nm values in the presence of LRS3 or GP-LRS3(Rendon-Huerta, Juarez-Flores, Pinos-Rodriguez, Aguirre-Rivera, & Delgado-Portales, 2011). A live Bifidobacterium oral preparation was obtained from Livzon Pharmaceutical Group Inc. (Guang dong, China), and kept at 4 °C. Before the fermentation experiments, bifidobacteria were incubated for 48 h in 15 ml of fermentation medium with added GLU in anaerobic bags (each bag with a volume of 2.5 L or less, Mitsubishi Gas Chemical, Tokyo, Japan) to produce inoculum. The incubation of bifidobacteria was performed in an anaerobic incubator at 37 °C (W-zipper Stangding-Pouch, Mitsubishi Gas Chemical, Tokyo, Japan). The resulting inoculum (16%, v/v) was transferred to 15 ml of fermentation medium containing GLU, LRS3, GP-LRS3, or HAMS as the carbon source. The carbon source concentration was 1.25, 2.5, 5, 10, 20, or 40 g/L. After 48 h of fermentation, the OD600 nm values of the fermentation media were analysed. Fermentation experiments were conducted in triplicate. 2.10. Statistical analysis The data of structural characteristics were analysed using the DPS 9.05 system (Science Press, Beijing, China) or Peakfit v4.12 software. All of the graphs were plotted using OriginPro 8.0. Data of bifidobacteria proliferation were analysed by one-way analysis of variance (ANOVA). Significant differences were determined using the DPS 9.50 system. Statistical significance was set to p < 0.05.

The molecular weight and polydispersity index values of lotus seed native starch, NP-LRS3, GP-LRS3, ZP-LRS3, and HAMS were showed in Table 1. Table 1 indicates that the Mw values of lotus seed native starch, NP-LRS3, GP-LRS3, ZP-LRS3, and HAMS were 1.307  106, 0.102  106, 0.014  106, 0.025  106, and 6 4.097  10 Da, respectively. The Mn and Mw values of HAMS were significantly higher than those for lotus seed native starch and resistant starches. This indicated that HAMS was composed of highly polymerized amylose. The observation that Mn and Mw values of LRS batches were lower than those of native starch indicated that native starch might have higher amylose content than the other types of starch analysed. Nonetheless, native starch differed from HAMS insofar as native starch was composed of low polymerized amylase. Given that NP-LRS3 may contain some straight-chain or branched-chain starch molecules with a high molecular weight, it showed higher Mn and Mw values than the other types of starch analysed. The Mw/Mn ratios of lotus seed native starch, NP-LRS3, GP-LRS3, ZP-LRS3 and HAMS were 4.118, 1.689, 1.044, 1.254, and 1.830, respectively. The larger the value of the polydispersity index, the wider the molecular weight distribution. When the value of the polydispersity index is 1, the molecular weight of the polymer is completely homogeneous. The fact that the Mw/Mn ratio of GP-LRS3 and ZP-LRS3 were close to 1 confirmed that LRS was degraded during the process of autoclaving, formed molecular chains with a low degree of polymerization, and provided a number of chain ends able to easily form double helical structures. Given that the capacity of LRS molecules to move relative to each other was enhanced, they were thus more likely to form a stable double helix structure that increased the overall orderly level of the starch molecules. 3.2. Polarization cross of lotus seed resistant starch Starch granules generally form part of a spherical crystal system with polycrystalline structures that include an amorphous region and have a crystalline structure. Whereas the starch molecules in the amorphous region are disordered, starch molecules in the crystalline region arrange in precise double helices. Therefore, when the polarized light passes through the starch granules, differences in starch density and the refractive index cause birefringence, which appears as the polarization cross on the umbilical point of starch granules (Błaszczak et al., 2011). However, this polarization cross disappears when the ordered crystalline regions are destroyed (Ratnayake, Hoover, & Warkentin, 2002). Hence, changes of the polarization cross reflect changes in the crystal structure of the starch granules. Fig. 1 shows the polarization cross results following the analysis of lotus seed native starch, LRS type 3, and HAMS (obtained from Beststarch Creatmaterial Co., Ltd, Fujian, China). Birefringence is a symbol of the average radial orientation of helical structures. A pronounced birefringence was displayed at

Table 1 Molecular weights of native starch, resistant starch prepared from lotus seeds, and HAMS.

Native starch LRS3 GP-LRS3 ZP-LRS3 HAMS

Mn (105 Da)

MW (106 Da)

MW/Mn

3.173 0.605 0.134 0.197 22.390

1.307 0.102 0.014 0.025 4.097

4.118 1.689 1.044 1.254 1.830

314

Y. Zhang et al. / Food Chemistry 155 (2014) 311–318

Fig. 1. Polarized light microscopy of native starch, LRS, and HAMS: (A) Native starch, 400; (B) NP-LRS3, 40; (C) GP-LRS3, 40; (D) ZP-LRS3, 40; and (E) HAMS, 400.

the center of native starch molecules (Fig. 1A). Given the complete destruction of granules of LRS, the larger particles analysed failed to exhibit birefringence (Fig. 1B–D). For HAMS (Fig. 1E), a loss of birefringence was apparent at the center of the granule, but not at its periphery. The thermal energy associated with autoclaving process promotes the mobility of the double helices, which reduces their radial orientation and causes a loss of birefringence of resistant starch. These results were in agreement with previous report (Li et al., 2011). The decrease in birefringence became more pronounced at the center of mung bean starch as the processing temperature increased. The above results suggest that the original crystal region of LRS was destroyed during processing. 3.3. Scanning electron microscopy The shapes and surface characteristics of native and resistant starch are shown in Fig. 2. Whereas native starch granules were oval and had a smooth surface, the structure of LRS was completely destroyed, and the original oval shape disappeared to form a compact block structure. These observations are consistent with those reported for chickpea starch (Polesi & Sarmento, 2011) and corn starch (Kim et al., 2010). The surface of NP-LRS3 was smooth, but also revealed the presence of layered strips. By contrast, flaky and gully shapes were observed on the rough surfaces of GP-LRS3 and ZP-LRS3, which

formed more compact structures than those formed by NP-LRS3. This observation could be attributed to the leaching of amylose, the loss of the amylopectin crystalline region during autoclaving, and re-association of the starch chains within the granules. Therefore, the expansion of starch granules to form tight structures within amorphous regions might reduce the susceptibility of LRS to enzymatic digestion. The gully shapes observed on the surfaces of GP-LRS3 and ZP-LRS3 might have been left by the hydrolysis of amylose on the surface of the weak tissue structure. HAMS has an irregular shape, and the sizes and surfaces of HAMS structures resembled those of the structures formed by native starch. Notwithstanding the high amylose content of HAMS, given the absence of a re-crystallization process, its structure differed substantially from that of resistant starch (Sha et al., 2012). 3.4. X-ray diffraction pattern and crystallinity The intensity of the X-ray diffraction peak and the crystallinity of starch reflect changes in amorphous regions and crystalline regions. A sharp diffraction peak is characteristic of starch with an extended crystal line and complete crystal surface, whereas a wide diffraction peak denotes an amorphous structure formed when the starch contains a small crystal line and incomplete crystal. The X-ray diffraction patterns of native starch, LRS, and HAMS are shown in Fig. 3A. According to the analysis of Peakfit v4.12

Y. Zhang et al. / Food Chemistry 155 (2014) 311–318

315

Fig. 2. Scanning electronic micrographs of native starch, LRS, and HAMS: (A) Native starch; (B) NP-LRS3; (C) GP-LRS3; (D) ZP-LRS3; and (E) HAMS.

software, native lotus seed starch exhibited a C-type crystalline pattern with characteristics of the A- and B-type crystalline pattern, four strong peaks at 17.63°, 19.72°, 21.11°, and 26.91° 2h, and two broad peaks at 25.49° and 28.20° 2h. The degree of crystallinity is a useful indicator of the properties of starch and starch-containing products. Lotus seed resistant starch had a Btype crystalline pattern with more intense diffraction peaks at 19.99° 2h than were observed for native starch. Moreover, LRS showed doublet peaks between 25.60° and 27.94°, with the peak near 17.43° disappearing and becoming the dispersion peak with characteristics similar to those of the sub-crystal structure (Vermeylen, Goderis, & Delcour, 2006). The change in crystalline pattern is consistent with a previous report that during the preparation of resistant starch from maize starch, the appearance of a peak at 19.77° 2h might contribute to the disappearance of an amorphous region based on amylopectin, which might be caused by the retrogradation that occurs at a low temperature (Frost, Kaminski, Kirwan, Lascaris, & Shanks, 2009). HAMS exhibited A-type and V-type crystalline patterns and showed intense peaks at 15.11°, 19.66°, 22.99°, and 26.31° 2h. This means that a combination of amylose and lipid in HAMS might influence the structure. The degrees of crystallinity of native starch, LRS and HAMS shown in Table 2A were determined according to the analysis method described by Mutungi (Mutungi, Passauer, Onyango, Jaros, & Rohm, 2012). Table 2A indicates that the relative size of the crystalline region in LRS was significantly smaller than in native starch. This might be because the peaks of crystalline phase and dispersion diffraction peaks of amorphous phase are always superposed

together, which reduces the content of the crystalline region in LRS. This result is consistent with the report by Jian, Gao, and Liang (2003) that LRS contains a lot of primary crystallite, and that the peaks of primary crystallite (sub-crystalline region) and dispersion diffraction peaks of the sub-crystal phase stacked up together. However, the compactness of these primary crystallites was not less than that of the advanced crystallite (crystalline region). The relative size of the sub-crystalline region was higher in NP-LRS3, GP-LRS3, and ZP-LRS3 than in native starch and HAMS, with LRS3 showing the highest number. Given that the primary crystallite in GP-LRS3 and ZP-LRS3 might lead to inconspicuous peaks, the crystalline region did not generate diffusion diffraction peaks associated with the sub-crystal structure. The crystallinity of LRS suggests that it contains a higher content of ordered double helix structure than HAMS. It seems possible that amorphous amylose was transformed to a double helix structure during the autoclaving and cooling processes, and that this in turn resulted in a more orderly crystallite. There was no significant difference in either the crystalline type or crystallinity between ZP-LRS3 and GP-LRS3. This means that these two purified methods have no influence on the crystal structure of resistant starch.

3.5. Fourier-transform Infrared Spectroscopy Infrared spectroscopy enables analysis of the molecular structures of starch granules in terms of short-range level differences, with changes in peak intensity indicating characteristics of the

316

Y. Zhang et al. / Food Chemistry 155 (2014) 311–318

amplitude (Flores-Morales, Jiménez-Estrada, & Mora-Escobedo, 2012). The absorption region of 800–1200 cm1 reflects C–C, C–OH and C–H stretching vibration (Zhang, Chen, Liu, & Wang, 2010), changes in starch polymer conformation, and the process of hydration (Wei et al., 2011). The absorption intensity of LRS in the band from 800 to 1200 cm1 was weaker than that of either native starch or HAMS; this indicated that certain conformation changes had occurred in LRS. The IR spectra in this region are characterized by three main modes, with maximum absorbances at 995, 1022, and 1047 cm1. The absorption bands at 995 and 1047 cm1 are associated with the ordered structure of starch. The absorption band at 1022 cm1 is associated with the amorphous structure. Given the sensitivity of the changes of starch conformation, the ratios of the bands (1047/1022 and 995/1022 cm1) could be used to quantify the degrees of order in the starch samples. A higher ratio indicates a higher proportion of crystalline region in the starch granules. What is more, the ratio of the absorbance (995/1022 cm1) provided a better way to characterize the internal changes in double helix than that of the absorbance (1047/1022 cm1). Analysis of the IR spectrum indicated that the ratios of absorbance (995/1022 cm1) of native starch, NP-LRS3, GP-LRS3, ZP-LRS3 and HAMS were 0.87, 1.01, 1.00, 1.01, and 0.97, respectively. The proportion of crystalline region in LRS was higher than in either native starch or HAMS, and this was confirmed using X-ray diffraction. This demonstrated the existence of a microcrystalline area formed by the ordered starch chains. Moreover, as shown in Fig. 3B, the absorption band at 1367.22 cm1 (corresponding to the bending vibration of –CH2OH) was narrower than that of either native starch or HAMS. Changes in the combinations of hydrogen bonds generated during the formation of resistant starch might contribute to this difference. 3.6. Solid state

Fig. 3. X-ray diffraction patterns, FT-IR spectra and 13C CP/MPS NMR spectra of native starch, LRS, and HAMS: (A) X-ray diffraction patterns; (B) FT-IR spectra; and (C) 13C CP/MPS NMR spectra.

conformational change in the starch. Fig. 3B shows FT-IR spectra of native starch, LRS, and HAMS. The failure of FT-IR to depict any changes in chemical groups indicated that LRS has undergone physical modification. The amplitude of LRS in the absorption bands between 3100 and 3700 cm1 (corresponding to the –OH group) was different from that of native starch and HAMS. During the gelatinization and retrogradation process, a more precise bonding combination of internal and intermolecular hydrogen might cause the change of

13

C nuclear magnetic resonance spectroscopy

Given that both X-ray diffraction and FT-IR spectra failed to indicate a significant difference in the crystal structures of GP-LRS3 and ZP-LRS3, GP-LRS3 was selected for NMR scanning to study the internal changes in resistant starch. Fig. 3C shows the 13 C CP/MPS NMR spectra of native starch, LRS, and HAMS. A carbon chemical shift for starch has been identified in 106– 96 ppm for C1; in 70–79 ppm for C2, C3, and C5; in 80–84 ppm for C4 and in 59–62 ppm for C6. The multiplicity of the C1 position of the glucose units provides information about the crystallinity of starch and the double helix symmetry. Whereas an A-type crystal presents three peaks at 102, 101, and 100 ppm, in a B-type conformation, the C1 resonance exhibits two peaks at 101 and 100 ppm; this corresponds to the non-identical sugar residues of amylose and amylopectin (Flores-Morales et al., 2012). The C2, C3, and C5 regions mainly showed the B-type double helices resulting from the residues of free amylose, whereas the C4 region represents information on the amorphous state (Fan et al., 2013). The two broad resonance peaks at 82 ppm for C4 might arise from the amorphous domains. Table 2B depicts the chemical shifts and relative crystallinity of native starch, LRS, and HAMS. Fig. 3C and Table 2B indicate that native starch shows multiplicity in the resonance peak for C1, which presents characteristics of both the A-type and B-type crystal structures. GP-LRS3 shows a typical B-type crystal structure and three peaks appear near 100 and 102 ppm in HAMS, which presents characteristics of the A-type crystal structure. All of the crystal structure types were consistent with the results obtained in X-ray diffraction tests. In addition, GP-LRS3 displays a higher relative crystallinity compared with those of native starch and HAMS, indicating that recrystallization was an important step to obtain a more stable structure during the formation of resistant starch. The resonance peaks at 102 ppm in NP-LRS3 result from a simple helix organized in the

317

Y. Zhang et al. / Food Chemistry 155 (2014) 311–318 Table 2A Crystallinity of native starch, LRS, and HAMS.* Crystalline type Native NP-LRS3 GP-LRS3 ZP-LRS3 HAMS

Proportion of crystalline region (%)

Proportion of crystalline region (%)

a

C-type B-type B-type B-type A + V-type

d

52.39 ± 1.24 29.08 ± 0.87c 46.47 ± 2.63b 44.34 ± 2.18b 44.98 ± 1.70b

Crystallinity (%) 80.14 ± 0.97a,b 79.60 ± 1.23a,b 81.65 ± 2.03a 83.95 ± 2.14a 76.42 ± 1.56b

26.80 ± 0.11 50.88 ± 1.60a 35.63 ± 2.10b,c 39.98 ± 1.22b 31.80 ± 1.21c

*

Results are expressed as mean ± standard deviation; values represent the average of duplicate analyses, Lower case letters within the same column are significantly different (p < 0.05).

Table 2B Chemical shifts and relative crystallinity of native starch, LRS, and HAMS. Chemical shifts (ppm)

Native

NP-LRS3

GP-LRS3

HAMS

a

Relative crystallinity (%)

PPAa (%)

C1

C4

C2,3,5

C6

96.69 98.96 100.56 101.92 96.72 99.94 102.68 99.35 101.09

81.61

71.71 75.29

61.08 61.63

77.60

5.11

82.10

61.23 61.41

79.95

6.04

61.52

89.91

2.52

96.85 99.10 99.34 102.13

82.10

72.05 73.02 74.69 71.22 73.88 76.19 71.83 73.26 74.63

61.26 61.36 61.55

88.04

4.63

82.10

The proportion of C4 peak fitting area relative to the total area of the spectrum.

Table 3 Effect of different concentration of substances on the number of Bifidobacterium cells (OD600 nm value). Concentration of substances (g/L) 0.625 1.250 2.500 5.000 10.000 20.000 40.000

GLU

LRS3 d

0.018 ± 0.006 0.026 ± 0.007d 0.040 ± 0.017c,d 0.172 ± 0.017c 0.665 ± 0.016b 1.031 ± 0.030a 0.904 ± 0.141a

GP-LRS3 f

0.033 ± 0.017 0.056 ± 0.011f 0.178 ± 0.018e 0.801 ± 0.030d 1.054 ± 0.028b 1.152 ± 0.045a 0.930 ± 0.037c

HAMS f

0.127 ± 0.058 0.338 ± 0.019e 0.506 ± 0.034d 1.047 ± 0.021c 1.253 ± 0.026b 1.360 ± 0.003a 1.117 ± 0.067c

0.023 ± 0.020e 0.036 ± 0.010e 0.113 ± 0.032e 0.301 ± 0.120d 0.739 ± 0.019c 1.106 ± 0.024a 0.919 ± 0.046b

Values are means ± standard deviations (n = 3). For each detection, means in the same column and the same letter are not significantly different (p < 0.05).

crystalline phase, or dispersed in an amorphous phase. Therefore, the relative crystallinity was lower than that of GP-LRS3 and HAMS, whereas it was higher than that of native starch. The results suggested that NP-LRS3 still contained part of the amorphous phase, which was easily hydrolysed. Atichokudomchai reported a method used to analyse the content of the double helix structure (Atichokudomchai et al., 2004). It was said that the amorphous phase decreased with the increase of double helix structure, whereas the proportion of the amorphous phase was related with the proportion of C4 peak fitting area relative to the total area of the spectrum (PPA). In Table 2B, it was found that the PPA of GP-LRS3 was significantly lower (p < 0.01) than other samples. This means that GP-LRS3 contained a higher proportion of double helix structure, and it was demonstrated that resistant starch was a stable crystalline structure that involved the formation of a double helix with the recombination of amylose. Although HAMS contains a higher amylose content than any of the other types of starch tested, it showed a lower double helix because of a higher PPA; this indicated that there was no definite relationship between amylose content and the extent of double helix formation. NP-LRS3 showed a higher PPA than native starch. This might be caused by the free amylose leaching out during the

gelatinization which has not yet formed as an ordered double helix. However, additional analysis of the crystal structure and combination and distribution of starch chains is required. 3.7. Proliferation of bifidobacteria The proliferation of bifidobacteria in the media with different carbon sources is shown in Table 3. For each determination, means in the same column and the same letter are not significantly different (p < 0.05). When the concentration of carbon source was in the range from 0.625 to 20 g/L, the OD600 nm value, which is indicative of the number of bifidobacteria, increased with an increase in the levels of the different carbon sources tested. When the concentration of carbon source was 20 g/L, the OD600 nm value reached the highest value after 48 h of fermentation: 1.03 in the GLU medium, 1.11 in the HAMS medium, 1.15 in the LRS3 medium, and 1.35 in the GP-RS3 medium. However, the OD600 nm value of fermentation media decreased when the concentration of the carbon source was 40 g/L. This phenomenon might be attributed to high concentrations of carbon sources in the medium that might result in a high osmotic pressure that negatively affect bacteria. At certain concentrations, LRS3 and especially GP-LRS3 could enhance the growth of

318

Y. Zhang et al. / Food Chemistry 155 (2014) 311–318

bifidobacteria. Among these carbon sources, the number of bifidobacteria in the medium containing GP-LRS3 was the highest. It was reported that HAMS is a better prebiotic than oligosaccharides (Bird et al., 2009). In this study, LRS3 and GP-LRS3 had greater beneficial effects on the growth of bifidobacteria than HAMS. LRS3 and GP-LRS3 could be regarded as prebiotics, which might be more beneficial than some types of oligosaccharides. 4. Conclusions Analysis of the structural characteristics and crystalline properties of LRS batches showed that the Mw of lotus seed native starch, NP-LRS3, GP-LRS3, ZP-LRS3, and HAMS were 1.307  106, 0.102  106, 0.014  106, 0.025  106, and 4.097  106 Da, respectively. The Mw values of LRS batches were lower than those of native starch. The Mw/Mn ratios of lotus seed native starch, NPLRS3, GP-LRS3, ZP-LRS3 and HAMS were 4.118, 1.689, 1.044, 1.254, and 1.830, respectively. The Mw/Mn ratios of GP-LRS3 and ZP-LRS3 were close to 1; this confirmed that LRS was more likely to form a stable double helix structure. The polarization cross disappeared and granules became irregular and showed a rough surface compared with native starch and HAMS. Lotus seed resistant starch showed B-type crystal structure and an increase in molecular order than any of the other starch types analysed. Lotus seed resistant starch is a physically modified starch. Analysis using NMR indicated an increase of the double helix and higher crystallinity in the LRS relative to the other starch samples analysed. Moreover, the results also failed to demonstrate any difference between purified resistant starches obtained by the different purification methods. In addition, LRS can increase the proliferation of bifidobacteria compared with growth in the presence of GLU or HAMS. This effect may be related to its stable double helix structure, high crystallinity, and rough surfaces with a scale-like or gully-like appearance. Acknowledgement This study is financially supported by the National Natural Science Fund (No. 31301441) and the Natural Science Fund in Fujian Province (No. 2011J01080). References Atichokudomchai, N., Varavinit, S., & Chinachoti, P. (2004). A study of ordered structure in acid-modified tapioca starch by 13C CP/MAS solid-state NMR. Carbohydrate Polymers, 58(4), 383–389. Barczynska, R., Slizewska, K., Jochym, K., Kapusniak, J., & Libudzisz, Z. (2012). The tartaric acid-modified enzyme-resistant dextrin from potato starch as potential prebiotic. Journal of Functional Foods, 4(4), 954–962. Bird, A. R., Vuaran, M., Crittenden, R., Hayakawa, T., Playne, M. J., Brown, I. L., et al. (2009). Comparative effects of a high-amylose starch and a fructooligosaccharide on fecal bifidobacteria numbers and short-chain fatty acids in pigs fed Bifidobacterium animalis. Digestive Diseases and Sciences, 54(5), 947–954. Błaszczak, W., Bidzin´ska, E., Dyrek, K., Fornal, J., Michalec, M., & Wenda, E. (2011). Effect of phosphorylation and pretreatment with high hydrostatic pressure on radical processes in maize starches with different amylose contents. Carbohydrate Polymers, 85(1), 86–96. Brouns, F., Kettlitz, B., & Arrigoni, E. (2002). Resistant starch and ‘‘the butyrate revolution’’. Trends in Food Science & Technology, 13(8), 251–261. Chen, L.-C., Chiang, W.-D., Chen, W.-C., Chen, H.-H., Huang, Y.-W., Chen, W.-J., et al. (2012). Influence of alanine uptake on Staphylococcus aureus surface charge and its susceptibility to two cationic antibacterial agents, nisin and low molecular weight chitosan. Food Chemistry, 135(4), 2397–2403. Ciolacu, D., Kovac, J., & Kokol, V. (2010). The effect of the cellulose-binding domain from Clostridium cellulovorans on the supramolecular structure of cellulose fibers. Carbohydrate Research, 345(5), 621–630. Eerlingen, R., & Delcour, J. (1995). Formation, analysis, structure and properties of type III enzyme resistant starch. Journal of Cereal Science, 22(2), 129–138.

Escarpa, A., Gonzalez, M., Morales, M., & Saura-Calixto, F. (1997). An approach to the influence of nutrients and other food constituents on resistant starch formation. Food Chemistry, 60(4), 527–532. Fan, D., Ma, W., Wang, L., Huang, J., Zhang, F., Zhao, J., Zhang, H., & Chen, W. (2013). Determining the effects of microwave heating on the ordered structures of rice starch by NMR. Carbohydrate Polymers, 92(2), 1395–1401. Flores-Morales, A., Jiménez-Estrada, M., & Mora-Escobedo, R. (2012). Determination of the structural changes by FT-IR, Raman, and CP/MAS 13C NMR spectroscopy on retrograded starch of maize tortillas. Carbohydrate Polymers, 87(1), 61–68. Frost, K., Kaminski, D., Kirwan, G., Lascaris, E., & Shanks, R. (2009). Crystallinity and structure of starch using wide angle X-ray scattering. Carbohydrate Polymers, 78(3), 543–548. Jian, H., Gao, Q., & Liang, S. (2003). Study on crystallinzation property of resistant starch. Food Science, 24(7), 44–47. Kim, N.-H., Kim, J.-H., Lee, S., Lee, H., Yoon, J.-W., Wang, R., et al. (2010). Combined effect of autoclaving-cooling and cross-linking treatments of normal corn starch on the resistant starch formation and physicochemical properties. Starch – Stärke, 62(7), 358–363. Li, S., Ward, R., & Gao, Q. (2011). Effect of heat-moisture treatment on the formation and physicochemical properties of resistant starch from mung bean (Phaseolus radiatus) starch. Food Hydrocolloids, 25(7), 1702–1709. Lin, J. H., Wang, S. W., & Chang, Y. H. (2008). Effect of molecular size on gelatinization thermal properties before and after annealing of rice starch with different amylose contents. Food Hydrocolloids, 22(1), 156–163. Mutungi, C., Passauer, L., Onyango, C., Jaros, D., & Rohm, H. (2012). Debranched cassava starch crystallinity determination by Raman spectroscopy: Correlation of features in Raman spectra with X-ray diffraction and 13C CP/MAS NMR spectroscopy. Carbohydrate Polymers, 87(1), 598–606. Onyango, C., Bley, T., Jacob, A., Henle, T., & Rohm, H. (2006). Influence of incubation temperature and time on resistant starch type III formation from autoclaved and acid-hydrolysed cassava starch. Carbohydrate Polymers, 66(4), 494–499. Polesi, L. F., & Sarmento, S. B. S. (2011). Structural and physicochemical characterization of RS prepared using hydrolysis and heat treatments of chickpea starch. Starch – Stärke, 63(4), 226–235. Ratnayake, W. S., Hoover, R., & Warkentin, T. (2002). Pea starch: Composition, structure and properties—A review. Starch – Stärke, 54(6), 217–234. Regmi, P. R., Metzler-Zebeli, B. U., Ganzle, M. G., van Kempen, T., & Zijlstra, R. T. (2011). Starch with high amylose content and low in vitro digestibility increases intestinal nutrient flow and microbial fermentation and selectively promotes bifidobacteria in pigs. Journal of Nutrition, 141(7), 1273–1280. Rendon-Huerta, J. A., Juarez-Flores, B. I., Pinos-Rodriguez, J. M., Aguirre-Rivera, J. R., & Delgado-Portales, R. E. (2011). Effects of different kind of fructans on in vitro growth of Lactobacillus acidophilus, Lactobacillus casei and Bifidobacterium lactis. African Journal of Microbiology Research, 5(21), 3404–3408. Rodríguez-Cabezas, M. E., Camuesco, D., Arribas, B., Garrido-Mesa, N., Comalada, M., Bailón, E., et al. (2010). The combination of fructooligosaccharides and resistant starch shows prebiotic additive effects in rats. Clinical Nutrition, 29(6), 832–839. Sha, X. S., Xiang, Z. J., Bin, L., Jing, L., Bin, Z., Jiao, Y. J., et al. (2012). Preparation and physical characteristics of resistant starch (type 4) in acetylated indica rice. Food Chemistry, 134(1), 149–154. Siew-Wai, L., Zi-Ni, T., Karim, A. A., Hani, N. M., & Rosma, A. (2010). Fermentation of Metroxylon sagu resistant starch type III by Lactobacillus sp and Bifidobacterium bifidum. Journal of Agricultural and Food Chemistry, 58(4), 2274–2278. Tan, I., Flanagan, B. M., Halley, P. J., Whittaker, A. K., & Gidley, M. J. (2007). A method for estimating the nature and relative proportions of amorphous, single, and double-helical components in starch granules by 13C CP/MAS NMR. Biomacromolecules, 8(3), 885–891. Tester, R. F., Karkalas, J., & Qi, X. (2004). Starch—Composition, fine structure and architecture. Journal of Cereal Science, 39(2), 151–165. Vermeylen, R., Goderis, B., & Delcour, J. A. (2006). An X-ray study of hydrothermally treated potato starch. Carbohydrate Polymers, 64(2), 364–375. Wang, J., Jin, Z., & Yuan, X. (2007). Preparation of resistant starch from starch–guar gum extrudates and their properties. Food Chemistry, 101(1), 20–25. Wei, C., Qin, F., Zhou, W., Xu, B., Chen, C., Chen, Y., et al. (2011). Comparison of the crystalline properties and structural changes of starches from high-amylose transgenic rice and its wild type during heating. Food Chemistry, 128(3), 645–652. Wu, J.-Z., Zheng, Y.-B., Chen, T.-Q., Yi, J., Qin, L.-P., Rahman, K., et al. (2007). Evaluation of the quality of lotus seed of Nelumbo nucifera Gaertn from outer space mutation. Food Chemistry, 105(2), 540–547. Yuan-yuan, L., Bao-dong, Z., Shao-xiao, Z., Fan, Z., & Shu-zheng, W. (2011). Effects of edible lotus seed starch coating on quality of fresh-cut pineapple. Journal of Fujian Agriculture and Forestry University (Natural Science Edition), 2, 018. Zeng, S.-X. (2007). Studies on qualitative characteristics of lotus-seed (Nelumbo nucifera Gaertn) starch and its application. Fuzhou: Fujian Agriculture and Forestry University. Zhang, J., Chen, F., Liu, F., & Wang, Z.-W. (2010). Study on structural changes of microwave heat-moisture treated resistant Canna edulis Ker starch during digestion in vitro. Food Hydrocolloids, 24(1), 27–34. Zhang, H., & Jin, Z. (2011). Preparation of resistant starch by hydrolysis of maize starch with pullulanase. Carbohydrate Polymers, 83(2), 865–867.