Enbo Xu, Hongyan Li, Zhengzong Wu, Fang Wang, Xueming Xu, Zhengyu Jin, and Aiquan Jiao

Abstract: Volatile compounds in enzymatic extruded rice, produced under different conditions of varying barrel temperature (BT), α-amylase concentration (AC) and moisture content (MC), were extracted and identified by headspace solid phase microextraction (HS-SPME) and gas chromatography-linked mass spectrometry (GC-MS). Statistical analyses reflected that the Maillard reaction could be inhibited both by the mild extrusion conditions and the enhanced hydrolysis caused by thermostable α-amylase. Relative amounts of total volatiles in enzymatic extruded rice were far less than those in severe processed extruded rice. Reverse-phase high-performance liquid chromatography (RP-HPLC) showed that the amino acids (AAs) involved in Maillrad reaction were utmostly preserved in extruded rice with highest amylase concentration by comparison of total AA content of different extrudates. These results suggest that enzymatic extrusion liquefaction is an effective way to control the generation of volatiles from extruded rice for Chinese rice wine production. Keywords: Chinese rice wine, enzymatic extrusion liquefaction, extruded rice, Maillard reaction, volatiles

Although enzymatic extrusion liquefaction is a popular and worldwide treatment for grains, only a few and partial studies have been focused to the knowledge of the volatile composition of these extrudates. However, the characteristic profile of enzymatic extruded rice suggests that lower contents of the Maillard-style volatiles are generated in the rice material for Chinese rice wine fermentation.

Practical Application:

Introduction Enzymatic extrusion liquefaction, or enzymatic hydrolysis in extrusion cooking, of cereals is used widely in the fields of oil and protein extraction, starch saccharification, and fermentation (Govindasamy and others 1997a; Lee and others 2009, 2010; De Mesa-Stonestreet and others 2012). The extruder runs as a bioreactor to liquefy starch or protein in feedstock with heat-stable enzymes addition at mild extrusion conditions of low temperature (70 to 110 °C) and high moisture (30% to 65%) generally (Govindasamy and others 1997b; Tomas and others 1997; Li and others 2012). The advantages of such a composite processing unit over conventional extrusion cooking or other multistep heat-treated methods are economic and efficient in terms of reduction in energy and the usage of raw materials. In addition, higher degradation rates of rice starch, protein, and lipid, which have been proven to extremely improve the fermentation rate and efficiency for alcohol or beverages manufacture, are reported under controlled operating conditions (Choi and others 2013; Li and others 2013). Chinese rice wine, fermented from glutinous rice with “wheat Qu” and yeast, is 1 of the 3 most ancient drinks in the world. This alcoholic beverage is extremely popular in China for centuries because of its abundant nutrition, sweet aroma, and characteristic “yellow” color (Yu and others 2007; Luo and others 2008; Fu MS 20140754 Submitted 5/4/2014, Accepted 10/19/2014. The authors are with The State Key Lab of Food Science and Technology, School of Food Science and Technology, Jiangnan Univ., Wuxi 214122, China, and with Synergetic Innovation Center of Food Safety and Nutrition, Jiangnan Univ., Wuxi 214122, China. Direct inquiries to author Jin (E-mail: [email protected]) and author Jiao (E-mail: [email protected]).

R  C 2014 Institute of Food Technologists

doi: 10.1111/1750-3841.12719 Further reproduction without permission is prohibited

2011). Many high-starchy cereals that are processed by extrusion pretreatment instead of traditional steam cooking—an intensive water, energy-consuming, and low-throughput process for modern industry—are used as fermenting materials of Chinese rice wine production in the last 2 decades (Xu and others 2014). By comparison, the content of amino acids (AAs) in the enzymatic extrusion-processed rice wine was approximately 1.5 times higher than that in traditional Chinese rice wine (Li and others 2012), but reduced to only three-quarter times when no enzymatic hydrolysis was performed in extrusion (Lu and others 2002). The AAs are not only the primary nutrients in Chinese rice wine but the precursors of the Maillard reaction products as well. So the Maillard reaction that happens in conventional extrusion processing may be responsible for the decrease of free AAs in rice wines. Furthermore, the hydrolytic action of enzyme can be an important factor during extrusion. For instance, Lee and others (2012) demonstrated that enzymatic modification of wheat gluten by deamidation as well as hydrolysis could affect its flavor properties by changing both volatile and flavor-enhancing compounds such as glutamic acid. The Maillard reaction comprises a complex network system of chemical reactions between AAs reducing carbohydrates and their respective degradation products (Shibamoto 1989; Ikan 1994; Tsutsumiuchi and others 2011). This nonenzymatic browning is influenced by many factors, especially temperature, water activity, and pH in conventional extrusion without enzyme. The Maillardderived volatiles contribute to the overall aroma in heat-treated foods at low concentrations because of their extremely low odor threshold values (Lee and others 2012). In spite of the flavor contribution of the Maillard reaction products to baked and roasted foods, these volatiles are poor at enhancing the flavor of Chinese rice wine. The validated aroma contributors in this wine

Vol. 00, Nr. 0, 2014 r Journal of Food Science C1

C: Food Chemistry

Influence of Enzymatic Extrusion Liquefaction Pretreatment for Chinese Rice Wine on the Volatiles Generated from Extruded Rice

Volatiles of enzymatic extruded rice . . .

C: Food Chemistry

are mostly of alcohols, acids, esters, aldehydes, ketones, aromatic compounds, AAs but seldom Maillard-derived volatiles (Cao and others 2010; Mo and others 2010; Yu and others 2012; Chen and others 2013). During the Chinese rice wine making, a trace of the complicated pyrazines are formed from fermentation process as flavor materials (Fan and others 2007) and antioxidants (Sun and others 2013). However, some furans and simple pyrazines, mainly from the thermal pretreatment of raw rice and detected in majority of finished rice wine (Luo and others 2008), show almost little effect on the overall aroma except for a caramel-like flavor (Yu and others 2012; Chen and others 2013; Wang and others 2013). To the best of our knowledge, although there are lots of sensory and instrumental analyses of flavor compounds in cereal extrudates, such as oat (Parker and others 2000), maize (Bredie and others 1998), wheat (Bredie and others 2002), millet (Liu and others 2012a), and lots of physicochemical property analyses of enzymatic extrudates under different operating conditions (Reinikainen and others 1986; Tomas and others 1997; De MesaStonestreet and others 2012), the characterization of volatiles in enzymatic extrudates has not been studied yet. So the objectives of the current work were to clarify the total composition of the volatiles generated from rice flour processed by simultaneous extrusion cooking and enzymatic hydrolysis as a novel pretreatment of Chinese rice wine manufacture, and to determine if the Maillard reaction during extrusion would be influenced by the thermostable α-amylase, or merely by the changes of operating conditions.

Materials and Methods

lower moisture content (MC) than enzymatic extrusion samples coded E1 to E9, according to the severe extrusion pretreatment of conventional extrusion-processed rice wine manufacture (Lu and others 2002). All samples were blended for 10 min to equally distribute the moisture and amylase, respectively. The hydrated flour was stored at 4 °C overnight (12 h) in sealed plastic bags for equilibration until 2 h before extrusion. The enzyme activity prior to extrusion can be negligible as described by Oliveira and others (Tomas and others 1997).

Extrusion processing The extrusion processing was conducted in a laboratory scale twin-screw extruder (TSE 24 MC, Thermo Scientific, Waltham, U.S.A.) with a length to diameter ratio of 40:1. Combinations of 4 process temperatures (90, 100, 110, and 140 °C), 3 amylase concentrations (0.2‰ , 1.0‰ , and 3.0‰ , starch dry weight basis), and 2 MCs (20% and 40%, wet basis) were used, giving 11 processing conditions. The screw speed of the extruder was steady at 100 rpm and the feed rate was 1.5 kg/h. The screw configuration gave a residence time of 35 s for all extrusions. The barrel contained 4 zones. During the extrusion processing, the temperature of zone IV was adjusted by experimental requirement (Table 1) and the temperatures of barrel zone I (60 °C), zone II (70 °C), and zone III (80 °C) were kept constant (Li and others 2013). The extrusion sequences of different samples with or without enzyme were performed as in Tomas and others (1997). At the discharge, a 6-mm dia circular die was used. After extrusion, α-amylase activity in the extruded rice was inhibited by pH adjustment (approximately 3.5) and water batch as a modified method by Vasanthan and others (2001). All extrudates were then dried at 40 °C, packed into sealed plastic bags with low oxygen permeability, and stored at –20 °C until analysis. The rice flours with enzyme generally had low energy inputs for given extrusion conditions except for SE1, which was processed under more severe operating extrusion and exhibited similar energy inputs as SE2. The process responses: torque, product temperature, and die pressure were detected and recorded by the probe placed in last screw section. Specific mechanical energy (SME) values were calculated (in kJ/kg) with a modified equation (Baks and others 2008):

Materials and reagents Sticky rice (535 nuo), which had been debranned, was obtained from CPFCO Co., Ltd. (Wuxi, China). The rice was analyzed for their major components according to AOAC methods (AOAC 2003), and contained starch 75.21% ± 0.86%, protein 8.64% ± 0.23%, lipid 1.17% ± 0.09%, ash 0.49% ± 0.0%, and moisture 12.43% ± 0.47%, respectively. Thermostable α-amylase used was from Bacillus licheniformis (Termamyl 120 L, Novozymes, Denmark). The enzyme had an optimum pH of 6 to 8, a density of 1.2 g/mL, and an activity of 120 KNU/g (1 KNU = Kilo Novo alpha-amylase unit, defined as the amount of enzyme that hydrolyzes 5.26 g of starch per h). Analytical grade standards 2π(N/60)τ SME = (1) for headspace solid phase microextraction (HS-SPME) and gas F chromatography-linked mass spectrometry (GC-MS) were purchased from Sigma-Aldrich Co., Ltd. (Shanghai, China). The sol- where N is the screw speed (rpm), τ is the torque (Nm), and F is vents (HPLC grade) for RP-HPLC analysis were obtained from the total feed rate of the extruder (kg/h). Sigma-Aldrich Co., Ltd. All other chemicals and reagents were of reagent grade. HS-SPME method Solid phase microextraction (SPME) is a rapid, simple, versatile, Sample preparation and solvent-free extraction technique developed by Pawliszyn and Total 11 sets were broadly divided into 2 extrusion modes: en- coworkers in 1989 (Verbeek and others 2011; Liu and others zymatic extrusion (E1 to E9, processed under mild conditions) and 2012b; Xiao and others 2014). As an equilibrium method, SPME conventional extrusion (SE1 to SE2, processed under severe con- depends on analyte partitioning between the sample and the fused ditions) (Table 1). The rice was milled (dry-grind process) to pass silica fiber coated with a solid sorbent. Mar´ıa Teresa Tena and through a 0.6-mm sieve. Taking the initial moisture of rice flour others (Ezquerro and others 2003) had proved that multiple SPME into account, the amylase concentration of rice flour was adjusted was useful to the direct quantification of solid samples removing to the desired level (Table 1). The thermostable α-amylase was any matrix effect, with water as reliable solvent to prepare the premixed with distilled water, and this enzyme solution was added calibration solutions. Using the water as dilutors of samples for HSto rice flour with a ribbon blender to give the enzymatic extrusion SPME-GC-MS analysis was also validated by Mes´ıas and others samples. Appropriate pure water was put into the rice flour with- (2013). out thermostable α-amylase addition to prepare the sample coded In our present study, a 75-μm carboxen/poly (dimethylsiloxane) SE2. This nonenzymatic extrusion feedstock was adjusted to a (CAR/PDMS) coated fiber (Supelco, Inc., Bellefonte, Pa., U.S.A.) C2 Journal of Food Science r Vol. 00, Nr. 0, 2014

Volatiles of enzymatic extruded rice . . . Table 1–Extrusion processing parameters for the extruded rice flour.

Sample code E1 E2 E3 E4 E5 E6 E7 E8 E9 SE1 SE2

Process responses

Target barrel temp (°C)

Moisture content (%, wb)

Thermostable α-amylase (‰ , starch dry weight basis)

90 100 110 90 100 110 90 100 110 140 140

40 40 40 40 40 40 40 40 40 20 20

0.2 0.2 0.2 1.0 1.0 1.0 3.0 3.0 3.0 3.0 0.0

Tempa

(°C)

89.7 102.1 110.2 90.5 98.9 108.7 91.5 100.3 112.0 140.5 138.6

Product temp (°C)

Die pressure (Mpa)

SMEb (kJ/kg)

Product responses Maillard-derived volatilesc (ng/10 g)

95.46 107.21 116.37 94.84 105.72 115.29 96.65 106.10 118.67 147.92 144.38

2.52 2.69 3.21 2.30 2.57 2.65 2.08 2.26 2.54 3.72 3.96

90.27 78.55 68.32 38.92 29.65 25.13 23.74 18.80 16.79 363.08 487.43

111.0 (10.2)d ce 97.3 (7.9)c 32.0 (2.7)c 60.7 (5.9)c 63.7 (5.2)c 19.5 (1.6)c 84.8 (7.5)c 77.5 (6.6)c 19.7 (1.4)c 1691.6 (115.7)b 2835.7 (122.5)a

a

Temperature was measured by the probe placed in last screw section before the die channel in contact with the melt fluid. mechanical energy. Total quantity of Maillard-derived rolatiles, including 2-ethylfuran, 2-pentylfuran, 1H-pyrrole, pyrazine, 2-methyl-1H-pyrrole, methylpyrazine, 2,5(6)-dimethylpyrazine, ethylpyrazine, 3(2H)-pyridazinone. d Average of 2 replicates (standard deviation). e Numbers with different letters are significantly different (p < 0.05). b Specific c

was used for the volatile compounds extractions from the sample headspace (Ezquerro and others 2003; Lin and others 2010). The fiber was conditioned before use and thermally cleaned between analyses according to Dong Cao and others (Wei and others 2013). Each extrudate flour sample (10 g) was diluted with deionized water (30 mL) and shaken gently. The total mixture, to which 1 μL of the internal standard 1,2-dichlorobenzene (1.306 mg/mL) in ether was added, was put into a 50 mL glass bottle. Then the glass bottle was tightly capped with a silicon septum. The samples were equilibrated at 60 °C for 15 min and extracted for 30 min at the same temperature under stirring in a multipurpose sampler with SPME capability (MPS2, Gerstel, Germany). Immediately following extraction, the fiber was inserted into the injection port of GC (250 °C) for 5 min in split mode to desorb the analytes (Mo and others 2010). All the samples were carried out in duplicate.

mass spectral integration report, with reference to the internal standard. LRIs were calculated for each component, with respect to the retention times of a homologous series of standard n-alkanes (C7 -C25 , J&K Chemical Co., Ltd., Shanghai, China) under the same GC-MS conditions. Wherever possible, positive identifications were obtained by comparing the LRI with those of authentic compounds or in the literature. When no reference spectra were available, tentative identifications were made by comparison with the MS data. Relative abundance (RA) analysis was useful for pairwise comparison of identified compounds in different sets as described by Ames and others (2001). The RAs of analytes were carried out by a modified equation (Luo and others 2008):

GC-MS analysis Identification was carried out on a 1200L GC/MS-MS system equipped with a 1079 middle injector and a mass selective detector (MSD). The injection was done in type 1 Electronic Flow Control (EFC) split mode for 5 min using an inlet of 0.53 mm i.d., which improved the GC resolution. The temperature of both the injector and detector was at 250 °C. The separation of volatile compounds was performed using a fused silica capillary column (50 m × 0.32 mm i.d., coated with BPX-5 at 0.5 μm film thickness, SGE Ltd.) with an oven temperature program of 40 °C (held for 0 min), 20 °C/min to 100 °C (held for 0 min), 6 °C/min to 230 °C (held for 0 min), 10 °C/min to 250 °C (held for 7 min). The carrier gas was helium with a flow-rate of 15 μL/s and a split ratio of 10 mL/min. The MSD was used for identification of unknown compounds. Mass spectra were recorded in the centroid scan mode at an ionization voltage of 70 eV and a source temperature of 200 °C. Electron impact (EI) mass spectra were recorded in the 33 to 450 amu range with a scan time of 0.400 s. The total time required for analyzing a single sample was approximately 33 min. Volatiles were identified by comparison of their mass spectra and linear retention indexes (LRI) values with the mass spectra from MS libraries (NIST/WILEY/REPLIB/MAINLIB, 2005) and those from authentic compounds, or published elsewhere. Semiquantitative data of those volatiles were obtained from the

where Ac is the peak area of analyte; Ais is the peak area of internal standard; cis is the final concentration of internal standard in the sample, ng/10 g; and ct is the total concentration of volatiles in sample, ng/10 g.

RA (%) =

Ac c is × × 100% Ai s ct

(2)

Free AA content analysis Three representative samples of E2, E8, and SE2 were selected for free AA content analysis. Each extrudate (10 g) was placed in a 100 mL beaker with added 5% trichloroacetic acid (50 mL). The slurry was shaken by ultrasonic processing by a KQ-250B ultrasonic cleaner (Kun Shan Ultrasonic Instruments Co., Ltd., China) at 15 kHz for 20 min (total of twice, with 10 min between each treatment) until the solvent was completely absorbed. Then the solution was poured into a funnel through a double-layer filter, and the filtrate (1 mL) was clarified by centrifugation (10000 rpm for 10 min). Finally, the supernatant (20 μL) was injected in a chromatographic column by an automatic sampler. The AA analysis was performed by an Agilent 1100 series HPLC (Agilent Tech Inc., Santa Clara, Calif., U.S.A.) combined with a precolumn derivatization of o-phthaldialdehyde (OPA) and 9fluorenylmethyl chloroformate (FMOC). The reverse-phase highperformance liquid chromatography (RP-HPLC) system consisted of an online degasser, a quaternary gradient pump, an automatic sampler, and a diode array detector (DAD). Chromatographic Vol. 00, Nr. 0, 2014 r Journal of Food Science C3

C: Food Chemistry

Extrusion conditions

Volatiles of enzymatic extruded rice . . . Table 2–Elution gradient used in the RP-HPLC method. Time (min)

C: Food Chemistry

0 27.5 31.5 32 34 35.5

Eluent A (%, v/v) 92 40 0 0 0 92

Eluent B (%, v/v) 8 60 100 100 100 8

Table 3–Correlation coefficients between product responses, process responses, and input parameters.

Flow rate (mL/min) 1.0 1.0 1.5 1.5 1.0 1.0

separation was performed on an ODS HYPERSIL C18 column (250 mm × 4.6 mm i.d., 5 μm, Agilent Tech Inc.) fitted with guard column (4 mm L × 3 mm i.d., Agilent Tech Inc.) packed with the same material (Topuz and others 2005). The column temperature was maintained at 40 °C. Mobile phase A: 99.48% sodium acetate (0.08 M), 0.02% triethylamine, 0.5% tetrahydrofuran; and mobile phase B: 20% sodium acetate (0.2 M), 40% acetonitrile, and 40% methanol (Shen and others 2010). The elution gradient and the flow rate are shown in Table 2. The column eluate was monitored with a DAD set at 338 nm and 262 nm (proline) for the identification of free AAs (Fu Liang and Xu 2011).

Statistical analysis Statistical analysis was performed using SPSS 16.0 for Windows (SPSS Inc., Chicago, Ill., U.S.A.). Majority of data were analyzed by analysis of variance (ANOVA), and a statistically significant difference was identified at the 95% confidence level. Multivariate of general linear model and correlation analysis were conducted to determine the significant differences of the input parameters, process responses, and product responses of enzymatic extrudates. All process treatments and analyses were conducted in duplicate or triplicate and the mean values were reported.

Results and Discussion Extrudates Eleven extrudates were produced under different combinations of barrel temperature, MC, and amylase concentration (Table 1), using sticky rice flour as feedstock material. Product temperatures were within approximately 7 °C of the corresponding target value. Table 1 also showed different values of SME and relative amounts of Maillard-derived volatiles as process and target responses, respectively. Effects of temperature and α-amylase concentration (AC) on SME and Maillard reaction SME was significantly affected by enzyme level at mild operating conditions in the extruder (Table 3). The negative correlation between temperature and SME was not very obvious in sets of E1 to E9, but undoubtedly SME had an inverse response to increasing temperature, as can be inferred both from the SME values of severe extruded rice (Table 1) and the significant difference (p = 0.025) between them (Table 4). As temperature increased, the rice flour melt was less viscous because of the intense effect of the enhanced thermal energy combined with the hydrolysis of enzyme, and did not require as much energy to flow through the barrel as compared to the melt extruded at low temperature. The decline in SME with increasing AC up to 3 ‰ under mild extrusion conditions can be attributed to the reduced viscosity brought by the action of α-amylase on the rice starch fraction. Without enzyme added, a high SME (487.43 kJ/kg) showed in sample SE2 was calculated C4 Journal of Food Science r Vol. 00, Nr. 0, 2014

BTa BT AC PT DP SME QM

1

ACb < 0.001 1

n.s.f

PTc

DPd

0.993∗∗∗

0.671∗

SME

− 0.220 n.s. 0.055 n.s. − 0.653 n.s. − 0.810∗∗ 1 0.645 – 0.225 n.s. 1 0.492 n.s. 1

QMe − 0.802∗∗ − 0.164 n.s. − 0.779∗ – 0.405 n.s. 0.514 1

∗ Significant at P < 0.05. ∗∗ Significant at P < 0.01. ∗∗∗ Significant at P < 0.001. a

Barrel temperature. concentration. Product temperature. d Die pressure. e Total quantity of Maillard-derived volatiles. f n.s., not significant. b Amylase c

by Eq. (1) for a high torque input. A similar high SME was also obtained in SE1 with enzyme addition. It may be partly due to the reduced enzymatic hydrolysis caused by the inactivation of αamylase at high temperature over approximately 130 °C (Likimani and others 1991), and partly by the role of low moisture level on the texture changes of feedstock. Several authors also reported that SME lessens with increasing temperature and AC during enzymatic extrusion processing of starches and cereals (Govindasamy and others 1997a; Baks and others 2008; De Mesa-Stonestreet and others 2012). The relative quantity of Maillard reaction products (QM) was significantly influenced by temperature (R = –0.802, p < 0.01) as described by the correlation analysis based on E1 to E9 (Table 3). The inverse relationship between QM and temperature was similar to the relationship of SME with temperature. The role of enzyme for Maillard reaction was also important according to the corrected model (Table 4). However, an opposite tendency between QM/total volatiles and temperature was discovered in other reports. For example, Ames and others (2001) demonstrated that the amounts of volatile compounds in glycine/glucose model systems increased with temperature from 120 to 180 °C at all pH values. On the other hand, Maillard reaction products were found in the most severely processed extrudates of oat flours (high temperature of 150 or 180 °C, low moisture of 14.5% or 18%), and the amount of total volatiles in OG2 (the only debranned oat flour sample of Germany) decreased with temperature at the fixed MC level (Parker and others 2000). In our present study, the probable explanations can be made that the degree of the Maillard reaction or other reactions during enzymatic extrusion could be lower than in other thermal treatments due to its less severe operating conditions (90 to 110 °C at a high water content of 40%) and the significant influence of added enzyme on SME (R = – 0.810, p < 0.01) (Table 3). As SME increased at lower temperature levels, the starch, protein, and lipid in rice flour was degraded to a greater extent, making it more accessible to enzyme attack, especially in high AC (Duque and others 2013). Thus more flavor precursors were released to form volatile compounds. Moreover, the difference of QM levels between enzymatic extruded rice (E9) and conventional extruded rice (SE1) were extremely significant at the same enzyme concentration, but those between enzymatic extrudates themselves (E1 to E9) were not significant. It indicated that the degree to which QM was affected by AC was small in comparison to the effect of operating conditions of temperature and MC.

Volatiles of enzymatic extruded rice . . .

Source

Response

df

Sum of squares

Mean squares

F value

P value

R2 /Adj R2

Corrected model

SME QM Total volatilesa SME QM Total volatiles SME QM Total volatiles SME QM Total volatiles

4 4 4 2 2 2 2 2 2 4 4 4

6238.71 8532.54 8.63 E6 308.41 6961.04 7.54 E6 5930.30 1571.49 1.09 E6 57.29 367.29 0.65 E6

1159.68 2133.13 2.16 E6 154.21 3480.52 3.77 E6 2965.15 785.75 0.54 E6 14.32 91.82 0.16 E6

108.91 23.23 13.30 10.77 37.91 23.24 207.05 8.56 3.36

< 0.001 0.005 0.014 0.025 0.003 0.006 < 0.001 0.036 0.139

0.991/0.982

Temperature Amylase concentration Error

a

0.959/0.917 0.930/0.860

Total quantity of flavor compounds in extruded rice.

GC-MS analysis A total of 62 volatile compounds, including 16 alcohols, 14 aldehydes, 7 ketones, 5 acids, 10 furans, and 10 nitrogen-containing compounds, were identified in the extruded rice flour samples (Table 5). Thirteen of these volatiles were found exclusively in the severe processed extrudates with or without enzyme, and most of them were furans and pyrazines. The compounds generated from mildly enzymatic extruded rice of E1 to E9 were seldom Maillard reaction products except for furans. It is appropriate to judge these volatiles in relation to their mode of formation from the lipid degradation even if the lipid content detected in rice flour was very low, because the Maillard reaction seems to be greatly inhibited by the mild conditions of simultaneous extrusion and enzymatic hydrolysis. Some compounds remained unidentified, but these constituted only a tiny part of the total ion chromatogram. Numerous identified compounds have been reported previously in extruded cereals and beverages (Parker and others 2000; Ames and others 2001; Liu and others 2012a; Chen and others 2013; Wang and others 2013). The kinds and amounts of the Maillard-derived volatiles generally increase as the extrusion conditions become more severe, and this tendency has already been reported in previous work carried out on wheat, maize, and oat flours (Bredie and others 1998; Parker and others 2000; Bredie and others 2002). These volatiles are often associated with inoperative or undesirable flavor characteristics in Chinese rice wine (Luo and others 2008; Mo and others 2010; Chen and others 2013; Wang and others 2013). In the present study, a similar effect of severe extrusion conditions was observed for Maillard volatiles of SE1 and SE2, and relative low levels of each category of volatiles were found in extruded samples of E1 to E9 used as Chinese rice wine fermenting material. The pattern shown for total volatiles of these enzymatic extrudates in Figure 1 was typical. Furthermore, as the effect of enzyme on the Maillard reaction could not be made clear only by comparison in mildly enzymatic extrusion samples, the RAs of Maillard Ncontaining compounds, alcohols, alkenals, and ketones between SE1 and SE2 was pairwise compared for reference (Figure 2a). Maillard-derived volatiles analysis. Nitrogen-containing compounds were present in all extruded rice. The high levels of the N-containing compounds generated during the Maillard reaction, such as pyrazines and pyrroles, were found both in the severe processed extruded rice with or without enzyme (Table 5), but barely in the mildly enzymatic extruded rice. The structures of identified pyrazines were relatively simple, such as pyrazine, methylpyrazine, 2,5(6)-methylpyrazine, and ethylpyrazine, which

was common to other extruded cereals and model systems (Bredie and others 1998; Parker and others 2000). The pyrazines were responsible for imparting the aroma described as roasted, nutty, and popcorn to some cereals and barbecue made at high temperature (Ames and others 2001; Herrera-Jim´enez and others 2007). According to the flavor wheel of Chinese rice wine, however, there is little contribution made by simple pyrazines to the main sensory flavor characterizations described as mellow, caramel, fruity, herbal, fragrant, wheat Qu, and honeyed for wine products (Wang and others 2013). The reaction mechanism for pyrazines formation had been proposed and evaluated (Hwang and others 1995; Van Lancker and others 2010; Guerra and Yaylayan 2012), which was based on the consumption of free amino group and carbonyl group in reaction substrates. The majority of multisubstituted pyrazine derivatives, together with thiophenes, thiazoles, and miscellaneous S compounds, were formed in cooked cereals under more severe conditions (temperature of 150 to 180 °C and MC of 15% to 18%) than the conventional extrusion of SE1 and SE2, let alone the mildly enzymatic extrusion (Parker and others 2000; Bredie and others 2002; Fan and others 2007). So it seems that the severe heat pretreatments, even including conventional extrusion cooking, are not applicable to Chinese rice wine brewing at the cost of nutrients and flavor precursors in fermenting material. Moreover, there was an overall decrease in both the number and relative amount of pyrazines in sets of E1 to E9 as the conditions became milder (Table 5). It was not suitable for confirming the influence of enzyme on Maillard reaction at similar low levels of QM. Therefore an obvious distinction of relative amounts of pyrazines found between severe extruded rice with or without enzyme was valuable (Figure 2a), which may be explained by the action of enzymatic hydrolysis on relieving the severe extrusion environment. Only 2 pyrroles, 1H-pyrrole and 2-methyl-1H-pyrrole, were detected under severe extrusion conditions in this study. Parker and others (2000) found that 1H-pyrrole was the most abundant species in extruded oat and its derivatives were only present in the extreme extrudates. Similar to the extruded oat, the levels of pyrroles in extruded rice reduced as temperature decreased. Several furans were identified as the Maillard-derived volatiles in the extruded rice with the same trends observed for the levels of pyrazines, except for 2-ethylfuran and 2-pentylfuran derived from lipid degradation. The similar high levels of furans were found in severe extruded rice of SE1 and SE2, with RAs of 28.76% and 28.59%, respectively (Figure 2a). During extrusion cooking, furans can be formed by 2 possible ways: (1) the thermal degradation of carbohydrates such as glucose, lactose, and fructose; (2) the thermal degradation of some AAs such as serine, alanine, and threonine, Vol. 00, Nr. 0, 2014 r Journal of Food Science C5

C: Food Chemistry

Table 4–Multivariate of general linear model of barrel temperature and amylase concentration on the process and product response of enzymatic extruded rice coded from E1 to E9.

C: Food Chemistry

Alkenals Acetaldehyde Pentanal 2,4-Pentadienal Hexanal 2-Hexenal Heptanal 2-Heptenal Octanal 2-Octenal Nonanal (E or Z)-2-nonenal Decanal Benzaldehyde (E,E)-2,4-decadienal Total alkenals Alcohols 3-Buten-1-ol 1-Pentanol 1-Hexanol 2-Heptenol 6-Methyl-6-hepten-4-yn-3-ol Z-linalool oxide 1-Octen-3-ol 1-Heptanol (E or Z)-2-Octen-1-ol 1-Octanol (E or Z)-2-nonen-1-ol 1-Nonanol 1-Decanol Phenol 2-Phenoxy-ethanol 2,4-bis(1,1-dimethylethyl)-phenol Total alcohols Furans 2-Ethylfuran Furfural Dihydro-2(3H)-furanone 2-Furanmethanol 2-Pentylfuran 2(5H)-furanone 5-Hydrxoymethylfurfural 5-Pentyl-2(5H)-furanone Dihydro-5-pentyl-2(3H)-furanone 2,3-Dihydro-benzofurane Total furans

Compound MS MS + LRI ms + lri MS + LRI MS + LRI MS + LRI MS + LRI MS + LRI MS + LRI MS + LRI MS + LRI MS + LRI ms + lri ms + lri MS MS + LRI MS + LRI MS + LRI MS MS MS + LRI MS + LRI MS + LRI MS + LRI MS + LRI MS + LRI ms + lri MS + LRI MS MS MS MS + LRI MS MS + LRI MS + LRI MS + LRI MS MS MS MS

Influence of enzymatic extrusion liquefaction pretreatment for Chinese rice wine on the volatiles generated from extruded rice.

Volatile compounds in enzymatic extruded rice, produced under different conditions of varying barrel temperature (BT), α-amylase concentration (AC) an...
517KB Sizes 0 Downloads 6 Views