Cake of Japonica, Indica and glutinous rice: Effect of matcha powder on the volatile profiles, nutritional properties and optimal production parameters

Highlights

• •Matcha addition modified the crystal structure of rice cakes, and reduced the relative crystallinity.
• •Matcha addition decreased the eGI of the rice cakes in a dose dependent manner.
• •Rice cakes with 1.6% matcha are with good sensory acceptance.

Abstract

Matcha addition decreased the relative crystallinity and provided with a refreshing flavor for all types of rice cakes. Matcha also significantly enhanced the phenolic content and the oxidant defense of cakes. Compared with the other two types of rice cakes, the one made of glutinous rice are with the lowest starch digestibility. Adding matcha to rice cakes inhibited the in vitro starch digestion, and a significant decrease in the expected glycemic index (eGI) and an increase in resistant starch (RS) were observed. Besides, according to the results of sensory evaluation, an optimized formulation of matcha rice cake was expected to contain 1.6% matcha, 82% water and steamed for 39 min. These findings suggest that matcha could be a favorable food additive to improve both the flavour and nutritional value of steamed rice cake.

1. Introduction

Steamed rice cake has been a traditional snack in Asian countries and is often consumed to celebrate the Lunar New Year (Chen et al., 2010). Due to the highly processed white rice flour used to make rice cake, a large amount of easily digested carbohydrate exists, which, when consumed in excess, can cause postprandial hyperglycemia. Thus, steamed rice cake may not be a suitable food choice for those seeking to reduce or control their blood glucose level. Similarly, sweet dumplings, mochi, zongzi and other starchy foods are not much friendly to hyperglycemia patients as well. Therefore, developing low glycemic index (GI) foods is necessary.

Matcha is a unique form of green tea that is grown under shade and processed by grinding tender whole tencha tea leaves (Hasegawa et al., 2016). Matcha contains numerous bioactive constituents, including polyphenols, amino acids, caffeine, and chlorophyll, all of which contribute to its soft green color and unique flavor. In addition to its traditional use consumed as a beverage, matcha powder more recently has been added to foods such as bread, cake, confectionery, yogurt and ice cream (Dietz et al., 2017). Matcha has also been shown to exert a number of favorable physiological functions, like oxidant defense, improvement of blood glucose regulation, and reduction of perceived stress, making it increasingly popular in the food market (Kochman et al., 2020, Unno et al., 2018, Unno et al., 2019). The addition of matcha powder to cooked rice and rice noodles is reported to refresh flavors and reduce starch digestibility (Fu et al., 2020, Li et al., 2021). However, the effects of matcha powder on the steamed rice cake, especially the ones made of different types of rice remain unclear. In the present study, matcha powder was added as a functional ingredient to rice cakes, and the volatile profiles, oxidant defense attributes, and the in vitro starch digestibility were evaluated.

2. Materials and methods

2.1. Materials

Matcha was purchased from Zhejiang Camel Transworld Organic Food Co., Ltd (Hangzhou, China), and Japonica rice (J), Indica rice (I) and glutinous rice (G) were gifted by the same company. The amylose contents of each rice are 0.83 ± 0.05% (G), 16.72 ± 0.58% (J) and 25.54 ± 1.97% (I), respectively. Pepsin, porcine pancreatic α-amylase, aspergillus niger amyloglucosidase and n-alkanes (C7-C40) standard solution were purchased from Sigma-Aldrich (MO, USA). Catechin standards were purchased from YuanYe Bio-Technology (Shanghai, China). Chlorophyll assay kits were purchased from Jiancheng Bioengineering institute (Nanjing, China). All other chemical solvents were at analytical grade, and purchased from Sinopharm Group Co., Ltd (Shanghai, China).

2.2. Chemical composition of matcha

Moisture content was measured by a moisture analyzer (MB23, OHOUS, NJ, USA) following the manufacturer’s protocol. Dry matcha powder (3 g) were tested by infrared heating. The total phenolic, free amino acids, and soluble sugar contents were analyzed according to the methods of Wang et al (Wang et al., 2018). Chlorophyll content was assayed following the kit manufacturer's instructions.

Catechins and caffeine contents were analyzed using a high performance liquid chromatography (HPLC) system (Shimadzu, Japan) with ultraviolet–visible detector (UVD) according to the methods of Xu et al (Xu et al., 2019), with some modifications. The analysis was performed on a Shim-pack ODS column (150 × 4.6 i.d. mm, Shimadzu, Japan), packed with 5 μm C18 silica. Mobile phase A and B were prepared with different ratios of acetonitrile, acetic acid (≥99.5%) and water (3:0.5:96.5 for A, 30:0.5:69.5 for B, v:v:v), respectively. The gradient program was established so that the mobile phase B increased from 30% to 85% by a linear gradient during the first 35 min of elution, then held at 85% for a further 5 min. Before each run, the column was re-equilibrated for 20 min using the initial solvent composition. The flow rate was constantly maintained at 1.0 ml/min and absorbance was tested at 280 nm.

2.3. Rice cake preparation

Japonica rice (J), Indica rice (I) and glutinous rice (G) were used to prepare three types of rice cakes, with or without matcha. Each of the three types of rice were first soaked in distilled water overnight, then separately ground with a household blender (MR6501, Morphy Richards, Australia). The wet rice powders were dried in an oven overnight and fractions smaller than 40-mesh were collected for final production. The dried rice powders (moisture content 3%-4%) were stored in sealed packages at 4 °C. For rice cake preparation, a planetary mixer (HMJ-A35M1, Bear, China) was used to blend each of the dry rice powders, matcha (2%, w/w) with distilled water (70% v/w) for 10 min at low speed. The resulting dough was divided into 20 g portions and molded, then steamed in a preheated electric cooker at 120 °C for 40 min. The resulting rice cakes were cooled at room temperature, vacuumed in plastic bags and preserved at 4 °C for analysis.

2.4. X-ray diffraction (XRD) spectroscopy

Samples were lyophilized and milled to collect fractions smaller than 150-mesh. The XRD spectroscopy was obtained from an X-ray diffractometer (D2 Phaser, Bruker, Germany) with a diffraction angle (2θ) between 4 and 40◦ at a rate of 2◦/min. The crystallinity was calculated according to the eqn (1) with Origin 9.0 software (OriginLab, MA, USA).

$$\mathit{Crystallinity}=\frac{\mathit{Area}\mathit{of}\mathit{crystalline}\mathit{peaks}}{\mathit{Area}\mathit{of}\mathit{all}\mathit{peaks}\left(crystalline+amorphous\right)}\times 100$$ (1)

2.5. Fourier transform infrared (FTIR)

A FTIR spectrometer (IR Prestige-21, Shimadzu, Japan) was used to analyze the 150-mesh sieved rice samples. Spectra were scanned from 4,000 to 400 cm⁻¹ at a resolution of 4 cm⁻¹. Data was processed using Origin 9.0 software (OriginLab, MA, USA).

2.6. Scanning electron microscopy (SEM)

Rice samples passing through 150 mesh sieves were mounted on a specimen stub and coated with gold prior to analysis at 1,200 × and 3,000 × magnification using a scanning electron microscope (TM3030, Hitachi, Japan).

2.7. Determination of volatile compounds

Volatile compounds were determined for each of the three rice cake samples by headspace solid phase micro-extraction (SPME) coupled with gas chromatography/mass spectrometry (GC/MS; Agilent, CA, USA). Samples (3 g) were placed in 20 ml vials, sealed with a silicone-teflon cap and then the extraction of volatile compounds was conducted by exposing the SPME fiber coated with Divinylbenzene/Carboxen/Polydimethylsiloxane (DVB/CAR/PDMS) (Supelco, Bellefonte, PA, USA) in the headspace at 80 °C for 30 min. The SPME fiber was then quickly transferred to a GC injection port operating in the split-less mode. The port temperature during the injection is 250℃. Volatile compounds were separated on a HP-5MS column (30 m × 0.25 mm × 0.25 μm, Agilent, USA). Helium was used as the carrier gas with a linear velocity of 1.0 ml/min. Chromatographic separation was conducted as follows: oven temperature was held for 5 min at 40 °C, then increased at a rate of 10 °C/min to 150 °C and kept for 5 min. Finally, the temperature was raised to 220 °C at the rate of 10 °C/min, and held for 22 min. The scanning range of the MS was 35–500 m/z. Data were acquired and individual peak identification was based on the comparison of their mass spectra with reference mass spectra from the Mass Spectral Libraries (NIST/Wiley Libraries, MA, USA). The mass spectra of volatile compounds were confirmed by comparing linear retention indices (LRI) calculated relatively to the standard alkanes (C7-C40). The quantities of volatile compounds were expressed as percentages of the total identified signal with 2-Octanol as the internal standard.

2.8. Oxidant defense capacity analysis

Ground samples of 500 mg each (150 mesh) were mixed with 25 ml methanol and boiled in distilled water for 45 min. After cooling, the mixture was then centrifuged at 1,000 rpm for 5 min, and the supernatant was collected. The radical scavenging activity of each rice powder was assessed using 2,2-diphenyl-1-picrylhydrazyl radicals (DPPH) and 2,2′-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid (ABTS) with a spectrophotometer (752 N, INESA, China) according to the method of Wang et al (Wang et al., 2018).

2.9. In vitro starch digestion

The in vitro starch digestion analysis was conducted following the method of Go et al (Goni et al., 1997), with some modifications. Briefly, 1 g of each rice cake was mixed with 20 ml phosphate buffered saline (PBS) and homogenized for 2 min using a household blender (MR6501, Morphy Richards, Australia). Next, a two-stage gastrointestinal digestion was performed according to the methods of Dartois et al (Dartois et al., 2010). After adjusting the pH to 1.50 ± 0.01, 5 ml pepsin solution (10 mg/ml) was added to the mixture and maintained in a circulatory water bath at 150 rpm and 37 ± 1 °C. After 30 min, the pH was increased to 6.80 ± 0.01 to inactivate the pepsin. Then, 10 ml of enzyme liquid (300 U/ml porcine pancreatic α-amylase and 260 U/ml amyloglucosidase) was added and stirred in a water bath at 150 rpm and 37 ± 1 °C for 3 hrs. Aliquots were withdrawn at 20, 60, 90, 120, and 180 min, and enzymes were inactivated by incubating in boiling water for 5 min. After centrifugation at 3000 rpm for 10 min, the supernatant was collected. A dinitrosalicylic acid (DNS) method was performed to measure the glucose concentration of the incubated mixture using an ultraviolet spectrophotometer (752 N, INESA, China) with absorbance measured at 540 nm (Li et al., 2021). Results were calculated as the starch digestibility (%), and the starch digestibility-time curves of each kind of rice cake were simulated and plotted with the Origin 9.0 software (OriginLab, MA, USA).

For each of the rice cakes, three starch nutritional fractions were determined according to the starch digestion rate by eqn (2): rapidly digestible starch (RDS, digested within 20 min), slowly digestible starch (SDS, digested between 20 min and 120 min), and resistant starch (RS, undigested after 120 min).

$$\mathit{Starch}digestibilityrate(\%)=\frac{\mathit{Content}ofhydrolyzedglucose\times 0.9}{\mathit{Total}starchweight}\times 100\%$$ (2)

The kinetics of starch hydrolysis and the digestion rate constant (K, min-1) were generated under a first-order equation model by eqn (3), where t represents the digestion time, Ct (%) is the percentage of digested starch at a specific time, and C∞ is the estimated percentage of digested starch at the end of the reaction.

$$C_t=C_{\infty }(1-e^{-kt})$$ (3)

The hydrolysis index (HI) and the expected glycemic index (eGI) were calculated according to the methods of Go et al (Goni et al., 1997).

2.10. Experimental design for response surface methodology

Rice cake made from Japonica rice was tested in this experiment. Under the Box-Behnken design, the water ratio (A), matcha ratio (B), and steaming time (C) were selected as three independent variables, and the effect of these three on the sensory score was regarded as the assessment index (Y). The range of each variable was set based on preliminary tests for the addition of 1%, 2%, or 3% matcha (w/w of the dry rice powder); water ratios of 60%, 80%, or 100% (w/w of the dry rice powder), and 20 min, 40 min, or 60 min of steaming time. Levels of variables in the experimental design were coded as −1, 0 and 1 and shown in Table S1. A total of 17 combinations were incorporated in the experimental design, and displayed in Table S2. A second order polynomial quadratic regression equation was generated to calculate the effects of these three independent variables on the responses (Y). The optimal levels for A, B, C were then predicted, and the most appropriate formulation was recommended. Data was processed and analyzed using Design Expert 8.0 software (State-East, MN, USA).

2.11. Preference test for matcha rice cake

Sensory evaluation was conducted by an experienced group of 40 panelists (20 males and 20 females aged 18–50) who were previously trained in sensory attributes and the scoring system. For each of the three types of rice cakes, they were removed from the steamer and cooled at room temperature for 10 min, then cut into equal sized slices and served on a white enamel plate. Panelists were instructed to rate samples on four main attributes: color, odor, taste, and appearance. A 10-point Hedonic Scale (0 = dislike extremely and 10 = like extremely) was used to assess preference for each of the four parameters, as well as the overall acceptability.

2.12. Statistical analysis

Data were obtained from at least three independent experiments and are presented as mean ± standard deviations for assays. One-way analysis of variance (ANOVA) and the Turkey test were performed to evaluate the differences among groups using SPSS software 20.0 (SPSS Inc., IL, USA). p < 0.05 was regarded as statistically significant.

3. Results

3.1. Chemical components of matcha powder

The chemical composition of the matcha powder used on this study is shown in Table 1. The free amino acid and soluble sugar contents were are 5.57 ± 0.15% and 1.22 ± 0.25%, respectively, mainly contributing to the fresh and sweet flavor. Caffeine, source of the bitter taste, was present at 2.35 ± 0.03%. The total chlorophyll content was 0.23 ± 0.01%, which was provided by the a and b isomers at 0.15 ± 0.01% and 0.08 ± 0.01%, respectively. Polyphenols contribute significantly to the astringency of matcha and were present at 20.26 ± 0.11%, with the catechins contributing approximately 80% of the total polyphenol amount. Among catechin monomers, epi-gallocatechin-3-gallate (EGCG) was with the highest ratio as present at 9.24 ± 0.53%.

Table 1.

Chemical components of matcha green tea.

Components	Contents (%)
moisture	5.55 ± 0.05
polyphenols	20.26 ± 0.11
free amino acids	5.57 ± 0.15
soluble sugars	1.22 ± 0.25
caffeine	2.35 ± 0.03
chlorophyll	0.23 ± 0.01
chlorophyll-a	0.15 ± 0.01
chlorophyll-b	0.08 ± 0.01
GC	0.96 ± 0.01
EGC	3.24 ± 0.03
C	0.10 ± 0.00
EC	0.56 ± 0.01
CG	0.28 ± 0.01
ECG	1.42 ± 0.01
GCG	0.32 ± 0.04
EGCG	9.24 ± 0.53

Volatile compound	LRI	Flavor composition (%)
Aldehydes
Pentanal	700	3.16 ± 0.61
Hexanal	798	6.13 ± 0.45
Heptanal	897	4.98 ± 0.13
Nonanal	1105	4.82 ± 0.62
Decanal	1203	2.31 ± 0.35
2-Heptenal, (Z)-	942	2.63 ± 0.08
Benzaldehyde	950	4.85 ± 0.71
2,4-Heptadienal, (E, E)-	1007	14.51 ± 1.74
Alcohols
1-pentanol	766	4.75 ± 0.13
1-Octen-3-ol	980	3.16 ± 0.25
1-Octanol	1069	4.11 ± 0.21
3-Hexen-1-ol, (Z)-	850	1.12 ± 0.22
2-ethyl-1-hexanol	1040	3.19 ± 0.57
Heptanol	979	1.34 ± 0.15
2-Nonen-1-ol	1081	6.15 ± 0.15
Alkanes
Dodecane	1200	2.12 ± 0.24
Tridecane	1300	1.34 ± 0.31
Tetradecane	1400	0.79 ± 0.06
Octadecane	1800	0.59 ± 0.1
Pentadecane	1500	0.53 ± 0.04
Hexadecane	1600	0.32 ± 0.05
Ketones
2-Heptanone	890	2.41 ± 0.02
6-methyl-5-hepten-2-one	982	0.49 ± 0.03
3,5-Octadien-2-one	1076	9.94 ± 0.56
Acetophenone	1051	0.22 ± 0.04
Others
2,5-Dimethylpyrazine	910	1.48 ± 0.22
p-Xylene	857	2.03 ± 0.25
1-Ethyl-1H-pyrrole	812	0.41 ± 0.05
Phytol	2092	0.97 ± 0.31

GC, (+)-gallocatechin; EGC, (-)-epigallocatechin; C, (+)-catechin; EC, (-)-epicatechin; CG, (-)-catechin gallate; ECG, (-)-epicatechin gallate; GCG, (-)-gallocatechin gallate; EGCG, (-)-epigallocatechin gallate. Note: % of dry tea weight (w/w).

LRI, retention index, which was calculated refereed to the retention time of C7-C40 n-alkanes under the same conditions.

Then volatile compounds in matcha powder were identified through the GC–MS. The key odorants consist of aldehydes (8 components), alcohols (7), ketones (4), alkanes (6) and others (4). Aldehydes predominated in matcha, with 2,4-heptadienal, (E, E)- being the principal one (14.51%), which provides with a green flavor. Next followed by hexanal, heptanal, nonanal and benzaldehyde. Alcohol compound 2-nonen-1-ol is contributory to the sweet melon aroma, and presented the highest content (6.15%) in alcohols. 1-pentanol is the second most abundant alcohol component, which contributes to a fruity and flora profile of matcha. Besides alcohols, ketones are previously considered as another source of green and fresh aroma of Japanese green tea (Tan et al., 2019). 3,5-octadien-2-one accounted for a significantly high content in ketones, which was characterized with a green and grassy notes. 2-heptanone has a fruity aroma, and had a relatively low level (2.41%). Other key odourants detected in matcha are 2,5-dimethylpyrazine (roasted, nutty), p-xylene (plastic, green, pungent), 1-ethyl-1H-pyrrole (burnt, roasted) and phytol (floral, powdery, waxy), playing together in the formation of the specific odor characterization of matcha.

3.2. Molecular structure of matcha rice cake

The crystallization, chemical bonds and surface frameworks of matcha rice cakes made by three types of rice were then evaluated. Similar diffraction peaks were noted for all types of rice cakes with or without matcha (Fig. 1-A). All samples displayed a well-defined peak around 17° with a typical B-type diffraction pattern. As indicated in other studies, the B-type peak is formed by the crystallization of amorphous starch during retrogradation, especially the amylopectin fractions (Zhang et al., 2017). A V-type diffraction pattern was found in the Japonica rice cake (JRC) and Indica rice cake (IRC) samples, with a peak at approximately 20°, which was usually formed by amylose and lipid. Notably, the diffraction intensity for all three types of rice cakes with matcha were slightly weakened. For example, the relative crystallinity decreased from 16.12 ± 1.94% to 11.88 ± 1.41% of the glutinous rice cake (GRC) after matcha addition. These results are consistent with the other studies showing that tea products could disintegrate the crystallization region, thus retard the retrogradation behavior of gelatinized starch and improve the deterioration of starchy food such as firmness (Wu et al., 2009, Fu et al., 2021). To further explore the effect of matcha on the chemical bonds of rice cake, FTIR was conducted (Fig. 1-B). Matcha supplementation did not generate new covalent bonds of rice cakes, as no other characteristic peaks appeared. However, matcha addition reduced the band intensity at around 3400/cm, which represents inter- or intramolecular hydrogen-bond hydroxyl groups (Fu et al., 2021). The strength of peaks around 1640/cm and 1400/cm (C—O—O stretching vibration) were also weakened with matcha supplementation (Han et al., 2012). Besides, as shown in Fig. 1-C, the number of the amorphous voids for all types of rice cakes with matcha increased, compared to the ones without matcha, as observed by SEM.

Fig. 1.

Crystal structure and molecular structure of rice cakes with or without matcha. (A) X-ray diffraction patterns. (B) FTIR spectra. Samples share same characteristic peaks between rice cake with or without matcha, indicating that there was no new chemical bond formed. Intensity of peak around 3400/cm weakened, which represents hydrogen-bond hydroxyl groups. (C) SEM images of rice cakes with or without matcha. Matcha addition changed the surface microstructure of rice cakes with hole density increased, as arrows indicated. Representative images at 1,200x and 3,000x were shown. JRC, Japonica rice cake; JRC-M, Japonica rice cake with 2% matcha; IRC, Indica rice cake; IRC-M, Indica rice with 2% matcha; GRC, glutinous rice cake; GRC-M, glutinous rice cake with 2% matcha.

3.3. Volatile compounds

The overall aroma and flavor of the rice cake was contributed by various volatile compounds. As shown in Table 2, the aromas from the three types of matcha rice cakes were primarily aldehydes, alcohols, esters, ketones, and alkanes. The primary group of compounds in all three rice cakes was the aldehydes. Adding matcha to the rice cakes decreased the amount of hexanal (green, leaf-like) and nonanal (fat, citrus, green) significantly (p < 0.05) compared to the normal ones. Specifically, the level of hexanal was greatly reduced from 36.88 ± 1.23%, 34.05 ± 2.70%, and 59.36 ± 2.33% in the normal JRC, IRC and GRC to 22.65 ± 5.76%, 27.72 ± 1.71%, and 38.97 ± 5.53% in their corresponding ones with matcha, respectively. Conversely, the amount of pentanal (almond, malt, pungent) and decanal (orange, herbal) increased in all three types of the rice cakes with matcha. 2-heptenal (Z) (green, fresh), and benzaldehyde (almond, burnt sugar) were only detected in the matcha rice cakes. In terms of alcohols, matcha addition decreased the content of 1-hexanol (fresh, green) in all three types of rice cakes, whereas increased the content of 1-pentanol (floral, fruity) and 1-octen-3-ol (mushroom), compared with the un-supplemented ones. 1-octanol (citrus, fatty) and 1-decanol (sweet, fruity, floral) were specifically detected in the GRC with matcha. Matcha addition to the rice cakes also increased levels of (Z, Z)-9,12-octadecadienoic acid-methyl ester, hexadecanoic acid-methyl ester, and (Z)-9-octadecenoic acid-methyl ester. Compared with the other two types of rice cakes, GRC with matcha produced four more esters, including oxalic acid-isobutyl hexadecyl ester, propanoic acid-nonyl ester, 2-thiopheneacetic acid-3-tridecyl ester and 11-octadecenoic acid-methyl ester. Regarding ketones, matcha addition enhanced the contents of 2-heptanone (fruity, cheesey, ripe), and 6-methyl-5-hepten-2-one (pepper, rubber), whereas reduced the abundance of 2-octanone (woody, yeast), in all three types of rice cakes.

Table 2.

Volatile components of rice cake with or without matcha.

Volatile compound (%)	LRI	Cc	JRC	JRC-M	IRC	IRC-M	GRC	GRC-M
Pentanal	699	Ald	3.12 ± 0.5	7.32 ± 1.14 *	2.07 ± 0.12	9.22 ± 0.64 **	2.36 ± 0.76	3.73 ± 0.51
Hexanal	798	Ald	57.87 ± 0.81	39.35 ± 9.17 *	52.34 ± 3.91	46.45 ± 2.41 *	74.94 ± 0.82	48.68 ± 7.37 **
Heptanal	899	Ald	1.84 ± 0.15	1.94 ± 0.53	2.07 ± 0.09	2.90 ± 0.33	1.07 ± 0.09	1.09 ± 0.11
Nonanal	1105	Ald	11.12 ± 0.53	9.67 ± 0.09 *	16.47 ± 1.6	8.56 ± 0.54 **	4.26 ± 0.31	2.78 ± 0.34 *
2-Nonenal, (E)-	1165	Ald	1.33 ± 0.09	1.79 ± 1.26	1.55 ± 0.17	n.d.	0.58 ± 0.13	0.42 ± 0.1
Decanal	1204	Ald	1.59 ± 0.01	2.71 ± 1.08 **	2.19 ± 0.37	2.46 ± 0.17	0.71 ± 0.02	0.76 ± 0.17
2-Heptenal, (Z)-	940	Ald	n.d.	2.13 ± 0.18	n.d.	1.12 ± 0.19	n.d.	0.93 ± 0.07
Benzaldehyde	950	Ald	n.d.	1.38 ± 0.31	n.d.	1.34 ± 0.1	n.d.	1.15 ± 0.13
1-Pentanol	763	A	0.97 ± 0.15	1.43 ± 0.36	0.89 ± 0.04	1.8 ± 0.06 **	0.73 ± 0.24	1.49 ± 0.24 *
1-Hexanol	863	A	1.76 ± 0.06	1.14 ± 0.38 *	n.d.	n.d.	0.74 ± 0.05	0.68 ± 0.11
1-Octen-3-ol	975	A	0.88 ± 0.07	1.99 ± 0.73	1.15 ± 0.10	1.98 ± 0.23 *	1.38 ± 0.05	1.86 ± 0.18 *
1-Octanol	1071	A	n.d.	n.d.	n.d.	n.d.	0.32 ± 0	0.42 ± 0.06
2-ethyl-1-Decanol	1783	A	n.d.	n.d.	n.d.	n.d.	n.d.	7.13 ± 1.76
9,12-Octadecadienoic acid (Z,Z)-, methyl ester	2112	E	0.89 ± 0.04	1.04 ± 0.08	n.d.	1.33 ± 0.27	0.35 ± 0.01	0.44 ± 0.09
Hexadecanoic acid, methyl ester	1977	E	1.18 ± 0.02	1.35 ± 0.61	0.79 ± 0.03	0.83 ± 0.15	0.39 ± 0.07	0.56 ± 0.05 *
9-Octadecenoic acid (Z)-, methyl ester	2118	E	0.92 ± 0.03	1.68 ± 0.49	0.86 ± 0.05	1.25 ± 0.07 **	0.45 ± 0.05	0.59 ± 0.04 *
Oxalic acid, isobutyl hexadecyl ester	1815	E	n.d.	n.d.	n.d.	n.d.	n.d.	1.21 ± 0.6
Propanoic acid, nonyl ester	1844	E	n.d.	n.d.	n.d.	n.d.	n.d.	1.4 ± 1.16
2-Thiopheneacetic acid, 3-tridecyl ester	1871	E	n.d.	n.d.	n.d.	n.d.	n.d.	1.23 ± 0.13
11-Octadecenoic acid, methyl ester	2117	E	n.d.	n.d.	n.d.	n.d.	n.d.	0.61 ± 0.12
2-Heptanone	889	K	1.17 ± 0.08	1.36 ± 0.32	1.02 ± 0.19	1.32 ± 0.1	1.61 ± 0.15	1.61 ± 0.29
6-methyl-5-Hepten-2-one,	980	K	1.05 ± 0.05	1.64 ± 0.39 *	0.83 ± 0.07	1.6 ± 0.15 **	0.41 ± 0.02	0.77 ± 0.06 **
2-Octanone	1003	K	12.13 ± 2.35	11.83 ± 3.12	13.43 ± 0.19	12.01 ± 0.33	6.67 ± 0.82	4.38 ± 0.42 *
Tridecane	1300	Alk	1.05 ± 0.01	1.22 ± 0.21	0.94 ± 0.04	n.d.	0.45 ± 0.01	n.d.
Tetradecane	1400	Alk	0.93 ± 0.08	1.54 ± 0.53	0.75 ± 0.06	n.d.	0.92 ± 0.13	0.52 ± 0.13 *
Octadecane	1800	Alk	0.99 ± 0.04	0.99 ± 0.25	0.95 ± 0.06	n.d.	n.d.	0.41 ± 0.02
Dodecane	1200	Alk	n.d.	1.14 ± 0.22	1.01 ± 0.13	n.d.	0.38 ± 0.03	n.d.
Pentadecane	1500	Alk	0.89 ± 0.03	1.71 ± 0.13 **	1.22 ± 0.15	n.d.	n.d.	n.d.

Results are presented as mean ± standard deviations. LRI, retention index, which was calculated refereed to the retention time of C7-C40 n-alkanes under the same conditions. Cc, chemical classes of volatile compounds; Ald, Aldehydes; A, Alcohols; E, Esters; K, Ketones; Alk, Alkanes; n.d., not detected; JRC, Japonica rice cake; JRC-M, Japonica rice cake with 2% matcha; IRC, Indica rice cake; IRC-M, Indica rice with 2% matcha; GRC, glutinous rice cake; GRC-M, glutinous rice cake with 2% matcha. Value with * (p < 0.05) or ** (p < 0.01) indicate significantly different with its corresponding one without matcha.

3.4. In vitro oxidant defense capacity of matcha rice cake

The total polyphenol content (TPC) of rice cakes with different ratios of matcha was shown in Fig. 2-A. As anticipated, incorporating matcha into rice cakes significantly improved its TPC and in vitro antioxidant activity (Fig. 2-B). Notably, though JRC fortified with matcha didn’t show a higher TPC level, it displayed higher DPPH or ABTS scavenging capacity, compared with the other two types of rice cakes. Addition of 1% matcha to those three types of rice cakes resulted in a six to seven folds increase of the DPPH scavenging activity (p < 0.01). Moreover, rice cakes with matcha also exhibited an enhanced ABTS scavenging ability, remarkedly higher than the ones without matcha (p < 0.01). The DPPH and ABTS scavenge rate of matcha alone was also tested (Fig. S1), and antioxidant capacity of matcha increased in a dose dependent manner. However, after adding the same amout matcha to the rice cakes, the antioxidant capacity of matcha decreased, suggesting that the interaction with rice starch could partly affect the oxidant defense capacity of matcha.

Fig. 2.

Effects of matcha on the antioxidant capacity of rice cakes. (A) Total polyphenol contents of matcha rice cakes. (B) Matcha increased the DPPH and ABTS scavenging rate of rice cakes. *p < 0.05, **p < 0.01 vs. control. JRC, Japonica rice cake; IRC, Indica rice cake; GRC, glutinous rice cake.

3.5. In vitro starch digestibility analysis of match rice cake

In the present study, the effects of matcha on starch digestion of rice cakes were investigated by in vitro enzymatic hydrolysis tests. Matcha markedly reduced the starch digestibility of all three types of rice cake, as shown in Fig. S2. In general, the starch digestibility of IRC is highest, followed by the JRC. At 120 min, compared with the one without matcha (33.85 ± 1.70%), 1%, 2% and 3% matcha decreased the starch digestibility of JRC to 18.67 ± 2.45%, 8.69 ± 2.23% and 5.48 ± 1.13%, respectively. GRC is with lowest starch digestibility, and its total digestibility is below 23%.

As expected, matcha reduced the ratio of RDS and SDS, while effectively increased the content of RS in all three types of rice cakes (Table 3). 3% matcha enhanced the content of RS to around 95% in the JRC and IRC, and 90% in the GRC. Consistent with the starch hydrolysis profile, matcha significantly decreased the glucose release rate (K) values and the expected glycemic index (eGI) for all three types of rice cakes.

Table 3.

Ratios of starch fractions and kinetics parameters of rice cake with or without matcha in vitro digestion.

Sample	matcha ratio	RDS (%)	SDS (%)	RS (%)	K × 10−2 min−1	R2	HI (%)	eGI
JRC	0%	13.65 ± 1.66	20.19 ± 0.96	66.15 ± 0.96	2.34 ± 0.08	0.998	41.59	62.54
	1%	4.43 ± 1.01*	14.24 ± 1.46	81.33 ± 1.46**	1.17 ± 0.10**	0.996	21.1	51.29
	2%	2.28 ± 1.12**	6.41 ± 0.89**	91.31 ± 0.89**	1.12 ± 0.14**	0.990	9.32	44.83
	3%	0.94 ± 0.70**	4.55 ± 1.84**	94.51 ± 1.84**	0.80 ± 0.17**	0.984	6.39	43.22
IRC	0%	14.07 ± 1.13	28.86 ± 2.33	57.08 ± 2.33	1.94 ± 0.18	0.990	51.79	68.14
	1%	4.24 ± 0.98**	17.93 ± 0.26*	74.83 ± 0.26**	0.69 ± 0.11**	0.994	25.9	53.93
	2%	1.66 ± 0.46**	8.77 ± 0.76**	89.58 ± 0.76**	0.66 ± 0.13**	0.991	11.62	46.09
	3%	1.14 ± 0.06**	4.13 ± 0.66**	94.73 ± 0.66**	0.28 ± 0.15**	0.986	5.57	42.77
GRC	0%	13.10 ± 1.16	10.29 ± 2.31	76.61 ± 2.31	3.82 ± 0.15	0.997	29.9	56.13
	1%	3.53 ± 0.02**	8.93 ± 0.22	87.54 ± 0.23*	1.60 ± 0.07**	0.998	14.41	47.62
	2%	2.63 ± 0.74**	9.23 ± 1.61	88.13 ± 1.61*	1.45 ± 0.17**	0.989	13.25	46.98
	3%	2.09 ± 0.29**	7.49 ± 1.40	90.43 ± 1.43*	1.31 ± 0.15**	0.990	10.63	45.55

Results are presented as mean ± standard deviations. JRC, Japonica rice cake; IRC, Indica rice cake; GRC, glutinous rice cake; RDS, ready digestible starch; SDS, slowly digestible starch; RS, resistant starch; K, kinetic constant of starch hydrolysis; HI, value of hydrolysis index; eGI, expected glycemic index. Values with * (p < 0.05) and **(p < 0.01) are significantly different vs. the corresponding rice cake without matcha.

3.6. Optimization of response surface methodology for matcha rice cake

Response surface methodology (RSM) was utilized to show the effect of all three factors (matcha ratio, water ratio and steaming time) on the rice cake quality. A quadratic polynomial regression model was established for predicting the Y response (sensory score) based on the three independent variables after a total of 17 experiments were performed (Table S2).

According to the sensory score of groups, the model proposed for the response of Y fitted equation is generated as follows: Y = 7.04 + 0.22A-0.45B + 0.027C + 0.20AB-0.25AC + 0.15BC-0.70A²-0.54B²-0.45C². The ANOVA analysis of the quadratic regression model are described in Table 4. Notably, the model created from the responses was shown to be significant with p = 0.0005, and there was no significance lack of fit (p = 0.1163 > 0.05) in all the responses. The satisfactory coefficient R² = 0.9586, and the R_adj² = 0.9053, suggesting that the model generated is a good fit to predict responses correctly. Besides, the results indicated that matcha amount had the most significant effect on the sensory score (F = 36.38, p = 0.0005), followed by the water ratio (F = 8.96, p = 0.0201), while the steaming time didn't exhibit a significant influence on the final quality.

Table 4.

ANOVA analysis of the response surface quadratic model.

source	sum of squares	df	mean square	F-value	p-value	significance
model	7.10	9	0.79	18.00	0.0005	**
A-water ratio	0.39	1	0.39	8.96	0.0201	*
B-matcha ratio	1.60	1	1.6	36.38	0.0005	**
C-steaming time	0.00595	1	0.00595	0.14	0.7235
AB	0.16	1	0.16	3.73	0.0947
AC	0.26	1	0.26	5.91	0.0454	*
BC	0.09	1	0.09	2.05	0.1951
A2	2.06	1	2.06	47.01	0.0002	**
B2	1.21	1	1.21	27.59	0.0012	**
C2	0.86	1	0.86	19.62	0.0030	**
residual	0.31	7	0.044
lack of fit	0.23	3	0.076	3.77	0.1163
pure error	0.08	4	0.020
Cor total	7.41	16

p-value less than 0.05 and 0.01 indicate that the parameter is significant and very significant to the model, and marked as * and **, respectively.

Response surface plots of the effects of the variables on the sensory score was displayed in Fig. S3 to aid visualization. To predict the most welcome mixture with a high sensory score, numerical optimal criteria of each variable was set individually. As a result, the most appropriate formulation was predicted as follows: water ratio 82.23%, matcha amount 1.60% and steaming time for 38.63 min. Under such conditions, the sensory score is predicted to be 7.14. Considering about the practical possibility, the optimized technological conditions have been adjusted to water ratio 82%, matcha amount 1.60% and steaming time of 39 min. Verification of the optimized formula for producing matcha rice cake was performed. The obtained rice cake was with moderate hardness, bright green color and unique tea flavor.

4. Discussion

In this study, the feasibility of adding matcha powder to rice cakes made from three different types of rice was assessed. Steamed rice cakes, as a starchy food, easily become rigid and have poor water retention during storage. This process is known as starch retrogradation, and it occurs along with the gelatinized starch recrystallizes through hydrogen bonds (Wu et al., 2014). In this study, matcha addition decreased the relative crystallinity of the three types of rice cakes. According to the FTIR results, rice cakes supplemented with matcha had a lower peak intensity at 3400/cm, indicating that matcha could interact with the hydrogen-bond in starch. During gelatinization, hydrogen bonds are formed between matcha polyphenols and starch, resulting in a decrease in free starch content and delay in starch retrogradation (Xiao et al., 2013). According to the SEM images, irregular erosions on the sample surface were detected, which were primarily generated by ice crystals during sample freezing. Matcha disrupted the reassociation of starch chains, causing more small voids to appear in the rice cakes with matcha, especially in the GRC, an amylopectin-rich starch. According to other studies, the effect of tea on the crystal structure of starch may also hinder the interaction between starch and digestive enzymes, thus reducing starch digestibility (Li et al., 2021, Fu et al., 2021).

Matcha enhanced the flavor of the rice cakes with its vegetal and malty aroma (Tan et al., 2019). Aldehydes are mostly detected in all three types of rice cakes, and contributes greatly to the fruit aroma and sweet and grassy flavor of cooked rice, regardless of rice cultivar (Setyaningsih et al., 2019). Matcha supplementation increased the level of pentanal and decanal, two prominent compounds always found in green tea when compared to those without. Almond flavored 2-heptenal (Z) and benzaldehyde were only generated in matcha supplemented rice cakes, compared with the one without matcha. It is reported that matcha contains the highest level of benzaldehyde, compared to Sencha, Gyokuro, and Hojicha (Tan et al., 2019). Benzaldehyde was always liberated from prunasin (a cyanoglycoside) hydrolysis with β-glucosidase, an endogenous enzyme that is only released when plant tissues are ruptured against damages from herbivores and pathogens (Guo et al., 1998, Wang et al., 2021). Hexanal and nonanal are the two low boiling aroma-active compounds with a fatty, grassy and citrus-like aroma, that are always formed during lipid peroxidation (Buttery et al., 1999, Kaarina Viljanen et al., 2011, Bryant, 2010). Though hexanal and nonanal are with relatively high level in matcha, after adding matcha to rice cakes, the level of these two compounds decreased instead. Among alcohols, matcha addition increased the level of 1-pentanol and 1-octen-3-ol, either with a green, floral and fruity green tea aroma or a mushroom note (Ni et al., 2017, Ming et al., 2015). 1-octanol and 1-decanol, which are mainly found in baked green tea (Xing et al., 2006), were only detected in GRC with matcha. Similarly, when compared to the other two types of rice cakes, GRC with matcha contained more volatile esters, indicating a more fruity and floral aroma (Yu et al., 2012). In terms of ketones, matcha promoted the generation of 2-heptanone, 5-hepten-2-one and 3-buten-2-one, all of which have a fruity flavor and may contribute to the enjoyment of green tea by consumers (Kim et al., 2016). Matcha addition, on the other hand, reduced the level of 2-octanone, a unique aroma compound that is always formed during rice fermentation (Yu et al., 2019).

Matcha inclusion greatly enhanced the anti-oxidant capacity of rice cakes. In consistent with our study, fresh instant rice mixed with matcha powder displayed higher DPPH and ABTS scavenging capacity and had a better effect than black tea (Fu et al., 2021). As indicated in other studies, ingredients enriched with polyphenols could also effectively enhance the in vitro oxidant defense capacity of different types of foods (Blanco Canalis et al., 2020). However, the antioxidant capacity of matcha decreased after adding it to rice cakes. JRC, mainly made up of amylose, displayed higher antioxidant activity than IRC and GRC after supplementing with matcha, indicating that structure of different types of rice starch could affect the antioxidant capacity of matcha to varying degrees.

As a starchy food, ingestion of rice cakes easily increases the postprandial blood glucose level. Polyphenols have been reported to inhibit starch digestion, effectively controlling the GI (Sun and Miao, 2020). Subjects who consumed polyphenol-and fiber-rich dried fruits with green tea had lower postprandial blood glucose level due to inhibition of α-amylase and α-glucosidase (Nyambe-Silavwe and Williamson, 2016). This enzyme inhibition could be explained by the interaction of the hydroxyl group of tea polyphenols with the catalytic residues of digestive enzymes (Sun et al., 2016). In this study, regardless of the type of rice cake, matcha significantly slowed starch digestion. Retrograded amylopectin-rich starch is believed to have a higher ordered physical structure and is always more resistant to enzyme hydrolysis (Zhu and Liu, 2020). As a result, when compared to the other two types of rice cakes, GRC with or without matcha has the lowest starch digestibility, followed by the JRC. Matcha also significantly increased the content of resistant starch (RS), which always resists digestion in the upper gastrointestine and is fermented in the colon to produce metabolites such as short chain fatty acids, potentially acting as a prebiotic (Zaman and Sarbini, 2016).

5. Conclusions

The addition of matcha powder to steamed rice cakes, no matter which types of rice used, interfered with the reassembly of starch chains and promoted the formation of polyporous structures. The volatile profiles were markedly enhanced by matcha powder, with 2-heptenal (Z) and benzaldehyde generated in all types of rice cakes. Matcha increased the TPC level along with significantly enhancing the in vitro oxidant defense potential of rice cake compared to the non-supplemented ones. Matcha strongly inhibited the in vitro starch digestibility for all types of rice cakes by increasing the ratio of RS, and reducing the HI and eGI. The sensory evaluation results showed that rice cake enriched with matcha is acceptable, as the consumer panel indicated that rice cake processes with 1.6% matcha, 82% water and steams for 39 min will be with a high quality. Consequently, the addition of matcha as an ingredient to rice cake, or other foods made with rice, would enrich the flavor, provide oxidant defense and potentially serve to help within blood glucose regulation. Future studies in humans are warranted.

Funding

This study was supported by the funds from the Agriculture and Social Development project, Hangzhou (2020ZDSJ0460), Zhejiang Education Department (Y202044943), and Zhejiang Agriculture and Forestry University (2020FR049).

CRediT authorship contribution statement

Ran Wei: Conceptualization, Methodology, Investigation, Writing – original draft. Lisheng Qian: Software, Formal analysis, Writing – original draft. Kayama Kayama: Writing – review & editing, Visualization. Fenghua Wu: Data curation. Zhucheng Su: Resources, Supervision. Xingquan Liu: Project administration.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

No data was used for the research described in the article.