Abstract
This study aimed to distinguish between Korean and Chinese red pepper powder (RPP) using inorganic elemental analysis data combined with orthogonal partial least squares-discriminant analysis (OPLS-DA). Elemental concentrations were obtained for 31 Korean and 31 Chinese RPP samples that were collected in Korea. Energy dispersive X-ray fluorescence spectroscopy detected 11 elements in these samples. Rb and Cl concentrations were selected as the variables which best allowed distinguishing between Korean and Chinese RPP using an S-plot from OPLS-DA. Rb and Cl concentrations in the Korean RPP samples were ≤ 1.6 mg/100 g (measured by inductively coupled plasma-optical emission spectroscopy) and ≤ 215 mg/100 g, respectively. A blind trial demonstrated that Korean RPP containing ≥ 50 g/100 g of Chinese RPP could be identified by applying predetermined ranges of Rb and Cl concentrations, suggesting that analysis of these two elements is a possible approach to distinguish between Korean and Chinese RPP.
Keywords: Red pepper powder, Geographic origin, Cl, Rb, Orthogonal partial least squares-discriminant analysis
Introduction
Red pepper (Capsicum annuum L.) belongs to the family, Solanaceae. It is an annual herbaceous plant. Red pepper is widely cultivated worldwide, including in Asia, northern America, southern and central Europe, and tropical and subtropical Africa (Thampi, 2003). In Korea, red pepper is typically used in a powdered form. Red pepper powder (RPP) is used as the most crucial and indispensable ingredient in Korean foods and dishes. In Korea, RPP is conventionally prepared from raw materials by grinding after drying (Kwon et al., 1990). However, imported RPP distributed in Korea is mostly manufactured from frozen red peppers imported at lower tariffs compared to the raw and dried products (Choi et al., 2020). The annual domestic production level of dried red peppers in Korea was approximately 80,000 tons during the last 5 years (2014–2018) (Korean Statistical Information Service, 2020). During the same period, approximately 203,000 tons of frozen red peppers were imported annually. The highest quantity (~ 94%) of frozen red peppers was imported from China (Korea Ministry of Food and Drug Safety, 2020).
An issue that has become of public concern worldwide is that of falsifying statements regarding the geographic origin of food products to gain unfair profits (Hong et al., 2017). False statements regarding the geographic origins of RPP have been a cause for concern in Korea. Chinese RPP and Korean RPP blended with Chinese RPP, deliberately labeled as “cultivated in Korea,” frequently appear in Korean markets because the retail price of Korean RPP is seven times as high as that of Chinese RPP (Lee et al., 2020). Therefore, there is a need to develop analytical procedures to accurately determine the geographic origin of RPP distributed in Korea.
The geographic origin of food products can be determined by analyzing the inorganic element composition in foods because element compositions are influenced by geographic factors (e.g., soil type) (Kelly et al., 2005) and agronomic conditions (e.g., fertilizer type) (Molina et al., 2009). Therefore, elemental analyses have been applied to confirm the geographic origins of several kinds of foods, including oil seeds (e.g., sesame seeds) (Choi et al., 2016; Choi et al., 2017), grains (e.g., rice, soybean, and conventional wheat, barley, faba beans, and potatoes) (Kelly et al., 2002; Laursen et al., 2011; Otaka et al., 2014), vegetables (e.g., garlic) (Camargo et al., 2010), dairy (e.g., milk and cheese) (Sacco et al., 2009; Suhaj and Koreňovská, 2008), meats (e.g., beef and chicken) (Franke et al., 2008a; 2008b), and alcoholic beverages (e.g., wine) (Galgano et al., 2008; Greenough et al., 1997). These published studies have employed X-ray fluorescence spectroscopy (XRF) (Choi et al., 2016; Otaka et al., 2014), inductively coupled plasma-optical emission spectroscopy (ICP-OES) (Laursen et al., 2011; Sacco et al., 2009), ICP mass spectrometry (Choi et al., 2017; Franke et al., 2008a; 2008b; Galgano et al., 2008; Greenough et al., 1997; Kelly et al., 2002; Laursen et al., 2011), atomic absorption spectroscopy (Galgano et al., 2008; Suhaj and Koreňovská, 2008), or neutron activation analysis (Camargo et al., 2010) to perform elemental analyses followed by an appropriate multivariate statistical method. The statistical methods include discriminant analysis (DA) (Choi et al., 2016; Franke et al., 2008a; 2008b; Galgano et al., 2008; Kelly et al., 2002; Otaka et al., 2014; Sacco et al., 2009; Suhaj and Koreňovská, 2008), principal component analysis (Camargo et al., 2010; Choi et al., 2017; Laursen et al., 2011; Otaka et al., 2014; Sacco et al., 2009; Suhaj and Koreňovská 2008), or cluster analysis (Galgano et al., 2008; Greenough et al., 1997). However, to the best of our knowledge, there have been no published studies that determined the geographic origin of RPP using elemental analyses. It should also be noted that in the previously published studies, instrumental analysis of a halogen element, that is, Cl, was preferentially conducted using energy dispersive XRF (EDXRF) analysis (Choi et al., 2016; Otaka et al., 2014).
EDXRF analysis is a simple, rapid, inexpensive, and non-destructive technique compared to ICP-based measurements (Nečemer et al., 2003). It enables the simultaneous measurement of multiple inorganic elements (ranging from Na to U) in a sample. Furthermore, EDXRF is more suited to analyzing halogen elements including Cl, than conventional ICP-OES instruments, because the latter have limited wavelength ranges (165–800 nm), and are unable to detect the halogen elements (Kunselman et al., 2006). However, EDXRF is not suited to detecting elements present in trace quantities in a sample because it has a relatively lower sensitivity compared with ICP-based techniques.
The aim of this study was to distinguish between Korean and Chinese RPP currently distributed in Korea by analyzing their inorganic element compositions, followed by DA. The element concentration data were obtained using EDXRF and ICP-OES. The variables that could best allow differentiation between the Korean and Chinese RPP samples were selected from the EDXRF data, using orthogonal partial least squares-DA (OPLS-DA), a supervised multivariate statistical method. Among the selected variables, the accurate range of a variable contained in lower concentrations was established, based on the ICP-OES analysis. A blind trial was then conducted to evaluate the predictive discrimination power of the selected variables.
Materials and methods
Samples
Samples of Korean RPP (n = 31) were collected from 11 sites in Korea during the 2017 and 2018 harvests. Samples of Chinese RPP (n = 31) imported between 2015 and 2018 were purchased from local grocery stores in Korea. The Korean and Chinese RPP samples described above were used to determine the best variables for discerning the geographic origins of RPP using the OPLS-DA technique. Blind samples (n = 20) were provided by the National Institute of Food and Drug Safety Evaluation of the Ministry of Food and Drug Safety, Korea (http://www.nifds.go.kr). The blind samples consisted of three Korean RPP samples, three Chinese RPP samples, and 14 samples of Korean RPP to which Chinese RPP was added at levels of 10, 20, 30, 50, 70, and 90 g/100 g, respectively. These blind samples were used to determine whether their geographic origins could be correctly identified by applying the range of variables found in the Korean RPP samples.
EDXRF analysis
EDXRF analysis was performed using an EDX-7000 instrument (Shimadzu, Kyoto, Japan) fitted with a 50 W Rh X-ray tube and a silicon drift detector with a resolution of 125 eV at Mn Kα 5.90 keV. The RPP samples were finely ground to powder samples using an EV-GB 5000 blender (Everhome, Seoul, Korea). The powder (~ 1.5 g) was placed in a sample cup (SPEX, Metuchen, NJ, USA) wrapped with a 5-μm thick polypropylene film (SPEX). The sample cup was transferred to a 12-sample turret positioned inside the irradiation chamber. The measurements were conducted in a helium atmosphere. The monitored spectral region ranged from 1 to 40 keV. Measurements of Mg Kα (1.25 keV), P Kα (2.01 keV), S Kα (2.31 keV), Cl Kα (2.62 keV), K Kα (3.31 keV), Ca Kα (3.69 keV), Mn Kα (5.90 keV), Fe Kα (6.40 keV), Cu Kα (8.05 keV), Zn Kα (8.64 keV), and Rb Kα (13.40 keV) characteristic X-rays were performed in triplicate. The EDXRF was controlled using the PCEDX software (version 2.01, Shimadzu). A certified reference material (CRM) of powdered infant formula [SRM1849a, National Institute of Standards and Technology (NIST), Gaithersburg, MA, USA] with certified values (per 100 g) of Cl (701.0 mg), Ca (525.3 mg), Mn (4.96 mg), Fe (17.56 mg), Zn (15.1 mg), Mg (164.8 mg), P (399.0 mg), K (922.0 mg), and Cu (1.98 mg) was analyzed for 14 measurements spread over 11 months to evaluate the accuracy and precision of the EDXRF measurements. The mean recovery of Cl [94.9%; measured value (per 100 g) = 666 ± 12 mg], Ca (102%; 534 ± 12 mg), Mn (95.2%; 4.7 ± 0.2 mg), Fe (96.2%; 16.9 ± 0.8 mg), and Zn (93.5%; 14.1 ± 0.5 mg) was within the Association of Official Analytical Collaboration (AOAC) acceptable ranges of 92–105% at concentrations of 1000 mg/100 g, 90–108% at 100 mg/100 g, and 85–110% at 10 mg/100 g. Whereas, the mean recoveries of Mg [62.0%; measured value (per 100 g) = 102 ± 30 mg], P (66.0%; 263 ± 7 mg), K (107%; 985 ± 20 mg), and Cu (225%; 4.5 ± 0.2 mg) were outside these ranges (AOAC 2002).
Rb analysis by ICP-OES
ICP-OES analysis of the Rb concentration was performed using an ICP-OES instrument (Varian, Palo Alto, CA, USA). RPP (~ 0.25 g) was weighed directly into Teflon vessels after the addition of concentrated nitric acid (10 mL). Samples were then subjected to MARS 6 microwave heating (CEM, Matthews, NC, USA) with a maximum power of 1800 W (170 °C). The microwave temperature was increased from room temperature to 100 °C in 10 min and maintained at this temperature for 10 min prior to carbonization. The temperature was then increased to 180 °C in 10 min, held at 180 °C for 10 min, and cooled to room temperature in 15 min. After cooling, the digests were passed through Whatman filter paper no. 40. The volumes of the filtrates were adjusted to 25 mL in volumetric flasks with deionized water. Measurement of Rb was performed in triplicate. The operating conditions for ICP-OES were as follows: power, 1.5 kW; plasma gas flow, 15 L/min, auxiliary gas flow, 1.5 L/min, nebulizer gas flow, 0.7 L/min, pump rate, 15 rpm; and stabilization delay, 30 s. The best analytical wavelength for Rb was fixed as 780.026 nm. The CRM of powdered pine needle (SRM1575a, NIST) with a certified value (per 100 g) of 1.650 mg was analyzed thrice to evaluate the accuracy and precision of the ICP-OES measurements. The mean recovery of 103% [measured value (per 100 g) = 1.69 ± 0.08 mg] was within the AOAC acceptable range of 80–115% at the 1 or 10 mg/100 g level (AOAC, 2002).
Statistical analysis
The two-tailed Student’s t-test was used to compare the elemental concentrations of the Korean and Chinese RPP samples (p < 0.05, p < 0.01, or p < 0.001). All statistical analyses were performed using IBM SPSS (version 23) software (SPSS Inc., Chicago, IL, USA). Using inorganic element composition data, OPLS-DA was performed using SIMCA-P + (version 15) software (Umetrics, Umea, Sweden). UV scaling was applied to the inorganic element composition datasets prior to OPLS-DA to reduce the masking effect from the elements occurring in higher concentrations. The OPLS-DA model was validated using a cross-validated analysis of variance (ANOVA). The S-plot generated by OPLS-DA was used to select significant variables that could best distinguish between Korean and Chinese RPP samples.
Results and discussion
Inorganic element composition analyzed by EDXRF
Table 1 summarizes the inorganic element compositions of 31 Korean RPP samples and 31 Chinese RPP samples obtained from EDXRF analyses. Also included are their production site and production year information. EDXRF analysis detected 11 different elements, namely Mg, P, S, Cl, K, Ca, Mn, Fe, Cu, Zn, and Rb, in RPP samples at levels from below 1 mg/100 g to over 3000 mg/100 g depending on the specific element. Seven elements were quantified, except for Mg, P, K, and Cu, which did not show acceptable recovery in the CRM analysis. The concentrations per 100 g of the elements observed in the Korean RPP samples were compared to those of the Chinese RPP samples. The Mn (mean = 1.9 mg, range = 0.3–4.3 mg) and Zn concentrations (mean = 2.5 mg, range = 1.7–3.1 mg) in Korean RPP were significantly (p < 0.01) higher than those in Chinese RPP (Mn concentration, mean = 1.5 mg, range = 0.7–2.4 mg; Zn concentration, mean = 2.1 mg, range = 1.4–4.5 mg), whereas the Fe concentration in the Korean RPP (mean = 5.7 mg, range = 4.3–17.8 mg) were significantly (p < 0.001) lower than those in the Chinese RPP (mean = 12.4 mg, range = 5.5–51.4 mg). The S, Cl, and Rb levels were also significantly (p < 0.001) lower in the Korean RPP (S concentration, mean = 208 mg, range = 190–239 mg; Cl concentration, mean = 154 mg, range = 114–215 mg; Rb concentration, mean = 0.0 mg, range = 0.0–0.0 mg) than those in the Chinese RPP (S concentration, mean = 226 mg, range = 191–279 mg; Cl concentration, mean = 299 mg, range = 177–405 mg; Rb concentration, mean = 1.1 mg, range = 0.0–1.7 mg). However, no significant (p > 0.05) difference was found in the Ca concentration between the Korean and Chinese RPP.
Table 1.
Measured concentration (mg/100 g) of inorganic elements in Korean and Chinese red pepper powder (RPP) samples
| Sample no | Production site | Production year | Sampling period | Analyzed by EDXRF | Analyzed by ICP-OES | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| S | Cl | Ca | Mn | Fe | Zn | Rb | Rb | ||||
| Korean RPP samples | |||||||||||
| 1 | Yeongwol | 2018 | 02/2019 | 208 | 129 | 84.8 | 3.5 | 5.9 | 2.7 | n.d.a | 1.1 |
| 2 | Yeongwol | 2018 | 02/2019 | 212 | 133 | 89.0 | 1.8 | 5.4 | 2.9 | n.d | 1.0 |
| 3 | Yeongwol | 2018 | 02/2019 | 211 | 149 | 79.9 | 3.3 | 5.0 | 2.3 | n.d | 1.1 |
| 4 | Yeongwol | 2018 | 02/2019 | 195 | 131 | 79.7 | 1.6 | 4.5 | 2.6 | n.d | 1.1 |
| 5 | Yeongwol | 2018 | 02/2019 | 207 | 135 | 79.2 | 1.9 | 5.1 | 2.7 | n.d | 1.0 |
| 6 | Yeongwol | 2018 | 02/2019 | 196 | 127 | 82.4 | 1.7 | 4.6 | 2.6 | n.d | 1.0 |
| 7 | Yeongwol | 2018 | 02/2019 | 194 | 127 | 67.9 | 1.6 | 4.8 | 2.5 | n.d | 1.2 |
| 8 | Yeongwol | 2018 | 02/2019 | 198 | 131 | 70.0 | 1.6 | 4.8 | 2.1 | n.d | 0.8 |
| 9 | Yeongwol | 2018 | 02/2019 | 211 | 122 | 76.5 | 3.6 | 4.5 | 2.3 | n.d | 1.1 |
| 10 | Yeongwol | 2018 | 02/2019 | 190 | 137 | 86.1 | 1.2 | 4.7 | 1.8 | n.d | 1.0 |
| 11 | Yeongwol | 2018 | 02/2019 | 217 | 179 | 104 | 2.1 | 6.9 | 2.5 | n.d | 1.0 |
| 12 | Chuncheon | 2018 | 02/2019 | 201 | 162 | 93.7 | 2.0 | 4.3 | 2.6 | n.d | 0.7 |
| 13 | Chuncheon | 2018 | 02/2019 | 202 | 148 | 97.9 | 4.3 | 5.7 | 2.5 | n.d | 1.6 |
| 14 | Chuncheon | 2018 | 02/2019 | 217 | 184 | 98.2 | 1.9 | 6.0 | 2.4 | n.d | 0.9 |
| 15 | Chuncheon | 2018 | 02/2019 | 203 | 173 | 95.8 | 2.4 | 4.5 | 2.7 | n.d | 0.7 |
| 16 | Chuncheon | 2018 | 02/2019 | 219 | 181 | 122 | 2.1 | 7.0 | 2.6 | n.d | 1.0 |
| 17 | Chuncheon | 2018 | 02/2019 | 195 | 154 | 90.1 | 1.9 | 6.1 | 2.5 | n.d | 0.9 |
| 18 | Jeongseon | 2018 | 02/2019 | 198 | 117 | 77.5 | 1.8 | 4.8 | 2.5 | n.d | 0.8 |
| 19 | Jeongseon | 2018 | 02/2019 | 211 | 140 | 79.6 | 1.5 | 4.7 | 1.7 | n.d | 0.8 |
| 20 | Yeongyang | 2017 | 01/2018 | 214 | 161 | 63.1 | 1.3 | 5.7 | 2.4 | n.d | 0.5 |
| 21 | Yeongyang | 2017 | 01/2018 | 239 | 178 | 78.7 | 0.3 | 5.6 | 3.1 | n.d | 1.0 |
| 22 | Yeongyang | 2017 | 01/2018 | 203 | 199 | 80.4 | 2.0 | 17.8 | 2.4 | n.d | 0.5 |
| 23 | Pyeongtaek | 2017 | 01/2018 | 210 | 151 | 99.3 | 1.4 | 4.5 | 2.6 | n.d | 1.1 |
| 24 | Pyeongtaek | 2017 | 01/2018 | 217 | 215 | 113 | 0.8 | 4.8 | 2.4 | n.d | 1.0 |
| 25 | Andong | 2017 | 01/2018 | 202 | 200 | 79.3 | 2.3 | 6.1 | 2.4 | n.d | 0.5 |
| 26 | Yeongju | 2017 | 01/2018 | 210 | 207 | 84.6 | 2.2 | 6.3 | 2.4 | n.d | 0.5 |
| 27 | Gochang | 2017 | 01/2018 | 209 | 129 | 84.6 | 1.7 | 6.0 | 2.4 | n.d | 1.1 |
| 28 | Jeongeup | 2017 | 01/2018 | 208 | 119 | 70.8 | 1.8 | 6.0 | 2.4 | n.d | 0.3 |
| 29 | Chuncheon | 2017 | 01/2018 | 210 | 141 | 91.2 | 1.5 | 5.3 | 2.7 | n.d | 1.0 |
| 30 | Wonju | 2017 | 01/2018 | 215 | 114 | 88.3 | 1.3 | 4.7 | 2.8 | n.d | 1.3 |
| 31 | Ganggyeong | 2017 | 01/2018 | 226 | 188 | 64.2 | 1.1 | 5.7 | 2.3 | n.d | 0.8 |
| Mean ± SD | 208 ± 10 | 154 ± 29 | 85.5 ± 13.3 | 1.9 ± 0.8 | 5.7 ± 2.4 | 2.5 ± 0.3 | n.a.b | 0.9 ± 0.3 | |||
| Range | 190.3–239.4 | 114.1–214.7 | 63.1–121.6 | 0.3–4.3 | 4.3–17.8 | 1.7–3.1 | n.a | 0.3–1.6 | |||
| Chinese RPP samples | |||||||||||
| 1 | n.a | 2018 | 03/2019 | 198 | 343 | 64.6 | 1.2 | 12.0 | 1.5 | n.d | 0.9 |
| 2 | n.a | 2018 | 03/2019 | 232 | 272 | 85.6 | 1.5 | 7.9 | 1.8 | n.d | 1.1 |
| 3 | n.a | 2018 | 03/2019 | 235 | 247 | 97.1 | 1.9 | 10.3 | 2.3 | n.d | 1.5 |
| 4 | n.a | 2018 | 03/2019 | 224 | 301 | 73.3 | 1.4 | 20.8 | 4.1 | 1.7 | 2.3 |
| 5 | n.a | 2018 | 03/2019 | 238 | 281 | 84.2 | 1.8 | 9.4 | 1.9 | 1.3 | 1.7 |
| 6 | n.a | 2018 | 03/2019 | 215 | 314 | 79.1 | 1.4 | 9.3 | 1.8 | n.d | 1.6 |
| 7 | n.a | 2018 | 03/2019 | 237 | 295 | 68.2 | 1.6 | 7.6 | 2.0 | 1.5 | 2.3 |
| 8 | n.a | 2018 | 03/2019 | 227 | 341 | 70.6 | 1.5 | 10.8 | 2.0 | 1.4 | 2.0 |
| 9 | n.a | 2018 | 03/2019 | 232 | 326 | 70.5 | 1.5 | 13.9 | 1.9 | 1.4 | 2.1 |
| 10 | n.a | 2018 | 03/2019 | 220 | 284 | 64.5 | 1.4 | 6.7 | 1.9 | 1.4 | 2.0 |
| 11 | n.a | 2018 | 03/2019 | 227 | 283 | 75.3 | 1.5 | 9.7 | 1.7 | 1.3 | 2.1 |
| 12 | n.a | 2018 | 03/2019 | 234 | 278 | 68.5 | 1.4 | 9.1 | 1.8 | 1.3 | 1.7 |
| 13 | n.a | 2018 | 03/2019 | 211 | 177 | 77.7 | 1.8 | 10.2 | 1.9 | n.d | 1.2 |
| 14 | n.a | 2018 | 03/2019 | 217 | 283 | 87.3 | 1.6 | 15.6 | 1.7 | 1.3 | 1.9 |
| 15 | n.a | 2017 | 03/2018 | 210 | 321 | 65.5 | 0.7 | 6.5 | 1.7 | 1.4 | 2.3 |
| 16 | n.a | 2017 | n.a | 254 | 293 | 164 | 1.5 | 9.1 | 1.8 | 1.2 | 1.9 |
| 17 | n.a | 2017 | n.a | 279 | 270 | 187 | 2.4 | 22.1 | 2.0 | 1.6 | 2.8 |
| 18 | n.a | 2017 | 03/2018 | 191 | 285 | 49.8 | 1.1 | 6.0 | 1.4 | 1.2 | 1.9 |
| 19 | n.a | 2017 | 06/2018 | 221 | 381 | 76.1 | 1.5 | 21.0 | 3.1 | 1.5 | 2.2 |
| 20 | n.a | 2017 | 06/2018 | 208 | 312 | 64.5 | 1.5 | 7.5 | 1.8 | 1.6 | 2.4 |
| 21 | n.a | 2017 | 06/2018 | 220 | 258 | 66.8 | 1.5 | 10.4 | 1.8 | 1.1 | 1.4 |
| 22 | n.a | 2017 | 06/2018 | 237 | 302 | 71.3 | 1.7 | 9.8 | 1.9 | 1.3 | 1.6 |
| 23 | n.a | 2017 | 06/2018 | 212 | 379 | 75.5 | 1.3 | 19.9 | 3.0 | 1.5 | 2.1 |
| 24 | n.a | 2017 | 06/2018 | 230 | 280 | 70.6 | 1.7 | 8.1 | 2.0 | 1.3 | 1.7 |
| 25 | n.a | 2016 | n.a | 245 | 300 | 126 | 1.7 | 9.7 | 1.9 | 1.3 | 1.8 |
| 26 | n.a | 2015 | n.a | 211 | 298 | 72.8 | 1.3 | 9.1 | 1.7 | 1.1 | 1.5 |
| 27 | n.a | 2015 | n.a | 237 | 327 | 105 | 1.8 | 51.4 | 4.5 | 1.4 | 1.7 |
| 28 | n.a | 2015 | n.a | 231 | 269 | 85.5 | 1.5 | 5.5 | 1.5 | 1.4 | 1.9 |
| 29 | n.a | 2015 | n.a | 214 | 405 | 81.3 | 1.4 | 15.1 | 2.9 | 1.5 | 1.9 |
| 30 | n.a | n.a | n.a | 232 | 246 | 76.4 | 0.9 | 8.9 | 1.9 | 0.6 | 1.4 |
| 31 | n.a | n.a | n.a | 213 | 329 | 73.0 | 0.7 | 11.6 | 2.0 | 1.7 | 2.5 |
| Mean ± SD | 226 ± 17*** | 299 ± 44*** | 83.1 ± 28.4 | 1.5 ± 0.3** | 12.4 ± 8.5*** | 2.1 ± 0.7** | 1.4 ± 0.2*** | 1.9 ± 0.4 | |||
| Range | 191.2–278.9 | 177.1–404.8 | 49.8–186.6 | 0.7–2.4 | 5.5–51.4 | 1.4–4.5 | n.d.–1.7 | 0.9–2.8 | |||
aNot detected
bNot available
**Significantly different from Korean RPP, p < 0.01; ***Significantly different from Korean RPP, p < 0.001
Variables to distinguish between Korean and Chinese RPP
OPLS-DA is a supervised multivariate statistical method that can determine important differences between two comparison groups. We performed OPLS-DA to obtain information on the differences in the inorganic element compositions between the Korean and Chinese RPP. The OPLS-DA model generated from the data set had a cross-validated ANOVA p-value of 1.09 × 10–23. The R2Y and Q2Y values of the model were 0.875 and 0.860, respectively, indicating that the model could provide a very high degree of goodness-of-fit and goodness-of-prediction. All the Chinese RPP samples were clearly distinguished from the Korean RPP samples by this model, except for one sample, as shown in the score plot (Fig. 1). An S-plot was constructed from the established model. In the plot, the X- and Y-axes represent the contribution and confidence of each variable, respectively. Thus, the statistically significant variables that could allow differentiation between the Korean and Chinese RPP samples were identified (Fig. 2). The closer a variable lies to the upper- right or lower-left corner of the graph, the more strongly the variable contributes to the difference between the two groups and the more significant is its contribution. Among the 11 different elements, the Cl and Rb constituents in the upper-right corner were regarded as variables whose concentrations differed markedly between the Korean and Chinese RPP samples. Accordingly, the Rb and Cl concentrations were selected as the variables that could best allow differentiation between the Korean and Chinese RPP.
Fig. 1.
Score plot generated by orthogonal partial least-squares discriminant analysis for the elemental composition data obtained from Korean and Chinese red pepper powder (RPP) samples using energy dispersive X-ray fluorescence spectroscopy. (filled circle): Korean RPP; (white circle): Chinese RPP
Fig. 2.
S-plot generated by orthogonal partial least squares discriminant analysis of inorganic element composition data obtained from Korean and Chinese red pepper powder (RPP) samples. Element composition data were obtained using energy dispersive X-ray fluorescence spectroscopy. I, Korean RPP; II, Chinese RPP
Although the mechanisms underlying the absorption of Rb from soils by vascular plants are not yet well understood yet, several published studies have demonstrated that edaphic conditions (e.g., soil acidity) and plant physiology (e.g., vitality of plants) might affect the absorption of Rb, thereby influencing the Rb concentration in plants (Drobner and Tyler 1998; Tyler 1983). A soil with high acidity and low concentrations of alkali and alkaline earth metals (e.g., Na, K, Mg, and Ca) in the rhizosphere favors the uptake of Rb by plants. One of the major factors that can acidify farmland soil is nitrogen fertilizer application (Hoyt and Hennig, 1982). For example, when nitrogen fertilizer containing ammonium sulfate ((NH4)2SO4) is applied to the soil, nitrogen is taken up by plants in the form of ammonium cations (NH4+), whereas the sulfate anions (SO42−) remaining in the soil are converted to sulfuric acid (H2SO4), thereby causing soil acidification (Malhi et al., 1998). According to a survey (Guo et al., 2010), farmland soil in China has been acidified since the 1980s because of the overuse of large quantities of nitrogen fertilizer. A comparison was made of the annual consumption of nitrogen fertilizers per unit of agricultural area in Korea and China during the last 5 years (2014–2018). The statistical data were obtained from the government agencies of the two countries: the Korean Ministry of Agriculture, Food and Rural Affairs (http://www.mafra.go.kr) and the National Bureau of Statistics of China (http://www.stats.gov.cn). The mean consumption of nitrogen fertilizers in China (341 kg/ha) was approximately 2.3 times greater than that in Korea (147 kg/ha). Thus, the higher Rb concentrations observed in the Chinese RPP were closely related to the higher acidity of farmland soil in China, which might have arisen from the overuse of nitrogen fertilizers in the country. Other published studies have also observed a tendency for greater Rb concentrations in Chinese plant foods such as sesame seeds (Choi et al., 2017) and milkvetch roots (Kwon et al., 2014) compared to those in Korean plant foods. In our recent study, a similar trend for the Rb concentration was found in Korean and Chinese perilla seeds (unpublished results). However, the underlying mechanisms or processes for the phenomenon of the higher Cl concentration in the Chinese RPP than that in the Korean RPP could not be elucidated in the present study.
Rb concentration analyzed by ICP-OES
The limit of detection in a conventional EDXRF analysis is ~ 0.1 mg/100 g for most elements (Fiamegos and de la Calle Guntiñas, 2018). However, in the present study, Rb was not detected by EDXRF analysis in an RPP sample containing ~ 1.6 mg/100 g of Rb (which was measured using ICP-OES). To accurately determine the Rb concentrations in the Korean and Chinese RPP, the samples were analyzed by ICP-OES. The Rb concentration was then compared. The results are presented in Table 1. The Korean RPP samples had significantly (p < 0.001) lower Rb concentrations per 100 g (mean = 0.9 mg, range = 0.3–1.6 mg) than the Chinese RPP samples (mean = 1.9 mg, range = 0.9–2.8 mg). These results suggest that elemental analysis is an appropriate analytical method for distinguishing between Korean and Chinese RPP. Table 2 summarizes the ranges of the Rb and Cl concentrations in the Korean RPP samples used in this study.
Table 2.
Cl concentration analyzed by EDXRF and Rb concentration analyzed by ICP-OES in Korean and Chinese red pepper powder (RPP) samples
| Variable | Korean RPP | Chinese RPP | ||
|---|---|---|---|---|
| Mean ± SD | Range | Mean ± SD | Range | |
| Rb (mg/100 g) | 0.9 ± 0.3 | 0.3–1.6 | 1.9 ± 0.4*** | 0.9–2.8 |
| Cl (mg/100 g) | 154 ± 29 | 114–215 | 299 ± 44*** | 177–405 |
***Significantly different from Korean RPP, p < 0.001
Blind trials
A blind trial using 20 RPP samples was conducted to accurately evaluate the predictive discrimination power of the selected variables by removing any bias that will arise from the investigators’ awareness of the treatment group (Table 3). The geographic origins of the blind samples were determined by applying the ranges of the Rb and Cl concentrations (per 100 g) found in the Korean RPP samples as follows. If the measured Rb concentration was ≤ 1.6 mg and the measured Cl concentration was ≤ 215 mg, the sample was a Korean RPP. However, if the measured Rb concentration was > 1.6 mg or the measured Cl concentration was > 215 mg, the sample might be a Chinese RPP. The blind samples consisted of three samples of Korean RPP (nos. 1–3), three samples of Chinese RPP (nos. 18–20), and 14 samples of Korean RPP adulterated with Chinese RPP (nos. 4–17). Finally, 12 of the 20 blind samples were correctly classified via their Rb and Cl concentrations.
Table 3.
Classification of geographic origins of blind red pepper powder (RPP) samples
| Sample no | Ingredient | Rb (mg/100 g) | Cl (mg/100 g) | Classification result |
|---|---|---|---|---|
| 1 | KR 1 | 1.1 | 159 | Ca |
| 2 | KR 2 | 0.6 | 167 | C |
| 3 | KR 1 + KR 2 (50:50, g/g) | 0.9 | 152 | C |
| 4 | KR 1 + CN 1 (90:10, g/g) | 1.1 | 165 | Ib |
| 5 | KR 2 + CN 2 (90:10, g/g) | 0.7 | 187 | I |
| 6 | KR 1 + CN 1 (80:20, g/g) | 1.2 | 181 | I |
| 7 | KR 2 + CN 2 (80:20, g/g) | 0.9 | 183 | I |
| 8 | KR 1 + CN 1 (70:30, g/g) | 1.4 | 200 | I |
| 9 | KR 1 + CN 2 (70:30, g/g) | 1.2 | 197 | I |
| 10 | KR 2 + CN 1 (70:30, g/g) | 1.2 | 213 | I |
| 11 | KR 2 + CN 2 (70:30, g/g) | 1.0 | 198 | I |
| 12 | KR 1 + CN 1 (50:50, g/g) | 1.6 | 219c | C |
| 13 | KR 2 + CN 2 (50:50, g/g) | 1.2 | 218c | C |
| 14 | KR 1 + CN 1 (30:70, g/g) | 1.8c | 236c | C |
| 15 | KR 2 + CN 2 (30:70, g/g) | 1.4 | 242c | C |
| 16 | KR 1 + CN 1 (10:90, g/g) | 2.0c | 275c | C |
| 17 | KR 2 + CN 2 (10:90, g/g) | 1.6 | 269c | C |
| 18 | CN 1 | 2.2c | 281c | C |
| 19 | CN 2 | 1.6c | 282c | C |
| 20 | CN 1 + CN 2 (50:50, g/g) | 1.9c | 279c | C |
KR Korean RPP, CN Chinese RPP
aCorrectly classified
bIncorrectly classified
cEntries are outside the range of values of Korean RPP samples used in the present study
We investigated the discrimination limits (i.e., the lowest quantity of the Chinese RPP contained in the Korean RPP adulterated with Chinese RPP that could be distinguished from the unadulterated Korean RPP) of the above procedure using the results obtained from the Korean RPP samples to which different quantities of Chinese RPP were added. All the samples of the Korean RPP (nos. 1–3) were correctly classified as being Korean RPP, whereas the adulterated samples [at a level of 10, 20, or 30 g/100 g (nos. 4–11)] were not classified as being Korean RPP. However, the samples (nos. 12–20) of Korean RPP adulterated with Chinese RPP at a level of 50, 70, 90, or 100 g/100 g were correctly classified as being Chinese RPP because they had Rb or Cl concentrations outside the predetermined ranges of these variables in the Korean RPP samples. Consequently, the discrimination limit of our procedure was determined to be 50 g/100 g for the Korean RPP adulterated with the Chinese RPP.
In conclusion, this is the first study to report the ability to discern the geographic origin of RPP using the elemental analyses of ICP-OES and EDXRF. In particular, this study distinguishes between Korean and Chinese RPP. This is because China provides most of the imported RPP distributed in Korea. The Rb and Cl concentrations were markedly higher in the Chinese RPP than those in the Korean RPP. The higher Rb concentration observed in the Chinese RPP might be closely related to the higher acidity of farmland soil in China compared to that in Korea. Further investigations are needed to identify the underlying cause of the higher Cl concentration found in the Chinese RPP. Korean RPP adulterated with ≥ 50 g/100 g of Chinese RPP could be distinguished by setting the minimum Rb concentration to ≤ 1.6 mg/100 g (analyzed by ICP-OES) and the minimum Cl concentration to ≤ 215 mg/100 g (analyzed by EDXRF) for the unadulterated Korean RPP samples used in this study. Our results suggest that the analysis of the Rb and Cl concentrations is a possible and effective approach to distinguish between Korean and Chinese RPP.
Acknowledgements
This research was supported by a Grant (17162MFDS065) from the Ministry of Food and Drug Safety, Korea, in 2017.
Declarations
Conflict of interest
The authors declare no conflict of interest.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Jung Eun Lee, Email: jungeun9179@naver.com.
Eunji Choi, Email: choieunji@sookmyung.ac.kr.
Cheol Seong Jang, Email: csjang@kangwon.ac.kr.
Hyang Sook Chun, Email: hschun@cau.ac.kr.
Sangdoo Ahn, Email: sangdoo@cau.ac.kr.
Byung Hee Kim, Email: bhkim@sookmyung.ac.kr.
References
- AOAC (Association of Official Analytical Chemists). AOAC Guidelines for single laboratory validation of chemical methods for dietary supplements and botanicals (2002). Available from: https://www.aoac.org/aoac_prod_imis/AOAC_Docs/StandardsDevelopment/SLV_Guidelines_Dietary_Supplements.pdf. Accessed Feb. 22, 2021
- Camargo AB, Resnizky S, Marchevsky EJ, Luco JM. Use of the Argentinean garlic (Allium sativum L.) germplasm mineral profile for determining geographic origin. Journal of Food Composition and Analysis. 2010;23:586–591. doi: 10.1016/j.jfca.2010.01.002. [DOI] [Google Scholar]
- Choi YH, Hong CK, Kim M, Jung SO, Park J, Oh YH, Kwon JH. Multivariate analysis to discriminate the origin of sesame seeds by multi-element analysis inductively coupled plasma-mass spectrometry. Food Science and Biotechnology. 2017;26:375–379. doi: 10.1007/s10068-017-0051-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Choi YH, Hong CK, Park GY, Kim CK, Kim JH, Jung K, Kwon JH. A nondestructive approach for discrimination of the origin of sesame seeds using ED-XRF and NIR spectrometry with chemometrics. Food Science and Biotechnology. 2016;25:433–438. doi: 10.1007/s10068-016-0059-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Choi E, Hwang J, Lee D, Jun H, Yoon JA, Chun HS, Ahn S, Kim BH. Discrimination of red pepper powder (Capsicum annuum L.) with added seeds using inorganic element and fatty acid profiles in combination with canonical discriminant analysis. Journal of the Korean Society of Food Science and Nutrition. 2020;49:716–728. doi: 10.3746/jkfn.2020.49.7.716. [DOI] [Google Scholar]
- Drobner U, Tyler G. Conditions controlling relative uptake of potassium and rubidium by plants from soils. Plant and Soil. 1998;201:285–293. doi: 10.1023/A:1004319803952. [DOI] [Google Scholar]
- Fiamegos Y, de la Calle Guntiñas MB. Validation strategy for an ED-XRF method to determine trace elements in a wide range of organic and inorganic matrices based on fulfilment of performance criteria. Spectrochimica Acta Part B: Atomic Spectroscopy. 2018;150:59–66. doi: 10.1016/j.sab.2018.10.009. [DOI] [Google Scholar]
- Franke BM, Hadorn R, Bosset JO, Gremaud G, Kreuzer M. Is authentication of the geographic origin of poultry meat and dried beef improved by combining multiple trace element and oxygen isotope analysis? Meat Science. 2008;80:944–947. doi: 10.1016/j.meatsci.2008.03.018. [DOI] [PubMed] [Google Scholar]
- Franke BM, Haldimann M, Gremaud G, Bosset JO, Hadorn R, Kreuzer M. Element signature analysis: its validation as a tool for geographic authentication of the origin of dried beef and poultry meat. European Food Research and Technology. 2008;227:701–708. doi: 10.1007/s00217-007-0776-8. [DOI] [Google Scholar]
- Galgano F, Favati F, Caruso M, Scarpa T, Palma A. Analysis of trace elements in southern Italian wines and their classification according to provenance. LWT-Food Science and Technology. 2008;41:1808–1815. doi: 10.1016/j.lwt.2008.01.015. [DOI] [Google Scholar]
- Greenough JD, Longerich HP, Jackson SE. Element fingerprinting of Okanagan Valley wines using ICP-MS: relationships between wine composition, vineyard and wine colour. Australian Journal of Grape and Wine Research. 1997;3:75–83. doi: 10.1111/j.1755-0238.1997.tb00118.x. [DOI] [Google Scholar]
- Guo JH, Liu XJ, Zhang Y, Shen JL, Han WX, Zhang WF, Christie P, Goulding KWT, Vitousek PM, Zhang FS. Significant acidification in major Chinese croplands. Science. 2010;327:1008–1010. doi: 10.1126/science.1182570. [DOI] [PubMed] [Google Scholar]
- Hong E, Lee SY, Jeong JY, Park JM, Kim BH, Kwon K, Chun HS. Modern analytical methods for the detection of food fraud and adulteration by food category. Journal of the Science of Food and Agriculture. 2017;97:3877–3896. doi: 10.1002/jsfa.8364. [DOI] [PubMed] [Google Scholar]
- Hoyt PB, Hennig AMF. Soil acidification by fertilizers and longevity of lime applications in the Peace River region. Canadian Journal of Soil Science. 1982;62:155–163. doi: 10.4141/cjss82-017. [DOI] [Google Scholar]
- Kelly S, Baxter M, Chapman S, Rhodes C, Dennis J, Brereton P. The application of isotopic and elemental analysis to determine the geographical origin of premium long grain rice. European Food Research and Technology. 2002;214:72–78. doi: 10.1007/s002170100400. [DOI] [Google Scholar]
- Kelly S, Heaton K, Hoogewerff J. Tracing the geographical origin of food: The application of multi-element and multi-isotope analysis. Trends in Food Science & Technology. 2005;16:555–567. doi: 10.1016/j.tifs.2005.08.008. [DOI] [Google Scholar]
- Korea Ministry of Food and Drug Safety. Imported Foods Information Site (2020). Available from: https://impfood.mfds.go.kr/CFDAA01F01. Accessed Feb. 22, 2021
- Korean Statistical Information Service. Crop Production Survey (2020). Available from: http://kosis.kr/statHtml/statHtml.do?orgId=101&tblId=DT_1ET0291&conn_path=I3. Accessed Feb. 22, 2021
- Kunselman G, Brown P, Seeley C, Reed, A. Determination of the halogen elements in aqueous, organic, and solid samples using ICP-OES. Spectroscopy. Special issues Oct. 1 (2006)
- Kwon YK, Bong YS, Lee KS, Hwang GS. An integrated analysis for determining the geographical origin of medicinal herbs using ICP-AES/ICP-MS and 1H NMR analysis. Food Chemistry. 2014;161:168–175. doi: 10.1016/j.foodchem.2014.03.124. [DOI] [PubMed] [Google Scholar]
- Kwon DJ, Jo JH, Kim HK, Park MH. Establishment of long-term storage condition of fresh red pepper paste. Korean Journal of Food Science and Technology. 1990;22:415–420. [Google Scholar]
- Laursen KH, Schjoerring JK, Olesen JE, Askegaard M, Halekoh U, Husted S. Multielemental fingerprinting as a tool for authentication of organic wheat, barley, faba bean, and potato. Journal of Agricultural and Food Chemistry. 2011;59:4385–4396. doi: 10.1021/jf104928r. [DOI] [PubMed] [Google Scholar]
- Lee D, Kim M, Kim BH, Ahn S. Identification of the geographical origin of Asian red pepper (Capsicum annuum L.) powders using 1H NMR spectroscopy. Bulletin of the Korean Chemical Society. 2020;41:317–322. doi: 10.1002/bkcs.11974. [DOI] [Google Scholar]
- Malhi SS, Nyborg M, Harapiak JT. Effects of long-term N fertilizer-induced acidification and liming on micronutrients in soil and in bromegrass hay. Soil and Tillage Research. 1998;48:91–101. doi: 10.1016/S0167-1987(98)00097-X. [DOI] [Google Scholar]
- Molina M, Aburto F, Calderón R, Cazanga M, Escudey M. Trace element composition of selected fertilizers used in Chile: phosphorus fertilizers as a source of long-term soil contamination. Soil and Sediment Contamination. 2009;18:497–511. doi: 10.1080/15320380902962320. [DOI] [Google Scholar]
- Nečemer M, Kump P, Rajčevič M, Jačimović R, Budič B, Ponikvar M. Determination of sulfur and chlorine in fodder by X-ray fluorescence spectral analysis and comparison with other analytical methods. Spectrochimica Acta Part B: Atomic Spectroscopy. 2003;58:1367–1373. doi: 10.1016/S0584-8547(03)00057-0. [DOI] [Google Scholar]
- Otaka A, Hokura A, Nakai I. Determination of trace elements in soybean by X-ray fluorescence analysis and its application to identification of their production areas. Food Chemistry. 2014;147:318–326. doi: 10.1016/j.foodchem.2013.09.142. [DOI] [PubMed] [Google Scholar]
- Sacco D, Brescia MA, Sgaramella A, Casiello G, Buccolieri A, Ogrinc N, Sacco A. Discrimination between Southern Italy and foreign milk samples using spectroscopic and analytical data. Food Chemistry. 2009;114:1559–1563. doi: 10.1016/j.foodchem.2008.11.056. [DOI] [Google Scholar]
- Suhaj M, Koreňovská M. Study of some European cheeses geographical traceability by pattern recognition analysis of multielemental data. European Food Research and Technology. 2008;227:1419–1427. doi: 10.1007/s00217-008-0861-7. [DOI] [Google Scholar]
- Thampi PSS. A glimpse of the world trade in Capsicum. In: De AK, editor. Capsicum: The genus Capsicum. 1. London: Taylor & Francis Group; 2003. pp. 16–24. [Google Scholar]
- Tyler G. Rubidium-availability and plant uptake in natural soils. Communications in Soil Science and Plant Analysis. 1983;14:1075–1089. doi: 10.1080/00103628309367433. [DOI] [Google Scholar]