Principal component (PC) analysis scores for 11 regions on PC1 and PC2. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
Principal component (PC) analysis scores for 11 regions on PC1 and PC3. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
Dendrogram of hierarchical cluster analysis. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
Click on author name to view affiliation information
Areca (Areca catechu L.) is one of the most important cash crops in China and is considered the fourth most widely used addictive substance. In addition, areca is widely used in traditional and herbal medicines. The major characteristics of the fruit are affected by its genetic background and growth environment. The growing environment in different regions will impact the quality of agricultural products and the processing quality. The quality of areca is not only the basis of its commercialization development and processing quality, but also is an important basis for the scientific planting of areca. Therefore, determining the quality of areca will provide evidence for scientific planting and more optimal applications. We evaluated the quality of areca by comparing the differences in physicochemical characteristics using principal component analysis (PCA) and hierarchical cluster analysis. A total of 165 arecas, in the same growth period, were collected from 11 main producing regions in Hainan Province. Our results illustrate that the physicochemical characteristics of areca in different regions were significantly different. The PCA was conducted using 10 quality indexes, and three principal components were extracted to reflect 80% of the original variables. The first principal component mainly reflected the fruit shape quality, the second principal component mainly reflected the hardness quality, and the third principal component mainly reflected the functional component quality. The relationship between each producing region and the principal component could be obtained intuitively from the principal component score plots. The arecas in Wanning and Wenchang were larger and their cellulose content was greater than in other areas, indicating that they were more suitable for processing. In contrast, the arecas in Baoting, Wuzhishan, Danzhou, Tunchang, and Dongfang had a greater arecoline content than the other areas, making them more suitable for use as medicinal materials. Hierarchical cluster analysis classified the 11 producing regions into five categories based on the measured parameters, which was consistent with the results of the PCA score plots. These results could provide information to improve the use of areca in China.
Areca (Areca catechu L.) is widely distributed throughout South and Southeast Asia, including China, India, Pakistan, Indonesia, Malaysia, the Philippines, and New Guinea (Heatubun et al. 2012; Verma 2011). In China, 95% of areca is planted in Hainan, of which 95% to 97% is processed into chewing products (Li et al. 2019; Sarode et al. 2013). During the past 5 years, Chinese areca production has increased steadily, with an annual output of 276,200 tons, and the annual output value of the areca industry has reached 10 billion US dollars. In the consumer market, the consumption of areca products has become the fourth largest worldwide, after cigarettes, alcohol, and coffee (Li and Zhang 2011). The major characteristics of the fruit are affected by its genetic background and growth environment. The growing environment in different regions will impact the quality of agricultural products and the processing quality (Liu et al. 2022; Wang et al. 2022). Therefore, the quality of areca is not only the basis of its commercialization development and processing quality, but also is an important basis for the scientific planting of areca. For example, the phenotypic traits of the areca fruit, such as weight, transverse diameter, and longitudinal diameter, can determine whether it is easy to eat. The cellulose, lignin, and pectin contents determine the hardness and feeling of chewing; hard fibers will cause oral mucosal damage. Alkaloids and polyphenols, which have refreshing effects, are the main active substances and medicinal ingredients in areca (Amudhan et al. 2012; Peng et al. 2015). These qualities affect sensory taste and consumer choice (Dai et al. 2021; Huang et al. 2021; Peng et al. 2021).
Limited studies have been conducted on the quality of areca. In this study, 165 arecas were collected during the same growth period from 11 main producing regions in Hainan Province. The quality was evaluated using PCA and hierarchical cluster analysis (HCA) and 10 indices, including weight, transverse diameter, longitudinal diameter, fiber layer thickness, cellulose, hemicellulose, pectin, and alkaloids. Consumers are paying increasing attention to the quality and safety of areca products. Therefore, this study provides information about areca quality to support its use, and improves the analysis of the impact of different regions on quality.
Arecas with a growth period of 150 d were selected from 11 producing regions in Hainan Province. Arecas that were green or dark green in coloration, free of any blemishes or deformities, and with a growth period of 150 d were selected. They included Wenchang (WC), Qionghai (QH), Wanning (WN), Baoting (BT), Sanya (SY), Dongfang (DF), Ledong (LD), Qiongzhong (QZ), Tunchang (TC), Danzhou (DZ), and Wuzhishan (WZS) (Table 1). Farmers planted the selected arecas. Fifteen arecas with appropriate maturity, no pests or diseases, and no mechanical damage were selected randomly from each region. Immediately after collection, the samples were transported to the laboratory for cold storage.
Fifteen arecas were selected randomly from each region, washed with clean water, and dried. Each index was measured in triplicate. All chemical tests used 1 g of an areca-treated sample—meaning, after drying, crushing, and passing through a 0.25-mm mesh sieve.
A single areca was weighed using an analytical balance, and its width and length were measured using a Vernier caliper. The areca was cut down the middle, its core was removed, and the thickness of the fiber layer was measured using a Vernier caliper.
The content of each lignocellulose component was determined using a cellulose tester. Determination of the content of neutral detergent fiber and acid detergent fiber in the sample, the acid-washing lignin content of the sample, was determined using a 72% concentrated sulfuric acid hydrolysis method, and the ash content was determined after dry ashing at 550 °C for 2 h in a muffle furnace. Last, the cellulose, hemicellulose, and lignin contents of the samples were calculated.
Pectin content was measured using spectrophotometry according to the Chinese Agricultural Industry Standard (Ministry of Agriculture of China 2008). We used a spectrophotometer to measure the absorbance at a wavelength of 525 nm. We then calculated the pectin content using the standard curve.
The arecoline content was measured using high-performance liquid chromatography (HPLC; HITACHI Chromaster, Japan). The chromatographic parameters were as follows: XB-SCX strong cation-exchange column, 4.6 mm × 250 mm × 5 μm; mobile phase, acetonitrile, 0.2% phosphoric acid (35:65, pH adjusted to 3.8 with ammonia); detection wavelength, 215 nm; column temperature, 25 °C; injection volume, 20 μL; and flow rate, 1.0 mL⋅min–1.
The tannin content was measured using HPLC (HITACHI Chromaster). The chromatographic parameters were as follows: chromatographic column, C18 column of 4.6 mm × 150 mm × 5 μm; mobile phase, methanol, 0.1% phosphoric acid (25:75); detection wavelength, 275 nm; column temperature, 30 °C; injection volume 10 μL; and flow rate, 1.0 mL⋅min–1.
Excel and the Statistical Package for the Social Sciences (SPSS) were used for statistical analysis. Single factor analysis of variance, least significant difference (LSD), and Waller–Duncan methods were used for the analysis of variance and multiple comparisons (α = 0.05); correlation analysis, PCA, and HCA were performed on the data of the measured samples. The original 2019b software (OriginLab, Northampton, MA, USA) was used for drawing, and all experiments were repeated in triplicate.
The weight, transverse diameter, and longitudinal diameter are the most important physical characteristics that determine the economic value of areca (Ren et al. 2020). As shown in Table 2, the weights, transverse diameters, and longitudinal diameters in the different regions were significantly different (P < 0.05). The weight, transverse diameter, and longitudinal diameter were the greatest in WN (28.40 g, 32.06 mm, and 55.40 mm, respectively), which were significantly greater than those in the other regions. The thickness of the fiber layer determines the chewing quality of the areca. A thick fiber layer increases the hardness of the areca, which damages the oral mucosa during chewing. Wanning had the smallest fiber layer thickness, with an average thickness of 5.36 mm, which was significantly less than that in the other regions (P < 0.05).
The chewing sensation is an important indicator of areca processing. Lignin determines the softness and hardness of chewing, whereas cellulose, hemicellulose, and pectin determine the elasticity and toughness of chewing (Bauer et al. 1973). Lignin is the main factor affecting the hardness of areca fibers (Fahlén and Salmén 2005; Jiang et al. 2018). During plant growth, lignin acts as a matrix to fill the remaining space in the cell wall, increasing the mechanical strength and enhancing the hardness of the plant (Li et al. 2019). Therefore, areca with a low lignin content should be selected for processing. Because of the hard texture of areca fibers, long-term chewing areca will lead to wear and tear and abrasion of the teeth, longitudinal fracture of teeth roots, and, in serious cases, it will damage the oral tissue, resulting in oral submucosal fibrous lesions (Arakeri and Brennan 2013; Brandt et al. 2013; Santos et al. 2020). Pectin plays important roles in cell wall hydration, adhesion, growth, extensibility, and elasticity (Cao et al. 2019; Ferreira-Lazarte et al. 2018; Li et al. 2021; Liao et al. 2021). Cellulose, hemicellulose, and pectin give areca products elasticity and toughness; therefore, fruit with high contents of cellulose, hemicellulose, and pectin should be used primarily when processing areca.
As shown in Table 3, the lignin, cellulose, and hemicellulose contents of areca in the different regions were significantly different (P < 0.05), but there was no significant difference in pectin content. In terms of lignin content, there was a decreasing trend from north to south, with the greatest lignin content in WC (115.80 mg⋅g–1) and the least in SY (102.75 mg⋅g–1). In terms of cellulose, there is a growth pattern from north to south, with the greatest cellulose content in SY (214.34 mg⋅g–1) and the least in QH (191.28 mg⋅g–1). Wanning had the greatest hemicellulose content (252.58 mg⋅g–1), which was significantly greater than those in the other regions.
Alkaloids and polyphenols, which have antioxidative, antiviral, antitumor, antibacterial, and antidepressant effects, are the active components of areca (Papke et al. 2015; Yamada et al. 2015), and alkaloids are the main source of its stimulating effects (Bhat et al. 2017; Mehrtash et al. 2017). Therefore, areca, which has high alkaloid and polyphenol contents, is suitable for traditional and herbal medicines. As shown in Table 3, the areca alkaloid contents in the different regions were significantly different (P < 0.05); however, there was no significant difference in polyphenol content. The content was the greatest in the eastern region (5.42 mg⋅g–1), which was significantly greater than that in the other regions.
Principal component analysis is a statistical analysis method that simplifies multiple indicators into a few comprehensive indicators and uses a few variables to reflect the information of the original variables as much as possible (Gao 2005). It can be seen from Table 4 that the eigenvalue of the first four principal components were greater than one. The cumulative variance contribution rates of the first three principal components were 80.00%, which integrated most of the area information, and three principal components were extracted.
Table 5 lists the information contents of the different principal components. The first principal component (PC1) mainly synthesized the information on weight, longitudinal diameter, and fiber layer thickness, and the three indicators all presented a positive distribution on the first principal component. Therefore, PC1 mainly explained the fruit shape of the areca, which can be referred to as the fruit shape index, and the second principal component (PC2) mainly synthesized the information of cellulose and lignin. Lignin presented a negative distribution in PC2; cellulose presented a positive distribution. The second principal component mainly explained the hardness of areca, which can be referred to as the hardness index, and the third principal component (PC3) mainly synthesized information on alkaloids and polyphenols, in which alkaloids are distributed in a negative direction. Hence, PC3 was named the functional component index.
In our study, PC1 and PC2 contained 40.91% and 25.97% of the original information, respectively. Researchers have used PCA score charts to reflect the relationships among indicators (Baraldi et al. 2007; Karlsen et al. 1999; Patras et al. 2011). The relationship between each region and PC1 vs. PC2 is shown in Fig. 1. Wanning and WC were located in the first quadrant, indicating arecas were larger and the cellulose content was greater. Qiongzhong, BT, and WZS, in the second quadrant, illustrated that arecas in these three regions were smaller and the cellulose content was greater. Six regions (LD, DF, SY, DZ, TC, and QH) were in the negative range of PC2, indicating the arecas in these six regions had a greater lignin content and were unsuitable for processing. Qiongzhong, BT, and WZS were in the negative range of PC1 and the positive range of PC2, explaining why the cellulose content of areca was greater and the fruit were smaller, and thus were more suitable for fresh consumption.
Citation: HortScience 58, 6; 10.21273/HORTSCI17125-23
Principal component 3 comprised 13.1% of the original information. Figure 2 shows the relationship between each region and PC1 and PC3. Baoting, WZS, DZ, TC, and DF were in the negative range of PC3, indicating that the arecoline content in these four regions was greater and the arecas were more suitable for use as medicinal materials.
Citation: HortScience 58, 6; 10.21273/HORTSCI17125-23
Based on physicochemical quality, the HCA of areca in 11 regions was determined. According to the HCA dendrogram (Fig. 3), samples from the 11 regions were divided into three clusters (distance between classes = 15).
Citation: HortScience 58, 6; 10.21273/HORTSCI17125-23
The first cluster included WC and WN, which had larger fruit, higher cellulose content, and lower lignin content, and were suitable for processing. The second cluster contained BT, QZ, WZS, SY, DF, and LD. This group mainly included areca with smaller fruit, which are suitable for fresh consumption. Baoting, QZ, and WZS were smaller and had greater cellulose and arecoline contents, making them suitable for fresh consumption and medicine. The third cluster, including QH, TC, and DZ, had a higher lignin content and lower cellulose content, which is not recommended for fresh consumption and processing. In summary, the results of HCA and the principal component comprehensive evaluation were relatively consistent, indicating that both methods could be used to analyze the quality indicators of areca and to determine whether the arecas was suitable for fresh consumption, processing, or medicine.
Principal component analysis and HCA are used increasingly by researchers in the quality analysis of multiple indicators of varieties (Geőcze et al. 2013; Keenan et al. 2012; Schnackenberg et al. 2010). In this study, 10 quality indices of areca from 11 producing regions were determined and investigated using statistics, PCA, and HCA. This can provide a foundation for areca use. For example, we can select regions in the third and fourth intervals of the PC1 and PC3 score charts for producing medicinal materials; fresh-consumption areca can be selected using the second and third intervals of the PC1 and PC2 score charts. Interestingly, the results showed that climatic conditions had an important effect on the quality of the arecas. Hainan’s climate is generally divided into three regions: the northern region, the central mountainous region, and the southern region (Chen et al. 2022). The PCA and HCA of the central mountainous region (BT, QZ, and WZS), southern region (SY, LD, and DF), and northern region (WC, WN, QH, TC, and DZ) reflected regional characteristics. In future studies, it will be necessary to consider whether the climatic characteristics of different regions lead to different accumulations of areca metabolites.
In this study, 10 quality indices were determined and analyzed for 165 arecas collected from 11 main areca-producing regions in Hainan. The results showed significant differences in eight quality indices of areca from different regions without pectin or polyphenols. Arecas in WN were the most suitable for processing because they had the largest fruit, the greatest hemicellulose content, and the thinnest fiber layer. The areca alkaloid content of DF was the greatest, making it suitable for medicinal use.
Principal component analysis was performed on the 10 quality indices, and three principal components were extracted to reflect 80% of the original variables. The first principal component mainly reflected the fruit shape quality, PC2 mainly reflected the hardness quality, and PC3 mainly reflected the functional component quality. The relationships among each production region and the principal component can be obtained intuitively from the principal component score plot. The arecas in WN and WC were larger, and the cellulose content was greater, making them more suitable for processing material. Baoting, WZS, DZ, TC, and DF had a greater arecoline content, which is more suitable for use as medicinal materials.
Based on the quality characteristics of arecas, the 11 producing regions were divided into three categories using HCA, and appropriate utilization fields were determined preliminarily. The results of the HCA and PCA were relatively consistent. The correlation between areca quality and the producing region was relatively significant. Principal component analysis and HCA can be used to analyze the quality index of areca and its appropriate use.
Principal component (PC) analysis scores for 11 regions on PC1 and PC2. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
Principal component (PC) analysis scores for 11 regions on PC1 and PC3. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
Dendrogram of hierarchical cluster analysis. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
Contributor Notes
This research was supported by the Finance Science and Technology Project of Hainan Province, China (KYYS-2021-09, 321QN0950, and ZDYF2022XDNY262) and the Scientific and Technological Innovation Team Projects of Hainan Academy of Agricultural Science, China (HAAS2022TDYD01).
S.W. and J.D. were responsible for writing, data curation, and investigation; X.K. was responsible for software and methodology; J.Z. was responsible for software and conceptualization; W.D. was responsible for writing editing, review, and supervision; J.J. was responsible for methodology and visualization.
J.D. is the corresponding author. E-mail: 1097074907@qq.com.
Principal component (PC) analysis scores for 11 regions on PC1 and PC2. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
Principal component (PC) analysis scores for 11 regions on PC1 and PC3. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
Dendrogram of hierarchical cluster analysis. BT = Baoting; DF = Dongfang; DZ = Danzhou; LD = Ledong; QH = Qionghai; QZ = Qiongzhong; SY = Sanya; TC = Tunchang; WC = Wenchang; WN = Wanning; WZS = Wuzhishan.
