Phenotypic and Genetic Uniformity in Three Populations of Panax notoginseng by Mass Selection  

Yun Yang1 , Junwen Chen1 , Ming Zhao1 , Cuiting Li1 , Zhengui Meng1 , Jianjun Wang1 , Zhongjian Chen2 , Guanghui Zhang1 , Shengchao Yang1
1. Yunnan Research Center on Good Agricultural Practice for Dominant Chinese Medicinal Materials, Yunnan Agricultural University, Kunming 650201, Yunnan, People's Republic of China
2. Institute of Panax notoginseng, Wenshan University, Wenshan 663000, Yunnan, People's Republic of China
Author    Correspondence author
Molecular Plant Breeding, 2013, Vol. 4, No. 21   doi: 10.5376/mpb.2013.04.0021
Received: 17 May, 2013    Accepted: 10 Jun., 2013    Published: 18 Jun., 2013
© 2013 BioPublisher Publishing Platform
This is an open access article published under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Preferred citation for this article:

Yang et al., 2013, Phenotypic and genetic uniformity in three populations of Panax notoginseng by mass selection, Molecular Plant Breeding, Vol.4, No.21 169-176 (doi: 10.5376/mpb.2013.04.0021)


Aim: Panax notoginseng is an important traditional Chinese medicinal plant. Although the species has been cultivated for more than 400 years, a certified variety is still not available. Natural populations (NP) exhibit high levels of morphological variation and genetic diversity, suggesting that mass selection may be used to improve P. notoginseng. In previous studies, we established 38 mass selection populations based on certain morphological traits (target characters) and eliminated undesirable individuals over five successive generations. The objective of the present study was to evaluate phenotypic and genetic uniformity of three of these populations: purple-stem (PSP), green-stem (GSP), and erect-type (ETP) populations. Methods: To assess phenotypic uniformity, 12 morphological traits were measured in NP and the three selected populations. Genetic uniformity of these four populations was also evaluated using inter-simple sequence repeat (ISSR) markers. Results: Phenotypic uniformity was only exhibited with respect to target characters, including stem color in PSP and GSP, and petiole-peduncle angle in ETP. Average coefficients of variation detected for most non-target characters in PSP, GSP, and ETP were similar or higher to those of NP. When these four populations were analyzed using 129 ISSR markers, the percentage of polymorphic bands detected was 72.27% in PSP, 76.40% in GSP, 58.34% in ETP, and 76.23% in NP. Genetic identities (I) in PSP, GSP, and ETP were 0.9058, 0.8663, and 0.8703, respectively, with a value greater than 0.7814 in NP. Conclusion: Mass selection is an efficient way to improve target characters and genetic uniformity in P. notoginseng. Nevertheless, selection of specific individuals exhibiting comprehensive phenotypic traits may be necessary to accelerate the breeding process.

Panax notoginseng; Mass selection; Phenotypic uniformity; Genetic identity; ISSR

Panax notoginseng (Burkill) F. H. Chen, an important traditional Chinese medicinal plant in the ginseng genus Panax L., is mainly distributed in the Wenshan region of Yunnan Province, China. Saponins in P. notoginseng roots are reported to have anti-inflammatory, cerebral-protective, hemostatic, and anti-tumor effects (Li and Chu, 1999; Jiang and Qian, 1995; White et al., 2000; Yin et al., 2012; Wang et al., 2012). Although the herb has long been widely cultivated commercially in the Wenshan region, a certified variety is unfortunately still not available.

Most medicinal plants are wild, with only a few having been domesticated and being in current cultivation (Singh et al., 2011). Panax notoginseng, however, has been cultivated for more than 400 years in a narrow habitat in the Wenshan region of Yunnan Province (Zheng and Yang, 1994; Guo et al., 2010). All known populations of this herb are cultivated ones domesticated by local inhabitants; wild populations have not been located (Guo et al., 2010). Surprisingly, given its very long cultivation history, a certified P. notoginseng variety is still lacking. The cultivated population ishighly heterozygous because non-selective seed collection is practiced; this not only gives rise to morphological variation, but also affects saponin content (Hong et al., 2005).
Genetic variation is the basis for species evolution and the raw material for breeding (Mousseau et al., 2000). Widevariationin morphology and saponin content, and high genetic diversity, is exhibited by members of the genus Panax. In P. notoginseng, morphological variability is exhibited with respect to characters such as root shape and color, stem color, inflorescence type, and fruit color (Xiao et al., 2008; Guo et al., 2010). Variation in saponin content has been detected between individual roots from different populations in P. notoginseng, P. ginseng,and P. quinquefolius (Hong et al., 2005; Schlag and McIntosh, 2006; Lee et al., 2011). Genetic diversity in cultivated populations of P. notoginseng has been demonstrated using fAFLP (fluorescent amplified fragment length polymorphism) and EST-SSR (expressed sequence tag- simple sequence repeats) markers (Wang et al., 2008; Zhang et al., 2011).
Genetic diversity within Panax can be exploited to improve ginseng plant breeding. An interspecific hybrid between P. ginseng and P. quinquefolius was produced as early as 1950 (Kuriyama 1950); the interspecific hybrid exhibited superior growth and higher ginsenoside levels, but could not be used for field cultivation because it was sterile (Susumu et al., 1997). Mass selection, which is an example of selection from biologically variable populations in which differences are genetic in origin (Acquaah, 2007), is still the primary and most effective approach to germplasm enhancement and new cultivar breeding of medicinal plants, including Panax L. In P. ginseng, mass selection based on root shape variation has been used to produce the high-yield variety Jishen 1 (Zhao et al., 1998).
In 1997, our research group began selecting different morphological forms of P. notoginseng, and have now established 38 basic populations that are based on morphological traits such as root color (purple, white, or yellow), root shape (radish-like or pimple-like), leaf shape (narrow or wide), stem color (purple or green), and fruit color (purple, yellow, or, most commonly, red) (Chen et al., 2001; Sun et al., 2003). After five generations of breeding, for example, stems of 80% of purple-stem population (PSP) individuals are purple. The aim of this study was to estimate phenotypic and genetic uniformity, based on 12 morphological traits and inter-simple sequence repeat (ISSR) markers, of three of these populations, namely, PSP, green-stem population (GSP), and erect-type population (ETP), with anatural population (NP) as a control. To our knowledge, this is the first report on mass selection breeding in P. notoginseng.
Results and Discussion
Morphological analysis
Variations in stem color, which are dependent on anthocyanin levels, are common in Panax. Panax ginseng normally has dark purple stems and bright red fruits (P-R), with only a few individuals exhibiting green stems and yellow fruits (G-Y); P-R is completely dominant to G-Y (Takahasi and Osumi, 1940). In contrast to P. ginseng, there are three stem color variations (bright purple, jade green, and purple-green) in P. notoginseng; the relative ratio of the three types varies between different populations (Sun et al., 2003; Wang et al., 2008). Furthermore, fruit color in P. notoginseng is normally red and only rarely yellow or purple. These morphological differences suggest a different mode of inheritance for stem and fruit color traits in P. notoginseng.
Stem color is the most obviously varying morphological character in P. notoginseng. We recently observed that a newly-described variety of P. vietnamensis, var. fuscidiscus, also has different stem colors, as does P. notoginseng (Zhu et al., 2003). According to our previous research, green-stem individuals have higher single root weights and yields than purple-stem individuals; the latter group, however, exhibited higher seedling survival rates (Chen et al., 2001), suggesting greater disease resistance. In the present study, green stems and purple stems were associated with greater than 82% of individuals in GSP and PSP, respectively, which were much more heavily skewed ratios than those observed in NP and ETP (Table 1). This result indicates that mass selection can increase the relative levels of green- or purple-stem individuals (Figure 1).

Table 1 Morphological characteristics of four P. notoginseng populations: purple-stem, green-stem, erect-type, and natural populations

Figure 1 Fifth generation of (A) purple-stem and (B) green-stem populations

Based on the angle between peduncle and petiole (APP), three plant types are found in P. notoginseng: erect (APP < 45°), semi-erect (60° > APP> 45°), and horizontal-spreading (APP > 60°). An erect-type population was developed with the idea that because erect individuals occupy only a small area, more plants can be cultivated in a given area, thus obtaining higher yields. In this study, about two-thirds of NP individuals were horizontal-spreading types (67.7%), whereas 23.4% of ETP individuals were horizontal-spreading and 76.6% were erect or semi-erect types (Table 1). Interestingly, higher percentages of erect and semi-erect plants were also found in PSP and GSP (Table 1), suggesting that mass selection for stem color also affects other traits in P. notoginseng. APP is a quantitative character controlled by multiple genes, and is also affected by many environmental factors (e.g., planting density and light intensity). For these reasons, APP was less uniform in ETP than was stem color in GSP and PSP (Table 1). 
Ten non-target characters were also measured in this study. CVs for most of these were higher in PSP, GSP, and ETP than in NP (Table 1), indicating that mass selection was unable to improve uniformity of the non-target characters in P. notoginseng. Among these non-target morphological traits, five were not significantly different among the four populations: petiole length, middle leaflet length, middle leaflet width, middle leaflet length-width ratio, and number of compound leaves (Table 1). A selective response was observed, however, for some non-target characters, especially stem and peduncle height. PSP, GSP, and ETP stem heights were all significantly lower — about 7 cm less — than in NP. While ETP also had a lower peduncle height than NP, peduncle height in GSP was higher than in the other populations (Table 1). In this study, plant height is defined as the combined height of stem and peduncle. Because peduncles and inflorescences were always removed to boost root yields, the results obtained here demonstrate that mass selection for these morphological traits may be used to breed a low-stem variety of P. notoginseng with lodging resistance.
Genetic uniformity
Sixteen ISSR primers, which produced clear and reproducible bands, were used in the ISSR analysis. The 16 ISSR primers generated 127 bands, of which 92 were polymorphic (72.20%). Each primer yielded 5–11 bands (average of 7.94), of which 3–10 were polymorphic (average of 5.75) (Figure 2). Genetic similarity among populations was determined by Nei’s genetic distance (D), Shannon’s information index (Ho), and genetic identity (I). Of the four populations, NP exhibited the greatest variability (D = 0.2186, Ho = 0.3873) and the lowest genetic uniformity (I = 0.7814), while PSP exhibited the lowest variability (D = 0.0941, Ho = 0.2953) and the greatest genetic uniformity (I=0.0941). GSP and ETP also had high levels of genetic uniformity (Table 2). These results demonstrate that mass selection can also increase genetic uniformity, and thus might be used to improve P. notoginseng.

Figure 2 PCR amplification profiles of purple-stem (A 1–15), green-stem (A 16–30), erect-type (B 31–45), and natural (B 46–60) populations in P. notoginseng using ISSR primer TR16

Table 2 Sample number, percentage of polymorphic bands, Nei’s genetic distance, Shannon’s information index, and genetic identity in control and selected populations

Mass selection in P. notoginseng
Genetic variation is indispensable to plant breeding. Because wild populations of P. notoginseng are extinct, variation in cultivated populations, with respect to aspects such as morphology, saponin content, and genetic diversity, has long been a focus of attention. Among these types of variation, morphological variation is the most important, not only from the viewpoint of phenotypic diversity, but because it also affects the quality of the final product (root weight). In P. notoginseng, phenotypic diversity is exhibited with respect to many morphological characters (e.g., in roots, stems, leaves, flowers, fruits, and seeds), and various characters display different CVs (Xiao et al., 2008). This vast phenotypic diversity is an important resource for plant breeding techniques, especially mass selection, the oldest phenotype-based breeding method.
In P. notoginseng, mass selection can be made more effective by selectingyield factor variations. Root weight is the most important yield factor for P. notoginseng (Chen, 2001) and P. ginseng (Zhao et al., 1998), but is not easily observed and selected. Mass selection might thus be performed instead by selecting other morphological characters positively correlated with yield and quality. Because leaf area, middle leaflet length, and middle leaflet width are all closely correlated with root weight (Chen et al., 2004), selection for large leaves is a primary P. notoginseng breeding objective. After five generations, however, average middle leaflet widths and lengths in the “large leaf” population were no larger than those in NP (data not shown), which suggests that more rigorous selection and additional generations are needed to enhance these traits.
Although this study is the first to reveal that mass selection can improve morphological and genetic uniformity of P. notoginseng, P. notoginseng breeding is still constrained by the species’ long growth duration and the paucity of information regarding its genetic inheritance patterns. Each generation bred through mass selection requires three years of growth: because P. notoginseng is commonly grown for one year as seedlings in nursery beds and two years in the field before harvesting, some morphological traits cannot be observed until the third year. In addition, 30–40% of individuals may die from diseases during cultivation. Except for some studies on P. ginseng (e.g., Takahasi and Osumi, 1940), there is little published research on trait inheritance in Panax. Panax quinquefolius is known to be highly self-fertile (Schluter and Punja, 2000), but the pollination characteristics of P. notoginseng are still not clear. This scarcity of data imposes restrictions on the application of pure line selection and cross breeding in P. notoginseng.
Materials and Methods
Plant materials and establishment of populations by mass selection
Mass selection in P. notoginseng was carried out beginning in 1997. Desirable individuals were selected from various fields in the Wenshan region, and mixtures of seeds from plants exhibiting similar phenotypes were sown at Miaoxiang Sanqi Technology Co., Wenshan, Yunnan, China. In the third year, inflorescences of unwanted individuals were removed during early stages of squaring, and the populations were isolated by covering with insect barrier fabric (Weinong Web Factory, Taizhou City, Zhejiang Province). The mature seeds of each population were harvested and re-sown repeatedly in the same manner for five generations (Figure 3). In the third year of the fifth generation, three populations — PSP, GSP, and ETP—established based on stem color or petiole-peduncle angle (Figure 4) were used to evaluate morphological traits and genetic diversity. Seeds purchased in the market in Wenshan City were sown as a control (natural population, NP).

Figure 3 Generalized steps for establishing basic populations in P. notoginseng by mass selection

Figure 4 Two target characters (stem color and plant spread) illustrated using (A) purple-stem, (B) green-stem, (C) erect-type, and (D) horizontal-spreading plants

Morphological traits
Ten quantitative and two qualitative characters were assessed across populations in the third year after flowering (Table 1). Characters were selected according to National Standards of the People’s Republic of China (Guidelines for the conduct of tests for distinctness, uniformity and stability Panax notoginseng, 2008). Data for each quantitative trait were scored from 40 randomly chosen plants. From these measurements, the mean, standard deviation (SD), and coefficient of variation (CV) were calculated for each morphological character. The two qualitative characters were measured in three different portions of 30 randomly selected individuals in the same population, the mean, SD, and CV were calculated as well.
ISSR-PCR amplifications
Leaf material was collected from 15 individuals in each of 4 populations (60 total individuals) and stored at -20℃ until DNA extraction was performed. Leaf tissues were ground into powder in liquid nitrogen. Genomic DNA was extracted following the cetyltrimethyl ammonium bromide (CTAB) procedure (Doyle, 1991). DNA concentrations were determined using a GeneQuant Pro visible spectrophotometer (Amersham, USA). DNA samples were diluted to 30 ng µL-1 for PCR amplification and stored at -20℃ prior to ISSR analysis. Sixty ISSR primers were designed (Beijing Genomics Institute, Beijing, China) and screened; 16 primers that produced clear and reproducible fragments were selected for further analyses (Table 3). PCR amplifications were carried out in 20 µL reaction volumes containing 30 ng DNA, 1× PCR buffer (Mg2+), 10 µM primers, 2.5 mM of each dNTP, and 5 units EasyTaq DNA polymerase (Beijing TransGen Biotech) using a 2720 Themal Cycler ( Applied Biosystems). The ISSR amplification program consisted of an initial denaturation step of 94℃for 5 min, followed by 45 cycles of 94℃for 30 s, 48–60℃for 45 s, and 72℃for 90 s, with a final extension step of 72℃ for 10 min. Primer annealing temperatures are shown in Table 3. The amplified products were separated on 2.0% agarose gels in 1× TAE buffer (20 mM Tris-HCl and 2 mM EDTA, pH 8.0), stained with ethidium bromide, and photographed under ultraviolet light using a Gene Genius Bioimaging System (Syngene).

Table 3 ISSR primer sequences and polymorphismNote: *R=(A, G); Y=(C, T)

Data analysis
Morphological analyses were carried out using univariate methods for calculating arithmetic means, SDs, and CVs in each population. Significance was tested based on Tukey’s test using SPSS18.0 software. ISSR bands were scored using binary characters of present (1) or absent (0). The resulting presence/absence data matrix was analyzed using POPGENE v.32 to estimate the percentage of polymorphic bands (PPB). Genetic identity (I), Nei’s genetic distance (D), and Shannon’s information index (Ho) were calculated using NTSYSpc 2.2.
This work was funded by a National “Twelfth Five-Year” Support Project of China (No. 2011BAI13B01) and an Important Specific Project of the Development and Reform Commission of Yunnan Province (No. 20112513).

Acquaah G., 2007, Principles of Plant Genetics and Breeding, Blackwell, United Kingdom
Chen Z.J., Cui X.M., Sun Y.Q., and Wang C.L., 2004, Correlation and path analysis of agronomic characters of Panax notoginseng. China Journal of Chinese Materia Medica, 29(1): 37-39
Chen Z.J., Wang Y., Ceng J., Li Z.Y., and Tian Y.L., 2001, Study on individual trait variations in Panax notoginseng and their effects on yield and quality, Chinese Traditional and Herbal Drugs, 4: 357-359
Chen Z.J., 2001, The main component factor of per plant yield in Panax notoginseng, Yunnan Agricultural Science and Technology, 2: 22
Doyle J., 1991, DNA protocols for plants –CTAB total DNA isolation. In: Molecular Techniques in Taxonomy (eds GM Hewitt & A Johnston), 283–293. Springer, Berlin, Germany
Guo H.B., Cui X.M., An N., and Cai G.P., 2010,Sanchi ginseng (Panax notoginseng (Burkill) F. H. Chen) in China: distribution, cultivation and variations. Genetic Resources & Crop Evolution, 57: 453-460
Hong D.Y.Q., Lau A.J., Leo C.L., Kiu X.K., and Yang Y., 2005, Genetic diversity and variation of saponin contents in Panax notoginseng roots from a single farm, J Agric Food Chem., 53: 8460-8467
Jiang K.Y., and Qian Z.N., 1995, Effects of Panax notoginseng saponins on post hypoxic cell damage of neurons in vitro, Acta Pharmacol Sin., 16:399-402
Kuriyama H., 1950, The cytological studies on the interspecific hybrid of Panax, Japanese Journal of Crop Science, 1-2: 209-213
Lee M.J., Choi J.S., Cha S.W., Lee K.S., Lee Z.W., Hwang G.S., Lee S.H., Kamal A.H.M., Jung Y.A., Seung N.S., and Woo S.H., 2011, Variation in the ginsenoside profiles of cultivated ginseng (Panax ginseng C.A. Meyer) landraces in Korea, Process Biochemistry, 1: 258-264
Li S.H., and Chu Y., 1999, Anti-inflammatory effects of total saponins of Panax notoginseng, Acta Pharmacol Sin., 20: 551-554
Mousseau T.A., Sinervo B., and Endler J.A., 2000, Adaptive genetic variation in the wild. Oxford University Press, Oxford
Schlag E.M., and McIntosh M.S., 2006, Ginsenoside content and variation among and within American ginseng (Panax quinquefolius L.) populations. Phytochemistry, 14: 1510-1519
Schluter C., and Punja Z.K., 2000, Floral biology and seed production in cultivated north American ginseng (Panax quinquefolius), J Amer Soc Hort Sci, 5: 567-575
Singh R.J., Lebeda A., and Tucker A.O., 2011, Medicinal plants—Nature’s pharmacy. 13-51. In: Singh RJ. Genetic Resources, Chromosome Engineering, and Crop Improvement: Medicinal Crops, Volume 6, CRC Press
Sun Y.Q., Chen Z.J., Li Y.G., Wang C.L., and Cui X.M., 2003, Observation of individuals traits variations in Panax notoginseng, Research and Practice of Chinese Medicines, supp: 16-17, 18
Susumu K., Manabu K., Takahiro T., Shuhichi T., Sizue Y., Youichi M., Yoshio N., Kazutoshi O., Shyuichi F., and Michio T., 1997, Ginsenoside production, tissue culture and induction of regeneration of interspecific hybrid ginseng : Panax ginseng × P. quinquefolium. Natural medicines 1: 1-3
Takahasi N., 1940, Osumi T., On the inheritance of the colour of stem and fruit in Panas ginseng, C. A. Meyer, Japanese Journal of Genetics, 16: 273-276
Wang D., Hong Deborah, Koh H.L., Zhang Y.J., Yang C.R., and Hong Y., 2008, Biodiversity in cultivated Panax notoginseng populations. Acta Phamacol Sin., 9: 1137-1140
Wang W., Zhang X., Qin J.J., Voruganti S., Nag S.A., Wang M.H., Wang H,. and Zhang R.W., 2012, Natural product ginsenoside 25-OCH3-PPD inhibits breast cancer growth and metastasis through down-regulating MDM2. PLoS One, 7: e41586
White C.M., Fan C., and Chow M., 2000, An evaluation of the hemostatic effect of externally applied notoginseng and notoginseng total saponins, J Clin Pharmacol., 40: 1150-1153
Xiao H., Jin H., Zhang J.Y., Xiao F.H., Duan C.L., Cui X.M., 2008, Study on phenotypic diversity of Panax notoginseng, Southwest China Journal of Agricultural Science, 1: 147-151
Yin L.M., Wang X., Qian X.D., Lin X.J., Chen X.H., and Gao R.L., 2012, Effects of Panax notoginseng saponins on proliferation and differentiation in NIH3T3 Cells, Chin J Integr Med., 8: 616-620
Zhang J.Y., Yang W.Z., Cui X.M., Jin H., Yu H., Chen Z.J., Shen T., and Yang T., 2011, Genetic diversity analysis of populations of Panax notoginseng using EST-SSR markers, Journal of Plant Genetic Resources, 2: 249-254
Zhao S.J., Li F.Y., Zhao Y.H., Liu Y.Z., Sun C.W., Huang Z.H., Wu L.J., and Jin Y.H., 1998, Studies on selection theory for high-yield variety of Panax ginseng and breeding of Jishen 1. Scientia Agriculture Sinica, 5: 56-62
Zheng G.Z., and Yang C.R., 1994, Biology of Panax notoginseng and its application, Beijing: Science Press

Zhu S., Fushimi H., Cai S.Q., Chen H.B., 2003, Komatsu K., A new variety of the genus Panax from Southern Yunnan, China and its nucleotide sequences of 18S ribosomal RNA gene and matK gene. The Japanese Journal of Botany, 78: 86-94

Molecular Plant Breeding
• Volume 4
View Options
. PDF(430KB)
. Online fPDF
Associated material
. Readers' comments
Other articles by authors
pornliz suckporn porndick pornstereo . Yun Yang
. Junwen Chen
. Ming Zhao
. Cuiting Li
. Zhengui Meng
. Jianjun Wang
. Zhongjian Chen
. Guanghui Zhang
. Shengchao Yang
Related articles
. Panax notoginseng
. Mass selection
. Phenotypic uniformity
. Genetic identity
. Email to a friend
. Post a comment