Article information

Korean J. Pl. Taxon. 2022;52(3):184-195

Publication date (electronic) : 2022 September 30

doi : https://doi.org/10.11110/kjpt.2022.52.3.184

Eun Su KANG, Han Gyeol KIM¹, Myoung Ja NAM², Mi Jung CHOI³, Dong Chan SON^,*

Division of Forest Biodiversity, Korea National Arboretum, Pocheon 11186, Korea

¹Conservation Center for Gwangneung Forest, Korea National Arboretum, Pocheon 11186, Korea

²The Society for Korean Peninsula Plants (Citizen Researcher), Anyang 14069, Korea

³Division of Variety Examination, National Forest Seed & Variety Center, Chungju 27495, Korea

강은수, 김한결¹, 남명자², 최미정³, 손동찬^,*

국립수목원 산림생물다양성연구과

¹국립수목원 광릉숲보전센터

²한반도식물연구회

³국립산림품종관리센터 품종심사과

Corresponding author Dong Chan SON, E-mail: sdclym@korea.kr

Received 2022 March 25; Revised 2022 September 05; Accepted 2022 September 20.

Trans Abstract

Abstract

INTRODUCTION

Brassicaceae Burnett comprises 351 genera and 4,000 species worldwide, mostly distributed across the temperate regions of the northern and southern hemispheres (Abdelhameed et al., 2020). The family generally has typical morphological characteristics: tetradynamous stamens, four cruciform or butterfly-like petals, and capsules with two locules divided by septa (Beilstein et al., 2006).

Brassicaceae members having conduplicate cotyledons and the presence or absence of simple hairs are classified into tribe Brassiceae, and some of those with dehiscent fruits at maturity are further subdivided into subtribe Brassicinae (Kadereit, 1994). Meanwhile, most members of Brassica have completely dehiscent fruits at maturity, whereas some plants with partially dehiscent fruits can even be classified as Coincya Rouy (Leadlay and Heywood, 1990; Kadereit, 1994; Al-Shehbaz et al., 2006a, 2006b; Nagpal et al., 2008; Al-Shehbaz, 2012). Owing to the highly variable morphological characters, however, taxonomy of some taxa within Brassica is debatable. Therefore, comprehensive studies for both morphology and molecular patterns are required to elucidate the taxonomic delimitations of some taxa.

In the present study, we report the invasive alien species Brassica tournefortii Gouan for the first time in South Korea. It was first detected in Gunsan-si of Jeollabuk-do in 2018, after which it was also found in Gochang-gun of Jeollabuk-do and Jeju-si of Jeju-do in 2020 (Fig. 1). Moreover, this plant has been treated as either B. tournefortii or Coincya tournefortii (Gouan) Alcaraz, T. E. Díaz, Rivas Mart. & Sánchez-Gómez (Alcaraz Ariza et al., 1989; The Plant List, 2013; POWO, 2022). Hence, its taxonomic position was reviewed by comparing its morphological characteristics with those of related species in South Korea and by using nuclear ribosomal DNA internal transcribed spacer (ITS) sequence data—which is a useful marker for phylogenetic analysis of the Brassicaceae (Álvarez and Wendel, 2003; Qi et al., 2007; Warwick et al., 2010; Salariato et al., 2013)—obtained from collected materials in Gochang-gun and registered in the National Center for Biotechnology Information (NCBI) database.

Fig. 1.

Distribution map of Brassica tournefortii Gouan in Korea. A–C. Gunsan-si, Gochang-gun, and Jeju-si.

The species’ potential distribution under climate change scenarios was also modeled based on the current species distribution information (GPS data) in Gunsan-si, Gochang-gun, and Jeju-si. Invasive alien plant tend to have high dispersal abilities, rapid growth with short generation times, and high tolerance of broad environmental conditions. Therefore, climate-related changes will almost certainly lead to changes in the distribution of alien plant species, as their populations respond to variability and changes in temperature, precipitation, and biotic interactions. Predicting how invasive species will respond under potential climate change scenarios is difficult but understanding alien plant species’ current and future distribution patterns under different climate scenarios may provide primary data that can be used for invasion risk assessment (Moran and Alexander, 2014).

The findings of the present study elucidate the identification of Brassica in South Korea, besides providing data on the distribution of B. tournefortii, which can be a fundamental resource for the invasion risk assessment and effective management of this alien plant species.

MATERIALS AND METHODS

Molecular genetic analysis

DNA extraction and PCR

Three samples of fresh leaves of B. tournefortii collected from Gochang-gun in 2021 were stored in silica gel and completely dried. A tissuelyser (Mixer Mill MM 200, Retsch, Haan, Germany) was used for pulverization, and total genomic DNA was extracted using the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany). The PCR mix consisted of 1 μL of DNA (extracted using the AccuPower PCR PreMix; Bioneer, Daejeon, Korea) with 10 μmol of ITS1 (forward) and ITS4 (reverse) primers with 1 μL and 17 μL of distilled water, respectively. PCR was performed in a DNA thermal cycler (Biometra, Whatman Co., Göttingen, Germany), with 35 cycles of 5 min at 94°C, 1 min at 94°C, 1 min at 54°C, and 1 min at 72°C, followed by 7 min at 72°C for fixation. PCR product purification and sequencing were performed by Macrogen, Co., Ltd. (Seoul, Korea).

Phylogenetic analysis

The ITS sequence data of B. tournefortii from Gochang-gun were compared with ITS sequences of related genera and species obtained from the NCBI database. In total, 20 species of Brassica and four of Coincya were included. Sisymbrium officinale (L.) Scop. and Barbarea orthoceras Ledeb. were used as outgroups. The sequences were aligned using MAFFT in Geneious ver. 8.0.5 (https://www.geneious.com). Maximum likelihood (ML) analysis was performed using 5,000 bootstraps and the STR + R2 model in the IQ-TREE 1.6.8 software (http://www.iqtree.org).

Potential distribution analysis

To predict the projection and change of potential distribution according to the current and climate change scenarios of B. tournefortii, the distribution coordinates of each B. tournefortii individual were collected from Gochang-gun, Gunsan-si, and Jeju-si. For predictive analysis, the SSDM package was used in the RStudio program based on the collected distribution coordinates (Schmitt et al., 2017). The Ensemble Species Distribution Models (ESDMs) fits nine algorithms, explores the prediction range across different species distribution models (SDMs), and then finds consensus among the SDM predictions (Table 1). The 23 environmental factors most commonly applied were used (e.g., mean annual temperature and precipitation) (Table 2).

Table 1.

The nine algorithms applied to the species distribution modeling of Brassica tournefortii Gouan.

Table 2.

The 23 abiotic variables used in the species distribution modeling of Brassica tournefortii Gouan and their respective descriptions.

To determine the potential distribution of B. tournefortii in South Korea under future climate trajectories, we applied the 4.5 and 8.5 scenarios of the Representative Concentration Pathway (RCP) for three different periods (current, 2050, and 2070). The RCP 4.5 scenario, as presented in the 5th IPCC report, is a stabilization scenario in which the total radiative forcing reaches 4.5W/m² by 2100 and stabilizes due to the employment of various technologies and strategies to reduce greenhouse gas emissions (Shrestha and Bawa, 2014). The RCP 8.5 scenario predicts that terrestrial mean temperature and precipitation will increase by 4.8°C and 6.0%, respectively, if there is a continued lack of efforts to reduce greenhouse gas emissions (IPCC, 2013).

By presenting only the potential distribution and the significantly correlated factors in our analysis according to the current and climate change scenarios, rather than comparing models, the focus was on predicting changes in the distribution of an alien species.

RESULTS

Morphological analysis

The basal leaves of young Brassica tournefortii have deep dissected lobes and are densely hirsute (Fig. 2B, F, G), closely resembling the radical leaves of young Cirsium japonicum Fisch. ex DC. var. maackii (Maxim.) Matsum. In addition, the flowering plant is similar to those of the species of genera Sisymbrium Burnett on Raphanus L. (Fig. 2B, I, J), but the characteristic of the fruit, i.e., dehiscence into two valves along the suture at maturity, confirms that the species belongs to Brassica (Fig. 2O) (Kadereit, 1994; Al-Shehbaz et al., 2006a, 2006b; Oh, 2007).

Fig. 2.

Brassica tournefortii Gouan in Korea. A. Habitat in Gochang-gun. B. Plant. C. Root. D. Basal stem E. Upper stem. F. Adaxial basal leaves. G. Abaxial basal leaves. H. Cauline leaves (left, adaxial; right, abaxial). I. Inflorescence. J. Flower. K. Sepals. L. Petals. M. Stamens. N. Nectar glands. O. Fruits. P. Seeds. Q. Seed surface.

Apart from these features, B. tournefortii can also be easily distinguished from the two other Brassica species, B. juncea (L.) Czern. and B. napus L., by many branches basally and distally on the stem, hairs on proximal stems, remain seeds in indehiscenced fruit beak, seeds with mucilaginous seed coats in the presence of moisture (Fig. 2B, D, P, Q) (Kadereit, 1994; Al-Shehbaz et al., 2006a, 2006b; Oh, 2007).

Molecular genetic analysis

In total, 28 ITS sequences were obtained from two accessions of B. tournefortii and the 24 related taxa (with two outgroups: Barbarea orthoceras and Sisymbrium officinale) (Fig. 3). The total alignment length of the ITS was 580 base pairs, and 74 variable sites were found. The identical sites were 340 base pairs (58.6%). The overall GC ratio was 53.5% for ITS.

Fig. 3.

Maximum likelihood (ML) tree of internal transcribed spacer sequence data of Brassica tournefortii Gouan and related species. Numbers above branches indicate bootstrap support values.

In ML trees, all 26 accessions of Brassica and Coincya species formed a monophyletic group (bootstrap value; BS = 97), excluding the outgroups (Fig. 3). Members of Brassica were separated into two clades, one of which formed a cluster with Coincya, even though overall node support was not robust (BS = 51). The identity of the Gochang-gun individual was B. tournefortii, as strongly supported by BS (BS = 100).

Potential distribution analysis

The potential distribution of Brassica tournefortii in the current climate condition was predicted, including the coastal regions where the plants were recorded (Gunsan-si, Gochang-gun, and Jeju-si), covering 2.2% of the total area of South Korea (Fig. 4A). In the RCP 4.5 scenario, the distribution was projected to reduce to 2.2% by 2050 and to 1.6% by 2070 (Fig. 4B, C). In both cases, bio 14 (precipitation of the driest month) was assessed as a significant factor (Fig. 5B, C). In this scenario, compared to the result that predicted the current climate condition, the probability of distribution of B. tournefortii decreased in Gochang-gun and Jeju-si but increased on the islands of the Yellow Sea. However, in general, the area of potential distribution decreased with time. In the RCP 8.5 scenario, the potential distribution was projected to increase to 1.5% by 2050, with the contribution of bio 3 (isothermality), and to 1.9% by 2070, influenced by bio 14 (Figs. 4D, E, 5D, E). In this scenario, the probability of distribution of B. tournefortii decreased in Gunsan-si and Gochang-gun and then increased in Gunsan-si, Gochang-gun, and Jeju-si by 2070.

Fig. 4.

Potential distribution of Brassica tournefortii Gouan under the current scenario and climate change scenarios. A. Present. B. RCP4.5 2050. C. RCP4.5 2070. D. RCP8.5 2050. E. RCP8.5 2070.

Fig. 5.

Relative contribution rates of bioclimatic variables used for creating the models. A. Present. B. RCP4.5 2050. C. RCP4.5 2070. D. RCP8.5 2050. E. RCP8.5 2070 (relative contribution of variables > 5).

DISCUSSION

Brassica tournefortii, unlike other Brassica species in South Korea, has pinnatisect basal leaves that are densely hirsute. Thus, at young stages, it is sometimes confused with either Sisymbrium or Raphanus species (Fig. 2B, F, G). However, they are very different in the presence of bracts on racemes, fruit shape, and cotyledon type of seeds (Kadereit, 1994; Al-Shehbaz et al., 2006a, 2006b; Oh, 2007). B. tournefortii and Raphanus species have racemes without bracts, globose seeds, and conduplicate cotyledons, whereas Sisymbrium species have racemes with bracts, flattened seeds, and incumbent cotyledons (Fig. 2B, P, Q) (Kadereit, 1994; Al-Shehbaz et al., 2006a, 2006b; Oh, 2007) (Fig. 2P, Q). Conversely, B. tournefortii can be distinguished from Raphanus species by morphological features such as yellow or light yellowish petals (vs. Raphanus, which has purple, pink, yellow, or white petals) and dehiscent fruits along the suture (vs. Raphanus, which has indehiscent fruits, often breaking transversely into segments fruits) (Fig. 2J, L, O) (Kadereit, 1994; Al-Shehbaz et al., 2006a, 2006b; Oh, 2007).

The two other Brassica species from South Korea, B. juncea and B. napus, have basal leaves that are not pinnatisect but lyrate to pinnatifid and are glabrous or sparsely hirsute; therefore, B. tournefortii can be easily distinguished from these two species. Moreover, both B. juncea and B. napus are branched at the upper part of the stem, have completely dehiscent fruits at maturity, with seed coats that do not turn mucilaginous when in contact with water, but B. tournefortii has stems branched basally and distally, partly indehiscent fruit beaks, and seed coats that become mucilaginous with moisture (Fig. 2B, D, P, Q) (Al-Shehbaz et al., 2006a, 2006b; Oh, 2007).

Meanwhile, one Brassicaceae member, the genus Coincya, shares a typical trait of partly indehiscent fruits; thus, some taxonomists consider this species as belonging to Coincya (Leadlay and Heywood, 1990; Kadereit, 1994; Al-Shehbaz et al., 2006a, 2006b; Nagpal et al., 2008; Al-Shehbaz, 2012). However, the existence of one vein on the fruit valves reveals that B. tournefortii really belongs to Brassica, as Coincya has three or more veins (Leadlay and Heywood, 1990; Al-Shehbaz et al., 2006a) (Fig. 2O). Furthermore, our ML tree showed that Coincya is not monophyletic because it clusters with the other group of Brassica, and B. tournefortii is more closely related to Brassica than to Coincya (Fig. 3). Therefore, we herein classify it as a Brassica member.

Brassica tournefortii is native to North Africa and countries near the Caribbean Sea, Central Asia, and Western India (Malusa et al., 2003; Dahlin et al., 2012; Abd El-Gawad, 2014; Curtis and Bradley, 2015; Winkler et al., 2019). The species was mainly known to have naturalized and spread to Australia and southern parts of the U.S.A. (Trader et al., 2006; Barrows et al., 2009; Dahlin et al., 2012; Abd El-Gawad, 2014; VanTassel et al., 2014). It was recently reported to have been introduced into China and Japan (Curtis and Bradley, 2015; Winkler et al., 2019; Kraus et al., 2020). In the U.S.A., B. tournefortii is presumed to have been introduced through date palm import from the Middle East in 1927 (Sánchez-Flores, 2007; Marushia et al., 2012; Winkler et al., 2019).

Brassica tournefortii usually grows on coastlines and along roads and railways (Trader et al., 2006; Lillian, 2017; Rahmani et al., 2020). As it shows strong resistance to drought stress, a desert ecosystem is suitable for its growth (Choudhary and Joshi, 2001; Abd El-Gawad, 2014; Winkler et al., 2018; Rahmani et al., 2020). The lifecycle of B. tounefortii is short, with two to three generations within a year; it produces thousands of seeds at a time, which is advantageous for dispersion; and the seeds have a high rate of gemination. These factors make this species highly invasive (Bangle et al., 2008; Dahlin et al., 2012; Abd El-Gawad, 2014; Curtis and Bradley, 2015). B. tournefortii seeds can remain dormant in an environment unfavorable for germination, and they can withstand flooding as the seed coats become mucilaginous when supplied with excessive moisture (Sánchez-Flores, 2007; Bangle et al., 2008; Dahlin et al., 2012; Abd El-Gawad, 2014; Berry et al., 2014; Curtis and Bradley, 2015). These mechanisms also allow the seeds to adhere to animals, further facilitating dispersion (Li et al., 2015; Winkler et al., 2018, 2019).

The spread of B. tournefortii is considered a chief factor in the reduction of biodiversity in desert ecosystems. In desert regions in the south of the U.S.A., the increase in annual precipitation and transport infrastructure has increased the spread of B. tournefortii as compared to the early distribution regions of the species (Malusa et al., 2003; Schiermeier, 2005; Barrows et al., 2009; Marushia, 2009; Dahlin et al., 2012; Marushia et al., 2012; VanTassel et al., 2014; Winkler et al., 2019), which in turn has caused it to monopolize resources previously used by native plants and disturb the habitats of small animals such as lizards (Trader et al., 2006; Sánchez-Flores, 2007; Bangle et al., 2008; Dahlin et al., 2012; Abd El-Gawad, 2014; Berry et al., 2014; Curtis and Bradley, 2015; Rahmani et al., 2020). The spread of this invasive alien plant was found to have also influenced the frequency of fires in the desert (Steers, 2008), prompting studies worldwide to identify the growth, ecological, and physiological characteristics of the species to control its dispersion and invasion (Chauhan et al., 2006; Trader et al., 2006; Marushia, 2009; Abd El-Gawad, 2014; Alfaro and Marshall, 2019). In particular, notable efforts have been made in the U.S.A., where B. tournefortii was designated as an invasive alien plant to prevent its spread and damage (Chauhan et al., 2006; Lillian, 2017; Kraus et al., 2020).

Brassica tournefortii, which inhabits in arid and semi-arid arears, showed narrow range of potential distribution according to the current environmental envelope in South Korea, and it was evaluated that the potential distribution area shown to not increase under the climate change scenarios (Fig. 4). However, the studies of Winkler et al. (2018, 2019) revealed that B. tournefortii exhibits population variability in germination and growth according to the local seasonal precipitation. Therefore, it is assumed that this species can adopt to trait of local environment, and increasing reproduction and competitiveness with native species. In Korea, ecological studies have not been conducted to date on B. tournefortii in the three aforementioned regions under human interference and specific environmental conditions (e.g., nutrient distribution; seed production; vegetation; competing species; soil; and the biological interaction network, including herbivores). Instead, considering the studies on the species’ dispersion in the U.S.A. and its growth and ecological characteristics, the probability of rapid spread in South Korea can be regarded as extremely low due to the lack of a broad prairie or desert area. The potential distribution of B. tournefortii was assessed that this species’ dispersion depends on spring and autumn precipitation, with little increase predicted for the concerned regions (Figs. 4, 5).

It is assumed that these results arise from B. tournefortii recent introduction to Korea (first detection in Gunsan-si in 2018) and main distribution along the coast. However, recently introduced alien plants ensure sustainability by forming a soil seed bank in a given region and spreading to surrounding regions, and these may be supported by macroscale climate factors and complex interactions in the ecosystem (Lim et al., 2020; Park and Choi, 2020). Under low conditions of inter- and intra-specific resource competition, B. tournefortii produces a high amount of seeds (Trader et al., 2006). The species’ characteristic onset in early spring, when no other species are growing, may enable the dispersion of B. tournefortii from the site of introduction in South Korea. This survival strategy of B. tournefortii may enable a continued invasion in future climate change scenarios through self-fertilization and propagation, and it is expected to compete with native species. In addition, future spread to island and coastal regions, cultivated lands, disturbance zones with sand-rich soil, and agricultural lands with environments similar to the current habitats is possible. Thus, preparation against the uncontrolled spread of the species, especially under anthropogenic pressure, is necessary. A possible measure for preventing the species’ migration toward inland regions is the removal of B. tournefortii plants along sand bed channels and dry lands at banks.

SDM analysis using distribution information of recently introduced alien plants revealed narrow predicted areas because a narrow environmental factor distribution range was used. Although the approach according to the current conditions and climate change scenarios has these limitations, it clearly has the advantage of rationally extracting uncertain distribution conditions at the beginning of the introduction of exotic plants. The results of this study will be helpful in various studies and vegetation management activities necessary for exploring the distribution of exotic plants and understanding species characteristics.

Article information Continued