Establishment and Application of a Monitoring Strategy for Living Modified Cotton in Natural Environments in South Korea

Abstract

Cotton (Gossypium hirsutum L.) is grown worldwide for its natural hollow fibers and is used as cattle feed. Living modified (LM) cotton is not cultivated in South Korea and must be imported for food, feed, and processing. From 2009 to 2013, the Ministry of Environment (MOE) and the National Institute of Ecology (NIE) conducted a natural environment monitoring and post-management initiative for living modified organisms (LMOs) in some areas to reduce the likelihood of harmful effects caused by unintentionally discharged LMOs during transportation and use. In this study, we adopted a new strategy to identify unintentionally released LM cotton plants nationwide from 2014 to 2018. A total of 451 suspicious cotton samples were collected from 3921 survey sites. Among them, we identified 255 LM cotton plants, of which approximately 72.2% had transgenic herbicide and insecticide traits. The majority of the samples were collected from the roadside along transportation routes and from stockbreeding farms. This study establishes an LMO safety management system to efficiently maintain conservation efforts in South Korea. Our findings suggest that these efforts may play a key role in safely transporting, using, and managing approved LMOs, as well as in regulating unintentionally released LMOs, in order to preserve the natural ecosystem of South Korea.

1. Introduction

Cotton (Gossypium hirsutum L.) is among the most important sources of natural plant fiber and seed oil [1,2]. Over the last five decades, the global cotton industry has seen dramatic changes, with production nearly quadrupling from 6.6 million tons in 1950/51 to a record 26.3 million tons in 2004/05. The average growth rate in global production has been approximately 2.5% per year. However, yields leveled off during the 1990s due to disease, pesticide resistance, and economic disruption. Yields began to rise again in the late 1990s as seed varieties improved and living modified (LM) cotton was introduced [3]. LM cotton is one of the major LM crops, along with soybean, maize, and canola. Cotton production is vulnerable to biotic stresses such as insect attacks and abiotic factors such as temperature, drought, and salinity [3,4,5,6,7]. LM crops are widely used in agriculture to improve crop quality by introducing new traits, including insect resistance, herbicide tolerance, higher yields, and resistance to abiotic and biotic stresses [8,9,10]. LM cotton with certain traits such as insect resistance and herbicide tolerance was also developed. Herbicide-resistant LM cotton employs epsps, bar, pat, dmo, and aad-12 genes to withstand herbicides that would otherwise destroy the crop along with the targeted weeds [11,12,13,14,15]. The use of insect-resistant LM cotton introduces Bacillus thuringiensis (Bt) toxins, such as crystal toxin (Cry) and vegetative insecticidal protein (Vip), which can reduce costs and environmental impact by discontinuing or decreasing the use of chemical insecticides [16,17,18,19]. In 2018, LM cotton plantations covered 24.9 million hectares (75.7%) of the 32.9 million hectares of global cotton cultivation, with LM cotton possessing traits for insect resistance, herbicide resistance, or both, accounting for 72.9%, 3%, and 24% of the global cotton cultivation, respectively [20].

Although the cultivation of LM crops is not permitted in South Korea, a total of 32 LM cotton events, including single genes and stacked traits, have been approved for food, feed, and processing (FFP). A total of 100,000 tons of LM cotton was imported in 2008, and since then, the rate of import has steadily increased every year [20]. However, there is growing concern about the unintentional release of LM cotton into the natural ecosystem in South Korea. As a result, the Ministry of Environment (MOE) and National Institute of Ecology (NIE) has been conducting an environmental monitoring and post-management initiative for living modified organisms (LMOs) since 2009. The MOE detected unintentionally released LMO plants in certain areas of the country from 2009 to 2013, and PCR analysis was performed on suspicious samples using the 35S promoter and nos terminator-specific primer pairs to determine the presence of transgenic events.

The aim of this study was to investigate unintentionally released LM cotton in natural environments in South Korea, classifying their collection sites and characterizing their transgenic traits using advanced analysis techniques. The study’s findings support the need for an LMO management system to reduce the risks posed by released LMOs and efficiently preserve the natural ecosystem in South Korea.

2. Materials and Methods

2.1. Survey of Natural Environments Using the Newly Developed LMO Monitoring Procedure

A nationwide survey of unintentionally released LM cotton was conducted in South Korea from 2014 to 2018, based on information from previous survey results. Natural environments were investigated using information about LMO import/transport routes. All investigators were given a guidance manual to help them identify suspicious cottons at survey sites. In accordance with the manual, samples were collected within a 100 m radius of each survey site, along with field photos and environmental information. The collection site of the suspicious sample was recorded with GPS coordinates using a Garmin Montana 680 (Garmin Inc., Olathe, KS, USA). Selected survey sites were investigated in each season, and suspicious LM cotton samples were collected, individually wrapped in paper bags, dried with silica gel, and stored at 4 °C. One leaf from each sample collected was used for LMO identification using an immunochemical strip kit and PCR analysis. A total of 451 suspicious cotton samples were collected nationwide.

2.2. LMO Trait Identification and Event-Specific PCR Analysis

Samples were analyzed using immunochemical strip kits for CP4 EPSPS, Cry1A, Cry 1F, Cry2A, PAT/bar, 2mEPSPS, DMO, and AAD-12 proteins (Envirologix Inc., Portland, ME, USA) to identify the presence of transgenic traits (herbicide or insect resistance), thus determining whether the cotton was LMO or non-LMO. Cotton leaves were ground in a 1.5 mL tube using a pestle, to which 0.5 mL of extraction buffer (Envirologix Inc.) was added until the sample was finely ground. Test strips were dipped in the mixture and bands appeared on the strips after approximately 10 min if kit-specific proteins were present. Each sample was classified as non-LMO (one band), LMO (two or more bands), herbicide-resistant LMO, and/or insect-resistant LMO.

Event-specific PCR analysis was performed to detect the presence of introduced LMO genes in the collected samples. Genomic DNA from cotton tissues was extracted using the Exgene Plant SV mini kit (GeneAll, Seoul, Korea), following the manufacturer’s recommendations. One oligonucleotide primer set for the adhC gene was used to confirm the stability of the PCR reactions. Specific PCR analysis using primer sets for 9 cotton events (T304-40, LLCOTTON25, MON1445, MON531, MON15985, MON88913, GHB614, GHB119, and 281-3006) was conducted to characterize the collected suspicious samples, as previously described (

Table 1. Oligonucleotide primers used in this study.

Target	Name	Sequence (5′-3′)	Product Size (bp)
GHB119	GHB119-F	CCAGTACTAAAATCCAGATCATGCA	327
	GHB119-R	ACTGATGGCTCAACGGTTACA
GHB614	GHB614-F	GCAGGCATGCAAGCTTTTAAA	226
	GHB614-R	TCAAGCCACAGTTTGATTGCA
MON88913	MON88913-F	GGCTTTGGCTACCTTAAGAGAGTC	299
	MON88913-R	GGCCTCCATTATTTGGCTTATTC
MON15985	MON15985-F	GTTACTAGATCGGGGATATCC	282
	MON15985-R	CAGTGGTGCCGATTCTCTTTC
LLCOTTON25	LLCOTTON25-F	CAGATTTTTGTGGGATTGGAATTC	231
	LLCOTTON25-R	TCATGAAACTTCTCACATTGGC
T304-40	T304-40-F	CCTAGATCTTGGGATAACTTGAAAAGA	277
	T304-40-R	GATCGTTCAAACATTTGGCA
MON1445	MON1445-F	GGAGTAAGACGATTCAGATCAAACAC	389
	MON1445-R	CGTGGCCTTCTATGACCGAAGTT
281-3006	281-3006-F	CTCATTGCTGATCCATGTAGATTTC	626
	281-3006-R	GCCTGAGTTGACTCTGGGGACCGGT
MON531	MON531-F	GGTACGGATGAGTAGGCCTACTCTT	629
	MON531-R	GTAGCCTCTACCTGGACAGACTCTAACC
AdhC	AdhC-F	TCCAGAGGCTCCACTTGAT	178
	AdhC-R	CCCACCCTTTTTTGGTTTAGC

) [21,22,23,24]. The certified reference materials that served as controls in the PCR reactions were purchased from the American Oil Chemist’s Society (AOSC; Urbana, IL, USA) and the Institute for Reference Materials and Measurements (IRMM; Geel, Belgium). The PCR reaction was prepared using the 2X EF-Taq PCR Pre-Mix (Solgent, Daejeon, Korea) in a reaction mixture of 30 µL total volume, containing 50 ng genomic DNA, and 1 µL per primer (10 pmol/µL). PCR amplification was performed using the ProFlex PCR System (Applied Biosystems, Waltham, MA, USA) under the following conditions: initial denaturation at 95 °C for 5 min; 33 cycles of denaturation at 95 °C for 30 s, annealing at 59 °C for 30 s, extension at 72 °C for 30 s, and a final extension at 72 °C for 7 min. The amplified products were observed by electrophoresis on 2.5% agarose gel with 1X TAE buffer for 25 min at 135 V. Bands were identified using the ChemiDoc XRS+ Imaging System (Bio-Rad, Hercules, CA, USA). The PCR-amplified products were then subjected to nucleotide sequencing by Biofact Inc. (Daejeon, Korea).

2.3. Classification of LM Cotton Collection Sites

After identification of an LM cotton sample using PCR analysis, the collection site was revisited and information about the surrounding environment was collected within a radius of 100 m, including field photos and characteristics of the site. Finally, based on the information gathered, the LMO collection site was classified as one of five types: port, roadside, stockbreeding farm, feed factory, or other.

3. Results and Discussion

3.1. Development of the New LMO Monitoring Strategy for LM Cotton Survey in Natural Environments

The “Transboundary Movement, etc., of Living Modified Organisms Act” gives seven ministries within the Korean government individual national authority over the use of LMOs. Accordingly, the MOE has monitored domestic LMO transport/use since 2009. However, the project procedures and management systems of the old LMO monitoring initiative need to be strengthened, in view of advancements in analysis techniques and performance efficiency. To enable efficient conservation of the natural ecosystem in South Korea, a new monitoring scheme was proposed. First, the new scheme covered South Korea nationwide, instead of conducting sample surveys in only some areas. Second, each suspicious sample was collected and individually analyzed, although multiple samples were collected from the same site. This enabled the calculation of the number of LMOs at each collection site. Third, event-specific PCR was performed to identify the introduced LMO gene, event name, and manufacturer of each confirmed LMO sample. This approach helped to estimate the putative risk to the receiving environment, including the soil fungal community and insect populations. Finally, classification of the LMO collection site was performed for further selection of LMO monitoring sites and post-management of released LMOs in the following year (Figure 1).

3.2. Selection of LMO Monitoring Sites and LM Cotton Survey

As shown in Figure 1, the MOE monitored and removed unintentionally released LMO plants in certain areas of the country from 2009 to 2013. However, there is a need to broaden the scope to protect all natural ecosystems in South Korea. Thus, nationwide monitoring sites for LM cotton were selected and surveyed in six administrative districts of South Korea from 2014 to 2018 (Figure 2A,B). Based on the monitoring results of the previous year, the number of survey sites was expanded from 2014 to 2018 and natural environments along LMO transport routes were investigated (Figure 2B). A total of 3921 survey sites were monitored and many suspicious samples were found in various natural environments (Figure 2). Over a five-year period, 451 suspicious samples were collected: 71 in 2014, 58 in 2015, 43 in 2016, 167 in 2017, and 112 in 2018 (Figure 2D), suggesting that the guidance manual provided to investigators played an important role in facilitating the identification of released LMOs at survey sites. Interestingly, the number of suspicious samples collected decreased from 2014 to 2016 (Figure 2D). However, the number of suspicious samples dramatically increased in 2017 and 2018 due to a massive spillage of LM cotton. In particular, unapproved LM cotton was accidentally cultivated at an exhibition site in 2017, which likely increased the number of suspicious cotton samples collected that year.

3.3. Recognition of Introduced LMO Proteins Using the Immunochemical Strip Kit Test during LM Cotton Survey

Transgenic crops expressing functional LMO proteins are being developed to reduce production costs and may improve product safety. However, the introduced proteins in LMOs may pose a variety of risk assessment and management challenges in LMO-released environments. Therefore, rapid recognition of LMO traits in collected samples during a survey may improve the efficiency of conservation efforts for the natural ecosystem. In the current study, introduced LMO proteins, such as Cry1A, Cry1F, Cry2A, Vip3A, EPSPS, PAT/bar, DMO, and AAD-12, were detected quickly in collected samples using immunochemical strip kits at the environmental survey sites. Over a five-year period, 255 and 196 cotton plants were identified as positive (LMO) and negative (non-LMO), respectively (Figure 3A). Assessing LMO traits with strip kits enabled the fast identification of unintentionally released LM cotton in South Korea as herbicide-resistant, insect-resistant, or both (Figure 3B). The percentages of LMO traits in the presence of stacked traits (more than one gene) for insect resistance, herbicide resistance, or both were 24.3%, 3.5%, and 72.2%, respectively (Figure 3B).

Event-specific PCR analysis was performed using the endogenous reference gene for cotton and event-specific introduced genes to identify the event name, manufacturer, and type of trait. As shown in

Table 2. Number of LM cotton plants, event name, and trait type by year.

Event	Trait	Number of LM Cotton
		2014	2015	2016	2017	2018
MON88913	Herbicide resistance	2	2	2	0	0
LLCOTTON25		0	1	0	1	0
MON1445		0	0	1	0	0
MON531	Insect resistance	0	4	8	44	4
MON15985		1	1	0	0	0
MON15985×MON88913	Herbicide/insect resistance	15	8	7	11	58
MON88913×281-3006		7	13	3	1	9
MON88913, MON15985, 281-3006 related stack		15	0	0	0	0
281-3006		0	3	6	2	0
MON1445×MON531		0	2	0	1	8
LLCOTTON25×MON15985×GHB614		0	6	0	1	1
MON15985×LLCOTTON25		3	0	0	0	0
GHB614×T304-40×GHB119		0	0	2	0	2
Total		43	40	29	61	82

, 13 event types were detected in 255 samples, and a variety of LM cotton events were identified each year: 43 in 2014, 40 in 2015, 29 in 2016, 61 in 2017, and 82 in 2018. Nucleotide sequencing analysis was used to investigate event information in the LM cotton samples, revealing that the number of LMO-positive populations and trait types matched exactly with the data in Table 2 and Figure 3. LM cotton was identified nationwide, and its abundance dramatically increased in 2018, which could be due to the expansion of survey areas for unapproved LMO cultivation following a number of unintentional releases of LM cotton at an exhibition site in 2017. Furthermore, based on classification analysis of LMO collection sites from 2014 to 2017, survey sites of natural environments near stockbreeding farms and transportation routes were expanded, and the number of LM cotton samples collected was substantially increased in 2018. Most LM cotton possessed herbicide- and insect-resistance traits. Except for 2014, the single event MON531 was detected in all periods, and stack events such as MON15985×MON88913 and MON88913×281-3006 were discovered every year. In particular, MON15985×MON88913 accounted for approximately 39% of all identified LM cotton (Table 2). These results could be helpful for managing LMO-released environments through physical elimination and/or herbicide or pesticide application depending on the specific LMO events identified in unintentionally released LMOs.

3.4. Classification Analysis of LMO Collection Sites

The classification of LMO-released locations is necessary for better monitoring performance and post-management in the following year. If the monitoring strategy in natural environments is to be made meaningful, it must be accompanied by risk assessment at LMO collection locations. Therefore, LMO collection sites were evaluated for 5 years for re-survey and post-management purposes. As shown in Figure 4A, the LM cotton plant population was spread across the country. If a stockbreeding farm or feed factory was located within 100 m of where the cotton was discovered growing in a natural environment, the collection site was classified as a stockbreeding farm or feed factory, respectively. Environments that did not fall into one of the four specific categories (port, roadside, stockbreeding farm, or feed factory) were classified as “other”, which included festival and planting sites. Over a five-year period, 255 collection locations were analyzed, among which 32.2% were roadside areas, 30.2% were stockbreeding farm areas, 9.8% were feed factory areas, and 27.8% were other areas (Figure 4B,C). Furthermore, analysis of the collection sites indicated that LM cotton plants were mainly released in and growing in natural environments near stockbreeding farms and transportation routes every year (Figure 4B). These findings may be useful for re-surveys, risk assessment, and post-management in all LM cotton release locations.

3.5. Establishment of LMO Safety Management System

From 2014 to 2018, the number of collected LM cotton plants increased, owing in part to advances in the LMO monitoring process. However, as time went on, the post-monitoring process became more challenging. This is because, as the numbers of management areas increased, the capacity to cover the increased areas decreased. The LMO monitoring project is important for the fulfillment of the LMO safety policy and ongoing research by the MOE. For this reason, it is necessary to establish an LMO safety management system for efficient maintenance of the conservation of the natural ecosystem in South Korea.

Based on the study results, we proposed an LMO safety management system that combined LMO safety policy and research with LMO monitoring practices (Figure 5). The new LMO monitoring strategy is divided into four stages: survey, LMO identification, data analysis, and post-management, which are individual parts of LMO policy and research by the MOE. The survey step affects LMO import/export/use, while the LMO identification step is part of the LMO policy’s biosafety plan. Furthermore, the analysis and post-management steps are linked to the LMO risk review and risk assessment/management in a sequential manner. If unintentional spills of LMOs into the natural environment occur during transportation and use, the MOE and the NIE remove the released LMOs to prevent further dispersal into the natural environment. Furthermore, the putative risk of the LMO collected environments is estimated, and it is determined whether the overall risk in the environments is acceptable and/or manageable. All information from the risk assessment is used for further LMO risk review in South Korea as part of the LMO approval process. Finally, the results of the survey, risk assessment, and review are used for the establishment of the next biosafety plan of the MOE to manage the import/export/use of LMOs in South Korea. To preserve the natural ecosystem in South Korea, this effort is critical for transporting, using, and managing approved LMO, as well as the post-management of released LMO into the natural environment.

4. Conclusions

This study confirms the importance of the LMO monitoring strategy based on a five-year survey and post-management of unintentionally released LM cotton in natural environments in South Korea. The NIE adopted a new strategy of LMO monitoring to cover South Korea nationwide, identify LMOs, and estimate the putative risk of the LMO collected environments. The application of this strategy revealed that our new strategy has higher accuracy, the convenience of a nationwide survey, and is more cost-effective than the old survey. This scheme has the potential to be an effective tool for environmental surveys and LMO post-management in a number of countries, including South Korea. The NIE then established an LMO safety management system, which integrated LMO safety policy and research with LMO monitoring practices. The new LMO monitoring scheme and proposed LMO safety management system by the NIE and the MOE may play an important role in reducing the risks posed by LMOs and preserving South Korea’s natural ecosystem.