1 Introduction

Gene flow refers to the transfer of genes or genetic material from one plant to another (Ellstrand 2003; Chandler and Dunwell 2008; Heuberger et al. 2011; Chen et al. 2016) and occurs via three routes; pollen, seed, and vegetative propagule-mediated (Qaim 2009; Kang et al. 2016). Pollen-mediated gene flow is influenced by physical and environmental conditions such as the distance between pollen donors and recipients, temperature, relative humidity and strength and direction of air flow (Husken et al. 2010). Gene flow is a major factor affecting the purity of crop cultivars (Dong et al. 2016) and therefore is assessed when conducting risk assessments of genetically modified (GM) plants (Chandler and Dunwell 2008; Jamal et al. 2012; Bohn et al. 2016).

In recent years, the increased use of GM plants has required the establishment of biosafety rules to ensure the low escape of exogenous genes from GM to non-GM crops (Kwit et al. 2011; DiFazio et al. 2012). Several previous studies have investigated pollen-mediated gene flow in the field (Goggi et al. 2007; Baltazar et al. 2015; Miroshnichenko et al. 2016); however, no data are available on the contamination risk via pollen under greenhouse conditions. When GM plants are cultivated in restricted spaces, such as a greenhouse, the risk of pollen contamination is lower than in external environment because temperature, wind and humidity are adjusted to restrict pollen dispersal. Yet, pollen contamination, in greenhouse conditions, may occur due to the narrow spacing between plants.

GM technology enables the production of compounds and molecules with beneficial properties in high amounts. Previously, we reported the potential of using plants to produce the vaccine candidate GA733, which is an epithelial cell adhesion molecule highly expressed in colorectal cancer cells (Mosolits et al. 1999). The recombinant protein GA733-FcK produced in plants was able to induce immunogenic responses in animal models (Lu et al. 2012; Lim et al. 2015). In this study, we investigated whether growth of tobacco transgenic plants expressing the GA733-FcK under greenhouse conditions resulted in contamination of wild type (WT) plants via pollen-mediated gene flow.

2 Materials and methods

2.1 Plant materials and experimental fields

WT and transgenic Nicotiana tabacum plants expressing the epidermal cell adhesion molecule GA733, which is a cancer vaccine candidate, were used (Kim et al. 2009; Lu et al. 2012). To monitor the flow of transgenes from transgenic to WT plants, both screening and selection were undertaken using GA733 fused to an IgG HC fragment (GA733-Fc) as target gene and the neomycin phosphotransferase II (nptII) gene as a selective marker (Lu et al. 2012). Tobacco plants were grown in greenhouse in 12 h light/12 h dark photoperiod. WT plants were placed around transgenic plants at distance intervals of 0.3, 1, 5, 10, and 20 m. Pollen-mediated gene flow from transgenic to WT plants assessed from plants having similar growth patterns and flowering times.

2.2 In vitro seed germination

Seeds were randomly collected from WT plants and were surface sterilized with in 20% ethanol and 10% sodium hypochlorite. Seeds were planted on Murashige and Skoog (MS) medium supplemented with 30 g L−1 sucrose, 6 g L−1 phyto agar, and 4.8 g L−1 MS B5 vitamin (Duchefa Biochemie, Haarlem, Netherlands), and with or without 100 mg L−1 kanamycin. Germinated seedlings were observed 4 weeks after seed sowing, and leaves were harvested for polymerase chain reaction (PCR) and immunoblot analyses.

2.3 Genomic DNA isolation and PCR analysis

The presence of transgenes in WT plant seeds was monitored using PCR. Genomic DNA was extracted from the leaves (100 mg) of 30-day-old transgenic and WT plant seedlings using DNA extraction kit (RBC Bioscience, Seoul, Korea) according to the manufacturer’s recommendations. Primer pairs used to amplify the GA733-FcK gene were 5′-CATGCCATGGATGGCTACTCAACG-3′ and 5′-ATCACAGAGCA TGAGAAGACGTTC-3′. Primer pairs used to amplify the nptII gene were 5′-CTCCCA ATCAGGCTTGATCCCCAG-3′ and 5′-CCTGCTAAGGTATATAAGCTGG TG-3′. The EF1α gene (Actin) was used as a reference gene. PCR analysis was conducted more than three times.

2.4 Immunoblot analysis

Total soluble protein was extracted from 100 mg of fresh leaves from transgenic and WT seedlings by cryomilling. The homogenized plant samples were mixed with 300 μL of 1X PBS (137 mM NaCl, 10 mM Na2HPO4, 2.7 mM KCl, and 2 mM KH2PO4). Leaf extracts were boiled with 5 × protein loading buffer (1 M Tris-HCl, 50% glycerol, 10% SDS, 5% 2-mercaptoethanol, and 0.1% bromophenol blue) for 10 min and then cooled for 2 min. Total soluble proteins (16 μL) were mixed with 4 μL loading buffer and were separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE; 10% SDS) before transferred to nitrocellulose membrane (Millipore, Billerica, MA). The nitrocellulose membranes were blocked in 5% skimmed milk (Sigma, St. Louis, MO) prepared in 1X phosphate-buffered saline (PBS) for 2 h at room temperature. Proteins were detected with mouse anti-EpCAM and rabbit anti-nptII IgGs, which recognize the GA733 and nptII proteins, respectively. Blots were incubated with either horseradish peroxidase-linked goat anti-mouse IgG Fc or anti-rabbit IgG Fc antibodies (Abcam Inc., Cambridge, MA) at 1:5000 dilution in 1X PBS for 2 h. Immunoblot were developed using SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific, Rockford, IL) and visualized by exposing the membrane to X-ray film (Fuji, Tokyo, Japan). Leaf tissue extracted from non-transgenic plants was used as a negative control.

3 Results

To investigate whether transgene flow occurs under greenhouse conditions, WT plants were grown on soil pots and placed at 0.3, 1, 5, 10, and 20 m away from GA733-FcK expressing plants. Transgenic and WT plants were selected to be at similar developmental stages and flowering periods. All plants started flowering at 14 weeks after planting and retained their blossoms until 30 days after onset of flowering (Fig. 1b). Figure 2 presents the experimental design with transgenic plants being located at the center of the greenhouse and WT plants being spaced at different intervals with the longest being 20 m away (Fig. 2). During the flowering period, the average temperature during the day and night was 30 °C and 15 °C, respectively (Fig. 3). The average relative humidity was 60% at daytime while it dropped to 28% at nighttime (Fig. 3). To confirm whether GA733-FcK transgene flow occurred between the transgenic and WT plants, seeds were randomly collected from WT plants located at 0.3, 1, 5, 10, and 20 m distance intervals and were germinated on either kanamycin-supplemented or control media (Fig. 4). In spite of separation distances, all WT plant seeds produced four true leaves and healthy roots when grown on media without kanamycin, whereas growth of WT plants on media supplemented with kanamycin impaired the production of four true leaves (Fig. 4). By contrast, growth of transgenic plant seeds in control and selective media did not affect the production of four true leaves and roots (Fig. 4). Both GA733-FcK and nptII genes were expressed in transgenic plants (Fig. 5, top), whereas no gene expression was detected in WT plants (Fig. 5, middle). Actin transcripts were detected at similar levels in both transgenic and WT plants (Fig. 5, bottom). Immunoblot analysis of total soluble extracts confirmed the expression of GA733-FcK (~ 68 kDa) and nptII (~ 30 kDa) proteins in transgenic plants while no proteins were present in the WT plants (Fig. 6a, b).

Fig. 1
figure 1

Experimental design used to assess pollen-mediated gene flow between wild-type (WT) and transgenic tobacco plants. a Schematic diagram indicates the distance intervals of 0.3, 1, 5, 10, and 20 m between WT and transgenic plants. b Representative images of transgenic and WT tobacco plants grown under greenhouse conditions

Full size image
Fig. 2
figure 2

Positioning pattern of transgenic and wild type (WT) plants grown under greenhouse conditions. a Schematic diagram of the greenhouse showing the positions 0.3, 1, 5, 10, and 20 m, where WT plants were positioned relative to transgenic plants. b Representative images of flowering WT plants located at 0.3, 1, and 5 m. c Representative images of flowering WT plants located at positions 0.3, 1, 5, 10, and 20 m

Full size image
Fig. 3
figure 3

Daily fluctuations of temperature (°C) and relative humidity (%) in greenhouse, where transgenic and wild type (WT) plants were grown. Data were collected during the growing period between June and August

Full size image
Fig. 4
figure 4

Phenotype comparison of growth and germination of seedlings from transgenic and WT plants. In vitro seed culture of transgenic and wild type (WT) plants grown at 0.3, 1, 5, 10, and 20 m distance intervals; images show 30-day old seedlings grown on media supplemented with (+) or without (−) kanamycin

Full size image
Fig. 5
figure 5

Expression of GA733-FcK, nptII and Actin genes in transgenic (T) and wild type (WT) plant seedlings grown at distance intervals of 0.3, 1, 5, 10, and 20 m

Full size image
Fig. 6
figure 6

Identification of GA733-FcK and nptII proteins in transgenic and WT plants. Immunodetection of GA733-FcK (68 kDa) (a) and nptII (30 kDa) (b) carried out using mouse anti-EpCAM and rabbit anti-nptII IgGs, respectively. c Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analysis of leaf extracts from transgenic (T) and WT plants grown at 0.3, 1, 5, 10, and 20 m distance intervals

Full size image

4 Discussion

Here we demonstrated that pollen-mediated gene flow did not occur between transgenic plants expressing the recombinant GA733-FcK protein and WT plants when grown under greenhouse conditions. GA733-FcK has been considered to be a good candidate for colorectal cancer vaccination and has been successfully expressed in transgenic plants (Lu et al. 2012; Lim et al. 2015) and insect cell systems (Oh et al. 2011; Kim et al. 2015; Lee et al. 2016). Gene flow from transgenic to WT plants has often been considered to be a potential risk in the biotechnological applications (Snow 2002; Song et al. 2015). Thus, it is essential to investigate whether gene flow of transgenes occurs from transgenic to WT plants.

The current study was conducted using transgenic and WT tobacco plants grown in greenhouse environment. Growth during the late summer and early autumn was chosen to ensure the synchronous production of flowers from both transgenic and WT plants. During the experimental period, temperatures sometimes reached high values and humidity levels varied from 35% to 85%; despite these conditions, plants did not exhibit any physiological problems. In the present study, we did not observe any gene flow from transgenic to WT plants, at any of the separation distances (0.3, 1, 5, 10, and 20 m between WT and transgenic plants) under greenhouse conditions. It was particularly important to confirm that there was no gene flow from transgenic to WT plants at a distance of 30 cm apart. These results are not unexpected given the ability of tobacco plants to self-pollinate and the fact that greenhouse conditions restrict pollination by wind transport (Serrat et al. 2013). Tobacco pollen can be transferred between flowers by bees and other insects (Loureiro et al. 2016). However, in the present study, we did not detect any flow of transgenes. We speculate that insects were well controlled in the greenhouse, therefore preventing any transfer by insects of pollen from transgenic to WT plants. Overall, our current data have implications for growing transgenic tobacco plants in greenhouse conditions: that is, no gene flow occurs between transgenic and WT plants, if grown under the containment conditions outlines in this paper, with good control of insects.