| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a VTT Biotechnology, P.O. Box 1500, FIN-02044-VTT, Finland
b Boreal Plant Breeding Ltd, Myllytie 8, FIN-31600 Jokioinen, Finland
c Ministry of Agriculture and Forestry, P.O. Box 30, FIN-00023 Government, Helsinki, Finland
* Corresponding author (anneli.ritala{at}vtt.fi)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: nptII, gene coding for neomycin phosphotransferase II (NPTII) • GM, genetically modified
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The first transgenic barley plants were produced by particle bombardment (Hagio et al., 1995; Jähne et al., 1994; Ritala et al., 1994; Wan and Lemaux, 1994). Later on other techniques have also resulted in transgenic barley plants (Funatsuki et al., 1995; Matthews et al., 1997; Nobre et al., 2000; Salmenkallio-Marttila et al., 1995; Tingay et al., 1997; Zhang et al., 1999). The most reproducible gene transfer method in barley has been the bombardment of embryos of the variety Golden Promise (Wan and Lemaux, 1994). The majority of gene transfers aiming at commercial applications have been carried out by this method (e.g., Jensen et al., 1996; Nuutila et al., 1999).
When varieties with novel traits are being developed, safety questions should also be considered. In the release of transgenic plants, the effect of the gene is of prime importance, not the way in which it was introduced into the genome (EUCARPIA, 1989). The safety assessment should cover the whole chain from research and production of transgenic plants to cultivation, processing, and the final use of the product (e.g., Koivisto et al., 2001; Wolfenbarger and Phifer, 2000). General ecological factors deserve attention in the first stage. Breeding can be broadly defined as the modification of a cultivated organism's genetic material for human needs (OECD, 2000). Many crop plants are fully dependent on man for their existence (EUCARPIA, 1989). During the millenia of agriculture and plant breeding, the cultivated crops have been adapted to special cultivation conditions provided by man. The cultivated crops have been bred for higher yield and for lower production of naturally occurring toxic compounds harmful to man (Ames et al., 1990; Ames and Gold, 2000; Simmonds, 1979). Thus they would usually not be competitive in nature. Most traits enhancing product quality or yield would not be beneficial for a plant in nature, and therefore such traits tend not to occur in natural populations. The majority of genetic engineering projects involve the introduction of nonadaptive traits into ecologically incompetent hosts, and belong to low-risk categories (Regal, 1994). Nevertheless, the impacts of different genetically modified plants should always be considered on a case-by-case basis according to the actual trait and plant in question.
Gene flow between plants has occurred ever since the emergence of the plant kingdom, and it has contributed to the evolution of new cultivated plant species, varieties, and weeds during the millenia of the agricultural era (e.g., Simmonds, 1979). Provided that the gene in consideration does not change the flower structure and thereby the frequency of outcrossing, gene flow of transgenes does not differ from that of other genes and common population genetic principles apply in all cases. In the risk assessment, the potential of gene flow is crucial (EUCARPIA, 1989). Even if the trait were adaptive, the lack of related plants may block its spread in the nature. Much knowledge on relevant phenomena is already available in the literature and has accumulated from the practice of traditional breeding. For example, practical isolation distances applied in breeding of cultivated species have been composed on the basis of these early studies (e.g., Bateman, 1947). However, to confirm with the earlier results and gain more detailed estimates of gene flow in local cultivation conditions, further studies using modern methods with high resolution are considered desirable. Essential factors that deserve study in the cultivation of novel varieties are gene flow between fields, survival of seeds in the field after cultivation, occurrence of established natural populations, and potential exchange of genetic material with weeds or wild species through hybridization (e.g., Giddings, 2000; Saeglitz et al., 2000).
In the present study, we carried out a safety assessment of the cultivation of transgenic barley. We concentrated on pollen-mediated dispersal of genes via cross-fertilization in the cultivation of transgenic barley. The spread of transgenes through pollen flight involves crossing with cultivated barley or with wild relatives. Barley is considered to be a self-pollinating species and in Finland there are no wild relatives with which transgenic barley could hybridize. Nevertheless, there is a slight possibility of cross-pollination between adjacent barley fields. Our aim was to evaluate the distribution of viable pollen and the consequent potential for transgene flow via cross-fertilization in cultivation scale. The occurrence of transgenic volunteers in the field was followed but seed dispersal through animal feeding, seeding, harvest, transportation, handling, or storage technology was not estimated. Transgenic barley carrying the nptII gene was used as a pollen donor and the paternity was confirmed by molecular analyses. For a maximum resolution in catching the cross pollen, a male sterile line was utilized as the recipient.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Male-Sterile Recipient Barley
A cytoplasmically male-sterile barley line (H. vulgare L. cv. Agneta, provided by G. Persson and L. Lehmann, Svalöf Weibull) was sown as the male-sterile recipient. For best resolution in pollen capture, the recipient line was chosen to represent an openly flowering type of barley. From previous experiments it was known that male-sterility is high in the selected recipient line. Nevertheless, the male-sterility is not absolute and rare heads with pollen-producing stamens do occur. Such escaped heads usually produce a full seed set via inbreeding, as in ordinary barley (Lehmann, 1988).
Male-Fertile Recipient Barley
The maintainer line of the male-sterile barley (maintainer of H. vulgare cv. Agneta, also provided by G. Persson and L. Lehmann, Svalöf Weibull) was used as male-fertile control recipient. It was equivalent to the male-sterile line including the open-flowering habit, except for pollen production. The earlier cultivation experiments revealed that flowering of the donor barley occurred 4 to 6 d later than that of the recipient lines. Thus the flowering time of the recipient lines was synchronized with the transgenic donor barley by adjusting the sowing times appropriately.
Wild Barley
Two wild barleys from a Nordic collection representing species of H. murinum L. and H. jubatum L. (provided by R. von Bothmer, Swedish Agricultural University) were used in forced crossing experiments. Attempts to cross these wild barleys with the transgenic donor cultivar Kymppi were carried out by highly competent personnel in greenhouse conditions.
Experimental Set-Up for Field Trials
The field trials with transgenic barley were arranged in 1996 and 1997 at Boreal Plant Breeding Ltd, Jokioinen, Finland, according to the scheme presented in Fig. 1
. The small scale donor area of homozygous transgenic (nptII) pollen was 225 m2. In addition to the small scale donor plots, a cultivation scale donor area of 2000 m2 was used in 1997.
|
Male-fertile recipient plots were only included at distance of 1 m, because the amount of pollen produced outside the actual donor plot had to be kept at a minimum. Furthermore, the male-fertile plots, two in each border, were placed symmetrically at distance of 3.5 and 11 m from the male-sterile sectors in the small and cultivation scale experiments, respectively, in order to reduce the consequent contamination in the male-sterile recipient plots.
During the anthesis, heads reaching the opening stage were marked daily with colored tapes and, in the autumn, the heads were collected separately. The collected seeds were counted individually in the laboratory and the presence of the transgene was analyzed by PCR. Detailed weather measurements were continuously recorded at the plots during the experiments.
Seed Germination and DNA Extraction
Seeds were surface sterilized with 70% (v/v) ethanol for 5 min and with sodium hypochlorite (4% available chlorine) for 10 min, and then rinsed thoroughly with sterile water. The surface-sterilized seeds were germinated on 0.7% (w/v) water-agar, which was supplemented with 2.4 mg L-1 of the seed disinfectant Baytan [Bayer AG, Germany; active ingredient triadimenol—1-(4-chlorophenoxy)-3,3-dimethyl-1-(1H-1,2,4-triazol-1-yl) butanone]. After germination, total genomic DNA was isolated from leaf tissue of the plantlets by CTAB (hexadecyltrimethylammonium bromide) method (Murray and Thompson, 1980).
PCR Analysis
The plantlets were screened for the presence of nptII by PCR reactions containing 100 ng of plant DNA, 100 µM dNTPs, 2.5 mM CaCl2, 1.5 mM MgCl2, 1x Taq DNA polymerase buffer (Perkin-Elmer, Norwalk, CT) and 1.25 U Taq DNA polymerase (Perkin-Elmer) in a final volume of 50 µL. The primers 5'-ACA CGC TGA AAT CAC CAG TCT C and 5'-TCG CCC AAT AGC AGC CAG TC were designed for the nptII gene and amplify a 420-bp fragment (Ritala et al., 1994). Thirty-five cycles were performed under the following conditions: 30 s denaturation at 95°C, 1 min annealing at 60°C, and 1.5 min extension at 72°C. The amplified products were analyzed by electrophoresis on nondenaturing 5% (w/v) acrylamide gels and stained with ethidium bromide.
Monitoring of Transgenic Volunteers
The donor plot areas of transgenic barley were monitored during subsequent years for the occurrence of transgenic volunteers. The transgenic barley materials from donor plots were harvested in the autumn with an experimental scale harvester. Consequently, less seeds were scattered on the soil than in normal cultivation.
In 1996, the straw residue was collected and taken away from the trial area after harvesting. Half of the donor plot area was left untreated, whereas ploughing of soil was carried out for the other half. The occurrence of volunteers was studied in the following summer. The observed volunteers were dug up and planted in a greenhouse for seed production. The obtained seeds were germinated and the presence of the transgene (nptII) in the progeny was analyzed by PCR as described above.
In the 1997 trials, the donor plot area (225 m2) of the small scale trial was left untreated after harvesting except that the straw residue was collected and removed from the area. In the cultivation scale trial, after careful harvesting, the shredded straw residue was left on the soil. The donor plot area (2000 m2) was then divided into three parts (Fig. 2) . Part A was left untreated, for part B a reduced tillage was applied the following spring, and part C was ploughed in a normal way in the autumn. To provide better comparison of treatment effects, two 5-row strips with controlled amounts of nptII-barley seeds (10 seeds per meter) were sown after harvesting across all three parts (Fig. 2). The occurrence of volunteers and germination of transgenic seeds were studied in the following summer.
|
|
|
 | [1] |
Here k/n is the observed proportion, and g is the number satisfying the equation 
- 
= 
, in which () denotes the cumulative standardized normal distribution.
The standard errors for the mean number of seeds for each distance class in Fig. 5 were obtained by dividing the square root of the variance in the class in question by the square root of its number of observations.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Cross-Fertilization Frequency
The germination of the seeds obtained from male-sterile heads was low. This was considered to be due to the poor weather conditions of the summer of 1997 and consequent high microbial contamination present in the seeds. Without seed disinfectant, no germination occurred, and addition of 2.4 mg L-1 Baytan to the water-agar was needed. With this treatment some of the seeds germinated. The germination frequencies for seeds from male-sterile heads remained clearly lower than for seeds from male-fertile heads (Table 1). After germination all the plantlets were analyzed by PCR (Fig. 6)
.
|
|
Occurrence of Transgenic Volunteers
The transgenic donor areas were monitored during the following summers. After the 1996 preliminary trial, four barley volunteers were found from the ploughed area. The four volunteers were dug up and planted in a greenhouse for seed production. Each of them produced a normal yield of grains. The seeds obtained were germinated and the seedlings were analyzed by PCR for the presence of the nptII gene. The PCR analyses revealed that each of these four barley plants represented true transgenic volunteers. After the 1997 trials, not a single volunteer was found in 1998, not even from the arranged experiment with purposely sown seeds.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Transgene Flow
Transgene flow is species specific and depends also on genotype, place, and season. Attention should be paid to differences in pollination types when estimating the pollen-mediated dispersal of transgenes. According to the results of this study, in barley transgene flow was low. However, cross-fertilization with very low frequencies was recorded with enhanced experimental systems up to a distance of 50 m (Fig. 5). Rare gene transfer over the same distance has earlier been reported in wild barley on the basis of isoenzyme studies (Wagner and Allard, 1991). In normal male-fertile barley, the cross-pollination frequency was between 0 to 7% at the distance of 1 m from the transgenic donor area. In corn (Zea mays L.), a wind pollinated species, a cross pollination frequency of 1% between GM and non-GM corn at a distance of 30 m from the transgenic donor area was reported by Jemison and Vayda (2000). However, according to their results the cross pollination frequency dramatically decreased to 0.1 and 0.04% with distances of 37 and 43 m from the donor area, respectively. At a distance of 350 m, no cross pollination was observed.
In the present study, cross-fertilization frequency also clearly varied on a year-to-year basis because of differences in weather conditions during the growth season. The cross-fertilization frequency was lower in 1997 than in 1996. One explanation might be the lower overall wind speed in 1997, and the wind direction distribution was also different in the two summers. Furthermore, during the flowering period some very rainy days occurred in 1997.
In Finland, there exist no established wild barley populations with which transgenic barley could cross. The few occurrences of wild barley species are ephemeral and usually situated far away from barley cultivation areas (e.g., in harbors); thus, in practice, there exists no possibility for cross-pollination with cultivated barley. In general, only if the transferred trait would give a strong selective advantage to wild relatives (provided that such wild relatives occur) is it necessary to consider the consequences of low gene flow frequency. In our study, even forced pollination failed and no seeds were produced when forced crossing experiments were carried out in greenhouse conditions between transgenic barley and two wild barley species (H. murinum L. and H. jubatum.) of the Nordic collection (data not shown). Although it is the experience of plant breeders that cultivated barley is very difficult to cross with wild barley species (Göran Persson, 1994, personal communication; Baum et al., 1992), the crossing ability depends on genotype and these results may not be generalized to other materials.
Contaminating Nontransgenic Background Pollen
PCR analyses of the seeds collected from male-sterile plots confirmed the existence of contaminating nontransgenic background pollen. The few seeds produced at distances of 50 and 100 m resulted mainly from fertilization with nontransgenic pollen. This background pollen is likely to have originated from occasional formation of fertile pollen in some male-sterile heads. Such an anomaly is fairly common in genetically male-sterile barley lines. Some pollen leakage, at a frequency of 0 to 3%, depending on the line, is also reported to occur in cytoplasmically male-sterile barley lines (Lehmann, 1988). Furthermore, some of the nontransgenic pollen may have originated from the male-fertile control plots at a distance of 1 m (see Fig. 1). Theoretically, it is also possible that some pollen might have originated from a more distant barley field. Thus, molecular analyses are necessary for estimation of the maximum resolution theoretically possible. Nevertheless, the 100% cross-fertilization in male-sterile heads at a distance of 1 m compared to the cross-fertilization frequency of approximately 4% in male-fertile heads confirmed the suitability of male-sterile lines for enhancing resolution in risk assessment studies. Consequently, more relevant gene flow estimates can be obtained by applying male-sterile recipient lines combined with molecular analyses rather than by pollen capture records.
Pollen Leakage Distorts Cross Pollination Frequencies
The prediction of pollen flow on the basis of the seed-set in male-sterile heads was an overestimation (Fig. 5). The results showing high frequency of pollination by nontransgenic background pollen (Fig. 4) confirmed our prediction on the basis of general theoretical grounds: The more rare occurrences of fertilization events are scored, the greater will be the proportion of false positives in the overall number of apparently positive results. If the frequency of the actual phenomenon under study decreases, while the frequency of its mimicking events remains considerable, an estimate based on apparent occurrence will become more and more biased until it finally represents the frequency of the mimicking events. In our study, such falsely positive cross pollination frequencies were obtained from the seeds produced by nontransgenic background pollen. For accurate results in a high-resolution experiment, it is crucial to minimize the frequency of false positive cross pollination frequencies, and to be able to discriminate the remaining false positive records from true ones with reliable analyses. One option in population studies is to complement the phenotypic observations with the confirmation of the genetic origin (paternity/maternity) of the progeny by using informative DNA markers or known DNA sequences with appropriate molecular analyses, as in the present study.
Overestimation of the Cross-Fertilization Because of Experimental Set-Up
The results of the present study apply to a transgenic donor barley representing a closed type of flowering. However, because of the characteristics of the recipient barley used, the recorded gene flow was overestimated compared with the actual situation in ordinary barley fields. First, the recipient lines (both male-sterile Agneta and its corresponding male-fertile line), represented an open type of flowering to maximize the resolution in capturing the pollen. Similarly, the results of Wagner and Allard (1991) also referred to an openly flowering barley type. However, the barley varieties commonly grown in Finland represent a much more closed type of flowering. Typically, with a closed type of flowering, the cross-fertilization frequencies are only a fraction of the values reported for the openly flowering barley types (Hammer, 1977). In the closed type of flowering, self-pollen has an advantage and the ovules are already fertilized at the time of flower opening. The second reason for overestimation is that, within a barley field with ordinary barley, the air surrounding the recipient head is crowded with local barley pollen arising from the neighboring plants. Thus, the field efficiently dilutes the concentration of the few pollen grains migrating from other fields. The border of the field is less protected, but the borders can be efficiently buffered with companion crops when necessary. The predominance of local pollen in fertilization reflects its relative abundance in the pollen cloud surrounding the recipient heads and its effect may even be accentuated by the presumably lower fertilization capacity of pollen arriving from further away, although experimental evidence for the decline of pollen fertility during flight is still missing in barley.
Occurrence of Transgenic Volunteers
The transgenic donor plots were monitored for volunteers during subsequent summers. Only four volunteers of transgenic barley were found after the 1996 trial. Even the transgenic seeds sown deliberately failed to produce plants after the winter of 1997-1998. The barley varieties grown in Finland, including the transgenic donor used in this study, represent spring barley. Obviously, most of the seeds which had fallen from the harvester on the soil had already germinated in the autumn of the trial. At least the starch degradation had already started in the autumn and consequently hindered germination the next spring. Moistened grains will inevitably be killed by freezing during a cold winter. Probably this explanation is also valid for the seeds sown because the subsequent rainfalls may have caused starch degradation or germination in the autumn.
The location of Finland as the northernmost agricultural country in the world may not rule out entirely the possibility of survival of cereal grains through the winter. Therefore, to maintain the purity of the varieties, the same fields should not be used for the production of different varieties of the same species in consecutive years. It is known that in exceptionally dry autumns barley seeds may remain nongerminated and germinate in later years, especially if they are ploughed deep (20–25 cm) into dry soil. However, the dry conditions necessary for the survival of the seeds rarely occur because of the typical heavy rains in the autumns. Furthermore, reduced tillage is often preferred to ploughing because it causes less erosion.
Risk Management
In the study, our primary interest was in basic population science. The previous estimations of gene flow in a self-pollinating species such as barley were to be confirmed with modern, high-resolution experimental systems. In Finnish conditions, gene flow appeared low even in our exaggerated measurements. Consequently, it can be adequately controlled for nearly all practical purposes with existing good cultivation practices, e.g., isolation distances, crop rotation, and control of escapes. In most situations, there is no special need to demand extra purity, but a few percent of mixed materials are commonly allowed in the cereal trade, except in sowing seed material and in GM crops. Even if superior purity of the product were to be required in the future, e.g., for certain medical or high-tech applications, the existing control measures could probably be correspondingly enhanced on the basis of the present results. However, more general, simple and reliable strategies of blocking the escape of transgenes and transgenic plants have recently been developed and may control even with the most demanding situations with inconvenient traits, outcrossing species, or abundant wild populations (Kuvshinov et al., 2001).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
It is evident that gene transfer techniques can be of valuable assistance in barley breeding programs. As a precaution, the safety of the use of the new varieties is assessed before their release. Although the risks and benefits depend mainly on the actual trait under consideration, knowledge of transgene flow parameters is also needed for determining the good cultivation practices. On the basis of our study, cross-fertilization in male-sterile recipient barley is possible with very low frequency up to 50 m from the donor area. However, the frequency dramatically decreases with distance. Because of normal self-pollination, often favored by a closed flowering habit, the possibility of cross-fertilization remains even lower in normal, cultivated barley. Suitably designed risk management procedures implemented in farming will make it possible to transfer the benefits of genetic engineering into the production of high quality plant-based raw materials.
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 23, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
DNA, digested with PstI).