| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Coastal Farmers Co-operative, P.O. Box 1003, Umhlanga Rocks, KwaZulu Natal, 4320, South Africa
b Biotechnology Dep., South African Sugar Association Exp. Stn., Private Bag X02, Mount Edgecombe, KwaZulu Natal, 4300, South Africa
* Corresponding author (snyman{at}sugar.org.za)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONNOTESREFERENCES |
|---|
US $14.01) L-1. For Buster treatments to be cost effective for commercial growers, the price would have to be considerably lower. However, since hand-hoeing is deleterious to the cane, small-scale growers currently controlling weeds manually would find herbicide use an advantage. If they were to adopt transgenic sugarcane, Buster costs would need to be reduced only slightly for them to increase their returns.
Abbreviations: Bt, Bacillus thuringiensis • pat, phosphinothricin acetyltransferase • PPT, phosphinothricin • SASEX, South African Sugar Association Experiment Station • SCMV, Sugarcane mosaic virus • WCT, weed control treatment • ZAR, South African Rand
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONNOTESREFERENCES |
|---|
Commercialization of a transgenic crop usually requires stable expression of the introduced gene in a plant that retains its own genetic integrity and agronomic characteristics. Extensive field testing is essential. Stable inheritance of a transgene in field trials has been reported in transgenic sugarcane previously (Gallo-Meagher and Irvine, 1996; Arencibia et al., 1999; Sala et al., 1999), as well as in other graminaceous monocotyledons such as barley (Horvath et al., 2001) and rice (Jiang et al., 2000). However, the importance of thorough screening of transgenic plants that are propagated vegetatively was stressed in a paper on transgenic tall fescue (Bettany et al., 1998). In that work, loss of gene expression was observed in some tillers in the plant and stable transgene expression could only be determined and selected after three to four rounds of vegetative propagation, an effect the authors suggested might be due to environmental factors. Underlining the need for field data is the fact that the integrity and stability of the plant genome as a whole may be affected by the tissue culture and transformation process. Genomic change resulting from transgenesis has been documented in various plants including rice (Jiang et al., 2000; Labra et al., 2001), barley (Bregitzer et al., 1998; Choi et al., 2000) and sugarcane (Arencibia et al., 1999, 2000; Sala et al., 1999).
In this study, the effect of pat herbicide-resistance expression on the agronomic potential of sugarcane cultivar NCo310 was investigated using a novel field approach. NCo310 was bred in South Africa and has been used as the recipient genotype and model system in several sugarcane transformation programs (Gallo-Meagher and Irvine, 1996; Snyman et al., 1998) as it responds favorably to in vitro culture. A single transformed line (22.2), characterized by nine copies of the pat gene inserted at separate loci, was selected for the trial experiments as it showed significant resistance to the herbicide Buster [monoammonium 2-amino-4(hydroxymethylphosphinyl) butanoate] in glasshouse spraying trials.
The aims of the study were twofold. First, to characterize the transgenic line genetically and phenotypically by (i) establishing the stability of the gene across several ratoons, (ii) determining herbicide sensitivity at different application rates, and (iii) comparing morphological and agronomic characteristics such as stalk morphology, susceptibility to pests and diseases, and yields in relation to the untransformed parent cultivar NCo310. Second, to assess the economic significance and viability of herbicide-resistant sugarcane by measuring the efficacy of weed control using two different treatments involving the herbicide Buster and two weed control methods currently used in the South African sugar industry, namely hand-weeding and a cocktail of pre- and postemergence herbicides used by small- and large-scale commercial growers, respectively. To address these aims, two field trials were devised, one a simple small-scale dosage trial conducted across two ratoons, the other a more complex weed control trial on a larger scale.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONNOTESREFERENCES |
|---|
Trial to Establish Herbicide Dosage and Phenotype Stability
A small-scale field trial was performed to measure herbicide sensitivity of a single transformed line, and to determine the stability of the resistance gene across several rounds of vegetative propagation. Line 22.2 contains nine copies of the pat gene, and expressed herbicide-resistance to Buster (200 g L-1 glufosinate ammonium a.i.) in glasshouse–based spraying trials. Sugarcane stalk sections (setts) from line 22.2 and from untransformed NCo310 were grown in polystyrene seedling trays for approximately three months, then transplanted into 7- by 18-m rows, separated by concrete strips 0.5 m wide. Supplementary irrigation was provided. Once the plants had tillered, Buster was applied to the foliage using a lever-operated knapsack fitted with a 110° Teejet (Minneapolis, MN) nozzle. Rates from 1 to 7 L ha-1 of Buster were applied incrementally to Rows 1 to 7 using the same volume of aqueous solution in each case (200 L ha-1). Damage was assessed 3 wk after application. The plant crop was cut back after 6 mo and allowed to ratoon (the regrowth of the crop after harvesting). The two subsequent ratoons each received a single application of Buster at 7 L ha-1 after four months, resulting in one of the rows (Row 7) of transgenic cane having received a total of 21 L ha-1 from planting to second ratoon.
Trial to Compare Phenotypic Responses of Herbicide-Resistant and Control Sugarcane in Four Weed Control Programs
A large-scale field trial was conducted to compare transformed with untransformed NCo310, for (i) stalk morphology (height, diameter, population density) and fiber content, (ii) response to smut disease (causal agent Ustilago scitaminea Syd. & P. Syd.), Sugarcane mosaic virus (SCMV), and the insect pest eldana (Eldana saccharina Walker; Lepidoptera: Pyralidae), and (iii) yield (both cane biomass and sucrose production). During field preparation, grass seeds of Panicum maximum Jacq. (1.7 kg ha-1), Sorghum bicolor (L.) Moench (2.4 kg ha-1) and Rottboellia cochinchinensis (Lour.) Clayton (3.1 kg ha-1) were introduced to the soil to increase weed pressure to the level normally found under industry conditions. Sugarcane stalk sections (setts) were obtained from bulking plots grown for 12 mo, and were dipped in Panoctine fungicide [1% (v/v); a.i. guazatine 400 g L-1; Rhone-Poulenc Agrichem, Durban, South Africa] before planting. Fertilizer (N-P-K at 160, 52, and 160 kg ha-1, respectively) was applied in the planting furrows and as a top-dressing along the rows.
The following four weed control treatments were applied: (i) Repeated Buster application (WCT1); Buster (5 L ha-1) was applied to sugarcane once weed emergence was complete. Three applications in the plant crop and two in the first ratoon were necessary to keep weeds under control. (ii) Preemergence herbicides, followed by repeated Buster application (WCT2). The plant crop was treated with a preemergence herbicide cocktail comprising Visor {2 L ha-1; a.i. thiazopyr [3-pyridine carboxylic acid, 2-(difluoromethyl)-5-(4,5 dihydro-2-thiazolyl)-4-(2-methylpropyl)-6-(triflouromethyl)-methyl ester] 240 g L-1; Rohm and Haas, Philadelphia, PA}, Harness [1.5 L ha-1; a.i. acetochlor (2-chloro-2'-methyl-6'-ethyl-N-ethoxymethyl-acetanilide) 900 g L-1; Monsanto, St. Louis, MO] and Diuron 800 SC [2.5 L ha-1; a.i. 1-(3,4-dichlorophenyl)-3,3-dimethylurea 800 g L-1; Sanachem, Canelands, South Africa] followed by two Buster (5 L ha-1) applications. The first ratoon was hand-weeded once using eight labor units (one labor unit = one laborer d-1 ha-1) before receiving Har-i-cane [2 L ha-1; a.i. acetochlor (2-chloro-2'-methyl-6'-ethyl-N-ethoxymethyl-acetanilide) 960 g L-1; Monsanto] and Diuron 800 SC (2.5 L ha-1). One Buster application (5 L ha-1) was necessary in the first ratoon. (iii) Conventional weed control program (WCT3). The plant crop received a preemergence application of Visor (2 L ha-1), Harness (1.5 L ha-1) and Diuron 800 SC (2.5 L ha-1), followed by a postemergence application of MCPA (3 L ha-1; a.i. 2-methyl-4-chlorophenoxyacetic acid 400 g L-1; Sanachem), Gesapax [N-ethyl-N'-(1-methylethyl)-6-(methylthio)-1,3,5-triazine-2,4-diamine; 4 L ha-1; 500 g ametryn per L-1; Syngenta, Greensboro, NC] and Reverseal 10 [adjuvant; 0,5% (v/v); Revertex Chemicals, Mobeni, South Africa]. The plant crop was hand-weeded using four labor units. The ratoon crop was hand-weeded with eight labor units and then sprayed with Har-i-cane (2 L ha-1) and Diuron 800 SC (2.5 L ha-1). Finally, weeds were removed by hand-weeding with eight labor units two months later. (iv) Hand-weeding (WCT4). Five operations using hand-hoes and a total of 78 labor units were required to keep weeds under control in the plant crop. Three hand-weeding operations were necessary in the first ratoon crop (total of 34 labor units).
The trial was laid out in a split plot design, with eight replicates, four treatments as whole plots (WCT1-WCT4) and two subplots (transformed and untransformed). The herbicide treatments were randomized between the whole plots, and the allocation of the transformed and untransformed sugarcane was randomized independently within each whole plot. Each split plot comprised six rows, 7.5 m in length and spaced at 1.2 m. The two outermost rows served as guard rows. Analysis of variance was performed on the full model (i.e., treatment + transformed + treatment x transformed). Comparisons were made using first ratoon data, on the basis that Bailey and Bechet (1989) found that growth characteristics of tissue culture (callus)-derived plants compared more favorably to conventional sugarcane in the first ratoon than the plant crop.
Records were kept of labor usage per hectare, operational times, and amounts of herbicide used in the four treatments. The number of stalks was counted in two inner rows in each plot to give an estimate of stalk population density per hectare. Stalk height (measured from base to the top visible dewlap) and diameter (measured at the center of the stalk) were recorded in each of 20 stalks per plot. The incidence of smut and SCMV was determined by field inspection performed when the ratoon crop was 10 mo old, directly before harvest. Damage by the borer eldana was measured as percentage internodes bored in each of 25 stalks per plot, sampled from harvested material. These assessments were performed by SASEX inspection teams from Pathology and Entomology departments, respectively.
Planting was performed in September 1998 and the trial was harvested when the plant crop was 12.5 mo old and the ratoon 10 mo of age. The cane was cut by hand, weighed, and sampled for direct analysis of sucrose and fiber (Buchanan and Brokensha, 1974) at the SASEX mill room. The South African industry formula for the estimated recoverable crystal (ERC) % cane, used to calculate the recoverable value payment system, was calculated using the following formula:
 |
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONNOTESREFERENCES |
|---|
|
|
Stalk heights measured when the first ratoon was harvested ranged from 185.6 to 204.5 cm and there were no significant differences between untransformed and transformed cane in Treatments WCT2, WCT3, and WCT4 (Fig. 2B) (ANOVA SED = 4.63; LSD0.05 = 9.6). The shortest stalks (186–190 cm) were observed in WCT4, where weeds were removed by hand-hoeing. It is possible that mechanical damage to the roots caused by hoeing stunted the growth of the cane.
Only the treatment that involved preemergence herbicides followed by two Buster applications at 5 L h-1 (WCT2) showed a significant difference in stalk populations between transformed and untransformed sugarcane (Fig. 2C) (ANOVA SED = 399; LSD0.05 = 824). This is because the untransformed cane is susceptible to Buster and some stalks did not survive the herbicide application. Although transformed cane in WCT3 and WCT4 contained more fiber than the untransformed plants from the same treatment, the differences were not significantly different (Fig. 2D) (ANOVA SED = 0.22; LSD0.5 = 0.46).
Pests and Diseases
Varying levels of smut resistance have been observed in previous field studies performed on tissue culture-derived (Bailey and Bechet, 1989) and transgenic (Arencibia et al., 2000) sugarcane. Cultivar NCo310 is susceptible to both smut and SCMV (Bailey, 1979). Directly before harvesting of the first ratoon crop, 1.4 to 5.1% of plants were infected with smut and 1.3 to 5.9% with SCMV. However, there were no significant differences in the ratings of transformed and untransformed cane in any of the treatments for both smut (ANOVA SED = 1.67; LSD0.05 = 3.44) (Fig. 3A) and SCMV (ANOVA SED = 1.74; LSD0.05 = 3.59) (Fig. 3B).
|
Yield
Yields obtained under the four different weed control treatments were compared to see whether cane treated with Buster showed comparable yields with cane in the conventionally used weed control treatments. In addition, transformed and untransformed cane yields were examined to determine whether the growth and therefore yield capacity of the plants was affected by transgenesis. When comparing treatments not in same whole plots, there were no significant differences in any yield parameter tested (Table 1) (i.e., there was no effect of weed control treatment on yield). No significant differences in cane quality were observed between transformed and untransformed cane in any of the weed control treatments, indicating that overall C allocation and the proportion of sucrose stored in the sugarcane culm was not influenced by either transformation or method of weed control. This result is compatible with the lack of effect of these factors on fiber (Fig. 2D). However, some significant differences in biomass and sucrose yields of transformed and untransformed cane were observed. As would be expected, untransformed cane under Treatment WCT1 died because it was susceptible to Buster, and yields from untransformed cane under Treatment WCT2 were poor due to the phytotoxicity of Buster applications after initial use of other herbicides. The effect of the blank plots in the former treatment on other plots was counteracted by the presence of guard rows. The poorest performance was in transformed cane under the WCT4 treatment, suggesting that transgenic cane does not respond well to hand-weeding. This possibly reflects great variability of growth and vigor of any kind of cane under hoeing. It is also a possible result of transgenic cane having less sturdy stalks than the untransformed counterparts, although statistical analysis had shown that these differences were not significant (Fig. 2A).
|
Recent reports dealing with yields in genetically engineered crops have shown a range of responses, from reduced yields in PPT-resistant barley (Bregitzer et al., 1998), to no effect in bacterial leaf blight-resistant rice (Tu et al., 2000a), and increased yields in herbicide-resistant flax (McHughen and Holm, 1991) and Bt-rice (Tu et al., 2000b). This emphasizes the need for thorough assessment of agronomic characteristics of transgenic crops.
Weed Control Costs
Economically viable transgenic sugarcane relies on a financial and/or management advantage to growers. To assess the financial advantage of herbicide-resistant sugarcane, a comparison was made between transgenic and conventional sugarcane based on the value of the crop and the cost of the herbicide treatment. From comparisons of yields in the previous section, no significant differences exist between transgenic cane using Herbicide Treatment WCT2 (the likely scenario) and untransformed cane using conventional herbicide treatment (WCT3), so the profit margin will be determined by the cost of the weed control treatment, and more specifically, by the herbicide to which resistance has been genetically engineered. The current cost of Buster in South Africa is high (ZAR 150 L-1, US $14.01 L-1), but indications from previously commercialized herbicide-resistant crops in the USA is that herbicide prices decrease with higher demand (Dunwell, 1999; Herrera-Estrella, 2000; Barboza, 2001; Bullock, 2002).
A costing exercise was performed using three different prices of Buster: ZAR150 L-1, ZAR60 L-1, and ZAR30 L-1. The return per ZAR spent on weed control was calculated where the value of the crop and the cost of weed control were taken into account. The returns per ZAR spent on weed control in transgenic cane using the two Buster treatments most likely to replace current weed control methods (WCT1 and WCT2) were compared with those in untransformed cane using current industry methods of weed control (WCT3 and WCT4; Table 2). Calculations show that commercial grower returns using the standard industry herbicide treatment (WCT3) are not matched by returns using Buster-resistant cane with the herbicide at its current price of ZAR150 L-1. If transgenic cane were to be grown, Treatment WCT1 would begin to be cost effective if the cost of Buster was half, and very advantageous if the cost of Buster were to drop to one-fifth. Treatment WCT2 would not offer any great advantage even at a low herbicide price. By contrast, because of the damaging effect of hand-hoeing, small-scale growers currently employing manual weed removal (WCT4) would find use of herbicides an advantage. If they were to adopt transgenic sugarcane and use only Buster (WCT1) to control weeds, they would maintain the same level of return on their crop at the present Buster price. Buster costs would need to be reduced only slightly for them to increase their returns. In view of the fact that Buster is not competitively priced, herbicide-resistance to a cheaper herbicidal compound may be an economically more viable approach.
|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONNOTESREFERENCES |
|---|
Received for publication February 12, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONNOTESREFERENCES |
|---|
-endotoxin. Nat. Biotechnol. 18:1101–1104.[Web of Science][Medline]This article has been cited by other articles:
|
|
 |
|
  J. Li, A. M. Brunner, R. Meilan, and S. H. Strauss Stability of transgenes in trees: expression of two reporter genes in poplar over three field seasons Tree Physiol, February 1, 2009; 29(2): 299 - 312. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Gilbert, M. Gallo-Meagher, J. C. Comstock, J. D. Miller, M. Jain, and A. Abouzid Agronomic Evaluation of Sugarcane Lines Transformed for Resistance to Sugarcane mosaic virus Strain E Crop Sci., August 26, 2005; 45(5): 2060 - 2067. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
