|Home | About | Journals | Submit | Contact Us | Français|
The effects of a genetically modified cucumber mosaic virus (CMV)-resistant tomato on soil microbial communities were evaluated in this study. Soil position and environmental factors played more dominant roles than the tomato genotype in the variation of soil microbial communities.
Microbial communities are integral parts of soil processes such as organic matter decomposition and nutrient cycling (2). Interaction between plant species/genotypes and rhizosphere microorganisms has been revealed in several reports (8, 9, 15, 16, 19, 20). A genetically modified (GM) cucumber mosaic virus (CMV)-resistant tomato (Lycopersicon esculentum) was developed by AVRDC—The World Vegetable Center (Tainan, Taiwan). The Taiwan Department of Health has conducted a safety assessment (project code no. DOH95-FS031) of this CMV-resistant tomato and concluded it to be safe (18). Although no significant impact on soil processes caused by GM crop plants has been reported, case-by-case detailed study of accessible and relevant indicators of soil ecosystems is the most feasible strategy until we more fully understand soil ecosystems (4). Plant-microbe-soil nitrogen cycling is an essential part of ecological functions and processes in ecosystems. The influence of plants on nitrogen transformation comes from the interaction between plant roots and microbial communities, as microbes are key players in soil nitrogen processes (3). To evaluate the impact of the GM tomato plant on soil microbial communities, general denaturing gradient gel electrophoresis (DGGE) profiles of bacteria (23), fungi (24), and actinomycetes (11), as well as three functional bacterial communities involved in nitrogen cycling—free-living nitrogen-fixing bacteria (5), ammonium-oxidizing bacteria (13, 17), and nitrate-reducing bacteria (7, 14, 21, 25)—were investigated in this study.
The transgenic CMV-resistant tomato (line R8) and the corresponding original plant line L4783 were provided and planted by AVRDC—The World Vegetable Center. Each GM organism (GMO) test field (5 by 5 m) was an independent net house separated by bush, wire fence, and fosse barriers, with 9-m gaps between houses. Tomatoes were ridge-furrow cultivated with 120-cm gaps between plants. Ridges were covered with silver-black plastic sheets for weed prevention and control of soil temperature. Test fields were routinely managed through watering, pest control, and nipping.
Samples of soil (sandy clay loam) from two test fields, one with GM tomato plants and one with wild-type (WT) plants, were collected in April 2007. The tomato plants were approximately 100 days old at the time of sampling. Samples were collected from sites 10 to 15 cm below the surfaces of furrows and ridges; each ridge sample was obtained from directly below the tomato plant and contained some root material. Each test field was sampled at five positions, each of which yielded one furrow and one ridge sample. A total of 20 samples were analyzed. The total nitrogen content of the soil was determined by the Kjeldahl method (10). The total organic carbon (TOC) was determined by the Walkey-Black method (27). The moisture contents of soil samples were determined by measuring weight loss. Agar plate enumeration of total microbes, fungi, and actinomycetes was carried out using nutrient agar (BD Biosciences, Franklin Lakes, NJ), Rose Bengal agar (BD Biosciences), and glycerol-yeast extract agar, respectively. Univariate analysis of the collected data was performed using SPSS software (version 12.0; SPSS Inc.). Duncan's multiple-range test or Dunnett's T3 test was used in post hoc analyses according to the homogeneity of variances (29).
Soil DNA was extracted from 1-g samples by using an UltraClean soil DNA kit (Mo Bio Laboratories Inc., Carlsbad, CA). DGGE was performed using the Dcode system (Bio-Rad Laboratories, Hercules, CA). DGGE images were processed and converted into an unweighted binary pattern (absent, 0; present, 1) by using ImageJ software (1). Similarity metrics for microbial community profiles were generated using SPSS software with the Dice measure. Principal component analysis (PCA) of the similarity metrics was performed using GenStat Discovery edition 3 software (VSN International Ltd., Hemel Hempstead, United Kingdom). Cluster analysis of microbial community profiles was performed with Phylip 3.68 by the unweighted-pair group method using average linkages (UPGMA) (6). Box-whisker plots were generated using SigmaPlot 8.0 (SPSS Inc., Chicago, IL).
The total numbers of microbes, fungi, and actinomycetes in soil samples were approximately 106 to 107, 105 to 106, and 104 CFU/g of dried soil, respectively. Although the average number of soil microbes detected in samples from the GM tomato test field was higher, there was no significant difference in the number of soil microbes between the GM and WT tomato test fields. In addition, no significant difference in soil properties between the GM and the corresponding WT tomato test fields was observed (data not shown). DGGE profiles of soil samples have shown diverse patterns with few common bands regardless of the tomato plant type (data not shown). PCA and cluster analysis of DGGE patterns have revealed that the effect of soil position was stronger than the effect of the GM tomato plant on the soil microbial communities (Fig. (Fig.1).1). Minor correlations between the tomato plant and variations in 16S rRNA genes and ammonium-oxidizer communities in the furrow soils were present in the cluster analysis results. However, the correlations were weakened by the low level of similarity among individual soil samples from the same treatment area. A box plot showed that the similarities between DGGE profiles of 16S RNA genes and ammonium-oxidizing bacteria in ridge soils were relatively low, with an average Dice measure of approximately 0.5 (Fig. (Fig.22).
Tomato L4783 (the WT plant) belongs to a unique popular group representing a Taiwan cultivar whose fruit is harvested and consumed at the breaker phase. The use of a ridge-furrow plot with support sticks is a popular tomato cultivation method in Taiwan. The ridges are typically covered with silver-black plastic sheets that create a physical barrier for weed prevention and control of soil temperature. In contrast to ridges, furrows are exposed to frequent human activities including watering, fertilizing, nipping, and walking during the husbandry of tomatoes. The isolation created by the plastic covering of the ridges and the frequent human activity occurring at the furrows potentially contribute to the difference between ridge and furrow soils in the test fields. Because furrow soil is less relevant to the plant and directly exposed to the environment, we think that the minor correlations between plant type and furrow soil microbial communities in this study (Fig. (Fig.1)1) resulted from environmental factors such as the position of the field relative to the drain, the wind direction, and human activities. The effect of the tomato genotype on the variations in DGGE profiles was considered to be minor because the effects of soil position or environmental factors were stronger than the effect of the plant genotype. The results of this study show that the CMV-resistant GM tomato plant had only a minor effect on soil microbial communities (12, 22, 26, 28).
Published ahead of print on 26 March 2010.