Recipient Organization
UNIVERSITY OF FLORIDA
G022 MCCARTY HALL
GAINESVILLE,FL 32611
Performing Department
Microbiology and Cell Science
Non Technical Summary
This project addresses standard program area 4: "environmental effects of GE relative to non-GE organisms in the context of production systems". Our focus is on program area 4d as this project is a "comparative assessment of environmental impacts of agricultural production systems using organic and/or conventional methods with those involving plant, animal, or microbial biotechnology". Here genetically engineered tomato plants are compared with non-genetically engineered tomato mutansthat all have increased immunity from plant pathogens Our goal is to identify environmental hazards associated with the genetically engineered lines. Tomato lines will be cultured under agricultural conditions with four plantings over two years. In objective 1, plant and soil microbiome changes will be determined in field-grown genetically engineeredlines in the field compared to the other lines. Microbiome analysis will include full-length rRNA operon and shotgun metagenomic sequencing. In objective 2, the effects of GE lines on disease pressure, symptom ratings, yield, horticultural characteristics, soil nutrients, and beneficial microbial populations will be measured. In objective 3, all the data from objectives 1 and 2 will be integrated to learn the most from the data for regulatory decision making.Our hypothesis is that the environmental effects of the geneticaly engineeredlines will be minimalcompared to the standard tomato line used for production.
Animal Health Component
34%
Research Effort Categories
Basic
33%
Applied
34%
Developmental
33%
Goals / Objectives
This project addresses standard program area 4: "environmental effects of GE relative to non-GE organisms in the context of production systems". Our focus is on program area 4d as this project is a "comparative assessment of environmental impacts of agricultural production systems using organic and/or conventional methods with those involving plant, animal, or microbial biotechnology". Here GE-derived tomato plants are compared with CRISPR mutants of tomato that express the same phenotype, increased plant defense. Our goal is to identify environmental hazards associated with the GE lines. Tomato lines will be cultured under agricultural conditions with four plantings over two years. In objective 1, plant and soil microbiome changes will be determined in field-grown GE lines in the field compared to the other lines. Microbiome analysis will include full-length rRNA operon and shotgun metagenomic sequencing. In objective 2, the effects of GE lines on disease pressure, symptom ratings, yield, horticultural characteristics, soil nutrients, and beneficial microbial populations will be measured. In objective 3, all the data from objectives 1 and 2 will be integrated using multiple hazards and XGBoost-based models. Hazard ratios of genotype by environment interactions will be systematically compared to provide federal regulators with decision-making tools. Our hypothesis is that the environmental effects of the GE lines will be insignificant compared to the CRISPR mutant lines. Both the GE and CRISPR lines are expected to show similar decreased and increased levels of pathogens and beneficial microbes, respectively, in bulk soil, rhizosphere, and apoplast compared to the unmodified wild-type lines.Specific Objectives:Objective 1. Define changes to the microbiome of bulk soil, rhizosphere, phyllosphere, fruit, and apoplast in tomato with samples collected from four field plantings over 2 years when planted with the CRISPR mutant lines, GE lines versus their parent line. These analyses will identify changes in bacterial and fungal pathogen populations both in relative and absolute abundance.Objective 2. Assess impact of the GE and non-GE plant defense lines on disease pressure, disease symptom ratings, mycorrhiza populations, yield, and soil nutrient levels.Objective 3. An integrated analysis of data generated from the many variables measured in objectives 1 and 2 will be performed including Cox proportional hazards modeling to identify hazards associated with using the GE-derived tomato genotypes compared to the wild-type parent and CRISPR mutants.Approach: The transgenic tomato lines to be used will express the Arabidopsis genes NPR1, LecRK-I.8, LecRK-VI.2, or ELP4. These genes are all driven by the 35S CMV promoter. The CRISPR-generated lines have a single base mutation in NPR3 or NPR4 or both. These mutations result in increased NPR1 expression. The corresponding wild-type line of each GE or non-GE line will also be used. Changes to soil nutrient levels and to microbial communities on the phyllosphere, apoplast (endophytes), rhizosphere, and bulk soil will be assessed through metagenomics at three time points during the growing season. These analyses will identify changes in bacterial and fungal pathogen populations both in relative and absolute abundance. Yield will be measured at the end of the growing season. Over time, tomato pathogens may develop resistance to these GE-derived or CRISPR mutant lines with increased plant defense. Although this is considered unlikely since these genome alterations take place in genes deep within a signaling pathway, we plan to assess the frequency of spontaneous resistance of pathogens to these altered lines.
Project Methods
Tomato lines proposed for the study. Four transgenic lines and three CRISPR mutant lines will be used in this study. The transgenic lines are in the genetic background of the commercial tomato line called 'Moneymaker' and overexpress the Arabidopsis defense genes NPR1, LecRK-I.8, LecRK-VI.2, and ELP4, respectively. The LecRK-I.8, LecRK-VI.2, and ELP4 lines are available now. The NPR1 line in 'Moneymaker' has been ordered from the University of Nebraska transformation center. Seeds from the NPR1 transformed line will be available for the Fall 2022 planting, the first field planting of this work. The three CRISPR-generated tomato lines are all within the 'FL8000' genetic background. They are npr3 (A7-27), npr4 (A7-3), and npr3 npr4 (A7-33).Experimental field design. Field plots will be located in Hillsborough County, Florida near Balm at the UF/IFAS Gulf Coast Research and Education Center. This Center is the premiere location for tomato field research in the state. The same field experiment will be planted on the same plots for four straight tomato plantings, two per year. This allows an accumulation of the effects of the treatment over that period. There will be nine host genotypes as described above including both parent lines.There will be eight replicates per genotype. This number was chosen based on the significant disease improvement seen with eight replicates (Figure 2). To simulate tomato production in Florida, each genotype will be 150 feet long (45.7 m) with rows separated by 48 inches (1.2 m). The genotype replicates will be in a randomized complete block design. Five-to-six-week-old tomato seedlings will be planted 24 inches (0.6 m) apart on plastic mulched beds. The first planting will take place in early September 2022 with succeeding planting occurring every September and February with the last planting taking place in February 2024. There are four harvests for each planting. For February plantings, harvesting is done from July through early August. Fall planting takes place in September with the four harvests collected from December through early January. Plots will be arranged in a randomized complete block design that will remain the same over the four plantings so that any cumulative effects in the soils can be observed with the tomato genotypes.Sampling. Bulk soil will be collected from each plot at four randomly chosen locations along the 45.7m row prior to planting using a soil borer 2.5 cm wide and 10 cm long. Separate sterile borers will be used to sample each plot. Samples will be placed in plastic bags and immersed in ice packs immediately. Once collected, the samples will be placed in a -80ºC freezer after their arrival at the Triplett lab in Gainesville. The four samples from each treatment row are technical replicates that will all be analyzed to determine within-row variability. Bulk soil will also be collected in the same manner.Verification of NPR1 levels will be determined by ELISA. DNA extration will be done using Power Soil extraction kits from Qiagen.Microbial composition will be done using Illiuma MiSeq paired-end, barcoded 16S rRNA sequencing, Metagenomic sequencing will be done using barcoded Oxford Nanopore sequencing in order to obtain long reads. From the metagenomes, genes involved in functions such as nitrogen fixation, flagella, and antimicrobial resistance will be mined and quantified.A similar approach will be used to mine whole genome sequences as needed with metaFlye as the assembler.Soil texture, pH, total N, micronutrients, and soluble Pwill be made by the University of Florida's Soil and Water Sciences Diagnostic Laboratories.Several plant growth and development parameters including plant height, canopy diameter, flowering time, fruit number, and fruit size/weight will be recorded at the time of each first harvest for each planting. Furthermore, several fruit quality parameters including lycopene, vitamin C, total sugar, total acid, and total phenol will be measured to evaluate the effect of the plant lines on fruit quality. Fruit size and weight will be determined for each of the four harvests from each planting.Categorical data will be compared by the chi-square test and Bonferroni post hoc test. Continuous data with or without a normal distribution will be compared by Student's t test or the Mann-Whitney U test, respectively. The association between genotypes, covariates, and outcome events (e.g. plant mortality, mild-severe disease, etc.) will be examined by using the Kaplan-Meier method and logrank test. Cox proportional hazards modeling will be applied to estimate the risk of death or disease. The proportional hazards assumption will be visually inspected by log-log survival curves. In case the disease or death event rate is relatively low, we will avoid overfitting the model by selecting relevant covariates in adjusted models. Spearman correlation will be applied to quantify relationships between pathogen population sizes and environmental factors. The degree of pathogen composition differences between genotypes will be visualized using non-metric multidimensional scale (NMDS). Two-tailed p-values <0.05 will be considered statistically significant.