2011 Annual Report
1a.Objectives (from AD-416)
Objective 1: Using an insect-pollinated crop system, elucidate principles and data requirements for better predictions of gene flow via pollen in insect-pollinated crops.
Objective 2: Using squash with transgenic resistance to viral pathogens as a model system, develop a methodology to assess the impact of this transgene recently introduced into the genome of a wild species.
1b.Approach (from AD-416)
The number, types and acreages planted to transgenic crops are increasing. Consequently, there is a need to predict the likelihood of gene escape for different crops and a need to develop methodology to determine the impact of a transgene as it introgresses into wild populations. Because many crops benefit from insect pollination, part of our research investigates how distinct insect pollinators disperse pollen from plant to plant and ultimately among populations (gene flow). A better understanding of the impact of pollinator type on pollen dispersal would help us evaluate the differential risk of gene escape for distinct insect-pollinated crops while increasing our ability to select alternative pollinators for specific crops in the event of a major honeybee decline. On the one hand we study the impact of pollinator group on pollen dispersal and gene flow using the blue columbine as a model system. Information developed using this system will later be applied to different crops. On the other hand we examine the consequences of a disease resistance transgene that confers resistance to three economically important squash viruses as it introgresses into wild populations. We determine both the direct effects of the transgene on the fitness of free-living Cucurbita pepo (wild squash) and the indirect effects on diabroticite beetles (the primary non-target herbivore) and bacterial wilt (the major disease that these beetles transmit). In addition we measure gene flow among wild squash populations and gather basic information on their pollination biology and mating system. These types of data are critical to the efficient evaluation by regulatory agencies of the potential risk of transgenes introduced into wild plant populations.
We have started applying to alfalfa the knowledge on gene flow by insect pollinators gained from our model system, the Rocky Mountain columbine. The long-term goal of the alfalfa research is to develop a model of gene movement in insect-pollinated crops in order to make quantitative predictions on gene flow. This is relevant to Plant Biotechnology Risk assessment. It also contributes to our first objective of elucidating principles and data requirements for better predictions of gene flow via pollen in insect-pollinated crops. We examine how different pollinators and landscape features affect pollinator behavior and how this ultimately influences gene flow via pollen. The model can then be used to predict gene flow by various pollinators on different crops grown in distinct landscapes to help contain the flow of transgenes and control gene spread.
This year, we determined how the model should be constructed and started obtaining and analyzing empirical data to test some of the model's assumptions. For example, if pollinator movement between two flower clusters is not correlated to the previous movement, the process of pollen dispersal becomes easier to model. We also determined the directionality of movement for different pollinators in order to include directionality properly into the model. Non randomness of movement is an important aspect of insect pollinator behavior which can greatly influence gene flow and the modeling of the process. We also started looking at differences in patterns of movement between two pollinators, honey bees and bumble bees as we are interested in comparing and predicting gene flow for distinct pollinators. Finally, we set up two patches of alfalfa of different sizes to determine how patch size affects pollinator behavior and ultimately gene flow for different pollinators.
In addition to the alfalfa project, we are working on gene flow in carrots. We have collected leaves from a number of wild carrot populations within a five mile radius of a carrot breeding area to examine if cultivar genes could be detected in the wild populations. We will use genetic markers to determine if gene flow has occurred between the cultivars and the wild carrots.
Using empirical data to test assumptions of the theoretical model. In order to optimize the usefulness of a theoretical model and its future applicability, it is important to test the assumptions of the theoretical model with empirical data. We have done this for the theoretical model that we are developing to predict gene flow in insect-pollinated crops. This model is important to help predict and minimize gene flow for the increasing numbers of transgenic insect-pollinated crops planted in the U.S. and worldwide. We have determined that the distance traveled between two flower clusters is independent of the order of visit in a pollinator's foraging bout. A foraging bout represents the number of flower clusters visited by the pollinator between the time it enters and leaves the patch of plants. In other words, we found that the distance traveled between two flower clusters was independent of the order in which the clusters of flowers were visited by pollinators during their visit to a given patch of plants. This is a very important assumption for a model that links pollinator behavior to gene flow and such independence renders the model more tractable mathematically. Testing the assumptions of the model validates the mathematical structure of the model and improves its applicability. It also tests important aspects of pollinator behavior.
Developing a model of gene flow by insect pollinators. With the increasing acreages planted to transgenic crops and the increasing number of transgenes inserted into some crops, it is important to develop methodology to predict and try to minimize gene flow or the movement of genes from such crops. However, relative to wind pollination, the development of models to predict and help reduce gene flow in insect-pollinated crops is seriously lagging behind. Such models would also help minimize gene flow between varieties in seed production areas. This year we have determined the structure of the mathematical model needed to link pollinator behavior to gene flow. Our model will be useful to farmers as well as bioregulatory agents to help predict gene flow and to highlight planting practices that could help minimize gene flow from transgenic crops and between different varieties in seed production areas.