Parasitic plants are found in 13 angiosperm families and occupy a wide range of habitats. The most economically important parasitic plants are Striga and Orobanche species of the Orobanchaceae, a monophyletic group of root parasites with approximately 90 genera and more than 2000 species . The Striga genus is composed of 30–35 species, over 80% of which are found in Africa, while the rest occur in Asia and the United States. Among the five major Striga species, S. hermonthica (Del.) Benth. and S. asiatica Kuntze. are the most important cereal weeds, whereas S. gesnerioides (Willd.) Vatke parasitizes cowpea and other legumes and is a serious constraint to legume production.
The Striga life cycle is highly synchronized with that of the host and generally involves the stages of germination, attachment to host, haustorial formation, penetration, establishment of vascular connections, accumulation of nutrients, flowering and seed production . Germination of Striga seeds only take place in response to chemical cues, most commonly strigolactones, produced by the host and in some cases non host species [3, 4]. It is believed that host-derived chemical signals further guide haustorial formation and subsequent attachment to the host. After penetration of the cortex, haustorial cells undergo a remarkable differentiation process to form vessels that form a continuous bridge with the host xylem  that serve as a conduit for host derived nutrients and water.
Economic losses due to Striga are enormous. All of the cultivated food-crop cereals (maize, sorghum, millets, wheat and upland rice) are parasitized by one or more Striga spp . Overall, Striga infests two-thirds of the arable land of Africa and constitutes the biggest single biological cause of crop damage in Africa in terms of grain yield loss, estimated at 40% and worth $US 7 billion annually .
Control options for Striga are limited. These have generally included modified/improved cultural practices (e.g., crop rotation, intercropping/trap crops, different planting techniques, hand weeding, management of soil fertility), use of herbicide containing seed dressing, direct chemical treatment of soil to reduce seed levels in the soil, and identification of resistant (the ability of a host to prevent/limit Striga attachment/growth) and/or tolerant (the ability of a host to maintain biomass and yield in spite of Striga infection) germplasm for directed breeding .
Overall, Striga management practices are limited by our understanding of the biology of the parasite-host interaction. Such information is vital for development of appropriate management strategies using both genetic modification (GM) and non-GM approaches . With the ongoing parasitic plant genome project (http://ppgp.huck.psu.edu/), parasitic plants are fast entering the genomics era. These efforts will bring to light a large number of genes (including resistance genes) with unknown functions, underscoring the need for functional genomics tools for studying host-parasite interactions .
We hypothesized that many genes involved in Striga-host interactions are expressed in roots, thus a genetic transformation method that rapidly and efficiently generates a large number of transgenic host roots would provide an excellent system for studying the functions of genes involved in all aspects of Striga-host interactions.
The soil bacterium Agrobacterium rhizogenes is a naturally occurring plant pathogen  that can transfer T-DNA into the genomic DNA of plants. Infected plant cells that integrate a root inducing (Ri) plasmid-derived T-DNA from A. rhizogenes develop a large number of neoplastic, plagiotropic transformed ‘hairy’ roots . The feasibility of using A. rhizogenes in plant transformation has been demonstrated in a diverse array of plant families [11–15] for various applications e.g. production of stably transformed plants, [16, 17], gene analysis, [18–20] secondary metabolite production reviewed in , plant-microbe interactions  and plant-pathogen interactions .
Of the diverse range of A. rhizogenes mediated transformation applications, a key milestone was the development of ‘composite’ plants . The term ‘composite’ plant was coined to describe plants that have a wild type shoot and a transformed root stock. Composite plants present an ideal system for gene function studies of plants in association with other organisms. As such, they have been extensively used in analyses involving infection of legumes with rhizobia and nitrogen fixation [24, 25] as well as host plant associations with mycorrhiza . In general, composite plants offer the following advantages; (i) root biology can be studied in the roots of whole plants rather than in axenic cultures, (ii) since every transformed root is an individual event, multiple transgenic events can be obtained in a single transformation experiment, and (iii) they can be maintained outside of tissue culture after induction  so the amount of time required to generate transgenic plant tissue in transformation is greatly reduced.
Despite successful application of composite plants in elucidating plant-microbe interactions, the importance of maize as a model for genetics, the importance of Striga as a root parasite, and the enormous amount of host-parasite interaction data obtainable from composite hairy roots, no transgenic hairy root composite system has been developed for any of the Striga hosts. Here we show that A. rhizogenes can be used to efficiently produce transgenic hairy roots in maize. We further show that transgenic roots of composite maize plants can be infected by the parasitic plant S. hermonthica and that this system can be used to study Striga-maize interactions as a functional genomics tool.