Weed Science, 53:929–937. 2005

Potential model weeds to study genomics, ecology, and physiology in the 21st century Wun S. Chao

Corresponding author. U.S. Department of Agriculture—Agricultural Research Service, Biosciences Research Laboratory, Plant Science Research, Fargo, ND 58105-5674 [email protected]

Dave P. Horvath James V. Anderson Michael E. Foley

U.S. Department of Agriculture—Agricultural Research Service, Biosciences Research Laboratory Plant Science Research, Fargo, ND 58105-5674

Plant model systems have contributed greatly to the dramatic progress in understanding the fundamental aspects of plant biology. Using model weeds will also help facilitate focused funding and research in the weed science community. Criteria for developing model weeds require attention to weedy characteristics that impart economic losses and a wide geographic distribution, attributes that present the potential for political and scientific support. Expressed sequence tag (EST) databases for model weeds are the most practical approach to identifying new genes and obtaining data on the gene expression underlying weedy characteristics. Weeds such as Canada thistle, eastern black nightshade, johnsongrass, jointed goatgrass, leafy spurge, waterhemp, and weedy rice are proposed as model systems. Nomenclature: Canada thistle, Cirsium arvense (L.) Scop CIRAR; common waterhemp, Amaranthus rudis Sauer AMATA; eastern black nightshade, Solanum ptycanthum Dun. SOLPT; johnsongrass, Sorghum halepense (L.) Pers SORHA; jointed goatgrass, Aegilops cylindrica Host. AEGCY; leafy spurge, Euphorbia esula L. EUPES; red rice (weedy rice), Oryza sativa L. ORYSA; tall waterhemp, Amaranthus tuberculatus (Moq.) J. D. Sauer AMATU. Key words:

Value of Model Plants Model Systems and Approaches Used in Plant Biology During the past 25 yr, phenomenal gains in plant biology have been achieved. For example, studies that have identified signal transduction cross talk between hormones, metabolic and environmental signals, and development have shown how plants alter their physiology in response to certain cues (Bishop and Koncz 2002; Finkelstein et al. 2002; Nishimura et al. 2004; Richards et al. 2001; Turner et al. 2002; Wang et al. 2002). Key genes and processes that are required for plant development have been characterized (Bowman and Eshed 2000; Clark 2001), and new signaling mechanisms, such as the action of microRNAs in gene silencing, have been discovered (Dugas and Bartel 2004). These and other findings have led to critical insights into how plants evolved in form and function. Even critical questions in ecology, such as the molecular responses of plants to herbivores and the selection of mutations that respond to such pressures, are being answered (Weinig et al. 2003). The information gained has been so in-depth that it is no longer inconceivable to produce computer models that predict plant responses to virtually any stimuli or to engineer plants by altering the metabolism of entire pathways to produce specific and complex compounds (Minorsky 2003). Research in plant biology has blossomed primarily because of extensive and focused studies on a very small number of plant species. For example, plants such as Arabidopsis, pea (Pisium sativa), and snapdragon (Antirrhinum majus) have served as models for studies on hormone physiology and basic developmental processes that affect plant architecture; grasses such as rice (Oryza sativa) and corn (Zea mays) have been used extensively to study genome evolution; and trees such as poplar (Populus spp.) and loblolly pine (Pinus

Genomics, model systems, weedy characteristics, weed science.

taeda L.) are being used to study bud dormancy and wood formation. Even though these species have been widely used to determine fundamental processes in plant biology, very few carry traits for the study of weedy characteristics. Thus, the weed science community could benefit greatly by choosing and promoting a limited number of model weeds.

What Traits Make a Good Model Plant So, one might ask, why were these plants chosen as the models over all others? When Arabidopsis was put forth as a potential model plant, it was noted to have a small genome, a short life cycle, and simple genetics, and it could be easily transformed, pollinated, and manipulated in the laboratory. In addition, it was a member of the Brassicaceae family, which is related to several major food and oilseed crops (Meyerowitz 2001; Meyerowitz and Pruitt 1985). For others, such as maize, one might surmise it was chosen as a model plant because of its importance as a major food crop. However convincing these traits are, they are certainly not universal among model plants. Snapdragon, for instance, is neither genetically simple nor fast growing, nor is it even related to a major food crop, yet it has become a model plant for studying plant development. When asked what he did to help drive the development of Arabidopsis as a model system, Elliot Meyerowitz replied that he simply let everyone know about Arabidopsis and that he sent seeds and information to anyone who asked. Obviously, rumors of (leafy) him placing subliminal (spurge ) messages in all (a model ) of his (for) talks and lectures (bud ) were apparently completely (dormancy ) unfounded. Thus, in all likelihood, the primary reason these plants were chosen over others was that they possessed intrinsic qualities that made them amenable to study and that a reasonably well-organized core of scientists was able to convince others of the benefits. The fundChao et al.: Potential model weeds

•

929

ing for the development of new tools and databases could be justified only after the initial growth of a community that possessed some basic tools, resources, and information on the physiology and genetics of these organisms. However successful the use of model plants has been to the general fields of plant biology, not every scientific community has exploited the benefits derived from resources committed to the development of model plants. The weed science community has benefited only tangentially from information gained by the study of model plants. Yet research on model plants could provide useful information to the weed science community. For example, nearly 80% of the genes in Arabidopsis are likely to have orthologous genes (i.e., genes with similar sequence and function) in everything from barnyardgrass (Echinochloa crus-galli) to horseweed (Conyza canadensis) (Yu et al. 2002). It is certainly possible to study the physiological processes involved in herbicide resistance in Arabidopsis (Jander et al. 2003; Roux et al. 2004) or to identify genes involved in weed growth and development cloned by homology to their counterparts in other model plants (Horvath and Schaffer 2003). Also, model Solanaceae, such as tobacco (Nicotiana tabacum), tomato (Lycopersicon esculentum), and potato (Solanum tuberosum), are closely related to weeds such as tropical soda apple (Solanum viarum), horsenettle (Solanum carolinense), and black nightshade (Solanum nigrum). The same is true for many other plant families. Consequently, the genetic resources developed for some model plants could also be used to identify and clone genes that influence weedy characteristics. Such studies using sorghum species have identified genes involved in rhizogenous reproduction from johnsongrass (Skinner et al. 2003). Additionally, the information gained from sorghum was extended to identify similar genes in wild rice (Oryza longistaminata) (Hu et al. 2003).

Why Model Systems Are So Powerful Surprisingly, many of the extensively studied model plants are not major crops. Thus, what was it that allowed so much to be learned from such seemingly uninteresting species? Undoubtedly, the presence of significant funding for research on these plants is the primary reason these species are so well studied. However, it must be recognized that this funding was granted only after careful, forceful, and obviously accurate arguments that such studies would succeed in producing information of general interest and benefit to the scientific and agricultural communities. Substantial funding to study any single model plant provides the opportunity to develop extremely powerful tools that assist all studies thereafter. Vast and freely available collections of expressed sequence tags (ESTs) from multiple cDNA libraries, in combination with whole genome sequencing projects, provide researchers with convenient resources for genes of interest. ESTs are obtained by singlepass sequencing of randomly selected clones. Collections of ESTs from species-specific and tissue-specific libraries are defined as EST databases. These databases enhance the ability to identify genes that are thought to play roles, particularly in physiological processes, simply on the basis of their similarity to better-characterized genes from other organisms. In addition, EST databases can identify previously uncharacterized, novel, and unknown transcribed genes, i.e., transcripts. These resources make it possible to ask questions 930

•

Weed Science 53, November–December 2005

about the function of specific members of large gene families and answer the broader question of how these families evolved functionally. Once these gene resources (i.e., EST databases and genome sequences) are available, DNA microarrays can be designed to address the question of how any conceivable stimulus regulates global patterns of gene expression (transcript levels). Solely on the basis of the coordinated expression or clustering of known and unknown genes after specific treatments, the potential function of unknown genes can be hypothesized. Freely available data repositories of numerous microarray experiments provide information to answer questions concerning function and regulation for any given gene. The use of expression data, in combination with genome sequences, allows the detection of short transcription factor binding sites shared between clusters of coordinately regulated genes. This provides a powerful tool to identify the specific genes and signaling components responsible for sensing the environmental or developmental cues. Finally, the development of seed collections with insertional mutations in defined genes, identified through sequence databases, has made the previously difficult task of functional characterization of unknown genes a reasonably simple procedure. Funding alone, although required to support research and tool development, does not adequately explain the success of these model systems. Model plants allow researchers to connect seemingly disparate information from multiple lines of study and to share intellectual and material resources to expedite scientific discovery. Developing a pool of scientists working on the same model organism provides the opportunity to explain findings such as the identification of regulatory genes independently playing roles in different signaling pathways. Such chance observations have resulted in many novel insights, such as the interactions that involve abscisic acid and sugar signaling, as well as the role of both these signals in light responses and plant growth and development (Gibson 2004; Leon and Sheen 2003). The free flow of information and the development and extension of such tools have enfranchised new scientists, allowing them to enter the field and rapidly realize success in developing their novel ideas. Likewise, these tools and this information have provided established scientists the opportunity to use their intuition and experience from different plants to easily test their theories in model systems.

Potential for Model Weeds

Need for Model Weeds Although model systems could prove useful for studies in weed science, most model plants do not display multiple weedy characteristics that can be used for direct investigations. A recent review on the potential for using genomicbased research in weed science highlighted the need to incorporate these powerful techniques into the repertoire of tools used by weed scientists (Basu et al. 2004). However, attempting to build these genomic-based tools for the many weeds being studied by all weed scientists would be costprohibitive. Thus, it is timely to select several model weed species that will provide the weed science community with a focused approach to determine the nature and mechanisms involved in weedy characteristics. Such species should be carefully chosen to maximize model plant traits (Table 1),

TABLE 1. Relative comparison of desired traits for proposed annual and perennial model weeds.a Traits

Economic impactb Ecotypes Genomic resources Geographic distributionb Growth in limited area Mutants Political support Rapid cycling Related to food crop Simple genetics Small genome size Scientific communityc Transformablec Weedy characteristics a 2, poor; 1, fair; 11, good; b U.S. and Canada. c In subject or related species.

AEGCY

AMATA

ARATH

CIRAR

EUPES

ORYSA

SORHA

SOLPT

11 1 11 11 1 ? 11 1 111 1 2 11 1 11

11 11 2 11 11 ? 1 1 1 ? 11 11 ? 11

2 111 111 1 111 111 1 111 111 111 111 111 111 1

111 1 1 11 1 ? 11 1 1 11 1 11 1 11

11 11 11 11 11 ? 1 1 11 2 11 11 1 11

11 11 111 1 1 11 1 1 111 111 11 111 1 11

11 1 11 11 1 ? 11 1 11 1 1 11 1 11

11 1 11 1 1 ? ? 1 111 ? ? 11 11 11

111, excellent; ?, unknown.

represent important annual and perennial grass and broadleaf weeds, and manifest many weedy characteristics (Table 2). Since funding is likely to be limited, priority should be given to weeds for which preexisting tools and information are available. For example, resources already being developed for the wheat genome project could be used directly and immediately on jointed goatgrass (Foley 2002). Currently, genomic resources for sorghum (Sorghum bicolor), rice, and cassava (Manihot esculenta) are being used to assist studies on johnsongrass, weedy rice, and leafy spurge, respectively.

Characteristics of Interest to Weed Scientists The selection of a model system from a weed centric point of view depends on the characteristics and traits that make weeds particularly persistent and pernicious as they relate to human activities. Baker (1974) provided a list of weedy characteristics, and additional characteristics and traits (Table 2) have become evident through observation and research (Basu et al. 2004; Bennett et al. 1998; Gu et al. 2005). Baker (1974) proposed the idea of ‘‘general-purTABLE 2. Characteristics of weeds that might be considered in selection of a model experimental system. Crop mimicry in life cycle, morphology, or physiology Cross-pollination by wind or unspecialized organisms Dormancy in seeds or vegetative propagules Facultatively self-compatible Germination in many environments Resistance to control measures, e.g., herbicide, mowing Interspecific interference, e.g., competitive, allelopathic, parasitic Longevity of seeds and vegetative propagules Perennials deep rooted; hard to uproot Perennials readily regenerated from fragments Polyploid and low nuclear DNA content Propagules adapted for short and long distance dispersal Rapid growth to flowering Seed appendages for dispersal e.g. awns, barbs, pappus, etc. Seed output high under favorable environmental conditions Seed output under a variety of environmental conditions Seed production as long as growing conditions permit Seed shattering Vegetative reproduction, e.g., root/crown buds, rhizomes, tubers

pose genotypes’’ in which weeds evolved to encompass multiple weedy characteristics that impart tolerance to different environments and thus the potential for range expansion. Building on this idea, we propose that, for a given characteristic, different combinations of genes and alleles account for the diversity in phenotypic adaptation required for survival. Logically, general-purpose genotypes must result from the quantitative genetic control of characteristics that account for the tolerance and plasticity of weeds (Table 2).

Practical Realities In addition to considering the perceptible traits of model weeds (Table 1) and the characteristics that define weeds (Table 2), other realities must be considered in the development of model weed systems. The selection of model weeds should be based on a reasonably wide geographic distribution, serious economic impact, and expectation that research on them would gather broad organizational and political support. In addition, there must be the presence of scientific and administrative leadership with the educational outreach to gather and maintain grass roots support. Broad organizational support may already be in place if a weed is economically important and widely distributed. For example, Skinner et al. (2000) found 506 plant species were listed as noxious by 38 governmental bodies in the United States and parts of Canada. Of those, 45 were listed multiple times, ranging from 33 for Canada thistle (Cirsium arvense) to 6 for silverleaf nightshade (Solanum elaeagnifolium). Therefore, Canada thistle might have broad political support as a model weed. Once a consensus is reached in the scientific community to develop a particular weed as a model system, it will be important to communicate with administrators, stakeholders, and weed science students to convey the benefits and potential outcomes of research using model systems. As the process moves forward, one or two key administrators, with the knowledge and experience to direct programs and gather resources, need to be incorporated into the team. Those of us who have watched and participated in applied and fundamental research aimed at the integrated management of leafy spurge have witnessed the power that science, leadership, and politics have to move program goals Chao et al.: Potential model weeds

•

931

forward (Anderson et al. 2003). For many individual researchers and institutions, deviations from applied research aimed at an immediate customer’s needs can be risky. Yet without risk, long-term gains in understanding the biology and ecology of weeds will be few and far between. Therefore, the weed science community must spend the time, energy, and resources to communicate with a variety of partners, customers, and stakeholders that fundamental research and model systems are critical to solving some of the more intractable questions in weed science.

Current Model Weeds

Perennial Broadleaf Leafy spurge is currently being promoted as a model for the study of perennial broadleaf weeds. Leafy spurge is a member of the genetically diverse Euphorbiaceae family that includes important crop, horticultural, weedy, and endangered species. For example, cassava is one of the most important human food crops in the world (Anderson et al. 2004). Castor bean (Ricinus communis) is a major source of castor oil and has gained attention because of the escalating threat of bioterrorism resulting from the production of ricin. Other important members of the family include the rubber tree (Hevea brasiliensis), poinsettia (Poinsettia pulcherrima), and endangered species such as Akoka (Chamaesyce spp.) and telephus spurge (Euphorbia telephioides). Leafy spurge is a highly competitive and invasive plant found in at least 35 states and 6 Canadian provinces (CAB International 2004b). Some of the more important weedy characteristics that make this plant a good model system include vegetative reproduction, seed and bud dormancy, seed longevity, resistance to control measures, a deep and extensive root system, and early spring growth and flowering. In addition, variable susceptibility to certain biological agents could be a topic of investigation (Lym et al. 1996). Leafy spurge has also been reported to be tolerant to abiotic conditions such as cold and drought, the study of which could provide insight into novel methods for weed management (Anderson and Davis 2004). In addition, current reviews provide a body of knowledge on the biology and ecology of leafy spurge (Anderson et al. 2001; CAB International 2004b; Horvath et al. 2003). Leafy spurge is being used as a model system to study the growth and development of vegetative propagules because of the abundance of underground adventitious buds that form on the crown and roots (commonly referred to as crown and root buds) and because of its relatively fast growth rate under greenhouse and field conditions. Turnaround time from the propagation of meristem cuttings in the greenhouse to the development of visible crown and root buds can be as short as 2 mo. An average population of 2,800 leafy spurge plants can be maintained in a 10- by 10-m greenhouse. Leafy spurge is also easily maintained in garden plots, and portable potting systems have been designed that allow plants grown under field conditions to be transferred to controlled environments without major disturbances to the root system. Of the potential model perennial broadleaf weeds being considered, leafy spurge likely leads the way in genomics resources. A leafy spurge EST database is being generated from a normalized cDNA library constructed from whole932

•

Weed Science 53, November–December 2005

plant leafy spurge tissues subjected to multiple abiotic and biotic stresses, different dormancy states, and growth induction regimes. This EST database, being developed by the U.S. Department of Agriculture—Agricultural Research Service (USDA-ARS) and the University of Illinois at Urbana, currently contains approximately 46,000 high-quality sequences that have been predicted to represent in excess of 23,000 unique leafy spurge sequences. In addition to the normalized library, three cDNA libraries are available and have been used to generate additional EST resources. One cDNA library was developed from 3-d growth-induced root buds (Anderson and Horvath 2001) and was originally used to isolate ESTs for the development of preliminary microand macroarrays (Anderson et al. 2004). This library was designed to allow two-hybrid screening for protein–protein interactions. Two subtracted libraries (forward, preferentially expressed in growing buds, and reverse, preferentially expressed in dormancy buds) have been constructed and used to screen for preferentially expressed dormancy genes (Jia et al. 2005). A leafy spurge genomic library is available and serves as an important resource for identifying and characterizing cis-acting elements. However, two current drawbacks of leafy spurge as a model are its complex genetic makeup (Schulz-Schaeffer and Gerhardt 1987, 1989) and the unresolved transformation system. The chromosome number of leafy spurge varies from 2n 5 48 to 2n 5 60, with the hexaploid (2n 5 60) being the most prevalent in nature (CAB International 2004b). The hexaploid species of leafy spurge contains 2,069 Mbp per haploid genome. Leafy spurge research has mostly been supported through in-house and area-wide USDA-ARS and USDA-Cooperative State, Research, Education, and Extension Service (CSREES) Hatch and special project funds. In addition, USDA-National Research Initiative (NRI) competitive grant funds have been awarded to study the biology and ecology of leafy spurge. Another funding strategy for augmenting genomics research is through collaborative efforts with the international community. For example, previous research has shown that leafy spurge genomic resources can be used to identify differentially expressed genes in other Euphorbs such as cassava (Anderson et al. 2004). Thus, collaborations are under way with the International Institute of Tropical Agriculture (IITA) to develop Euphorbiaceae-specific microarrays for the scientific community. Such family-based genomic approaches are important for broadening partnerships within the scientific community (Forcella 2003) and maximizing resources and impact in a funding-limited environment.

Annual Grass The Poaceae family contains many serious weeds such as weedy rice, which is an important annual weed worldwide and, certainly, in parts of the southern United States. Its status as a model weed stems from the resources devoted to rice as a model system. Rice is the base genome for determining syntenic relationships with other cereal species such as wheat and maize (Chen et al. 2002), has a relatively small genome size (430 Mbp per haploid genome), and is nearly fully sequenced (Goff et al. 2002; Yu et al. 2002). Genomic resources include a rice EST database (Wu et al. 2002), commercially available microarray chips (Meyers et al. 2004; Rensink and Buell 2004), abundant genetic markers (Tem-

nykh et al. 2001), integrated genetic and physical maps (Chen et al. 2002), and a number of quantitative trait loci (QTL) for traits of interests, including some weedy traits (Gu et al. 2004; Ishimaru et al. 2001). In addition, wheat and barley EST databases afford the opportunity to investigate the genome evolution of domesticated and weedy grass species (Meyers et al. 2004). Weedy rice could be used to examine most of the characteristics in Table 2. Weedy rice has been used to study physiological, biochemical, and genetic aspects of seed germination, dormancy, and afterripening (Footitt and Cohn 1995; Gu et al. 2004; Leopold et al. 1988), crop interference (Diarra and Talbert 1985), allelopathy (Duke et al. 2003), gene flow between crops and weeds (Gealy et al. 2003), rhizomatousness (Hu et al. 2003), and other weedyrelated traits (Gu et al. 2005). Funding for most research on weedy rice in the United States comes from in-house and special USDA-ARS and USDA-CSREES project funds. However, some projects have been funded through USDANRI and the National Science Foundation (NSF) competitive grant programs. Internationally, limited funds for research on weedy rice are garnered through research that has a tangential relationship to cultivated rice. Topics of recent interest include domestication of rice (Bres-Patry et al. 2001), shattering and dormancy (Cai and Morishima 2000), and gene flow from cultivated rice to weedy rice (Chen et al. 2004). Our rationale for using weedy rice as a model system is that the map-based cloning of seed dormancy genes from other weeds such as wild oat is impractical (Foley 2002). Thus, we encourage all interested crop and weed scientists to work toward map-based cloning of rice genes to determine if orthologs regulate dormancy in other weedy grasses.

Perennial Grass Johnsongrass is a perennial and is another example of a serious weed within the Poaceae family. Holm (1969) listed johnsongrass as one of the 10 worst weeds in the world, because it can interfere with the production of crops such as cotton (Gossypium spp.), corn, sorghum, soybean, and sugarcane (Saccharum officinarum) (CAB International 2004a). Johnsongrass has a chromosome number of either 2n 5 20 (diploid) or 2n 5 40 (tetraploid). Johnsongrass, the tetraploid species found in the United States, contains 1,617 Mbp per haploid genome (Bennett and Leitch 2003) and was likely derived from naturally occurring crosses between Sorghum bicolor (cultivated sorghum, 2n 5 20) and Sorghum propinquum (a grassy weed, 2n 5 20) (CAB International 2004a). Consequently, johnsongrass makes a good model weed because of its close relationship to cultivated sorghum and because of the genomic efforts directed toward these species. For example, an NSF grant enabled scientists at the University of Georgia to contribute 110,000 ESTs to GenBank, representing a 15,000-unigene set for cultivated sorghum. Renewed funding will help increase the number of unigenes from cultivated sorghum to approximately 20,000 (http://fungen.org/Index.htm). In addition, NRI and NSF funding have allowed sequencing of 1,200 ESTs from a rhizome cDNA library of johnsongrass and 5,000 ESTs from a rhizome cDNA library of S. propinquum. Johnsongrass has many weedy traits. It reproduces through seeds and rhizomes. A single plant can produce .

28,000 seeds per year (Monaghan 1979), which remain viable and germinate intermittently for as long as 6 yr (Leguizamon 1986). One plant of johnsongrass can produce 8 kg of fresh weight and 70 m of rhizomes in a single growing season (Monaghan 1979), and this vigorous rhizome system can be spread effectively by tillage. Paterson et al. (1995) identified QTLs that affect rhizomatousness and tillering using progenies from the cross between S. bicolor and S. propinquum. To date, johnsongrass has evolved resistance to several groups of herbicides: ACCase and acetolactate synthase (ALS) inhibitors, dinitroanilines, and others (Heap 2004). The ability to overcome herbicidal control poses a great threat to crop production but provides opportunities for using johnsongrass to study herbicide resistance.

Annual Broadleaf Arabidopsis thaliana is a diploid (2n 5 10) and contains 171.5 Mbp per haploid genome (Bennett and Leitch 2003). Arabidopsis is usually not thought of as a weed, but it does carry several weedy traits. Seed dormancy has been studied and has resulted in the elucidation of several different signaling pathways that control these traits (Alonso-Blanco et al. 2003; Debeaujon et al. 2000; Debeaujon and Koornneef 2000; Koornneff et al. 2002; Steber and McCourt 2001). The potential for using Arabidopsis to understand weedy traits lies in the tools that have been developed to study biological and ecological processes and in knowing that the genes and signaling pathways controlling these processes exist in most other plants. Resources available on The Arabidopsis Information Resource (TAIR) website (www. arabidopsis.org, last accessed November 23, 2004) include links to sites with relative expression data for most of the genes from Arabidopsis under hundreds of experimental conditions, links to DNA and protein sequences, mutant collections, literature, and protocols and techniques for studying the molecular and genomic aspects of Arabidopsis. Therefore, if there is a question about how some environmental condition will affect the growth of a weed, chances are good that someone has already determined the effect of that environmental condition on the growth of Arabidopsis, determined why that environment altered plant growth, and made that information readily available in an easy-to-find and understandable format. Although Arabidopsis is currently considered the model plant and should serve as a template for the development of all other model systems, it is a poor example for annual broadleaf weeds (Table 2). Consequently, weed scientists should consider additional models for annual broadleaf weeds.

Potential Model Weeds

Canada Thistle Canada thistle is a competitive broadleaf perennial in the Asteraceae family. It is a noxious or prohibited weed in 43 states (http://invader.dbs.umt.edu/NoxiouspWeeds/, last accessed November 23, 2004). It is also a noxious weed in Canada and many other areas of the world (Holm et al. 1977). Canada thistle is a diploid (2n 5 34) with a small genome size (1,519 Mbp per haploid genome) (Bennett and Leitch 2003). This deep-rooted perennial weed is well adapted to a variety of environmental and edaphic condiChao et al.: Potential model weeds

•

933

tions in temperate regions and infests agronomic and horticultural crops, rangelands, turf and urban landscapes, riparian areas, and recreational and natural lands. Canada thistle also serves as an alternate host for insects and pathogenic microorganisms that attack various crops (Donald 1994). Because of this wide geographic distribution and impact on multiple ecosystems, further research on Canada thistle is likely to garner political support. Recently, the USDA-ARS and NSF have supported investigations on the population genetics of Canada thistle and related species, and a whole-plant normalized cDNA library has been developed and used to initiate a small collection of ESTs (J. V. Anderson, personal communication). Canada thistle reproduces from vegetative buds and from seed. Root fragments as small as 8 mm long by 3 mm diam are able to develop new plants (Moore 1975). Seed production is prolific, and seed can be easily spread long distances. Seed can remain viable for up to 20 yr in a soil seed bank (Madsen 1962). Debris of Canada thistle is reported to have allelopathic effects on surrounding plants (Donald 1994), and ecotypes with resistance to synthetic auxins have been found (Heap 2004). In addition, attempts at biological control have proven problematic, suggesting additional research is required (Jordon-Thaden and Louda 2003; Louda et al. 1997). Thus, this plant has a multitude of characteristics that can be studied to improve weed management (Table 2).

Waterhemp Waterhemp (Amaranthus spp.) is a troublesome annual broadleaf weed with a wide geographic distribution. Traditionally, waterhemp has been divided into two species, common waterhemp (Amaranthus rudis) and tall waterhemp (Sauer 1955). However, Pratt and Clark (2001) have questioned whether common and tall waterhemp are in fact one polymorphic species. Common waterhemp infestations are frequent from Nebraska to Texas, while tall waterhemp is prevalent from Indiana to Ohio. Both species are widespread in Iowa, Illinois, and Missouri (Hartzler 2003). Strong weedy characteristics and the development of herbicide-resistant biotypes have contributed to the increased frequency and severity of waterhemp infestations in corn and soybeans grown in the Midwest (Hartzler et al. 2004; Nordby and Hartzler 2004). Its rapid proliferation is not common in agriculture and has drawn the attention of weed scientists. Waterhemp has a relatively small genome size (657 Mbp per haploid genome in tall waterhemp) (Jeschke et al. 2003) and possesses many weedy characteristics relative to a potential model weed. It is a dioecious species that cross-pollinates (possibly between species?), leading to high genetic variability. Waterhemp can produce 3,000 to 300,000 seeds per plant under unfavorable or favorable conditions, respectively (Hartzler et al. 2004). The high seed production increases the probability of seeds being dispersed over long distances. These seeds display dormancy (Leon and Owen 2003) and can germinate under a wide range of environmental conditions throughout the growing season. Waterhemp seedlings grow very fast; the relative growth rate is 50 to 70% greater than that of other annual weedy species (Horak and Loughin 2000; Seibert and Pearce 1993). Most importantly, waterhemp has evolved resistance to ALS, photosystem II and protoporphyrinogen oxidase inhibitors 934

•

Weed Science 53, November–December 2005

(Heap 2004), and glyphosate (http://www.cropchoice.com/ leadstry080e.html?recid51292, last accessed December 2, 2004). These traits have made waterhemp a rapid invader in many crop production systems.

Nightshades Nightshades (Solanum spp.) are annual to short-lived perennials in the Solanaceae family (other examples include horse nettle, jerusalem cherry, Solanum pseudocapsicum, and tropical soda apple). Black nightshade is the best-known noxious weed among the nightshade species (Defelice 2003; Ogg et al. 1981) and is reported as a weed in . 37 crops and 61 countries around the world (Holm et al. 1991). Black nightshade is a hexaploid with a chromosome number of 2n 5 72; however, eastern black nightshade is common to the United States and is a diploid species with a chromosome number of 2n 5 24. Therefore, eastern black nightshade holds the best potential as a model for Solanum weed species. In addition, because of extensive studies on potato, tomato, and pepper (Capsicum annuum), numerous genetic resources are available and could prove useful for research in solanacious weeds. With the genomics resources available for these crop species, the development of genomic resources for nightshades would specifically allow evolutionary comparisons between crops and weeds in this family. Nightshade species, in general, reproduce only by seeds. Berries can contain 15 to 96 seeds, and a single plant can produce up to 30,000 seeds in a single season. Seeds remain viable after years in the soil seed bank and can germinate intermittently under favorable conditions (Defelice 2003). Nightshades also are toxic, compete with crops, impede harvest, and reduce crop quality by seed discoloration (Cooper and Johnson 1984; Lampe and McCann 1985). In addition, some nightshades have evolved resistance to photosystem II and ALS inhibitors and to bipyridiliums (Heap 2004). Because of these characteristics and relationships to other important horticultural plants worldwide, eastern black nightshade might serve as a suitable model for other solanaceous weeds.

Jointed Goatgrass Jointed goatgrass is a winter annual grass that spreads exclusively by seed. Jointed goatgrass is found in most major U.S. winter wheat (Triticum aestivum) production regions and has infested . 5 million acres of winter wheat cropland. Total losses due to jointed goatgrass infestation in the western United States annually exceed $145 million (Westbrooks 1998). Jointed goatgrass is an allotetrapoid and shares the D genome with winter wheat (Donald and Ogg 1991). This attribute allows the use of wheat genome resources (Anonymous 2004), which might reduce the cost of jointed goatgrass genomics research. Jointed goatgrass is a competitive weed that mimics the life cycle of wheat. Its seeds are similar in size and shape, making them difficult to separate. Seeds are dormant after shattering, require afterripening to germinate, and remain viable for 3 to 5 yr in the soil seed bank (Donald and Ogg 1991). Few, if any, herbicides can selectively control this weed in winter wheat because of species similarity (Seefeldt et al. 1998; Zemetra et al. 1998). These characteristics make crop protection extremely difficult; therefore, herbicide-re-

sistant transgenic wheat has been considered. However, interspecific hybridization between these species occurs and would likely result in herbicide-resistant jointed goatgrass (Zemetra et al. 1998). Nevertheless, interspecific hybridization could prove useful in a model system, as the genetics of weedy characteristics could more effectively be studied.

Conclusions and Recommendations Developing model weed systems will be an important step to ensure that focused funding and research occur in the weed science community. The benefits of such efforts to the plant science community have been highlighted in this review and provide the basis for developing model weed systems. However, no single weed will serve as a general-purpose model for all weedy characteristics. Since the Weed Science Society of America has recognized . 2,000 weedy species (Westbrooks 1998), it is crucial to select the most appropriate models that efficiently address the concerns of the weed science community. In this review, we propose current and potential models that represent important annual and perennial grass and broadleaf weeds and have multiple weedy characteristics, serious economic impact, wide geographic distribution, political support, and preexisting tools and information. The weeds proposed in this review may not represent a comprehensive list of potential models, but they provide the impetus for further discussion among weed scientists. Another way to select models might be to focus on the plant families that contain a majority of serious weeds. For example, we have not mentioned any members of the Cyperaceae family, which contains some of the world’s worst weeds (Holm 1977). Thus, other model weeds could be proposed on the basis of the criteria outlined in this review. It is not likely that the weed science community will receive the disproportional funding provided for model plants like Arabidopsis. Therefore, whole-genome sequencing is currently unrealistic. We recommend that resources be focused on developing models with a good potential for expression-based approaches (Chao 2002) that will aid in the identification of differentially expressed genes or proteins critical to the underlying blueprint for weedy characteristics. EST databases, such as those described for leafy spurge and johnsongrass, provide the most practical approach to identifying new genes, obtaining data on gene expression via DNA microarrays, and constructing genome maps. However, the community should not overlook the power and potential of developing maps for comparative genomics and of developing transformation systems and mutagenized lines. These tools and proposed model weeds are necessary to resolve the mechanisms and pathways that govern weedy characteristics and will provide the basis to develop and exploit new strategies for sustainable and biologically based weed management. Acknowledgment All authors contributed equally.

Literature Cited Alonso-Blanco, C., L. Bentsink, C. J. Hanhart, H.B.E. Vries, and M. Koornneef. 2003. Analysis of natural allelic variation at seed dormancy loci of Arabidopsis thaliana. Genetics 164:711–729.

Anderson, G. L., C. W. Prosser, L. E. Wendel, E. S. Delfosse, and R. M. Faust. 2003. The Ecological Areawide Management (TEAM) of leafy spurge program of the United States Department of Agriculture-Agricultural Research Service. Pest Manag. Sci. 59:609–613. Anderson, J. V., W. S. Chao, and D. P. Horvath. 2001. A current review on the regulation of dormancy in vegetative buds. Weed Sci. 49:581– 589. Anderson, J. V. and D. G. Davis. 2004. Abiotic stress alters transcript profiles and activity of enzymes involved in glutathione-metabolism in Euphorbia esula. Physiol. Plant. 120:421–433. Anderson, J. V., M. Delseny, M. A. Fregene, et al. 2004. An EST resource for cassava and other species of Euphorbiaceae. Plant Mol. Biol. 56: 527–539. Anderson, J. V. and D. P. Horvath. 2001. Random sequencing of cDNAs and identification of mRNAs. Weed Sci. 49:590–597. Anonymous. 2004. Aegilops cylindrical jointed goatgrass—Executive Summary of the National Jointed Goatgrass Research Program CSREESUSDA Special Grant. Available at http://www.jointedgoatgrass.org/. Baker, H. G. 1974. The evolution of weeds. Annu. Rev. Ecol. Syst. 5:1– 24. Basu, C., M. D. Halfhill, T. C. Mueller, and C. N. Stewart. 2004. Weed genomics: new tools to understand weed biology. Trends Plant Sci. 9: 391–398. Bennett, M. D. and I. J. Leitch. 2003. Plant DNA C-values database. www.kew.org/cval. Bennett, M. D., I. J. Leitch, and L. Hanson. 1998. DNA amounts in two samples of angiosperm weeds. Ann. Bot. 82:121–134. Bishop, G. J. and C. Koncz. 2002. Brassinosteroids and plant steroid hormone signaling. Plant Cell 14:S97–S110. Bowman, J. and Y. Eshed. 2000. Formation and maintenance of the shoot apical meristem. Trends Plant Sci. 5:110–115. Bres-Patry, C., M. Lorieux, G. Clement, M. Bangratz, and A. Ghesquiere. 2001. Heredity and genetic mapping of domestication-related traits in a temperate japonica weedy rice. Theor. Appl. Genet. 102:118–126. CAB International. 2004a. Sorghum halepense. In Crop Protection Compendium. 2004 ed. Wallingford, Great Britain: CAB International. [CD-ROM]. CAB International. 2004b. Euphorbia esula [original text by W. Chao and J. V. Anderson]. In Crop Protection Compendium. 2004 ed. Wallingford, Great Britain: CAB International. [CD-ROM]. Cai, H. W. and H. Morishima. 2000. Genomic regions affecting seed shattering and seed dormancy in rice. Theor. Appl. Genet. 100:840–846. Chao, W. S. 2002. Contemporary methods to investigate dormancy. Weed Sci. 50:215–226. Chen, L. J., D. S. Lee, Z. P. Song, H. S. Suh, and B. R. Lu. 2004. Gene flow from cultivated rice (Oryza sativa) to its weedy and wild relatives. Ann. Bot. 93:67–73. Chen, M. S., G. Presting, W. B. Barbazuk, et al. 2002. An integrated physical and genetic map of the rice genome. Plant Cell 14:537–545. Clark, S. E. 2001. Cell signaling at the shoot meristem. Nat. Rev. Mol. Cell Biol. 2:276–284. Cooper, M. R. and A. W. Johnson. 1984. Poisonous plants in Britain and their effects on animals and man. London: Her Majesty’s Stationery Office. 305 p. Debeaujon, I., K. M. Le´on-Kloosterziel, and M. Koornneef. 2000. Influence of the testa on seed dormancy, germination, and longevity in Arabidopsis. Plant Physiol. 122:403–413. Debeaujon, I. and M. Koornneef. 2000. Gibberellin requirement for Arabidopsis seed germination is determined both by testa characteristics and embryonic abscisic acid. Plant Physiol. 122:415–424. Defelice, M. S. 2003. The black nightshades, Solanum nigrum L. et al.— poison, poultice, and pie. Weed Technol. 17:421–427. Diarra, A.R.J. and R. E. Talbert. 1985. Interference of red rice (Oryza sativa) with rice (O. sativa). Weed Sci. 33:644–649. Donald, W. W. 1994. The biology of Canada thistle (Cirsium arvense). Rev. Weed Sci. 6:77–101. Donald, W. W. and A. G. Ogg, Jr. 1991. Biology and control of jointed goatgrass (Aegilops cylindrica), a review. Weed Technol. 5:3–17. Dugas, D. V. and B. Bartel. 2004. MicroRNA regulation of gene expression in plants. Curr. Opin. Plant Biol. 7:512–520. Duke, S. O., S. R. Baerson, F. E. Dayan, et al. 2003. U.S. Department of Agriculture—Agricultural Research Service research on natural products for pest management. Pest Manag. Sci. 59:708–717. Finkelstein, R. R., S.S.L. Gampala, and C. D. Rock. 2002. Abscisic acid signaling in seeds and seedlings. Plant Cell 14:S15–45.

Chao et al.: Potential model weeds

•

935

Foley, M. E. 2002. Weeds, seeds, and buds—opportunities and systems for dormancy investigations. Weed Sci. 50:267–272. Footitt, S. and M. A. Cohn. 1995. Seed dormancy in red rice (Oryza sativa). IX. Embryo fructose-2,6-bisphosphate during dormancy breaking and subsequent germination. Plant Physiol. 107:1365–1370. Forcella, F. 2003. U.S. Department of Agriculture—Agricultural Research Service research on pest biology: weeds. Pest Manag. Sci. 59:754–763. Gealy, D. R., D. H. Mitten, and J. N. Rutger. 2003. Gene flow between red rice (Oryza sativa) and herbicide-resistant rice (O-sativa): implications for weed management. Weed Technol. 17:627–645. Gibson, S. I. 2004. Sugar and phytohormone response pathways: navigating a signaling network. J. Exp. Bot. 55:253–264. Goff, S. A., D. Ricke, T.-H. Lan, et al. 2002. A draft sequence of the rice genome (Oryza sativa L. ssp. japonica). Science 296:92–100. Gu, X.-Y., S. F. Kianian, and M. E. Foley. 2004. Multiple loci and epistases control genetic variation for seed dormancy in weedy rice (Oryza sativa). Genetics 166:1503–1516. Gu, X.-Y., S. F. Kianian, and M. E. Foley. 2005. Seed dormancy imposed by covering tissues interrelates to shattering and seed morphological characteristics in weedy rice. Crop Sci. 45:948–955. Hartzler, R. G. 2003. Waterhemp—the perfect weed? www.weeds.iastate. edu/mgmt/2003/symposium.shtml. WSSA-North American Weed Management Association Invasive Plant Species Workshop, Kansas City, MO. Feb. 12–13. Hartzler, R. G., B. A. Battles, and D. Nordby. 2004. Effect of common waterhemp (Amaranthus rudis) emergence date on growth and fecundity in soybean. Weed Sci. 52:242–245. Heap, I. 2004. The International Survey of Herbicide Resistant Weeds. www.weedscience.com. Holm, G. L., D. L. Plucknett, J. V. Pancho, and J. P. Herburger. 1977. The World’s Worst Weeds: Distribution and Biology. Honolulu, HI: Hawaii University Press. Holm, L. 1969. Weed problems in developing countries. Weed Sci. 17: 113–118. Holm, L. G., D. L. Plucknett, J. V. Pancho, and J. P. Herberger. 1991. The World’s Worst Weeds: Distribution and Biology. Malabar, FL: Krieger. 609 p. Horak, M. J. and T. M. Loughin. 2000. Growth analysis of four Amaranthus species. Weed Sci. 48:347–355. Horvath, D. P., J. V. Anderson, W. S. Chao, and M. E. Foley. 2003. Knowing when to grow: signals regulating bud dormancy. Trends Plant Sci. 8:534–540. Horvath, D. P. and R. Schaffer. 2003. Urg1, a conserved and previously uncharacterized gene expressed preferentially in growing shoots of Arabidopsis and leafy spurge. Appl. Genomics Proteomics 2:169–179. Hu, F. Y., D. Y. Tao, E. Sacks, et al. 2003. Convergent evolution of perenniality in rice and sorghum. Proc. Natl. Acad. Sci. USA 100:4050– 4054. Ishimaru, K., M. Yano, N. Aoki, K. Ono, T. Hirose, S. Y. Lin, L. Monna, T. Sasaki, and R. Ohsugi. 2001. Toward the mapping of physiological and agronomic characters on a rice function map: QTL analysis and comparison between QTLs and expressed sequence tags. Theor. Appl. Genet. 102:793–800. Jander, G., S. R. Baerson, J. A. Hudak, K. A. Gonzalez, K. J. Gruys, and R. L. Last. 2003. Ethylmethanesulfonate saturation mutagenesis in Arabidopsis to determine frequency of herbicide resistance. Plant Physiol. 131:139–146. Jeschke, M. R., P. J. Tranel, and A. L. Rayburn. 2003. DNA content analysis of smooth pigweed (Amaranthus hybridus) and tall waterhemp (A. tuberculatus): implications for hybrid detection. Weed Sci. 51:1–3. Jia, Y., Y.-Q. Gu, D. P. Horvath, J. V. Anderson, and W. S. Chao. 2005. Subtractive hybridization as a tool for genes involved in dormancy and growth in underground adventitious buds of leafy spurge (Euphorbia esula). Annual Meeting of the Weed Science Society of America. Honolulu, HI. Feb. 7–10. Jordon-Thaden, I. E. and S. M. Louda. 2003. Chemistry of Cirsium and Carduus: a role in ecological risk assessment for biological control of weeds? Biochem. Syst. Ecol. 31:1353–1396. Koornneef, M., L. Bentsink, and H. Hilhorst. 2002. Seed dormancy and germination. Curr. Opin. Plant Biol. 5:33–36. Lampe, K. F. and M. A. McCann. 1985. AMA Handbook of Poisonous and Injurious Plants. Chicago: American Medical Association. 432 p. Leguizamon, E. S. 1986. Seed survival and patterns of seedling emergence in Sorghum halepense (L.) Pers. Weed Res. 26:397–403. Leon, P. and J. Sheen. 2003. Sugar and hormone connections. Trends Plant Sci. 8:110–116.

936

•

Weed Science 53, November–December 2005

Leon, R. G. and M.D.K. Owen. 2003. Regulation of weed seed dormancy through light and temperature interactions. Weed Sci. 51:752–758. Leopold, A. C., R. Glenister, and M. A. Cohn. 1988. Relationship between water content and afterripening in red rice. Physiol. Plant. 74:659– 662. Louda, S. M., D. Kendall, J. Connor, and D. Simberloff. 1997. Ecological effects of an insect introduced for the biological control of weeds. Science 277:1088–1090. Lym, R. G., S. J. Nissen, M. L. Rowe, D. J. Lee, and R. A. Masters. 1996. Leafy spurge (Euphorbia esula) genotype affects gall midge (Spurgia esulae) establishment. Weed Sci. 44:629–633. Madsen, S. B. 1962. Germination of buried and dry stored seeds III, 1934– 1960. Proc. Int. Seed Testing Assoc. 27:920–928. Meyerowitz, E. M. 2001. Prehistory and history of Arabidopsis research. Plant Physiol. 125:15–19. Meyerowitz, E. M. and R. E. Pruitt. 1985. Arabidopsis thaliana and plant molecular genetics. Science 229:1214–1218. Meyers, B. C., D. W. Galbraith, T. Nelson, and V. Agrawal. 2004. Methods for transcriptional profiling in plants. Be fruitful and replicate. Plant Physiol. 135:637–652. Minorsky, P. V. 2003. Achieving the in silico plant. Systems biology and the future of plant biological research. Plant Physiol. 132:404–409. Monaghan, N. 1979. The biology of johnsongrass (Sorghum halepense). Weed Res. 19:261–267. Moore, R. J. 1975. The biology of Canadian weeds. 13. Cirsium arvense (L.) Scop. Can. J. Plant Sci. 55:1033–1048. Nishimura, C. C., Y. Ohashi, S. Sato, T. Kato, S. Tabata, and C. Ueguchia. 2004. Histidine kinase homologs that act as cytokinin receptors possess overlapping functions in the regulation of shoot and root growth in Arabidopsis. Plant Cell 16:1365–1377. Nordby, D. E. and R. G. Hartzler. 2004. Influence of corn on common waterhemp (Amaranthus rudis) growth and fecundity. Weed Sci. 52: 255–259. Ogg, A. G., B. S. Rogers, and E. E. Schilling. 1981. Characterization of black nightshade (Solanum nigrum) and related species in the United States. Weed Sci. 29:27–32. Paterson, A. H., K. F. Schertz, Y.-R. Lin, S.-C. Liu, and Y.-L. Chang. 1995. The weediness of wild plants: molecular analysis of genes influencing dispersal and persistence of johnsongrass, Sorghum halepense (L.) Pers. Proc. Natl. Acad. Sci. USA 92:6127–6131. Pratt, D. B. and L. G. Clark. 2001. Amaranthus rudis and A. tuberculatus— one species or two? J. Torrey Bot. Soc. 128:282–296. Rensink, W. A. and C. R. Buell. 2004. Arabidopsis to rice. Applying knowledge from a weed to enhance our understanding of a crop species. Plant Physiol. 135:622–629. Richards, D. E., K. E. King, T. Ait-ali, and N. P. Harberd. 2001. How gibberellin regulates plant growth and development: a molecular genetic analysis of gibberellin signaling. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52:67–88. Roux, F., J. Gasquez, and X. Reboud. 2004. The dominance of the herbicide resistance cost in several Arabidopsis thaliana mutant lines. Genetics 166:449–460. Sauer, J. D. 1955. Revision of the dioecious amaranths. Madron˜o 13:5– 46. Schulz-Schaeffer, J. and S. Gerhardt. 1987. Cytotaxonomic analysis of the Euphorbia spp. (‘‘leafy spurge’’) complex. Biol. Zentbl. 106:429–438. Schulz-Schaeffer, J. and S. Gerhardt. 1989. Cytotaxonomic analysis of the Euphorbia spp. (‘‘leafy spurge’’) complex. II. Comparative study of the chromosome morphology. Biol. Zentbl. 108:69–76. Seefeldt, S. S., R. Zemetra, F. L. Young, and S. S. Jones. 1998. Production of herbicide-resistant jointed goatgrass (Aegilops cylindrica) 3 wheat (Triticum aestivum) hybrids in the field by natural hybridization. Weed Sci. 46:632–634. Seibert, A. C. and R. B. Pearce. 1993. Growth analysis of weed and crop species with reference to seed weight. Weed Sci. 41:52–56. Skinner, D. L., A. H. Paterson, W. K. Vencill, and H. Ma. 2003. Cloning and characterization of rhizome-specific genes. Poster 443, Plant and Animal Genomes XI Conference. January 11–15, 2003. Town & Country Convention Center, San Diego, CA. http://www.intl-pag.org/ 11/abstracts/P5dpP443pXI.html. Skinner, K., L. Smith, and P. Rice. 2000. Using noxious weed lists to prioritize targets for developing weed management strategies. Weed. Sci. 48:640–644. Steber, C. M. and P. McCourt. 2001. A role for brassinosteroids in germination in Arabidopsis. Plant Physiol. 125:763–769. Temnykh, S., G. DeClerck, A. Lukashova, L. Lipovich, S. Cartinhour, and

S. McCouch. 2001. Computational and experimental analysis of microsatellites in rice (Oryza sativa L.): frequency, length variation, transposon associations, and genetic marker potential. Genome Res. 11: 1441–1452. Turner, J. G., C. Ellis, and A. Devoto. 2002. The jasmonate signal pathway. Plant Cell 14:S153–164. Wang, K. L.-C., H. Li, and J. R. Ecker. 2002. Ethylene biosynthesis and signaling networks. Plant Cell 14:S131–151. Weinig, C., J. R. Stinchcombe, and J. Schmitt. 2003. Evolutionary genetics of resistance and tolerance to natural herbivory in Arabidopsis thaliana. Evol. Int. J. Org. Evol. 57:1270–1280. Westbrooks, R. L. 1998. Invasive plants, changing the landscape of Amer-

ica: fact Book. Federal Interagency Committee for the Management of Noxious and Exotic Weeds (FICMNEW). Washington, D.C. Pp. 4–5, 16–17. Wu, J. Z., T. Maehara, T. Shimokawa, et al. 2002. A comprehensive rice transcript map containing 6591 expressed sequence tag sites. Plant Cell 14:525–535. Yu, J., S. N. Hu, J. Wang, et al. 2002. A draft sequence of the rice genome (Oryza sativa L. ssp. indica). Science 296:79–92. Zemetra, R. S., J. Hansen, and C. A. Mallory-Smith. 1998. Potential for gene transfer between wheat (Triticum aestivum) and jointed goatgrass (Aegilops cylindrica). Weed Sci. 46:313–317.

Received December 29, 2004, and approved April 18, 2005.

Chao et al.: Potential model weeds

•

937