|
|
||||||||
International Rice Research Institute, APPA Division, P.O. Box 3127, MCPO, 1271 Makati City, Philippines, and Weed Science, Royal Veterinary and Agric. Univ., 10/2630 Agrovej Toastrup, Denmark
Corresponding author (mol{at}kvl.dk)
| ABSTRACT |
|---|
|
|
|---|
This paper discusses the progress made in recent years and suggests some direction for future research.
Abbreviations: QTLs, quantitative trait loci
| INTRODUCTION |
|---|
|
|
|---|
| About rice allelopathy |
|---|
|
|
|---|
There is Large Variation in Allelopathy among Rice Cultivars
To quantify allelopathy in rice, several laboratory and field screening procedures have been developed and tested (Fujii, 1992; Dilday et al., 1998; Navarez and Olofsdotter, 1996; Olofsdotter and Navarez, 1996). The conclusions from all these studies were that there are differences in allelopathic potential among rice cultivars, expressed as a reduction in root growth of test plants grown in association with rice in the laboratory or as a weed reduction (dry matter and density) in the field. For the field screening, however, it is important to note that allelopathy is difficult to distinguish from competition and therefore such screening must be compared with data where resource competition can be eliminated as a factor in the experiment. Olofsdotter et al. (1999) correlated screening results from the laboratory with a range of competition components, measured in the field, and claimed that allelopathy can explain 34% of the reduction in total weed dry weight 8 wk after seeding. Table 1 presents selected data from this study including barnyardgrass [Echinochloa crus-galli (L.) P. Beauv.] plant dry weight 8 wk after seeding, tillering, rice and weed height, the barnyardgrass root length in laboratory screening, and a cultivar ranking. Plant height is often described as one of the most important factors for total competitive ability of a crop and accounts for a similar percentage of total competitive ability (Gaudet and Keddy, 1988; Garrity et al., 1992).
|
Allelopathy Plays a Role under Field Conditions
The first observation of allelopathy in rice was made on field examinations in Arkansas (1988 cropping season) (Dilday et al., 1991). A seed-increase field with 5000 accessions sown in hills had a natural ducksalad [Heteranthera limosa (Sw.) Willd.] infestation and around some of the rice accessions was a weed-free area that could not be related to shading and appeared to be the result of chemical interaction. This led to a large field screening program in Arkansas and today 12000 accessions have been screened, with about 3.5% of the accessions rated as allelopathic against ducksalad (Dilday et al., 1998). These lines have not been tested in laboratory experiments, where other competition variables can be eliminated. The success in field screening against ducksalad can be explained by the weed biology. Ducksalad invades rice fields after irrigation at a time when the rice crop is already established, therefore making it easy to visually evaluate weed suppression. As the rice crop is already established when the weed problem occurs, these results also raise the question whether allelopathy requires older plants to be efficient.
Considering more important (globally) weed species suggests that it is harder to get a clear and direct picture of weed suppression under field conditions. In the Philippines, 111 rice cultivars have been evaluated for weed suppression capability against barnyardgrass under field conditions over three seasons (Olofsdotter et al., 1999). Weeds and rice were direct-seeded the same day, which resulted in similar emergence. During establishment and early growth of the crop and weeds, a range of variables were measured (e.g., tillering, height, and density) for rice and weeds. Finally, 8 wk after seeding, weed suppression was evaluated using barnyardgrass dry weight. The results showed that rice height was the most important factor influencing weed suppression but was closely followed by allelopathic potential as measured in laboratory screening using the relay-seeding technique for the same set of cultivars (Olofsdotter et al., 1999). The field results were confirmed in similar experiments conducted in Korea using some of the same cultivars as in the Philippines (Kim and Shin, 1998). The similarities in the results from the Philippines and Korea indicated that cultivar differences in weed suppression were relatively stable in different environments, which suggests that allelopathy was more influenced by genetics than environment. The experiments also indicated that maximum weed suppression was obtained in cultivars that have both allelopathic and competitive traits. Progress in field weed suppression is likely to be most efficient when both competition and allelopathy are optimized.
Allelopathic Rice Can Suppress Both Mono- and Dicot Weed Species
Both laboratory screening and field experiments reveal that rice allelopathy is active against both monocot and dicot weeds (Dilday et al., 1991; Fujii, 1992; Olofsdotter and Navarez, 1996; Hassan et al., 1998; Kim and Shin, 1998). However, there is no assurance for activity against both groups for a given rice accession. Therefore, more than one chemical is likely to be responsible for allelopathy in rice, and allelopathy is probably quantitatively inherited. One rice cultivar (Taichung Native 1) has shown activity against most of the weeds tested (e.g., barnyardgrass, desert horsepurslane [Trianthema portulacastrum L.], ducksalad, and toothcup [Ammannia coccinea Rottb.]) (Dilday et al., 1998; Olofsdotter and Navarez, 1996), and is therefore a suitable choice for both identifying allelochemicals and studying allelopathy genetics. Taichung Native 1 carries the semi-dwarf gene, included in most postGreen Revolution rice cultivars, and is used intensively in modern breeding programs. This could explain the surprisingly frequent occurrence of allelopathic capability in modern rice cultivars.
Progress Has Been Made in Identifying Rice Allelochemicals
The process of isolating and identifying rice allelochemicals has started but has not yet yielded chemicals that could explain the allelopathic effect (Rimando et al., 2001). None of the known chemicals released from rice can alone explain the weed suppression seen in the laboratory and fields. Phenolic acids have been identified in allelopathic rice germplasm (Rimando et al., 2001) and they have previously been described as allelochemicals (Inderjit, 1996; Mattice et al., 2001; Blum, 1998). However, concentrations of single phenolic acids and combinations of all phenolic acids measured in rice ecosystems do not approach phytotoxic levels (Tanaka et al., 1990; Olofsdotter et al., 2001). Identifying allelochemicals in rice is important for understanding allelopathy mechanisms and for use as markers in gene identification.
Quantitative Trait Loci Correlated with Allelopathy Have Been Identified
Evaluating a screening program from a breeder's point of view often starts by calculating the broad-sense heritability for the trait in question. The broad-sense heritability ranges from 0 to 1, and a higher score indicates that the genotype is described more closely by the phenotype measured. For allelopathy, data developed using the relay-seeding technique produced a broad-sense heritability for reduction of barnyardgrass roots of 0.85, which satisfies breeders' needs of screening program reliability (Courtois and Olofsdotter, 1998). Consequently, genetic studies of allelopathy in rice were started at the International Rice Research Institute in the Philippines. Initially, an existing population of recombinant inbred lines developed to study blast resistance, but found useful for allelopathy studies, was used for genetic analysis of allelopathy in rice. All phenotyping in these experiments was done using the relay-seeding technique. The cross IAC 165 x Co39 has one highly allelopathic cultivar, IAC 165, and one moderately allelopathic accession, Co39. Using this population, four QTLs have been identified that correlate with allelopathy expression, measured as barnyardgrass root reduction in laboratory screening (Jensen et al., 2001). This shows that allelopathy is quantitatively inherited. The work also emphasizes the importance of a good screening system.
IAC 165 is a highly allelopathic Brazilian cultivar and laboratory screening has shown that a large proportion of Brazilian upland rice germplasm is allelopathic. However, weed selectivity, autotoxicity, and residual effects from Brazilian rice germplasm are different from those of Asian germplasm (unpublished data). Therefore, it is important that genetic studies continue using crosses between cultivars other than Brazilian. Besides gene identification for allelopathy, this work might result in a better understanding of autotoxicity problems in Brazilian upland rice production.
| Future research |
|---|
|
|
|---|
Understanding the Mechanisms of Allelopathy
For optimal use of allelopathy under field conditions, the influence of environmental factors needs to be investigated. Although allelopathy seems to be genetically based, its expression needs to be optimized through management, especially in the light of reports showing that allelochemical production and release are influenced by abiotic and biotic factors, such as plant age, temperature, light and soil conditions, microflora, nutritional status, and herbicide treatments (Duke, 1985; Hoagland and Williams, 1985). In particular, the importance of and fluctuations with plant age need to be investigated in order to develop weed management strategies that use allelopathy at the right time of the season. Moreover, allelopathy needs to be expressed at an early stage of development when weed control is most important.
The nutrient availability and nutrient efficiency of a rice cultivar will probably interact with allelopathic strength. Production of secondary metabolites is often enhanced by nutrient stress (Hoagland and Williams, 1985). For rice, it has been shown that plants growing in aerobic P-deficient soil can solubilize phosphate in the rhizosphere and thereby increase their P uptake by excreting organic anions from their roots, particularly citrate (Kirk et al., 1999a,b). The synthesis and excretion of citrate and certain other organic anions increase under P deficiency. Comparing these studies with allelopathy research shows that phosphorus-efficient rice cultivars are often the same as the allelopathic ones. It is important to understand such interactions to evaluate the effect of introducing allelopathy into rice.
Evaluating Allelopathic Rice from an Agronomic Point of View
The interactions between managerial interventions in rice production, such as flooding time and depth, seeding or transplanting, and amount of fertilizer applied, and allelopathy also need to be investigated. Crop establishment will influence both the early vigor of the rice and weed establishment. Whether rice is transplanted, wet-seeded, or dry-seeded will create different environments for weed establishment and thereby affect early competition and probably allelopathic interference between rice and weeds. Timing and nutrient source might also play a role in allelopathy expression in a field situation.
Allelopathic rice cultivars have produced residual effects on weed emergence after the rice harvest (Dilday et al., 1998). Experiments in the Philippines, however, have shown that the residual effect might differ among allelopathic rice cultivars (unpublished data). This suggests that more than one allelochemical is involved and that they might have different persistence in soils. Understanding these residual effects, including autotoxicity, is important in planning crop cycles or cropping systems. Crop management strategies to enhance or decrease the residual effects need to be identified and studied further. Autotoxicity problems have occurred in Brazilian upland rice cropping systems, thus underlining that such autotoxicity risks must be studied before allelopathic rice is introdued, to ensure that these cultivars do not cause new agronomic problems in continuous rice production systems.
Finally, allelopathic selectivity against rice weeds needs to be investigated from both the viewpoint of additional management strategies and eventual resistance development in the weed population. Allelochemicals operate like herbicides, which means that tolerance in the weed populations is likely to occur if the same allelochemicals are constantly "applied" to the same weed population. It is also likely that long-term use of allelopathic rice would result in a shifting weed flora. Strategies to minimize these risks have to be developed.
Identifying Allelochemicals
Allelochemical identification in rice is being undertaken (Rimando et al., 2001; Mattice et al., 2001) and is of high priority if other allelopathy outcomes are to be achieved. After identifying allelochemicals in rice, research efforts should be directed toward understanding the allelochemical mode of action, allelochemical production and release in rice, uptake processes by weeds, soil interactions with allelochemicals, and joint action relations between different allelochemicals in rice.
Identifying Genes Encoding for Allelopathy in Rice
To be able to breed for plant interference potential in future breeding programs, it is important to know the genes coding for allelopathic potential and competitive ability generally. Competition, being dependent of several physiological and phenological traits, and allelopathy are both polygenic characters. Before an efficient breeding strategy can be formulated, it is necessary to know which genes are involved in plant interference. Molecular marker-aided genetics is presently the best tool for studying the genetic control of a quantitative trait, mapping the genes involved on the chromosomes with a reasonable level of precision, and analyzing the relationships between the trait of interest and other important agronomic traits (assessing pleiotropic effects and risks of linkage drag). This work is in progress (Jensen et al., 2001) and continued work on populations developed especially for allelopathy is essential. If the allelochemicals involved are identified and if they correspond to molecules whose coded genes are already sequenced, the genes can be used as candidate probes to confirm the role of the identified QTLs. Improving phenotyping with a more efficient screening technique is an important part of this process.
Evaluating Allelopathic Rice from an Environmental Point of View
No chemicals, synthetically or biologically produced, added to the environment can be considered safe before environmental evaluation has been undertaken. The effects of any given chemical depend on the dose in which it is occurring or applied. Using allelopathic rice cultivars will increase the frequency and probably also allelochemical concentrations in the environment. Ecotoxicological studies must therefore start as soon as the allelochemicals are identified. Environmental fate and degradation patterns of allelochemicals and degradation products are essential knowledge for implementing allelopathic rice cultivars. Not only do we need to study effects on nontarget organisms but, especially for rice, there is also a direct connection to water quality both in irrigation schemes and for consumption.
| Conclusions |
|---|
|
|
|---|
From the experience gained in rice allelopathy, it is also possible to identify a range of researchable areas that still need attention, such as understanding the mechanisms involved in allelopathy expression from both an agronomic and environmental viewpoint. Identifying the allelochemicals responsible for weed reduction in the field is also becoming urgent as well as their environmental and toxicological assessment. Finally, work on genetics needs to continue to enable the use of allelopathy in a breeding context. For this to be achievable, some general points need to be considered in the continuing research: (i) Laboratory screening is essential for all parts of the research, such as genetics, allelochemical identification, and studies of allelopathy mechanisms. Thus, further development of a more reliable screening technique is required. (ii) Neither laboratory experiments alone nor field experiments alone can quantify the importance of and reveal the interactions between different competition variables, such as physical and chemical interference, including allelopathy. Field, laboratory and greenhouse studies must therefore work hand in hand to get the full picture of rice allelopathy. It is my view that reductionistic research is a prerequisite for studies of complex plant interactions. This means that the initial studies have to be done in simple experimental setups in which other interactions can be eliminated, but, as knowledge and understanding increase, the experimental systems should become more and more complex to achieve a fuller picture of plant interactions. (iii) Rice scientists must work together across traditional research areas to make important progress when working on complex interactions such as plant competition, including allelopathy. (iv) It is important to state that allelopathy research on rice must continue in a focused direction. The information gained from such research will help us understand the complexity of weed competition in rice especially, but it will also direct research on weed competition in other crops.
| ACKNOWLEDGMENTS |
|---|
Received for publication November 30, 1999.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. B. Jensen, B. Courtois, and M. Olofsdotter Quantitative Trait Loci Analysis of Allelopathy in Rice Crop Sci., July 1, 2008; 48(4): 1459 - 1469. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||