Introduction

Herbicides are one of the most effective tools farmers can use in their battle against weeds. The use of these agrochemicals has increased considerably in recent years, especially in countries with more advanced agricultural systems. The appearence of weed populations resistant to the same herbicides that effectively controlled them in the past, is a serious concern for the herbicide industry and for the users of these products. This brief review presents a historical account of the evolution of resistance to herbicides, the importance and extent of the problem, the elements that influence the appearence of resistant populations, the physiological mechanisms responsible for this phenomenon, and some suggestions for the prevention or reduction of the incidence of weeds resistant to these agrochemicals.

Evolved resistance is understood as the ability of a plant population to avoid being affected by a herbicide as a result of continuous or frequent application of the herbicide for long periods. This type of resistance differs from that observed when a species is not controlled by a herbicide to which it has never been exposed (natural resistance). Evolved resistance will therefore refer to the population of a species that was susceptible before and that, after prolongued use of the herbicide, can no longer be controlled with normal rates or even overdoses. The first report on herbicide resistance dates from 1970, when Ryan (1970) documented that certain populations Senecio vulgaris in the state of Washington, U.S.A., could no longer be controlled even with massive doses of simazine. Pioneer studies by Ryan (1970) and by Radosevich & Appleby (1973a,b) indicated that these populations were not only resistant to simazine, but also were resistant to other triazines (cross resistance). In 1980, around 250,000 hectares in Washington were estimated to be infested with Senecio populations resistant to triazines (LeBaron & Gressel, 1982). Since then, reports on resistance to triazines have increased considerably. In Hungary, the biggest producer of corn of the Socialist Block, triazines have been discontinued (except in mixtures with other herbicides) due to the fact that more than 75% of the land designated for agricultural use, is infested with resistant species of Amaranthus spp. There are currently at least 55 estimated weed species in more than 20 countries that have evolved resistance to this group of herbicides (Table 1).

There are weeds resistant to practically all the families of important herbicides, as appreciated in Table 1.

Even when the number of herbicide resistance cases has been increased substantially , it has not yet reached the alarming levels attained with insecticides. In 1980, populations of at least 428 species of insects and mites resistant to insecticides had been documented. These populations were distributed in 14 orders and 83 families of agricultural and veterinary importance (Forgash, 1984).

Factors Determining the Selection of Biotypes Resistant to Herbicides

The main elements that influence the appearence of biotypes resistant to herbicides have been considered in detail by Gressel (1978) and by Gressel & Segel (1982). The most important factors are:

1. Selection Pressure

Selection pressure is the result of effective mortality, that is, the mortality rate in terms of seeds or disseminating agents present at the end of the growth period, not immediately after herbicidal treatment. If resistance genes are already found in a population; the greater the effective mortality, the greater the probability of selecting resistant individuals.

It is considered that genes coding for herbicide resistance are commonly present in weed populations even before the introduction of a herbicide and not due to the introduction of resistant individuals from other areas or created by the herbicide per se (LeBaron, 1984). It is just the selection pressure imposed by the herbicide what determines the enrichment of the population with resistant individuals.

2. Herbicide Persistence in Soil

Herbicides with long residual effect exert selection pressure on weed populations longer than herbicides that easily dissipate in soil. This is probably one of the reasons why weed populations resistant to a group of herbicides as new as the sulfonylureas have been found. If the herbicide is not very persistent, the seed bank in the soil may decrease the probability of proliferation of the resistant biotypes by keeping an elevated population of susceptible individuals that germinate and reproduce once the herbicide has lost its biological effect.

Herbicides that are inactivated immediately in the soil are not necessarily free of resistance problems. For example, there are weeds resistant to paraquat; these resistant biotypes occurred often regular and continuous applications of paraquat, a condition which is equivalent to a high selection pressure.

3. Biological Factors

Some of the characteristics of the weeds have great impact and actually delay the development of herbicide resistance. As mentioned before, even though the herbicide resistance problem has become worse in recent years, it is certainly not as serious as resistance to fungicides and insecticides. Weeds usually exhibit a great reproductive ability, which is maintained even when environmental conditions are not ideal for plant development. When weed populations decrease in the field as a result of some control strategy, the few individuals that escape herbicidal treatment are capable of producing enough seed to ensure the permanence of a seed bank in the soil allowing the perpetuation of the species. This plasticity of the populations greatly influences the "dilution" of resistant weeds that have become selected by the continuous use of a particular herbicide. Seed' longevity and dormancy in the soil also delay the selection of biotypes resistant to herbicides.

Finally, a lack of fitness or decreased competitive ability of resistant weeds is frequently mentioned, as compared to susceptible populations, as yet another factor that delays the development of herbicide resistance. When a gene (the one that confers resistance) substitutes another (the one responsible for the susceptibility) in a population, the new individual is commonly less fit because it has some physiological disadvantage in relation to the original one (Holt, 1989). These disadvantages keep the new gene in a very low frequency within the population. Therefore, it is logical to conclude that if the resistant individuals were more fit, in the absence of selective pressure (herbicide), they would be the dominant type within the population, and the herbicide would never have effectively controlled the weed species in the first place. Thus fitness could be defined as the reproductive success or the proportion of genes that an individual leaves in the genetic pool of a population. Its two main components are survival and reproduction (Holt, 1989). In initial studies by Conard & Radosevich (1979) and Radosevich & Holt (1982), it was noted that triazine resistant biotypes produced less dry matter than the susceptible biotypes, both under competitive and non-competitive conditions. Resistant biotypes were also more affected by competition against susceptible biotypes than by competition amongst themselves. Similar observations also have been made from weeds resistant to other herbicides, including dinitroaniline-resistant Eleusine indica (Valverde, 1989). This reduction in the competitive ability is frequently explained in terms of the inability of resistant weeds to perform fundamental physiological processes, which were affected by the alterations suffered at the sites of activity of the herbicides and which are responsible for their resistance to the herbicides. Recent studies, however, indicate that triazine resistance is not always associated with a reduction in plant vigour (Jansen et al , 1986; Schonfeld et al , 1987).

Biochemical Basis of Herbicide Resistance

Differences in herbicide absorption, translocation and metabolism may be cited as probable mechanisms that explain the evolution of herbicide resistance. The better documented cases, however, indicate a prevalence of alterations at the site of action as responsible for this phenomenon. In the case of the triazines, their mechanism of action involves the inhibition of photosynthesis ( Fig. 1 ), as they prevent the transport of electrons within Photosystem II (PSII). The blocking occurs between the primary electron acceptor (Q) and the plastoquinone (PQ). Located in this region of the thylakoid membrane is a protein with a molecular weight of 32 kD, known as protein B or QB, with a binding site for which the triazines and the plastoquinone compete (Gressel, 1985). The substitution of an amino acid (serine for glycine) in this chloroplastic protein prevents binding of the triazines and will result in resistance to this group of herbicides (Hirschberg and MacIntosh, 1983). Also, at least one case, resistance to triazines was reported to occur because of a increased herbicide detoxication ability by the resistant biotype (Gronwald et al , 1989).

Alterations of the coupling site of the herbicide also figure as responsible for resistance to sulfonylureas, condition in which the acetolactate synthetase enzyme has lost its sensitivity (Thill et al , 1989), and for the resistance to dinitroanilines, where apparently a modified tubuline confers resistance in Eleusine indica (Vaughn, 1989).

In order to explain the resistance to paraquat, two mechanisms have been proposed. Paraquat exerts its phytotoxic action by forming free radicals. By means of the intervention of Photosystem I (PSI), dicationic paraquat is reduced to the monocation, which in turn reduces molecular oxygen (O₂) to the anionic superoxide radical (O₂-), thus regenerating the dicationic paraquat. The radical O₂- in turn produces hydrogen peroxide (H₂O₂) and the hydroxyl radical (OH-). The superoxide and hydroxyl radicals, especially, destroy cellular membranes by peroxidizing fatty acids (Harris & Dodge, 1972). The first resistance mechanism proposed is related to the presence in some species and in resistant biotypes, ofincreased levels of the enzymes (like the superoxide dismutase) capable of detoxifying free radicals produced by paraquat in the presence of light (Shaaltiel & Gressel, 1986). In the resistant biotypes of some species, including Hordeum glaucum and Conyza bonariensis, paraquat seems to have been excluded from its site of activity in the chloroplast, even if its final location has not been elucidated yet (Fuerst, 1989; Fuerst et al, 1985; Powles & Cornic, 1987). In H. glaucum, for example, no differences in the activity of the enzymes superoxide dismutase, catalase and peroxidase, between resistant and susceptible biotypes, were observed. Both biotypes were also similar in the permeability to paraquat of their plasmatic membrane or of the membranes of the chloroplast envelope (Powles & Cornic, 1987). The penetration of the cuticle by paraquat is similar in both biotypes; however, translocation of the herbicide from the surface of the leaf is insignificant in the resistant biotype compared to a certain translocation degree present in the susceptible biotype. Based on these results, it has been proposed (Bishop et al , 1987) that the resistance mechanism is based on the exclusion of the paraquat from the cytoplasm by means of its retention in the apoplast xylem, cell walls and intercellular space).

Prevention and Control of Resistance

Practically speaking, it is best to confront the problem of herbicide resistance with preventive measures. Those cultivation practices associated with high selection pressure must be avoided: use of very persistent herbicides, elevated doses, regular applications of only one herbicide, and mono-cultures that depend solely on chemical methods for the elimination of weeds. It is advisable to rotate the herbicides with others with a totally different mechanism of action than the one previously used, as well as applying them in mixtures. When selecting mixtures, the herbicides included must each have different mechanisms of action.

When the presence of resistant populations is verified at the farm level, the necessary steps must be taken to prevent their dissemination, taking special care to prevent sedd production and dispersion. The use of the herbicide to which resistance evolved must be suspended and other control measures must be carefully selected, considering that, generally, resistance to any herbicide in particular carries a cross resistance to herbicides of the same chemical family or even to herbicides not chemically related. The grower must always read the labels of the products and wisely select the herbicides he will use.

References

CONARD, S. G., and S. R. RADOSEVICH. 1979. Ecological fitness of Senecio vulgaris and Amaranthus retroflexus biotypes susceptible or resistant to atrazine. J. Appl. Ecol. 16:171-177.

BISHOP, T., S. B. POWLES, and G. CORNIC. 1987. Mechanism of paraquat resistance in Hordeum glaucum . II. Paraquat uptake and translocation. Aust. J. Plant Physiol. 14:539-547.

FORGASH, A. J. 1984. History, evolution, and consequences of insecticide resistance. Pestic. Biochem. Physiol. 22:178-186.

FUERST, E. P.; H. Y. NAKATANI; A. D. DODGE; D. PENNER and C. J. ARNTZEN. 1985. Paraquat resistance in Conyza. Plant Physiol. 77:984-989.

__________. 1989. Paraquat resistance mechanisms. Weed Sci. Soc. Am. Abstracts 29:131.

GRESSEL, J. 1978. Factors influencing the selection of herbicide resistant biotypes of weeds. Outlook Agric. 9:283-287.

__________ and L. A. SEGEL. 1982. Interrelating factors controlling the rate of appearance of resistance: The outlook for the future. In H. M. LeBaron, and J. Gressel (eds). Herbicide resistance in plants. John Wiley and Sons, New York, pp.325-347.

__________.1985. Herbicide tolerance and resistance: alteration of the site of activity. In Duke, S. O. (ed). Weed Physiology. Vol. II. Herbicide Physiology. CRC Press, Florida, pp.159-189.

GRONWALD, J. W.; R. N. ANDERSEN and C. YEE. 1989. Atrazine resistance in velvertleaf (Abutilon theophrasti) due to enhanced atrazine detoxification. Pestic. Biochem. Physiol. 34:149-163.

HARRIS, N. and A. D. DODGE. 1972. The effect of paraquat on flax cotyledon leaves: Physiological and biochemical changes. Planta 104:210-219.

HIRSCHBERG, J. and L. MACINTOSH. 1983. Molecular basis of herbicide resistance in Amaranthus hybridus. Science 222:1346-1349.

HOLT, J. S. 1989. Fitness and ecological adaptability of herbicide resistant biotypes. In Fundamental and practical approaches to combatting resistance. ACS Symposium Series. In Press.

JANSEN, M. A. K.; J. H. HOBE; J. C. WESSELIUS and J. J. S. VAN RENSEN. 1986. Comparison of photosynthetic activity and growth performance in triazineresistant and susceptible biotypes of Chenopodium album . Physiol. Vég. 24:475-484.

LeBARON, H. M. and J. GRESSEL, eds. 1982. Herbicide resistance in plants. John Wiley and Sons, New York, 401 p.

__________.1984. Triazine and other photosystem II herbicides. In J. P. Thornber, L. A. Staehelin, and R. B. Hallick (eds). Biosynthesis of the photosynthetic apparatus: molecular biology, development and regulation. Alan R. Liss, New York, pp. 351-356.

__________.1989. Report on herbicide resistant weeds. Pesticide Resistance Management (Michigan Sta. Univ., Michigan) 1:9-10.

POWLES, S. B. and G. CORNIC. 1987. Mechanism of paraquat resistance in Hordeum glaucum . I. Studies with isolated organelles and enzymes. Aust. J. Plant Physiol. 14:81-89.

RADOSEVICH, S. R. and A. P. APPLEBY. 1973a. Relative susceptibility of two common groundsel ( Senecio vulgaris L.) biotypes to six s-triazines. Agron. J. 65:553-555.

__________ and A. P. APPLEBY. 1973b. Studies on the mechanism of resistance to simazine of common groundsel. Weed Sci. 21:497-500.

__________ and J. S. HOLT. 1982. Physiological responses and fitness of susceptible and resistant weed biotypes to triazine herbicides. In H. M. LeBaron, and J. Gressel (eds). Herbicide resistance in plants. John Wiley and Sons, New York, pp.463-485.

RYAN, G. F. 1970. Resistance of common groundsel to simazine and atrazine. Weed Sci. 27:216-218.

SCHONFELD, M.; T. YAACOBY, O. MICHAEL and B. RUBIN. 1987. Triazine resistance without reduced vigor in Phalaris paradoxa. Plant Physiol. 83:329-333.

SHAALTIEL, Y. and J. GRESSEL. 1986. Multienzyme oxygen radical detoxification system correlated with paraquat resistance in Conyza bonariensis . Pestic. Biochem. Physiol. 26:22-28.

THILL, D. C.; C. A. MALLORY; L. L. SAARI; J. C. COTTERMAN and M. M. PRIMIANI. 1989. Sulfonylurea resistance - Mechanism of resistance and cross resistance. Weed Sci. Soc. Am. Abstracts 29:132.

VALVERDE, B. E. 1985. Apuntes sobre el modo de acción de los herbicidas. International Plant Protection Center Paper A-11, Corvallis, Oregon, 19 p.

__________. 1989. Dinitroaniline-herbicide resistance in goosegrass ( Eleusine indica ): Ecophysiological aspects. Ph.D. Dissertation. Oregon Sta. Univ., Corvallis, Oregon, 110 p.

VAUGHN, K. C. 1989. Dinitroaniline resistance - mechanism and characteristics of resistance. Weed Sci. Soc. Am. Abstracts 29:132.