Originally published in Aquatics magazine, Volume 28, No. 1, Spring, 2006
Tyler J. Koschnick, W.T. Haller and M.D. Netherland 1
University of Florida, Institute of Food and Agricultural Sciences, Center for Aquatic and Invasive Plants, Gainesville, FL 32653
Recently, aquatic plant managers have been introduced to the reality of weed resistance to aquatic herbicides. In Florida, hydrilla (Hydrilla verticillata (L.f.) Royle) has developed resistance to fluridone (MacDonald et al. 2001, Arias et al. 2005), and a species of duckweed (Landoltia punctata (G. Meyer) D.H. Les and D.J. Crawford) was identified that developed resistance to diquat (Koschnick 2005). Although the full extent of fluridone resistant hydrilla (FRH) is not known, it occurs in many of Florida's largest water bodies that have had historical hydrilla problems. Diquat resistant duckweed is less widespread. Weed resistance to aquatic herbicides is an emerging issue in aquatic plant management, and education and research are keys to managing this problem.
First, what is resistance? Resistance occurs in a plant species that was originally susceptible to an herbicide, but over time control is lost through the selection of an existing resistant individual or biotype. Think of it as a form of natural selection. There are slight genetic differences between plants in the same population. When the same herbicide is used repeatedly, a strong selection pressure is exerted for individuals with the genetic make-up that allows these plants to resist the herbicide and survive, and then increase their presence in the population. It is important to emphasize that the herbicide does not cause a mutation or create a super plant, and you can't visually discern the difference between a resistant versus susceptible individual. For example, hydrilla was initially susceptible to low use rates of fluridone, but over time a population was selected that was no longer controlled at these recommended use rates, and the appearance of the individual plant is the same. Additional applications of fluridone facilitated the spread or increased the proportion of a resistant biotype throughout the waterbody.
There are also concerns about cross-resistance, which is resistance to different herbicides with similar modes of action. This should not be confused with multiple resistance, which is resistance to multiple herbicides with different modes of action. Experimentally, we have shown cross-resistance under laboratory conditions. Diquat resistant duckweed is also resistant to paraquat because both these herbicides kill the plants by stopping the same biochemical process. In hydrilla, fluridone inhibits the enzyme phytoene desaturase. FRH is also resistant to norflurazon and several other herbicides that inhibit the same enzyme. There have been no cases of multiple resistance or resistance to at least two different modes of action by aquatic plants.
In contrast to resistance, tolerance is the term used to describe plants that have never been susceptible to a particular herbicide or class of herbicides at labeled use rates. For example, aquatic grasses tend to be tolerant of compounds such as 2,4-D and triclopyr. Likewise, a plant such as hygrophila has proven to be fairly tolerant of all currently registered aquatic herbicides. While the terms resistance and tolerance have often been used in the same context, they have very different meanings to those in the field of weed science. Resistance is the result of a trait that is selected for, whereas tolerance is an inherent ability to survive the herbicide application. Tolerance may be biochemical (e.g. metabolism), the result of reduced uptake (e.g. thick cuticle), or other means that allow some plant species to tolerate the herbicide.
In theory, "every" plant species has a biotype that is resistant to "every" herbicide. The question becomes: Has it been selected for yet? The chances of selecting for that "one" individual increases in areas with repeated use of the same herbicide and widespread weed populations. Resistance is not a new subject with herbicides, but it is new in aquatics. There are currently over 177 plant species (>295 biotypes) that have developed resistance to herbicides worldwide, with approximately 70 species in the US, with most occurring in agricultural systems (www.weedscience.org).
There are four main mechanisms of herbicide resistance in plants. Some herbicides target or prevent formation of a key enzyme. Resistant biotypes have an alteration at the site of action that prevents an enzyme-specific herbicide (e.g. fluridone, ALS inhibitors) from affecting the target site. Resistance can also result in biotypes that have greater ability to metabolize or detoxifiy the herbicide (e.g. substituted ureas). Herbicides can also lose their effectiveness due to being compartmentalized or bound-up prior to getting to the site of action, or due to reduced transport or movement of the chemical (e.g. glyphosate). Finally, resistant biotypes may have reduced uptake of the herbicide into the plant or movement to the site of action inside the cell.
There are certain characteristics of herbicides that can lead to an increase in the development of resistance. Some herbicides, such as copper and endothall, kill cells by destroying membranes and shutting down respiration and photosynthesis, essentially affecting several cell processes simultaneously. In contrast, the more specific (simple) the mode of action the greater the chance of selecting for a biotype with one of the 4 resistance mechanisms. Herbicide characteristics and use patterns that favor resistance include: 1) use of compounds with similar or single modes of action; 2) persistence in the environment; and 3) products that are commonly or repeatedly used (high market share) due to the lack of effective or cost effective alternatives.
How many duckweed plants in a 10-acre pond? Ten billion? That is not out of the question if you assume a frond is 0.125 inches long by 0.0625 inches wide and consist of a single layer of plants (~800 million per acre). Even if 0.0000001 % of the duckweed plants have one of these 4 resistance mechanisms (altered site of action, metabolism/ detoxification, reduced transport, or reduced uptake/movement to site of action) and 9,999,999,999 plants are killed by your treatment, 1 may survive. Dense infestations of hydrilla and duckweed are characterized by the presence of huge numbers of meristematic growing points in an aqueous environment. Moreover, this is also characteristic of numerous other aquatic plants.
Weed characteristics can also contribute to the development of resistance, especially characteristics that can increase genetic diversity in the weed population. These characteristics may include species with high reproductive rates (e.g. high seed production, asexual budding), short seed longevity, and species with naturally diverse genetic makeup. Also, once a species develops resistance, the resistant biotype must be able to compete and survive against susceptible biotypes in the absence of further selection pressure.
To reduce the chances of resistant populations developing in the aquatic environment the following practices are recommended: 1) alternate modes of action or use herbicide mixtures 2) utilize chemical, biological, and mechanical control options when feasible; 3) do not use herbicides with the same mode of action repeatedly, and 4) treat weeds when infestations are low. By following these recommendations, you will reduce the chances that a "single duckweed plant" will survive long enough to create a large population of resistant plants. The main key to weed resistance management in terrestrial systems has been alternating crops and herbicide modes of action. While we are limited in our ability to alternate our "weeds" in aquatic plant management, we can consider changing our herbicides or mixtures.
Aquatic weed control is conducted with very few herbicide choices, and managers are often heavily dependent on one or two standard herbicides for a particular weed species. Factors impacting these use patterns include cost effectiveness, use restrictions, and selective properties of the herbicide. This reliance, coupled with the limited number of herbicides registered in aquatics, surprisingly has not resulted in widespread development of more resistance issues. While techniques such as biocontrol and mechanical control are well known, herbicide programs are generally implemented when neither of these options is feasible due to the scale of the problem or the need to provide predictable management results. Moreover, issues such as crop rotation, herbicide rotation, and pest scouting that are familiar to traditional integrated pest management programs in terrestrial agriculture have not proved to be easily incorporated into aquatic plant management programs. Therefore, in aquatics we are unable to utilize many terrestrial weed recommendations for reducing the potential for resistance development.
Mueller et al. (2005) discuss proactive weed management versus reactive weed management as it pertains to resistance. Most people employ a reactive strategy, which means "don't do anything until resistance occurs", since it won't happen to me in "my lake". This is driven by economics and often we wait until weeds are widespread (crisis) in order to gain public support and funding for operations. It is difficult to switch to more expensive management methods due to the priority of controlling weeds at the lowest cost in public funds. The proactive strategy involves determining what you can do to delay the onset of resistance since it will eventually happen in "my lake", and try to protect the currently registered products. Rotate herbicides, don't treat every year with the same mode of action at the same site, and use herbicide mixtures. However, this strategy typically comes at a cost, and scientists have not yet determined the most practical means of accomplishing this.
The Agrichemical industry and state/ federal scientists are trying to bring new herbicides and tools to the market to give managers more options for managing aquatic plants. In the last 5 years, 3 new herbicides have been registered for aquatics (triclopyr, imazapyr, and carfentrazone). Currently, there are 4 additional herbicides with experimental use permits (EUP) granted by EPA or applied for (penoxsulam, imazamox, flumioxazin, and bispyribac sodium), and hopefully more will be submitted for EUP status in the near future. While these new EUP products typically have good toxicity profiles that will aid in the aquatic registration process (some are classified as reduced risk products), they also have a single site of action in plants, which increases the chances for resistance to occur.
For example, 3 of the herbicides currently being developed for hydrilla control are classified as acetolactate synthesis (ALS) inhibitors (penoxsulam, imazamox, and bispyribac sodium). ALS-inhibitors affect a single enzyme necessary for amino acid/protein synthesis in plants; acetolactate synthase, and there are about 50 ALS-inhibiting herbicides registered in the U.S. While most of these ALS compounds will likely prove active on hydrilla, resistance development to one of these products could lead to wide-scale cross resistance (Tranel and Wright 2002), or resistance to all 50+ ALS-inhibiting herbicides. Resistance to ALS inhibiting herbicides has occurred in terrestrial sites over a relatively short period of time (few years) compared to other herbicide families such as the triazines (10 to 20 years). The first documented case of resistance was only 5 years after ALS herbicides were commercialized in 1982. Today, there are more plant species resistant to ALS herbicides than any other herbicide, including the triazines, which have been used for approximately 20 years longer than the ALS herbicides.
There are numerous species of wetland plants [e.g. Limnophila sessiliflora (Vahl) Blume] that have developed resistance to ALS herbicides in rice, and over 16 plant families have representative species that have developed resistance to ALS inhibitors (Heap 2005). This suggests that ALS resistance will occur in submersed aquatic species, unless active steps are taken to prevent this from happening. While recognition of this potential is an important first step, it is also important that resistance management strategies be put in place prior to wide-scale use of these products.
Based on the experience with large-scale fluridone use and the proven ability of hydrilla to develop resistance, developing programs for resistance management are critical to protect the long-term viability of ALS herbicides. In addition to ALS chemistry, there is a strong need to identify an alternate mode of action that can be used in rotation with other management tools.
The number of herbicides or modes of action for use against hydrilla is limited. There are approximately 300 herbicides registered in the US representing 6 general modes of action (photosynthetic inhibitors, amino acid/ protein synthesis inhibitor, cell division/ growth inhibitors, cell membrane disruptors, pigment synthesis inhibitors, and growth regulators). Many of these compounds are too toxic for aquatic use (diuron, trifluralin, etc.), many do not control hydrilla (2,4-D, glyphosate, etc.), and many are off patent (dicholbenil, simazine, etc.), which greatly reduces the potential for incurring high registration costs. Decisions on registration and use of aquatic herbicides made in the next few years will determine managers' abilities to control aquatic weeds, particularly hydrilla, 20 years from now.
The situation in Florida for hydrilla control is particularly problematic because of the widespread occurrence of fluridone resistant hydrilla in many of the economically important large lakes of central Florida. If a cost-effective ALS-inhibitor is registered for use by 2007, there will be pressure for frequent use of this herbicide. If the ALS-inhibitors are used annually, will resistance to ALS-inhibitors also occur, making hydrilla resistant to both fluridone and ALS compounds? Then what? Ideally, to protect the use of ALS compounds in fluridone resistant hydrilla, we need another mode of action. The herbicide rotation should at least be ALS-new mode of action-ALS new mode of action. In waters where fluridone susceptible hydrilla occurs (in parts of Florida and rest of the U.S.) then registration of the ALS inhibitors will provide one more tool that can be rotated with traditional chemistries and other control techniques. In this way, the chances of developing fluridone or ALS resistance (or any herbicide mode of action) should be greatly reduced.
Currently, resistance to aquatic herbicides is isolated to Florida. There are no documented cases of resistant aquatic plant species outside Florida. Yet, resistance will not be a problem isolated to Florida, and duckweed and hydrilla are likely not unique in their ability to develop resistance. It is best to take a proactive strategy where and when you can to delay resistance. While this may result in incurring greater costs in the short-term, the loss of our limited aquatic herbicides is a much greater cost in the long run.
1US Army Engineer Research and Development Center.
Stationed at the University of Florida's Center for Aquatic and Invasive Plants, Gainesville, FL