Use of plant viruses as bioherbicides: the first virus-based bioherbicide and future opportunities

Until recently, only a few plant viruses had been studied for use as biological control agents for weeds, but none had been developed into a registered bioherbicide. This position changed in 2014, when the US Environmental Protection Agency granted an unrestricted Section 3 registration for tobacco mild green mosaic virus (TMGMV) strain U2 as a herbicide active ingredient for a commercial bioherbicide (SolviNix LC). It is approved for the control of tropical soda apple (TSA, Solanum viarum), an invasive 'noxious weed' in the United States. TSA is a problematic weed in cattle pastures and natural areas in Florida. The TMGMV-U2 product kills TSA consistently, completely, and within a few weeks after its application. It is part of the TSA integrated best management practice in Florida along with approved chemical herbicides and a classical biocontrol agent, Gratiana boliviana (Coleoptera: Chrysomelidae). TMGMV is nonpathogenic and nontoxic to humans, animals, and other fauna, environmentally safe, and as effective as chemical herbicides. Unlike the insect biocontrol agent, TMGMV kills and eliminates the weed from fields and helps recycle the dead biomass in the soil. Here the discovery, proof of concept, mode of action, risk analyses, application methods and tools, field testing, and development of the virus as the commercial product are reviewed. Also reviewed here are the data and scientific justifications advanced to answer the concerns raised about the use of the virus as a herbicide. The prospects for discovery and development of other plant-virus-based bioherbicides are discussed. © 2023 Society of Chemical Industry.

Host-virus interaction between tobacco mild green mosaic virus strain U2 and tropical soda apple resulting in systemic hypersensitive necrosis and the host range, survival, spread, and molecular characterization of the virus

BACKGROUND

Tobacco mild green mosaic virus strain U2 (TMGMV-U2) is a registered active ingredient in a bioherbicide to control tropical soda apple (TSA), Solanum viarum, an invasive weed. As required for registration, we developed empirical data on the host-virus interaction and the virus's host range, survival, spread, and genomic sequence.

RESULTS

TMGMV-U2 killed TSA plants by causing systemic hypersensitive necrosis (SHN). It elicited local lesions in inoculated leaves which was followed by the plant's wilting and death. It moved from inoculated terminal leaves through the vasculature to roots and then to newly developed leaves. Phloem death was implicated in wilting and plant death. The SHN response was attenuated in plants grown at constant 32 °C. TMGMV-U2 titer in TSA was low compared to a systemically susceptible tobacco. The virus remained infective for up to 6 months in infected dead TSA tissues and in soil in which infected plants had grown. Susceptible tobacco and pepper plants grown in soil that previously had infected dead TSA or in soil amended with the virus remained asymptomatic and virus-free. A susceptible pepper crop grown in a field block following two consecutive crops of TMGMV-U2-infected susceptible tobacco grew disease-free and virus-free and without yield loss. Purified TMGMV-U2 was infective for 1 year when stored at -20 °C or 5 °C and for 1 month at room temperature. No virus spread was found in the field. Genomic analyses confirmed the registered isolate to be a U2 strain and free of satellite TMV. The TMGMV-U2-susceptible species preponderantly belonged to the Solanaceae. A few hosts that were killed belonged to this family. Several new hosts to TMGMV-U2 were found. These data enabled registration of TMGMV-U2.

CONCLUSION

TMGMV-U2 can be used safely as a bioherbicide without risks to nontarget plants and the environment. © 2023 Society of Chemical Industry.

Ophiobolin E and 8-epi-ophiobolin J produced by Drechslera gigantea, a potential mycoherbicide of weedy grasses

Drechslera gigantea, a fungal pathogen isolated from large crabgrass (Digitaria sanguinalis) and proposed as a potential mycoherbicide of grass weeds, produces phytotoxic metabolites in liquid and solid cultures. Ophiobolin A and three minor ophiobolins i.e., 6-epi-ophiobolin A, 3-anhydro-6-epi-ophiobolin A and ophiobolin I were obtained from the liquid culture broths. Interestingly and unexpectedly, ophiobolins also appeared in cultures of this fungus and they were isolated together with the known ophiobolins B and J, and designed as ophiobolin E and 8-epi-ophiobolin J. They were characterized using essentially spectroscopic methods. It is noteworthy that D. gigantea produces such a plethora of bioactive organic substances. Some structure-activity relationship results are also discussed in this report.

Herbicidal potential of ophiobolins produced by Drechslera gigantea

Drechslera gigantea, a potential mycoherbicide of grass weeds, was isolated in Florida from naturally infected large crabgrass (Digitaria sanguinalis); it produces phytotoxic metabolites in liquid culture. The main metabolite was identified by spectroscopic methods and optical properties as ophiobolin A (1), a well-known phytotoxic sesterterpene produced by several phytopathogenic fungi of important crops and already extensively studied for its interesting biological activities. The other three minor metabolites proved to be related to ophiobolin A and were identified using the same techniques as 6-epi-ophiobolin A and 3-anhydro-6-epi-ophiobolin A (2 and 3) and ophiobolin I (4). Assayed on punctured detached leaves of several grass and dicotyledon weeds, ophiobolin A proved to be on average more phytotoxic as compared to the other related compounds. Some structural features appear to be important for the phytoxicity, such as the hydroxy group at C-3, the stereochemistry at C-6, and the aldehyde group at C-7. Furthermore, grass weeds usually proved to be more sensitive to the phytotoxins than dicotyledons, on which ophiobolin A caused the appearance of large necrosis even at the lowest concentration assayed. This is the first report about the production of ophiobolins from D. gigantea and of the proposed use as potential natural herbicides against grass weeds.

Influence of Epidemiological Factors on the Bioherbicidal Efficacy of Phomopsis amaranthicola on Amaranthus hybridus

Greenhouse experiments were performed to determine the effect of dew period temperature and duration, plant growth stage, conidial concentration, and the addition of adjuvants on the bioherbicidal efficacy of Phomopsis amaranthicola on Amaranthus spp., using Amaranthus hybridus as test plant. P. amaranthicola infected A. hybridus at 20, 25, 30, and 35°C but the disease level achieved at 20°C may not be sufficient to cause high plant mortality. Plant mortality was also significantly lower in plants that were exposed to 4 h of dew. Plants at less than two- to two- to four-leaf stage were more easily killed than older plants, and increasing conidial concentration from 10⁵ to 10⁶ or 10⁷ conidia ml^-1 did not result in higher mortality levels. Among the adjuvants tested, polyalkyleneoxide-modified heptamethyltrisiloxane, algal polysaccharide, hyrdroxyethyl cellulose, and octylphenoxy polyethoxyethanol reduced conidial germination. Conidia applied with invert emulsion caused the highest plant mortality (74%) but invert emulsion alone caused 33% plant death due to phytotoxicity. Results indicate that P. amaranthicola can infect and kill Amaranthus spp. under a range of temperature, dew period, and inoculum levels and, therefore, has good potential as a bioherbicide agent.

Evaluation of Agar and Grain Media for Mass Production of Conidia of Dactylaria higginsii

Growth and sporulation of Dactylaria higginsii were quantified on complex agar media containing biological materials (group 1) and chemically defined agar media (group 2), as well as on grains, and the inoculum produced on these various substrates was tested for virulence on Cyperus rotundus. The fungus grew well between 25 and 30°C on potato dextrose agar (PDA), with 27°C being the optimum temperature. Generally, conidial production was highly variable and lower on complex agar media than on chemically defined media. Addition of purple nutsedge leaves to PDA did not increase colony growth or conidial production when compared with una-mended PDA. Conidial production was lowest on brown rice compared with white rice or white rice with nutsedge leaves. Peak production on grain media occurred from day 12 in test 1 (2.4 × 10⁶ spores/g of grain) and on day 16 in test 2 (2.5 × 10⁶ spores/g of grain). Germination rate of conidia produced on white rice was 50% compared with the near 100% germination of conidia produced on PDA or on white rice amended with potato dextrose broth (PDB). Conidia produced on white rice or PDA, when tested fresh or after two washings, were less virulent on C. rotundus than conidia from white rice amended with PDB. After four washings, conidia from all three media produced the same level of disease severity. White rice supplemented with PDB and PDA in trays were suitable for mass production of conidia of D. higginsii.

Field Evaluation of Phomopsis amaranthicola, A Biological Control Agent of Amaranthus spp

There are approximately 60 species in the genus Amaranthus, of which seven are used as grains, leafy vegetables, or ornamentals. The majority of the remaining species are considered important weeds. A new fungal species, Phomopsis amaranthicola, isolated from stem and leaf lesions on an Amaranthus sp. plant, was found to be pathogenic to 22 species of Amaranthus tested. The efficacy of this fungus was tested in field trials using one or two postemergent applications of the fungus consisting of two concentrations of conidia or mycelial suspensions. Species tested for susceptibility in the field included Amaranthus hybridus, A. lividus, A. viridus, A. spinosus, and a triazine-resistant A. hybridus. The cumulative disease incidence data for each treatment within each species were plotted versus time using regression for lifetime data. Plant mortality was recorded 2, 4, and 6 weeks after inoculation. There were significant differences between the treatment effects in the control plots versus the plots treated with P. amaranthicola. The highest level of control was obtained in the first trial when the fungus was applied at 6 × 10⁷ conidia per ml. Final mortality of all species, except A. hybridus, reached 100% in inoculated plots 25 days earlier than in noninoculated control plots. Conidial suspensions were more effective in controlling the species in the second trial than were mycelial suspensions. Spread of the pathogen to noninoculated control plots was faster in the second year than in other years. High levels of plant mortality were achieved in plots of A. spinosus, A. lividus, and A. viridis. A. hybridus and the triazine-resistant A. hybridus were not effectively controlled in the second year due to the advanced stage of plant growth (8 to 10 true leaves) at the time of pathogen application. Results confirmed that P. amaranthicola is an effective biocontrol agent of some of the Amaranthus spp. tested.

Occurrence of Dactylaria higginsii on Purple Nutsedge in Florida

In September 1994, a population of severely diseased purple nutsedge (Cyperus rotundus L.) was found in Gainesville, FL. The symptoms were characterized by necrotic leaf spots, blotches, and foliar blighting. A fungal isolate was consistently recovered from symptomatic leaves and grown in pure culture. Based on the characteristics of conidia (28.6 × 6.6 μm) and conidiophores (45.2 × 7.0 μm at the broadest base), the fungus was identified as Dactylaria higginsii (Luttrell) M.B. Ellis. This fungus was first described as Piricularia [sic] higginsii from Georgia (4) and later redescribed as D. higginsii (2). A Pyricularia sp. and P. grisea (Cooke) Sacc. have been recorded on C. alternifolius L., C. papyrus L., and other Cyperus spp. from Florida, but species of Pyricularia or Dactylaria have not been reported from this state on purple nutsedge (1). Proof of pathogenicity (Koch's postulates) was established in repeated trials in a greenhouse; the disease symptoms were reproduced, and the fungus was reisolated from inoculated plants and confirmed to be the same organism used for inoculations. Four- to six-leaf-stage purple nutsedge and yellow nutsedge (C. esculentus L.) plants were sprayed with a suspension of 1 × 10⁶ conidia per ml amended with 0.02% Silwet L-77 (vol/vol). Control plants were sprayed with 0.02% Silwet L-77 only. Small, water-soaked lesions developed 4 days after inoculation. The lesions coalesced into larger necrotic blotches with grayish centers 8 days after inoculation. Most of the inoculated foliage was blighted within 15 days after inoculation. The disease did not kill nutsedge bulbs or tubers, but reduced shoot and tuber yields. None of the control plants developed any symptoms. The ability of D. higginsii to cause severe disease and reduce the yields of vegetative organs indicates that it has potential as a biological control agent for purple nutsedge and yellow nutsedge, two of the world's worst weeds. This is the first record of occurrence of D. higginsii outside of its original distribution in Georgia (3). References: (1) S. A. Alfieri, Jr., et al. 1994. Diseases and Disorders of Plants in Florida. Bull. No. 14. Division of Plant Industry, Gainesville, FL. (2) M. B. Ellis. 1976. More Dematiaceous Hyphomycetes. Commonwealth Mycological Institute, Kew, Surrey, England. (3) D. F. Farr et al. 1989. Fungi on Plants and Plant Products in the United States. American Phytopathological Society, St. Paul, MN. (4) E. S. Luttrell. Mycologia 46:810, 1954.

Plectosporium tabacinum, a Pathogen of the Invasive Aquatic Weed Hydrilla verticillata in Florida

Plectosporium tabacinum, the anamorph of Plectosphaerella cucumerina, was isolated in 1996 from Hydrilla verticillata (hydrilla), an invasive aquatic weed in Florida. P. tabacinum, applied as a suspension of conidia, was pathogenic to hydrilla shoots maintained in aqueous solutions in test tubes. Koch's postulates were fulfilled in several repeated experiments. Infected shoots became slightly chlorotic within 24 h after inoculation. Infected leaves remained intact and were supported by water pressure but collapsed upon removal from water. Histological studies of leaves stained with malachite green and acid fuchsin revealed fungal hyphae within plant cells. The disease developed over a range of temperatures from 15 to 30°C. At 25°C, symptoms were most severe in 5% Hoagland's solution, followed by river water, deionized water, 0.5% Hoagland's, tap water, and spring water. Disease severity increased as inoculum concentration was increased from 10⁵ to 10⁷ conidia ml^-1. This is the first report of P. tabacinum as a pathogen of hydrilla, a fully submerged aquatic plant species.

Dispersal of Spores of Fusarium culmorum in Aquatic Systems

ABSTRACT The dispersal of spores of Fusarium culmorum, a biological control agent for the aquatic weed Hydrilla verticillata, was investigated in aquatic systems. Macroconidia and chlamydospores that were applied to the surface of the liquid settled rapidly in deionized water, tap water, 5% Hoagland's solution, natural spring water, or river water held in glass containers. The rate of fall, as measured for 50% of the spores, was determined to be 9 cm h(-1). Rapid lateral dispersal of spores from a point source occurred in still water. This initial spore movement occurred at a rate of >9 m h(-1) (15 cm min(-1)), approximately 100 times faster than the rate of settlement. The spores attained an even lateral distribution in a still, closed system. Spores dispersed rapidly in moving water and were transported with the water current. Spores were determined to carry positive electrostatic charges as they migrated towards the negative pole during electrophoresis. The physical components of dispersal of F. culmorum spores were defined in a still aquatic system to consist of rapid lateral dispersal and sinking due to gravity. In moving water, the dynamics of water movement was superimposed over the other two factors.

First Report of Rhizoctonia solani Causing a Foliar Leaf Spot on Brazilian Pepper-tree (Schinus terebinthifolius) in Florida

The Brazilian pepper-tree (Schinus terebinthifolius Raddi) native to Brazil, recently has become an aggressive perennial weed in southern Florida. During a survey in December 1995, a foliar disease was observed on several pepper-tree plants in Palm Beach County. Disease symptoms consisted of dark, reddish-purple necrotic lesions, either with or without dry necrotic centers, that were distributed randomly over the leaf surface. Infected leaf samples from two separate sites were plated on potato dextrose agar (PDA; Difco) and water agar and incubated at 25°C in the dark. A fungus resembling a Rhizoctonia sp. was consistently recovered. To prove Koch's postulates, the fungus was grown on PDA for 10 to 14 days, and the cultures blended in a Waring blender. Metamucil (Procter & Gamble) was added to the mixture at the rate of 0.5% wt/vol, and the suspension was used to spray and inoculate 2- and 3-month-old Brazilian pepper-tree seedlings. Seedlings were sprayed until the inoculum dripped off the foliage and after inoculation were maintained at 100% relative humidity. After 48 h in the dew chamber the inoculated seedlings were moved to a greenhouse bench and examined for infection 5 and 10 days later. Inoculation was completed three times with the leaf lesions occurring 94 to 100%. A Rhizoctonia sp. was recovered from the lesions that appeared on the challenged plants. A determination of the anastomosis group was performed by plating it against the tester isolates of R. solani, AG1-1A, AG2-2IV, AG-3, AG-4, and AG-5. In two separate tests anastomosis (imperfect fusion) (1) was observed between the recovered Rhizoctonia sp. and tester strain AG2-2IV of R. solani. The fungus was identified as R. solani, and this is the first report of R. solani causing a leaf lesion of Brazilian pepper-tree in Florida. The potential of this R. solani as a biological control agent of Brazilian pepper-tree remains to be tested. Reference: (1) B. Sneh et al. Identification of Rhizoctonia Species. American Phytopathological Society, 1991.

Bostrycin and 4-deoxybostrycin: two nonspecific phytotoxins produced by Alternaria eichhorniae

Two crystalline red pigments with phytotoxic activity were isolated from culture filtrates of Alternaria eichhorniae, a pathogen of the water hyacinth Eichhornia crassipes. The pigments were present in the ratio of 4:1 and were identified as bostrycin and 4-deoxybostrycin, respectively. This is the first isolation of 4-deoxybostrycin from a natural source. Bostrycin, 4-deoxybostrycin, and their isopropylidene derivatives induced necrosis on tested plant leaves comparable to the A. eichhorniae-induced necrosis on water hyacinth. The lowest phytotoxic concentrations of crystalline bostrycin and 4-deoxybostrycin on water hyacinth leaves were about 7 and 30 microgram/ml, respectively. Both substances were inhibitory to Bacillus subtilis but were inactive against the fungus Geotrichum candidum.