Research Project: Developing Mitigation Strategies for Poisonous Plants in Livestock Production Systems

Location: Poisonous Plant Research

2025 Annual Report

Objectives
Objective 1: Develop science-based guidelines to reduce livestock losses on rangelands, including evaluating differences in toxin accumulation in poisonous plants. Sub-objective 1.A: Evaluate herbicides to determine efficacy in controlling Geyer larkspur (Delphinium geyeri) and determine if the toxicity of Geyer larkspur changes due to herbicide treatment. Sub-objective 1.B: Determine the effect of forage selenium concentrations on relative palatability, and subsequent productivity in livestock. Sub-objective 1.C: Characterize changes in norditerpene alkaloids in Delphinium species among geographical locations, plant parts, and over the growing season. Sub-objective 1.D: Screen herbarium specimens representing species of different genera (Delphinium, Zigadenus, Astragalus, Oxytropis, and Salvia) for suspected toxins. Objective 2: Enhance methods for analyzing plant and animal tissues for plant toxins, measuring toxicokinetics, assessing carcinogenic and genotoxic potential, as well as identifying plant toxin metabolites and biomarkers of toxicoses. Sub-objective 2.A: Evaluate the utility of earwax and hair as noninvasive specimens for diagnosis of livestock exposure to additional poisonous plants (death camas, locoweed, and lupine). Sub-objective 2.B: Characterize the carcinogenic potential of purified dehydro-pyrrolizidine alkaloids (DHPA’s) (lasiocarpine, seneciphylline, senecionine, heliotrine and their n-oxides) and compare these with known DHPA carcinogens. Sub-objective 2.C: Determine primary serum and rumen biomarkers for livestock poisoned by death camas. Sub-objective 2.D: Evaluate DNA metabarcoding technologies as a diagnostic method for poisoned animals and contaminated feeds. Objective 3: Develop improved diagnostic and prognostic procedures to reduce negative impacts of poisonous plants on livestock including early identification of poisoned animals, predicting poisoning outcomes, as well as best management and treatment options. Sub-objective 3.A: Compare the pathological changes in livestock poisoned by Ipomoea carnea with and without swainsonine. Sub-objective 3.B: Determine the effects of chronic exposure to excessive amounts of selenocompounds commonly found in supplements and forages, on spermatogenesis and sperm quality in sheep. Sub-objective 3.C: Develop drug treatments for livestock that are poisoned by poison hemlock. Sub-objective 3.D: Determine the potential toxicity and teratogenicity of hemp to livestock. Objective 4: Develop guidelines to aid producers and land managers in making evidence-based herd management decisions to improve livestock performance on rangelands infested with poisonous plants. Sub-objective 4.A: Determine if mineral supplementation can reduce plant-induced poisonings of livestock. Sub-objective 4.B: Characterize conditions that cause sheep to become poisoned when grazing on death camas-infested rangelands. Sub-objective 4.C: Compare the susceptibility of cattle that are native to larkspur-infested ranges to cattle that are naïve to larkspur-infested ranges. Sub-objective 4.D: Comparison of the susceptibility of taurine and indicine cattle to lupine toxicosis.

Approach
There are hundreds of genera of toxic plants, representing thousands of species. Plant poisonings occur worldwide including 333 million hectares infested with poisonous plants in China and 60 million hectares in Brazil. The livestock industry in the western United States loses over $500 million annually from death losses and abortions due to poisonous plants. Actual losses due to poisonous plants are much greater due to wasted forage, reduced animal performance, and increased management costs. The Poisonous Plant Research Laboratory (PPRL) provides numerous solutions to toxic plant problems using an integrated, interdisciplinary approach representing several scientific disciplines and continues to provide worldwide leadership in poisonous plant research to the livestock industry and the scientific community. The PPRL research team investigates plant poisonings in a systematic manner by identifying the plant, determining the toxin(s), evaluating the mechanisms of action, and describing the effects in animals. Our mission is to develop research-based solutions to reduce livestock losses from toxic plants. There are four coordinated objectives in this project plan providing guidelines for potential scientific-based management. The project focuses on several toxic plants including larkspur, locoweed, lupine, and dehydro-pyrrolizidine alkaloid-containing plants utilizing the various scientific disciplines of the staff. The products of this research will help to reduce livestock losses from plants and enhance the economic well-being of rural communities, improve rangeland health by combating invasive plant species, and help to provide safe animal products free from potential plant toxins for consumers.

Progress Report
This report documents the FY 2025 progress of project 2080-21500-001-000D, titled, “Developing Mitigation Strategies for Poisonous Plants in Livestock Production Systems”, which began in December 2023. In support of Sub-objective 1.B, research continues to examine if sheep will decrease consumption of forages as the concentration of selenium increases. Sheep were offered selenium continuing diets at 0.3, 10, 20, 40 and 120 ppm selenium and the consumption of the diets measured daily to determine feed and selenium intake. For Sub-objective 1.C, to further characterize changes in norditerpene alkaloids in larkspur species among geographical locations, plant parts, and over the growing season larkspur samples were collected, processed, chemically analyzed and tested. Two different larkspur species (Delphinium occidentale and Delphinium glaucescens) were collected from locations in Montana, Idaho, and Wyoming every three weeks representing the early vegetative, bud, flower, pod, and senesced stages of plant growth for two years at each location. Each larkspur species (D. occidentale or D. glaucescens) will be analyzed separately with each model including plant parts (leaf, stem, reproductive), phenological stage, and location and resulting interactions. In support of Sub-objective 2.A, earwax was collected from pregnant cattle grazed on lupine-infested rangelands and was analyzed. Two lupine species, Lupinus sericeus and Lupinus polyphyllus, were present on a rangeland. The teratogen, anagyrine, was detected in L. sericeus and the teratogen, ammodendrine, was detected in L. polyphyllus plants. Anagyrine was detected in the earwax of all 69 cows sampled. Ammodendrine, was detected in the earwax of 28 of the 69 cows sampled. Work will continue to evaluate the utility of earwax as noninvasive specimens for diagnosis of livestock exposure to additional poisonous plants. Under Sub-objective 2.B, experiments examining dehydropyrrolizidine alkaloid (DHPA) containing plants are underway. Rodent studies have been completed and tissues from the studies are being embedded, cut, and placed on a glass slide and stained for microscopic examination. All neoplastic or proliferative hepatic lesions will be identified, classified, and compared between the different treatment groups. For Sub-objective 2.C, the toxicity of foothill death camas attributed to esters of zygadenine was investigated. Experiments indicate that zygacine is metabolized to zygadenine by esterases mostly in the liver which suggests that zygacine undergoes first-pass liver metabolism, significantly reducing the toxin in the general circulation. Experiments using in vitro assays of rumen, plasma, and liver metabolism examined two other esters of zygadenine; 3-angeloylzygadendine, and 3-veratroylzygadenine have been completed. In support of Sub-objective 2.D, chemistry, genome skims, and metabarcoding were used to retrospectively describe the composition of contaminated alfalfa hay from a case of Salvia reflexa (lanceleaf sage) poisoning that killed 165 cattle. Halogeton (Halogeton glomeratus) and greasewood (Sarcobatus vermiculatus) were detected in the rumen contents of poisoned cattle and sheep using DNA metabarcoding. A clinical case of milkweed toxicity in cattle due to contaminated hay was reported where the cattle had clinical signs consistent with milkweed toxicity and milkweed was detected in the rumen contents of the poisoned cow using DNA metabarcoding. For Objective 3, with a goal to develop improved best management and treatment options for livestock impacted by poisonous plants, a further characterization of cattle that abort in the last trimester of gestation was performed. Ponderosa pine (Pinus ponderosa) needles are known to induce abortions in cows when consumed during the last trimester of pregnancy. Traditionally pine needle-induced abortions occur in areas where cattle are forced into a stand of ponderosa pine trees to seek shelter and feed during a winter storm that is either very cold, has heavy snow or high winds. However, we evaluated two incidences of pine needle-induced abortions in cattle that were unique. In one case, there was no weather-related event that forced the cattle into the stand of pine trees, rather the cattle went into the pine trees seeking the new growth of grasses near the pine trees. In the second incident, the abortions occurred in an area with no previous history of pine needle-induced abortions, which may be due to the relatively low concentration of abortifacient compounds in the needles from that area. However, in both cases, the resultant abortions, as well as the effects on the cows and calves, were consistent with pine needle-induced abortions. These two cases highlight the fact that when cows are allowed to graze in areas with ponderosa pine needles during the last trimester of gestation, there is always a risk for the cattle to consume enough pine needles for abortions to occur. Under Sub-objective 3.C, investigations of evidence-based therapies to manage the clinical signs of intoxication caused by toxic plants in livestock are underway. Work was completed to develop a drug-based intervention for the management of clinical signs of piperidine alkaloid intoxication in livestock. The actions of anabasine, coniine, gamma-coniceine, and two total alkaloid extracts from Lupinus sulphureus were compared in the presence and absence of the nicotinic acetylcholine receptor partial agonist varenicline. Varenicline had a protective effect in cell culture experiments with anabasine and coniine. Pretreatment with varenicline protected mice from anabasine, coniine, and gamma-coniceine toxicity. However, varenicline was not an effective means to manage poisoning by piperidine alkaloids in goats. In support of Sub-objective 3.D, industrial hemp (Cannabis sativa) production and cannabinoid extraction results in extracted plant material, or hemp byproducts, and these byproducts have been identified as a feed-source for livestock. However, there is a potential that cannabinoids may be teratogenic to ruminant livestock species. To evaluate the teratogenic potential of hemp pregnant ewes were dosed with hemp in the form of dried ground hemp plant material from gestational days 10 to 20. No adverse developmental effects were observed in the lambs from ewes fed hemp, suggesting that hemp, or hemp byproducts may be a suitable forage source for ruminant livestock species. In support of Objective 4, additional work has been conducted to characterize the effect of livestock consuming more than one poisonous plant. Death camas is a toxic plant often associated with poisoning cases in livestock, especially in sheep in the spring in mountain foothills. Consequently, when sheep are found dead in mountain foothills and death camas is present, it is often assumed to be the cause of death. However, there may be other plants that also contribute to, or are the cause of death in sheep. Scientists in Logan, Utah, were recently involved in diagnosing a case wherein approximately 200 sheep out of a band of 400 died within 24 hours after being dropped off on a mountain foothill in northern Utah. Death camas was present in the area, and the initial diagnosis was that death camas was the cause of the deaths. However, after further evaluation, even though death camas was present in this area, there was not enough death camas to kill that many sheep. Consequently, studies have been conducted to assess the effect of dosing multiple plants from the area together with death camas. Studies were also conducted to more definitively evaluate the toxicity of the chokecherry from that location. These additional investigations suggest that chokecherry may have been the cause of death in sheep in this area. For Sub-objective 4.B, ARS scientists in Logan, Utah, hypothesized that sheep that are held off feed and turned out to graze rangelands infested with death camas in a hungry state will consume more death camas than satiated animals resulting in more poisoned animals. Grazing studies have been conducted. Samples have been analyzed for nutrition content and toxin content. Data is currently being analyzed. A manuscript is currently being written. Under Sub-objective 4.C, a study that compared the susceptibility of cattle that are native to larkspur-infested ranges to cattle that are naïve to larkspur-infested ranges has been completed. Specifically, a grazing study has been conducted. Samples have been analyzed for nutrition and toxin content. Data is being analyzed. The manuscript is in preparation.

Accomplishments
1. Diagnosing livestock poisoning with DNA metabarcoding and chemistry. Poisonous plants on western U.S. rangelands cause substantial economic losses each year and accurate diagnosis of plant-induced livestock poisoning hinges on multiple, corroborative lines of evidence. ARS scientists in Logan, Utah, with collaborators, combined DNA metabarcoding with targeted chemical analyses to confirm that halogeton (Halogeton glomeratus) and milkweed (Asclepias spp.) were responsible for the poisonings two separate livestock-fatality events, and in a retrospective study of contaminated alfalfa hay that killed 165 cattle, researchers used genome skims alongside metabarcoding and chemistry to pinpoint Salvia reflexa as the toxic adulterant. Results from this research suggest that DNA metabarcoding enables high-throughput identification of multiple plant species from complex botanical samples (e.g., mixed hay, rumen contents) and provides an independent line of evidence when morphological ID is impossible. These advanced livestock forensic techniques significantly improve the diagnosis of field cases of acute livestock deaths when traditional necropsy and pasture surveys fall are of limited utility. In addition, DNA Metabarcoding can be used in assessing prepared feeds, silage, hay and mixed forages for accidental inclusion of toxic weeds. This research helps extension agents, veterinarians and producers to implement targeted weed management and feed screening protocols.

2. Dallisgrass and Japanese Yew Poisoning Cases in livestock. ARS scientists in Logan, Utah, with collaborators, investigated two significant cases of plant poisoning in animals: Dallisgrass poisoning in cattle and Japanese Yew poisoning in wild ungulates. Dallisgrass, a warm-season perennial forage common in southeastern U.S. pastures, can become infected by ergot fungi, producing tremorgenic indole-diterpene alkaloids and two outbreaks were reported, Mississippi: A 32-head herd experienced 65% morbidity and 6% mortality, and Oklahoma: A 42-head herd experienced 31% morbidity and 5% mortality. Affected cattle exhibited whole-body shaking, tremors, ataxia, and hyperexcitability, chemical analysis confirmed the presence of alkaloids in ergotized seed heads and rumen contents, linking ingestion to clinical signs. Japanese Yew species contain taxine alkaloids that disrupt cardiac function, and several cases of yew poisoning involving deer, elk, and moose occurred in Utah during the winter of 2022–2023, despite the anecdotally believed high tolerance of wild ungulates to yew. Diagnosis combined three methods: Visual identification of yew fragments in rumen contents; chemical detection of taxines in rumen and liver samples; and DNA metabarcoding to confirm Taxus species. This research indicates that ranchers and extension agents should employ combined diagnostic approaches such as visual, chemical, and molecular information when plant poisoning is suspected in livestock or wildlife.

3. Toxic potential and management strategies for larkspur (Delphinium spp.) in western U.S. rangelands. Larkspurs are a leading cause of cattle poisoning on western rangelands, resulting in significant economic losses. In larkspur plants, norditerpenoid alkaloids are the primary toxic agents, with over 18 alkaloid variants existing as mixtures in various larkspur species that exhibit up to 30-fold differences in toxicity. ARS scientists in Logan, Utah, standardized collection and analysis methods to compare alkaloid profiles across 20 Delphinium species, and this toxicity assessment provided a clear hierarchy of Delphinium species risk, enabling targeted grazing management. Concurrently, ARS scientists assessed herbicide efficacy against Geyer’s larkspur, focusing on application timing and soil moisture. These herbicide trials confirm that aminopyralid is a reliable control agent for Geyer’s larkspur when timed with soil moisture. Combining plant toxicity mapping and moisture-based herbicide application timing will minimize cattle losses and support sustainable rangeland stewardship, and this information will be beneficial for scientists, livestock producers, extension agents, veterinarians to evaluate the risk of grazing livestock in the various larkspur-infested rangelands of the western United States.

4. Determination of the primary toxin(s) in death camas. Death camas (Zigadenus species) is a common poisonous plant found throughout North America in diverse habitats. The toxic alkaloids in foothill death camas are zygadenine, and esters of zygadenine. ARS scientists in Logan, Utah, compared the acute toxicity of zygacine and zygadenine in both mice and sheep and evaluated the toxicity of angeloylzygadenine and veratroylzygadenine. The results indicated that zygacine is more toxic than zygadenine. However, zygacine is metabolized very quickly, thus if an animal that has been exposed to zygacine can live long enough to metabolize these compounds, the animal will survive. Additionally, the data suggest that the rank order of toxicity of the death camas alkaloids are veratroylzygadenine > angeloylzygadenine > zygacine > zygadenine. This data will be beneficial for scientists, livestock producers, extension agents, veterinarians to evaluate the risk of grazing livestock in death camas-infested rangelands of the western United States.

5. Earwax can be used as a diagnostic specimen to determine livestock exposure to poisonous plants. ARS Scientists in Logan, Utah, have developed an innovative, non-invasive method for monitoring cattle exposure to teratogenic lupine alkaloids and foothill death camas. In the lupine study, all 69 cows had detectable levels of anagyrine a teratogen produced by Lupinus sericeus while about 40% (28 out of 69) showed the presence of ammodendrine from Lupinus polyphyllus. In the death camas study, ARS Scientists detected death camas alkaloids in the earwax of sheep and cattle. This research demonstrates the potential of earwax as a noninvasive specimen for chemical analyses to aid in the diagnosis of livestock that may have been exposed to and poisoned by lupine or death camas, and this suggests that earwax sampling offers a practical means to verify exposure to multiple poisonous plant species simultaneously without invasive procedures, and for ranchers, knowing which animals have been exposed to these toxic compounds is valuable. Additional research will be needed to determine factors such as toxic thresholds, individual susceptibility, and environmental variables.

6. Characterization of the toxicity of death camas in goats. Livestock losses to death camas have been reported in numerous species including cattle and sheep, with the largest losses generally occurring in sheep. Clinical signs of poisoning are similar for all animal species studied. ARS scientists in Logan, Utah, compared the susceptibility of goats and sheep to the acute toxic effects of death camas. The data presented in this manuscript demonstrate that goats are more susceptible to the acute toxic effects of death camas. Consequently, ranchers that graze goats in death camas infested pastures should use as much caution, if not more than they would with sheep. Additionally, the data presented in the study, suggests that goats can be used as a small ruminant model to study the toxic effects of death camas.

Review Publications
Green, B.T., Lee, S.T., Welch, K.D., Cook, D., Stonecipher, C.A. 2024. The actions of varenicline on alkaloids from Conium maculatum (poison hemlock), Lupinus sulphureus (sulphur lupine) and Nicotiana glauca (tree tobacco). Toxicon. 252. Article 108184. https://doi.org/10.1016/j.toxicon.2024.108184.
Lee, S.T., Ozuna, G.C., Villasenor, A., Cook, D. 2024. A case of dallisgrass staggers in Oklahoma. Poisonous Plant Research. 7:29-33. https://doi.org/10.26077/256c-1d01.
Lee, S.T., Stonecipher, C.A., Welch, K.D., Cook, D. 2024. The evaluation of earwax as a noninvasive specimen to determine livestock exposure to death camas (Zigadenus paniculatus). Toxicon. 252. Article 108181. https://doi.org/10.1016/j.toxicon.2024.108181.
Avellaneda-Caceres, A., Lee, S.T., Ruiz, A., Sandoval, G.V., Colque-Caro, L.A., Cook, D., Aguirre, L.S., Uzal, F.A., Micheloud, J.F. 2025. Oxalate nephropathy in cattle associated with the consumption of Megathyrsus maximus in Argentina. Journal of Veterinary Diagnostic Investigation. https://doi.org/10.1177/10406387251336265.
Kelly, E.J., Lee, S.T., Brown, A., Cook, D. 2025. Halogeton and greasewood poisoning in cattle and sheep in Utah. Poisonous Plant Research. 8:3-6. https://doi.org/10.26077/bd54-998b.
Stegelmeier, B.L., Davis, T.Z., Panter, K.E., Welch, K.D., Knoppel, E.L. 2024. A review of intermittent poisoning to mitigate toxic plant-induced disease in livestock. Veterinary Sciences. 12(1). Article 13. https://doi.org/10.3390/vetsci12010013.
Odum, A.L., Willis-Moore, M.E., Callister, K.T., Haynes, J.M., Frye, C.C.J., Scribner, L.N., Legaspi, D.N., Santos Da Silva, D., Olsen, A.L., Truscott, T.T., Alden, P.T., Bevins, R.A., Leventhal, A.M., Lee, S.T., Gomer, B., Benninghoff, A.D. 2025. The rodent electronic nicotine delivery system: Apparatus for voluntary nose-only e-cigarette aerosol inhalation. Journal of the Experimental Analysis of Behavior. 123(2):337-354. https://doi.org/10.1002/jeab.70005.
Welch, K.D., Dietz, M.A., Edmonds, S.E., Gardner, D.R., Cook, D. 2025. Two unique cases of ponderosa pine needle-induced abortions. Poisonous Plant Research. 8:23-28. https://doi.org/10.26077/a65e-3a11.
Walck, R., Webb, B., Ensley, S., Cook, D. 2025. A case of milkweed poisoning in cattle in Colorado. Poisonous Plant Research. 8:31-36. https://doi.org/10.26077/13yr-nf86.
Ovelar, M.F., Garcia, J.A., Cook, D., Gardner, D., Stegelmeier, B.L., Diez de Ulzurrun, P., Tettamanti, A., Balbuena, D., Lita, E.V., Poo, J.I., Scioli, M.V., Canton, G.J. 2025. Senecio pampeanus poisoning in beef cattle: Case report and toxicological evaluation. Veterinary Research Communications. 49. Article 205. https://doi.org/10.1007/s11259-025-10768-y.
Welch, K.D., Gardner, D.R., Cook, D., Lee, S.T., Stonecipher, C.A., Green, B.T. 2025. An evaluation of the toxic norditerpenoid alkaloid content across Delphinium species, and their acute toxicities in a murine model. Chemistry and Biodiversity. Article e00779. https://doi.org/10.1002/cbdv.202500779.
Stonecipher, C.A., Welch, K.D., Derner, J.D., Gardner, D.R., Ransom, C., Cook, D. 2025. Evaluation of herbicides for Geyer larkspur (Delphinium geyeri). Weed Technology. 39. Article e72. https://doi.org/10.1017/wet.2025.35.
Cook, D., Kocurek, B., Stonecipher, C.A., Welch, K.D., Gardner, D.R., Mammel, M., Reed, E., Ramachandran, P., Erickson, D., Commichaux, S., Ottesen, A. 2025. DNA data (genome skims and metabarcodes) paired with chemical data demonstrate utility for retrospective analysis of forage linked to fatal poisoning of cattle. Toxicon. 256. Article 108285. https://doi.org/10.1016/j.toxicon.2025.108285.
Marin, R.E., Gardner, D., Cook, D., Armien, A.G., Fortunato, R.H., Riet-Correa, F., Uzal, F.A. 2025. Intoxication of sheep by Astragalus arequipensis in northwestern Argentina. Journal of Veterinary Diagnostic Investigation. 37(2):375-379. https://doi.org/10.1177/10406387241311815.
Jumper, W.I., Brown, C.C., Lee, S.T., Cook, D., Stilwell, J.M., Harvey, K.M. 2024. Case report: Investigating an outbreak of tremorgenic mycotoxicosis in beef cows on pasture in Mississippi due to ergot (Claviceps paspali) infection in dallisgrass (Paspalum dilatatum). Bovine Practitioner Journal. 58(2):59-68. https://doi.org/10.21423/bpj20249035.
Riet-Correa, F., Cook, D., Micheloud, J.F., Machado, M., Mendonca, F.S., Schild, A.L., Lemos, R.A. 2024. A review on mycotoxins and mycotoxicoses in ruminants and Equidae in South America. Toxicon. 247. Article 107827. https://doi.org/10.1016/j.toxicon.2024.107827.
Welch, K.D., Gardner, D.R., Lee, S.T., Stonecipher, C.A., Cook, D. 2024. Comparison of the acute toxicity of zygacine versus zygadenine. Toxicon. 248. Article 108037. https://doi.org/10.1016/j.toxicon.2024.108037.
Stonecipher, C.A., Lee, S.T., Welch, K.D., Valles, K.R., Cook, D. 2024. The use of earwax to determine livestock exposure to teratogenic lupine. Toxicon. 248. Article 108053. https://doi.org/10.1016/j.toxicon.2024.108053.
Davis, J.S., Scott, M., Cook, D., Gardner, D., Morse, G., Grillo, M. 2024. Extensive local geographic variation in locoweed toxin produced by a fungal endophyte. Journal of Chemical Ecology. 50:465-477. https://doi.org/10.1007/s10886-024-01529-3.
Lee, S.T., Kelly, J., Stout, V., Lamb, S., Baldwin, T.J., Cook, D. 2024. Japanese Yew (Taxus) poisoning of wild ungulates in Utah during the winter of 2022-2023. Toxicon. 246. Article 107779. https://doi.org/10.1016/j.toxicon.2024.107779.