Evaluation of the Toxicity of Adonis Aestivalis in Calves

Abstract

Toxicosis of Adonis aestivalis is well documented in horses, but little is known of its toxicity in cattle. A. aestivalis (summer pheasant's eye) was collected over multiple years, under different growing conditions, and at various stages of maturity, dried, and administered to calves to evaluate the toxicity of A. aestivalis in cattle. Four 300-lb Holstein, and 2 90-lb, preruminating Jersey calves were administered 1% body weight of ground A. aestivalis via a stomach tube and monitored for clinical signs for 2 weeks and 1 week, respectively. The Holstein calves were then fed 0.2 to 1% body weight A. aestivalis daily for 4 to 5 weeks. The Holstein calves had transient, mild cardiac abnormalities during the feeding trial. Mild, transient gastrointestinal and cardiac signs were noted in the preruminating calves. No gross or microscopic lesions were seen on necropsies performed at the end of the study. Based on the results of this study, cattle do not appear to be as susceptible to toxicosis from A. aestivalis as other species, such as horses and pigs.

Acute poisonings with Adonis spp. have been reported in a variety of animal species. 8,9,14,18,22 Reports of toxicoses in horses 9,16,22 and pigs 8 have been based on cases of natural exposure to Adonis sp. In horses, gastrointestinal stasis and myocardial degeneration occurred after ingestion of Adonis aestivalis-contaminated hay. 22 Feed refusal, vomiting, dyspnea, and death occurred in pigs fed a diet containing seeds of Adonis microcarpa. 8 In laboratory studies in mice 12 and cats, 3 intravenous administrations of Adonis-like glycosides and the aglycone strophanthidin resulted in acute toxicoses. An experimental feeding trial performed in 1929 demonstrated Adonis annua was lethal to sheep when fed 1.0 lb of fresh plant, seed-bearing mature stage of the plant and extracts of the partially dried plant. 14 In the same study, however, feeding cattle 2 to 6 lb daily for 36 days failed to elicit clinical signs and death. Based on these archived 1929 records from the New South Wales Department of Agriculture, cattle are under consideration as alternative consumers of A. aestivalis-contaminated hay. A more current study evaluating the potential toxicity of A. aestivalis in cows was needed and therefore conducted.

A. aestivalis (summer pheasant's eye) was collected from 3 different sites in Modoc County, California, with variable growing conditions over a 3-year period (2003–2005) and at different stages of maturation (early bloom, full bloom, and seed set) so that if a higher level of toxin is associated with a specific soil condition, stressor in different years, or growth stage, collection would include plants with an adequate level of toxin. Plants adjacent to the road side with poor soil conditions were collected in 2003 (early bloom) and 2004 (full bloom). Adonis plants collected in 2003 and 2004 were dried, boxed, and preserved under dry conditions in a warehouse for 2 years and 1 year, respectively. Combined dry weight of Adonis collected in 2003 and 2004 was 70 lb. In 2005, plants were collected from a horse pasture and a catfish farm during early bloom and from the same horse pasture at full bloom and again at seed set. Most of the plants collected in the horse pasture with dry volcanic soil were of short stature (2–3 feet high) with short internodes. Plants collected from a catfish farm with moist organic soil were 5 to 6 feet high with extended internode distance. Full plant cycle occurred over a 3-month period (May-July).

Whole plants were sun dried and ground in a hammer mill. Portions of whole plants and ground Adonis were analyzed for the strophanthidin aglycone using liquid chromatography mass spectrometry as previously described. 11 Strophanthidin is the aglycone of several cardenolides found in Adonis. 10 Due to the lack of commercially available analytical standards for the cardenolides present in Adonis, the analytical method that was used does not directly measure the concentrations of all the individual toxic constituents in the plants. It instead provides an estimate of the overall levels of the strophanthidin glycosides by estimating the concentration of the aglycone strophanthidin. The method detection limit for strophanthidin in plant material was 0.08 μg/g, whereas the limit of quantitation was 0.24 μg/g.

Concentrations of strophanthidin in stems from plants collected in 2004 from the roadside (full bloom), 2005 from the horse pasture (early bloom), and 2005 from the catfish pond (early bloom) were 36.3, 24.5, and 11.4μg/g, respectively. Leaves from plants collected in 2004, 2005 (horse pasture), and 2005 (catfish pond) contained 19.6, 6.42, and 7.98 μg/g, respectively. Only flowers from the plants collected from the catfish pond were analyzed from this group, which had 12.4 μg/g strophanthidin. Ground Adonis collected at full bloom in 2005 and a whole plant collected at full bloom in 2005 had 7.5 and 8.2 μg/g, respectively, and Adonis collected at seed set in 2005 contained 98 μg/g strophanthidin. Correlation of strophanthidin concentration and growing stage or growing conditions cannot be made, but general trends suggest stems have higher concentrations of strophanthidin than leaves and later growing stages of the plant contain higher concentrations, but more complete evaluation is warranted.

Four 300- to 350-lb Holstein bull calves (Nos. 1–4) were purchased from a local dairy (2 at a time) and housed in an outside corral/pen with weather covering at the University of California, Davis, feedlot. Animals were acclimated for 1 week prior to the feeding trial. The toxicity study was performed twice, 1 pair at a time, because of the size of the pen and availability of space. Calves were fed 2% body weight alfalfa hay and 5 lb of grain daily prior to and after dosings and water ad libitum. By the end of the feeding trials, the Holstein calves weighed 500 to 600 lb. Two 2-month-old, 80- to 90-lb, preruminating Jersey bull calves (Nos. 5 and 6) were obtained from the Department of Animal Science, University of California, Davis, and fed milk replacer daily prior to and after dosing. Animal studies were conducted in compliance with the Animal Welfare Act, the U.S. Public Health Service Policy on the Humane Care and Use of Laboratory Animals, and the Guide for Care and Use of Agriculture Animals in Agriculture Research and Teaching. Amendments were submitted to and approved by the University of California, Davis, Animal Care and Use Committee prior to adjusting the design based on results obtained during the study.

Holstein calves (Nos. 1–4) were dosed with a slurry of finely ground A. aestivalis in water (1% body weight) via stomach tube using a bilge pump. Inoculum was divided into 2 doses 16 hours apart. The amount of Adonis administered at 1 dosing was based on the amount of water required to maintain flow through the stomach tube and the degree of distension of the rumen. The Adonis administered was collected in 2004 at full bloom (Nos. 1 and 2; 36.3 μg/g of strophanthidin) and a mix of 2005 (Nos. 3 and 4; 44 μg/g of strophanthidin). Calves were observed for clinical signs for 2 weeks. Alfalfa hay and grain (2% body weight and 5 lb, respectively) were fed once daily during the 2-week period of observation.

A low-dose, multiweek feeding study was initiated when calves failed to develop clinical signs 2 weeks after dosing with 1% body weight. Calf Nos. 1 and 2 were fed up to 50:50 (volume), 30:70 (dry weight) finely ground Adonis: grain mixed with molasses/corn syrup for 4 weeks starting with 0.2% body weight Adonis gradually increasing to 0.7% body weight Adonis by week 2 and maintaining this percentage until the end of the feeding trial. Each calf was fed 5 to 8 lb Adonis/grain mix divided twice daily and 1 to 1.5 flakes of alfalfa hay (8–12 lb) once daily. Full bloom (2004; 36.3 μg/g) and early bloom (2003; not analyzed for strophanthidin) Adonis harvest were used for this feeding trial. Calf Nos. 3 and 4 were fed 100% alfalfa pellet (2% body weight) for 3 days and then increasing amounts of 25:75 Adonis:alfalfa pellets attaining 25% Adonis:alfalfa pellet (7.5 μg/g of strophanthidin) by 1 week. This pellet was then mixed with the 50% Adonis:alfalfa pellet (12 μg/g of strophanthidin) gradually increasing to 50% Adonis:afalfa pellet by 2 weeks (1% body weight). This concentration was maintained until the end of the 5-week feeding trial. Calves were supplemented with 1 flake of alfalfa daily. Adonis used for calf Nos. 3 and 4 was a mix from all growth stages of the 2005 harvest (early, full, and late bloom/seed set). Overall, strophanthidin in the plant material fed the calves in this study ranged from 7.5 to 98 μg/g strophanthidin. Different levels of the strophanthidin aglycone were detected in samples from the same inoculum of ground Adonis due to uneven mixing of plant material with variable levels of strophanthidin. Therefore, precise levels of strophanthidin administered could not be determined.

Two 80- to 90-lb Jersey calves (Nos. 5 and 6) were administered 1% body weight of finely ground Adonis (late bloom 2005 collection; 44 μg/g of strophanthidin) via stomach tube, divided into 3 administrations 8 to 12 hours apart, and monitored for clinical signs for 1 week. A 500-lb Holstein calf (negative control) was administered a slurry of 1% body weight ground alfalfa via a stomach tube. A 900-lb cow was housed under the same environmental conditions and fed ground alfalfa hay mixed with grain and a flake of alfalfa hay for several weeks as a negative control for the multiweek dosing study.

Calf Nos. 1 and 2 received precursory physical examinations prior to dosing. Whole blood was collected before inoculation and 24, 48, and 192 hours after inoculation for complete blood counts, and all values were within normal limits. At the end of the feeding trial just before necropsy, calves received complete physical and cardiac examinations. Physical and cardiac examinations were performed on calf Nos. 3 and 4 before the initial dosing, 48 hours after dosing, and at the end of the feeding trial before necropsy and in preruminating Jersey calves before dosing with Adonis and at 12, 24, and 48 hours after dosing. All calves were observed daily for clinical signs until the end of the toxicity study. Jersey calves were observed for clinical signs daily for 1 week after administration of Adonis, at which time they were necropsied. Pre- and 24 hours postdosing cardiac examinations were performed on the negative control cow that was administered 1% body weight ground alfalfa in water via stomach tube. This cow was observed daily for clinical signs and was not necropsied because no clinical signs were observed.

Oral dosing of Adonis via stomach tube did not result in clinical signs of acute toxicity in any of the 4 Holstein calves. In the multiweek feeding trial, 2 of the 4 Holstein calves (Nos. 1 and 3) consumed all the Adonis every day whether it was ground or provided in pellet form. Calf No. 2 left more than half of the Adonis/grain mix every third day, but at the end of the feeding trial, 1 day prior to necropsy, whole dried A. aestivalis was tested for palatability, and calf Nos. 1 and 2 readily ate the plant whole. Calf No. 4 exhibited feed refusal when initially fed the pellet, and then only ate the pellet every third day for the entire feeding trial. Calves were fed alfalfa hay for maintenance on days they refused feed. The Holstein calves exhibited no clinical signs during the multiweek feeding trials. The Jersey calves were lethargic and developed diarrhea on days 2 and 3 after dosing, but no clinical signs were noted for the remainder of the week. The 2 negative control animals exhibited no clinical signs, and cardiac examination of the negative control calf dosed with ground alfalfa by stomach tube was unremarkable. These calves were not necropsied.

Calves were loaded into a trailer and moved to the Veterinary Medical Teaching Hospital, University of California, Davis, for cardiac evaluations. Extensive cardiac examinations performed before, during, and at the end of the feeding trials included electrocardiography; ultrasonography; measurements of central venous pressure and arterial blood pressure; and plasma electrolytes, glucose, and blood gas analyses. Electrocardiograms were recorded using both orthogonal and augmented unipolar leads. a Cardiac 2-dimensional ultrasound examinations were conducted using a 6.0-mHz microconvex linear probe. b Evaluation of left ventricular contractility and quantitation of fractional shortening was calculated using described methods for cattle. 2 Consistent, sustained changes in cardiac activity were not observed in calves before and after dosing with Adonis or in the Holstein calves that were given alfalfa (negative controls).

Heart rates ranged between 43 and 84 beats per minute after calves were given Adonis and were 73 and 102 beats per minute before and after in the alfalfa-dosed controls. One Holstein Adonis-dosed calf (No. 4) developed a 48-hour postdosing heart rate of 43 beats per minute and simultaneously had a sinus arrhythmia with R-to-R intervals ranging between 0.9 and 1.4 seconds 48 hours after acute dosing. Heart rate was within normal limits the next day by auscultation, and these changes were not seen in this animal at the end of the feeding trial. Sinus arrhythmias were not detected in any other calves. Forty-eight hours after dosing with Adonis, fractional shortening was reduced by 61% from pre-exposure levels in preruminating calf No. 5 and by 40 and 28% in 2 ruminating calves (Nos. 3 and 4, respectively). The other preruminant calf (No. 6) had increased fractional shortening (18%) 48 hours after Adonis was administered.

At the time points of cardiac examination, anticoagulated c blood was collected for measurement of plasma concentrations of glucose, sodium, potassium, and calcium and of blood pH, bicarbonate, and total concentration of carbon dioxide using a Point of Care Clinical Analyzer. d Blood pressure was measured using a noninvasive oscillometry method with a 3-cm cuff on the coccygeal artery. e Central venous pressure was measured manometrically with the zero pressure point located at the level of the acromion. Plasma sodium concentrations ranged between 131 and 139 mEq/L in all calves, which were within expected age-adjusted reference ranges. 4 Other pre- and postdosing blood chemistry values were in the range of predicted normal values for young calves. 17 Ranges included 72 and 45 mg/dl glucose predosing and 121 and 41 mg/dL 24 hours postdosing. Blood bicarbonate concentrations ranged between 39 and 46 mM/L predosing and 39 and 44 mM/L postdosing. Concentrations of potassium, total carbon dioxide, and pH were within the reported normal ranges for calves. 4,17 Central venous pressures of all calves at all times ranged between 0.5 and 1 mm H₂0. Average blood pressure for 4 calves ranged between 66 and 111 mm Hg before exposure and between 71 and 110 mm Hg after exposure. All blood pressure measurements were within the expected normal range of age-adjusted values for calves. 1

Necropsies were performed on all dosed calves with subsequent histologic examination of select tissues. Portions of brain, tonsil, retropharyngeal lymph node, liver, kidneys, lungs, spleen, rumen, reticulum, omasum, abomasum, small and large intestines, adrenal glands, skeletal muscle, trachea, thyroid, diaphragm, and esophagus and 10 sections of heart (including the sinus node, atrioventricular node, common bundle, left and right crura and conducting fibers, each valvular region, each chamber, and great vessels) were collected from all calves, immersed in 10% buffered neutral formalin for 24 hours, sectioned, stained with hematoxylin and eosin, and examined by light microscopy. No lesions were noted on gross or microscopic examination of tissues from all 4 Holstein calves and the 2 Jersey calves except for mild superficial rumenitis in the Holstein calves that received the Adonis/grain mix (Nos. 1 and 2). No microscopic lesions were seen in the 10 standard sections of heart examined from each of the calves.

Strophanthidin is the aglycone of several cardenolides in A. aestivalis. 10 The direct effect elicited by these compounds is similar to other cardiac glycoside-containing plants and is due to inhibition of the sodium potassium adenosine triphosphatase enzyme system pump 15 resulting in hyperkalemia and degenerative changes in the heart. Hyperkalemia was not evident in any of these calves after dosing, and degenerative lesions were not seen in any of the calves on gross or microscopic examination. Calves exhibited only mild and transient changes when dosed with up to 1% body weight Adonis, a higher percentage than would be expected with consumption of contaminated hay. Therefore, cattle do not appear to be as susceptible to toxicosis from A. aestivalis as other species, such as horses and pigs. The mechanism underlying the variation of A. aestivalis toxicity in different livestock species is unknown and needs to be elucidated. It is well known, however, that the toxicity of many toxic plants (including cardiac glycoside-containing plants) varies in animal species. 13,15,19,20 For example, human and murine pharmacokinetic and metabolic differences have been demonstrated with oleandrin. When oleandrin is incubated with murine plasma, it is converted to its aglycone metabolite oleandrigenin, but conversion does not occur when incubated with human plasma. 19 A cellular basis for species differences in sensitivity to cardiac glycosides has also been reported, as well as differential effects on tumor cell lines from different animal species. 13,20

Cardiac glycosides increase vagal tone, which decreases the rate of sinoatrial node depolarization. 21 Electrocardiographic changes typically seen with cardiac glycoside toxicosis include bradycardia, varying levels of atrioventricular block, ventricular arrhythmias, and ventricular fibrillation. Electrocardiographic abnormalities seen in these calves were inconsistent and transient. One of the Holstein calves (No. 4) developed a postdosing bradycardia (43 beats per minute) and simultaneously had a sinus arrhythmia with R-to-R intervals ranging between 0.9 and 1.4 seconds. No corresponding gross or microscopic lesions were seen in the heart of this calf however. Causes of these changes may include a combination of direct activity of strophanthidin upon atrioventricular nodal conduction or indirect activities of the vagus nerve. 5,21 Adminstration of the quantity of ground plant material needed for the acute dose required excessive water to aid flow through the stomach tube. The dose was split into 2 allotments approximately 12 hours apart, and at the end of each dosing, the rumens were markedly distended. Severe distension of the rumen could have triggered vagal-related changes. 6,7 The alfalfa-gavaged negative control calf had normal postdosing heart rates, and sinus arrhythmias were not detected in this calf or in any of the other calves, so vagal stretch receptors were probably not the predominant cause of the bradycardia and sinus arrhythmia. Therefore, the vagotonic effects were likely initiated by a direct toxic effect.

The lack of significant clinical signs or lesions in the Holstein calves prompted a short feeding trial in the preruminating Jersey calves. Because monogastric animals (horses and pigs) are susceptible to toxicosis with Adonis, a functional rumen may be associated with the lack of susceptibility of calves to toxicosis. Sheep are apparently susceptible to toxicosis, however. 14 In the present study, the reduced fractional shortening was more significant in the postdosing, preruminating Jersey calf than in the weaned Holstein calves, suggesting cardiac glycosides were partially inactivated by rumen metabolism, reduced affinity of strophanthidin for the cardiocytes of older calves, or less likely a breed predisposition. Preruminating calves also developed a transient diarrhea. By 72 hours, however, the calves recovered, and no lesions were evident on necropsy 1 week after dosing. Therefore, the role of the rumen in the variable species susceptibility to Adonis remains elusive.

In contrast to calves, pigs exhibited feed refusal, vomiting, dyspnea, and death after consuming a diet containing seeds of A. microcarpa. 8 Three horses developed colic followed by cardiac arrhythmias, becoming moribund 24 to 96 hours after ingesting Adonis-contaminated hay. 22 Myocardial degeneration was seen on necropsy and microscopic examination of tissues from these horses, and strophanthidin was detected in gastrointestinal contents. Adonis plants found in the contaminated alfalfa hay ingested by the horses had 11.3 μg/g strophanthidin. Although it is difficult to estimate true percentage of the Adonis contaminant in the hay from small samples submitted to a diagnostic laboratory, the percentage of Adonis in the submitted sample from this original case appeared to be minimal.

At the dose administered in this study, dried A. aestivalis with 11 to 98 μg/g of strophanthidin produced only subtle and transient electrocardiographic changes and no gross or microscopic changes in the hearts of 90- to 500-lb calves. These results suggest that cattle are less susceptible than horses and pigs to cardiotoxic effects and sudden death after ingestion of relatively small quantities of A. aestivalis in contaminated feedstuffs. Therefore, alternative etiologies need to be fully investigated and ruled out in instances of sudden deaths in cattle in which exposure to A. aestivalis represents less than 1% body weight. Further studies need to follow this preliminary study to determine the toxicokinetics of Adonis cardiac glycosides in different animal species.

Acknowledgements. Project funding was provided by the Center for Food Animal Health, School of Veterinary Medicine, University of California, Davis, California. The authors thank Richard Plocher, Plocher Partnership, Cache Acres Holsteins, for livestock donations; Joe Moreo, Modoc County Agriculture Commissioner; and Jerry Johnson and Jason Campbell, Department of Animal Science, University of California, for their valuable assistance.