Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Evaluating digestibility and toxicity of native warm-season grasses for equines

Evaluating digestibility and toxicity of native warm-season grasses for equines †,1, ‡ † || † S.M. Ghajar , H. McKenzie , J. Fike , B. McIntosh , B.F. Tracy School of Plant and Environmental Sciences, Virginia Polytechnic Institute & State University, Blacksburg, VA 24061; Department of Large Animal Clinical Sciences, Virginia Polytechnic Institute & State University, || Blacksburg, VA 24061; and Mars Equestrian™, McLean, VA 22101 ABSTRACT:  Introduced cool-season grasses are Biomarkers for hepatotoxicity remained within dominant in Virginia’s grasslands, but their high acceptable ranges for all treatments. Apparent dry digestible energy and nonstructural carbohy- matter digestibility (DMD) did not differ among drate (NSC) levels pose a risk for horses prone to hays, ranging from 39% to 43%. NSC levels ranged obesity and laminitis. Native warm-season grasses from 4.4% to 5.4%, below maximum recom- (NWSGs) have lower digestible energy and NSC mended concentrations for horses susceptible to levels that may be more suitable for horses sus- laminitis. For the second objective, a grazing trial ceptible to laminitis. Although NWSGs have de- was conducted comparing IG, BB, and eastern sirable characteristics, they are novel forages for gamagrass (EG) (Tripsacum dactyloides) yields, horses. Little is known about NWSG intake or po- forage losses, changes in vegetative composition, tential toxicity to horses or how grazing by horses and effects on equine bodyweight. Nine, 0.1-ha may affect NWSG swards. The overall objectives plots were seeded with one of the three native of this research were to 1) assess voluntary intake, grass treatments, and each plot was grazed by one toxicological response, and apparent digestibility Thoroughbred gelding in two grazing bouts, one of NWSG hays fed to horses; and 2)  evaluate in July and another in September 2019. IG had the characteristics of three NWSG species under the greatest available forage, at 4,340 kg/ha, com- equine grazing. For the first objective, a hay feed- pared with 3,590 kg/ha from BB (P < 0.0001). EG ing trial using indiangrass (IG) (Sorghastrum plots established poorly, and had only 650  kg/ha nutans) and big bluestem (BB) (Andropogon gerar- available forage during the experiment. Grazing dii) was conducted with nine Thoroughbred geld- reduced standing cover of native grasses in IG ings in a replicated 3  × 3 Latin square design. and BB treatments by about 30%. Horses lost 0.5– Voluntary dry matter intake of IG and BB hays by 1.5  kg BW/d on all treatments. Findings suggest horses were 1.3% and 1.1% of BW/d, lower than IG and BB merit further consideration as forages orchardgrass (Dactylis glomerata), an introduced for horses susceptible to obesity and pasture-asso- cool-season grass, at 1.7% of BW/d (P = 0.0020). ciated laminitis. Key words: equine, forage, native grasses, pasture © The Author(s) 2020. Published by Oxford University Press on behalf of the American Society of Animal Science. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribu- tion, and reproduction in any medium, provided the original work is properly cited. Transl. Anim. Sci. 2021.5:1-13 doi: 10.1093/tas/txaa224 INTRODUCTION 1 More than half of horses in Virginia are over- Corresponding author: sghajar@vt.edu weight or obese (Thatcher et  al., 2008). Obesity Received September 18, 2020. puts a horse at risk for serious health issues such Accepted November 30, 2020. 1 Ghajar et al. as insulin resistance and laminitis (Geor, 2009). game species (Tompkins et  al., 2010). Common Laminitis is a disease characterized by an inflam- species native to much of the eastern United States matory response damaging the lamellar layer of include big bluestem (BB) (Andropogon gerar- the horse’s hooves, allowing the coffin bone to ro- dii), indiangrass (IG) (Sorghastrum nutans), and tate (Geor, 2010). It is painful, costly to treat, and switchgrass (Panicum virgatum). Though research may necessitate humane euthanasia of the horse. is lacking on the use of BB or IG as equine for- Although the mechanisms that precipitate ages, one study examining NSC in forages under laminitis are not fully understood, this inflamma- different light conditions found that NSC levels in tion in the hoof usually follows the consumption of BB never exceeded 12%, suggesting some NWSG large quantities of readily-fermented carbohydrates may have ideal NSC levels for horses susceptible to (Geor, 2010). When a horse with a predisposition to laminitis (Kephart and Buxton, 1996). obesity or insulin resistance consumes a large quan- Though the lower NSC concentration in tity of nonstructural carbohydrates (NSC)—carbo- NWSGs may be optimal for horses prone to meta- hydrates not forming the walls and membranes bolic disorders, questions remain regarding the of the plant, but rather starch and water-soluble safety of these grasses for horses. A  number of carbohydrates—it may prove more than the intes- studies have determined that the Panicum genus tine can readily digest, and the bolus of carbohy- causes hepatotoxicity in horses. Several species of drates can end up being fermented rapidly in the grass in that genus are common and widespread hindgut (McIntosh, 2006). This rapid fermentation throughout the eastern United States, including lowers hindgut pH, altering the microbiome, and switchgrass, fall panicum (Panicum dichotomi- causing the release of endotoxins into the blood- orum fl ), and many species commonly referred to as stream (Geor, 2010). For horses susceptible to “panic grasses” due to their large panicle seedheads. obesity and laminitis, it is recommended not to ex- After 14 horses at a boarding facility in Virginia fell ceed 10–12% NSC concentration in the diet (Geor, severely ill in 2004, the cause was determined to be 2010). Cool-season grass pastures can easily exceed their hay, which was largely comprised of fall pan- this limit throughout much of the year (McIntosh, icum (Johnson et al., 2006). A subsequent feeding 2006), and even grass hays commonly fed to horses trial of fall panicum to two research horses for 12 such as orchardgrass (Dactylis glomerata) can have d resulted in highly elevated biomarkers for hep- NSC concentrations above 12% (Martinson et al., atotoxicity in blood samples taken from the horses, 2012). As a result of the high NSC levels in many and histology revealed bile duct hyperplasia and common forage species, pasture-associated lamin- hepatocyte swelling (Johnson et al., 2006). itis may account for nearly half of all cases in the Other research found that switchgrass ingestion United States (Geor, 2009). by horses demonstrated hepatotoxicity as well, and Warm-season grasses, which are grasses with determined diosgenin, a steroidal sapogenin, to be higher optimal growth temperatures and often a the primary toxin (Lee et al., 2001). While the sap- C4 photosynthetic pathway, generally have lower ogenin is believed to be metabolized in a form that levels of NSC than cool-season forages, and no crystallizes in the liver of sheep affected by Panicum fructan, a simple carbohydrate thought to play a toxicity, the mechanism by which the chemical role in laminitis (Kagan et al., 2011). Staniar et al. damages the equine liver is not established, though (2010) found teff hay (Eragrostis tef) presented ad- it is thought to involve apoptosis rather than crys- equate nutrition and low carbohydrate levels for tallization (Johnson et al., 2006). horses. Kagan et al (2011) measured carbohydrate Aside from the documented toxicity of the levels of bermudagrass (Cynodon dactylon) at dif- Panicum genus for horses, we could find no stud- ferent stages of maturity, and at different times of ies linking NWSG to toxicity in horses. However, day, and found it to have suitably low carbohydrate one of the common native grasses mentioned levels to be considered safe for horses prone to earlier, IG, is related to the Sorghum genus. laminitis. While these studies on introduced forage Sorghum species, whether annual or perennial, species have provided alternatives to high-carbo- have been linked to cystitis ataxia in horses, a hydrate cool-season forages for horses, no stud- condition wherein hydrocyanic acid causes de- ies have examined the use of native warm-season generation of the nervous system (Morgan et al., grasses (NWSG) in equine forage systems. 1990). This results in loss of bladder control and NWSG were once abundant in the Piedmont of hind leg coordination, and is irreversible and Virginia as a result of Native American use of pre- often fatal (Adams et  al., 1969). The primary scribed fire to promote savannahs with abundant chemical that is hydrolyzed into hydrocyanic acid Translate basic science to industry innovation Native warm-season grasses as equine forage is dhurrin, which is also found in IG seedlings, medium, and low relative BW groups, and then one though in IG the concentration of dhurrin de- horse from each group was randomly assigned to clines as the plant matures (Gorz et  al., 1979). each of the three squares such that each square had Nevertheless, no cases of toxicity from IG have a similar mean BW. been documented in horses, and even-aged sor- Three hay types were used in the feeding trials. ghum hay has been found to be safe for horses, as NWSG hays were IG and BB donated in July 2018 the toxic compound does not survive prolonged by Ernst Conservation Seeds (Meadville, PA) from storage (Adams et al., 1969). pure stands normally used for seed production. In addition to the questions of safety and nu- A  common cool-season grass hay (orchardgrass tritional value of NWSGs for horses, we found cv ‘HLR’, Barenbrug) was used for comparison. no prior research on these grasses examining their Orchardgrass hay was produced on site in May response to grazing. As equine grazing can have 2018 at Virginia Tech’s Middleburg Agricultural different impacts on a pasture than cattle grazing Research & Extension Center. (Bott et al., 2013), it is necessary to conduct grazing The study consisted of a 10-d acclimation phase trials with horses rather than drawing conclusions and a 4-d digestibility trial. During the first 8 d of from research on cattle grazing and NWSGs to the acclimation phase (day 1 to day 8), horses were understand the potential impacts of horse grazing housed by a treatment group in three adjacent dry on NWSG swards. lots with access to run-in sheds for shelter, ad lib- This research evaluated IG and BB for use as itum white salt and water, and were fed their treat- equine forages. Specific objectives were to: ment hay ad libitum from round bales. On day 9, horses were moved to individual stalls (3.5 m × 3.5 1. Determine if feeding IG and BB hay causes hep- m). for the remainder of the period where they were atic insult to horses. fed their treatment hay and again had ad libitum ac- 2. Determine the voluntary intake of horses fed cess to clean water and white salt. On day 10, horses BB and IG hay as compared to a common hay were fitted with fecal collection harnesses (Equisan species, orchardgrass. Ltd, Australia) to ensure comfort and familiarity 3. Compare the nutritive value and apparent di- with the harness. The harness was also designed gestibility of these NWSG species to a common to collect urine; however urinary analyses were not cool-season grass species, orchardgrass, when conducted in this trial. From day 9 onwards, horses fed as hay. had group access to a dry lot for an hour per day 4. Compare forage productivity, nutritive value, for exercise and social time. The digestibility trial and short-term trampling effects in eastern began on day 11 and concluded on day 14. On day gamagrass (EG), BB, and IG swards grazed by 15, horses were turned out together into a mixed horses, and whether bodyweight gain differed cool-season pasture for a 2-wk washout between among forage types. experimental periods. Horses were weighed on a livestock platform scale on day 1, day 8, and day 15 of each period of the experiment. MATERIALS AND METHODS Blood samples were collected three times per Two experiments were conducted at the period per horse—once on day 1 (baseline), day Virginia Tech Middleburg Agricultural Research 8, and day 15 between 0700 and 0900 hours each & Extension Center in Middleburg, Virginia to day. Horses were not fasted prior to sampling. evaluate NWSG for use as hay and pasture spe- Samples were collected via jugular venipuncture cies for horses. Protocols for both experiments were into 10-mL vacutainer tubes, placed on ice, and approved by Virginia Tech’s Institutional Animal driven directly to Virginia Tech’s Marion DuPont Care and Use Committee. Scott Equine Medical Center in Leesburg, VA, for analysis. Plasma was analyzed for nine different markers of toxicity. The markers were selected Experiment I—Hay Feeding Trial based on past studies of Panicum toxicity and other A hay feeding trial was conducted in November common pasture-associated toxicities that caused and December of 2018 and January of 2019. A rep- elevated marker profiles in horse serum (Table  1) licated Latin square design with three treatments, (Curran et al., 1996; Johnson et al., 2006). Results three periods, and nine horses was used. The nine were forwarded to a veterinarian the same day to horses were Thoroughbred geldings (9–13 yr) and confirm that they were within acceptable ranges. 569  ± 38  kg BW. Horses were divided into high, Horses were also monitored daily for any changes Translate basic science to industry innovation Ghajar et al. Table 1.  Biomarkers assessed to detect potential block design with three replicates of three treat- hepatic insult to horses fed novel NWSG hays in ments: IG, BB, or gamagrass (GG). Plots were the study established on Fauquier-Eubanks and Purcellville- Tankerville soil series (fine, mixed, active mesic Albumin Typic Hapludults). Slopes at the study site ranged Alkaline phosphatase (ALP) from 7% to 15%. Aspartate aminotransferase (AST) Prior to the study, the site was managed as Bile acid Direct bilirubin cool-season pasture with tall fescue (Schedonorus Gamma glutamyl transferase (GGT) arundinaceus) as the dominant species. The site Sorbitol dehydrogenase (SDH) was sprayed with 4.7  L/ha glyphosate the third Total bilirubin week of April 2018. Two weeks later, prescribed Triglycerides fire was used to prepare a clean seedbed and en- sure fescue mortality. On June 1, glyphosate was in behavior that might have indicated an adverse re- applied again at 2.3  L/ha to kill a flush of weedy sponse to the novel hays being tested. species following the fire. The same week, IG and A digestibility trial was conducted in the last BB plots were seeded using a Truax FLEX-II no-till four days of each period. Bedding was removed drill at a depth of 6  mm and subsequently rolled from stalls and fecal collection harnesses were put with a water-filled roller to ensure adequate seed on each horse. Each horse was offered its treatment to soil contact. Gamagrass seeds were soaked in a hay at 2.5% BW dry matter based on the BW meas- 15% hydrogen peroxide (H O ) solution for 18 h to 2 2 ured on day 8 of the experimental period. Hay DM break seed dormancy (Klein et al., 2008), then were concentration was determined by taking approxi- drained, rinsed, and transferred to a Great Plains mately 20 cored samples per round bale being fed, 706NT seed drill. After a 2-d delay due to inclement drying the samples at 135 ºC for 2 h, and dividing the weather, the GG was seeded at a depth of 2 cm. dried weight by the original weight. Hay was split IG and BB plots were sprayed with imazapic into two daily feedings at 0800 and 2000 hours, and herbicide at a rate of 0.15 L/ha the week after seed- fed using hay nets. Orts were collected and weighed ing. GG plots were not sprayed with imazapic, as twice daily at 0700 and 1900 hours. Fecal collec- imazapic causes stunting and mortality in GG. In tion harnesses were emptied at least three times the second week of July, GrazonNextHL (active in- daily to ensure they did not become uncomfortable gredients: 2,4-D and aminopyralid) was applied to for the horse, at 0600, 1400, 2000, and if needed, all plots at a rate of 2.3 L/ha to control broadleaf 0000 hours. Feces were collected in tubes lined with weeds (BW). In mid-August the same year, another plastic bags which were kept shut to preserve mois- application of 0.3  L/ha imazapic was conducted ture, and total fecal output was weighed for each in IG and BB plots to control crabgrass (Digitaria 24-h period starting at 2000 hours the day prior to sanguinalis). No further herbicide applications were 2000 hours on the day of weighing. Two, 1-kg sub- applied to the plots. samples were collected after weighing and compos- The grazing trial began in July 2019. Horses iting each horse’s fecal output each day, and one of were first turned out into their plots on July 10, and these samples was dried at 55 °C until it reached a removed on July 24 when the majority of plots were constant weight to determine dry matter. The re- reduced to 20- to 30-cm stubble height. Nine (n = 9) maining sample was placed in a −20 °C freezer for Thoroughbred geldings aged 10–14 years (median: storage, and later thawed at room temperature for a 13)  and weighing an average of 550  ± 31  kg were day, then dried at 55 °C and sent to Equi-Analytical grouped by weight to reduce variation in grazing (Ithaca, NY) for chemical analyses. Grab samples and trampling pressure among treatments. Each of approximately 50  g were collected from each group was assigned to one treatment per grazing hay bale daily as they were fed to horses, and these bout, and then reassigned randomly to another samples were composited and submitted to Equi- treatment on the following bout. Each plot was Analytical for chemical analyses as well. grazed by one horse for the duration of the grazing bouts. Horses were each provided a shade structure for shelter from the weather, ad libitum access to Experiment II—Equine Grazing Trial water and white salt, and daily applications of fly The grazing trial was conducted from July repellant. through September 2019. Nine, 0.1-ha plots were Horses were then turned out into fescue-dom- established in May 2018 in a randomized complete inated pastures to allow NWSG plots to regrow Translate basic science to industry innovation Native warm-season grasses as equine forage until all plots had reached at least 46  cm height. 1 (baseline) and day 15. Changes in biomarkers of On September 4, horses were placed back on the toxicity were analyzed using mixed models with NWSG plots and removed on September 12, once treatment, period, and treatment × period as fixed some plots were estimated to have inadequate effects, and horse as a random effect. Mixed models forage to meet daily dry matter requirements of the with treatment and period as fixed effects and horse horses grazing them. as a random effect were used to analyze differences Plots were sampled for biomass and percent in apparent digestibility. Intake, BW changes, ap- cover using haphazardly-placed 20-cm × 50-cm parent digestibility, and biomarkers of toxicity are quadrats at the start of each grazing bout and presented as least squares means. shortly after the removal of horses. At the start Experiment II.  Biomass data were analyzed of the first grazing bout, nutritive samples were both separately by date, and in total using repeat- also collected from plots by compositing biomass ed-measures ANOVA. Available forage and forage subsamples and sending them to a commercial la- removed variables were analyzed with repeated boratory for chemical analyses (Equi-Analytical, measures ANOVA to determine differences among Ithaca, NY). Biomass samples (n = 5 per plot) were treatments. hand-clipped at ground level and separated into Percent cover was analyzed separately by date standing NWSG and weedy species, then dried in with ANOVA and pairwise comparisons made with a forced-air oven at 55 °C. NWSG biomass at the Tukey’s HSD. Cover variables analyzed include beginning of each grazing bout was categorized as NWSG standing cover, NWSG trampled cover, “available forage.” Trampled biomass was also har- GW, and BW. vested from within each quadrat; however, because Changes in equine weight for the two grazing of the difficulty in clipping fallen NWSG, these bouts were calculated on a per day basis to account data were discarded as unreliable. The difference for the differing lengths of each bout in the ana- between standing NWSG biomass at the start and lysis. Weight changes were compared with ANOVA. finish of each grazing bout was classified as “forage For all analyses, differences were considered signifi- removed” from the plot. Forage removal included cant when P < 0.05. both forage consumed by horses and forage that was trampled into the ground and thus “removed” RESULTS from the available forage pool. Percent cover was assessed visually as standing NWSG, grassy weeds Experiment I (GW), and BW (n = 10 quadrats per plot). Weeds Forage nutrient value. The three hays used in the were defined as any species not seeded in the plots. grazing trial were evaluated for 18 variables to es- In assessments made after the end of grazing bouts, timate nutritive value for horses. Digestible energy the percent cover of newly trampled NWSG was was greater in the IG hay at 1.9 Mcal/kg compared also assessed. to 1.8 for BB (P = 0.0345) (Table 2). However, NDF Horses were weighed on a digital livestock scale was also over 70 in both NWSG hays, while it was immediately before being turned out into the plots, 66.3 in the OG hay (P < 0.0001). Several nutrients and immediately upon removal. were also greater in the OG hay than in the NWSG hays, including Ca (P < 0.0001), P (P = 0.0009), Mg Statistical Analysis (P < 0.0001), K (P < 0.0001), and Fe (P = 0.0069). Experiment I.  Nutrient compositions for the Starch was greatest in the IG hay (P = 0.0049). three species of hay fed were compared by one-way Toxicological markers.  Biomarkers for toxicity ANOVA. If a difference was found, Tukey’s HSD stayed within parameters deemed acceptable by our was used for pairwise comparisons. Apparent di- veterinarian based on accepted normal ranges and gestibility was calculated by dividing the difference prior toxicological research on our research farm between average daily total nutrient intake and (Mercer et  al., 2020), with one exception. For one average daily nutrient excretion and dividing by horse, ALP, AST, bile acid, GGT, SDH, and total average daily total nutrient intake. Voluntary dry bilirubin were elevated past acceptable ranges on the matter intake (DMI) was compared using a mixed final sampling of period 1. The horse was immediately model with treatment, period, and treatment × removed from the study and was sampled frequently period as fixed effects, and horse as a random effect. to monitor biomarkers. The horse did not exhibit Biomarkers of toxicity were calculated as the overall clinical symptoms at any time. As no other horses change between values from samples taken on day were affected, and the horse’s biomarkers remained Translate basic science to industry innovation Ghajar et al. Table 2.  Mean nutrient composition values (±SE) for big bluestem, indiangrass, and orchardgrass hay treatments Treatment Variable BB IG OG P-value Dry matter, % 94.5 ± 0.5 95.0 ± 0.4 93.4 ± 4.8 0.6613 b a ab DE, Mcal/kg 1.8 ± 0.1 1.9 ± 0.0 1.8 ± 0.1 0.0345 b b b a CP , % 8.8 ± 1.7 7.4 ± 0.9 13.9 ± 1.0 <0.0001 c a b b ADF , % 49.4 ± 3.8 44.6 ± 2.7 41.3 ± 2.3 0.0013 d a a b NDF , % 76.1 ± 2.5 74.6 ± 2.2 66.3 ± 2.0 <0.0001 b a a Ca, % 0.2 ± 0.1 0.3 ± 0.0 0.4 ± 0.1 <0.0001 b b a P, % 0.2 ± 0.0 0.2 ± 0.0 0.3 ± 0.0 0.0009 b c a Mg, % 0.1 ± 0.0 0.1 ± 0.0 0.2 ± 0.0 <0.0001 b c a K, % 2.1 ± 0.3 1.6 ± 0.3 2.9 ± 0.1 <0.0001 b b a Fe, PPM 80.7 ± 27.3 97.6 ± 10.4 197.3 ± 90.7 0.0069 Zn, PPM 21.5 ± 4.5 19.2 ± 1.3 16.8 ± 3.1 0.0854 Cu, PPM 5.8 ± 1.2 6.0 ± 0.7 7.2 ± 0.8 0.0503 Mn, PPM 66.0 ± 21.1 84.4 ± 24.3 75.2 ± 16.5 0.3631 b a a Starch, % 0.4 ± 0.1 0.7 ± 0.1 0.6 ± 0.2 0.0049 WSC , % 4.0 ± 1.3 4.7 ± 0.7 4.4 ± 1.5 0.6642 ESC , % 3.2 ± 1.4 3.2 ± 0.9 2.2 ± 1.5 0.3223 NSC , % 4.4 ± 1.3 5.4 ± 0.8 5.0 ± 1.4 0.4275 Values with differing letters are significantly different at P < 0.05. Treatment: BB = big bluestem hay; IG = indiangrass hay; OG = orchardgrass hay. Crude protein. Acid detergent fiber. Neutral detergent fiber. Water-soluble carbohydrates. Ethanol-soluble carbohydrates. Nonstructural carbohydrates. elevated after removal from the hay and turnout on was more digestible for OG than in BB or IG cool-season mixed pastures, our veterinarian deter- (P < 0.0001). Starch was more digestible in OG and mined the elevated biomarkers most likely indicated IG than BB (P  =  0.0096). The IG treatment had either a response to an unknown insult or a recur- higher apparent digestibility of Ca (P  =  0.0002), rence of a past medical issue. The horse was on the Mn (P = 0.0065), and Cu (P = 0.0033). BB treatment; no other horse had a similar response to consuming the BB hay. Consequently, we omitted Experiment II that horse from toxicological analyses. In periods 2 and 3, an alternative horse was used. Forage characteristics.  Available forage differed Most biomarkers did not differ among treat- at the beginning of the grazing trial. BB and IG ments (Table 3). Of those that did, GGT increased standing biomass did not differ (about 3000 kg/ha), the most in the IG treatment (P = 0.0166) and SDH but the GG treatment had much lower available decreased the most (P = 0.0377). Triglycerides were forage (410 kg/ha) (P < 0.0001). Weedy species bio- lower in BB and IG treatments than in the OG mass also differed, with BB having the lowest weedy treatment (P = 0.0062). biomass at 130  kg/ha and GG having the most, Intake.  Voluntary DMI was greater for OG at 506  kg/ha (P  =  0.0002). At the end of the July than either of the NWSG hays, both in terms grazing bout, available forage did not differ among of mass-consumed (P  =  0.0042) or percent BW treatments, ranging from 386 kg/ha in the GG plots (P  =  0.002) (Table  4). Horses lost weight on both to 565  kg/ha in the BB plots. Weedy biomass dif- NWSG species tested, but gained weight on OG fered again, with GG having greater weedy biomass (P = 0.0357). For all measures of intake or change at 580 kg/ha than both the BB and IG plots, at 62 in BW, there was a treatment effect, but no period ef- and 295 kg/ha, respectively (P = 0.0003). fect or period × treatment interaction was detected. At the beginning of the September grazing Digestibility. Apparent DM digestibility did not bout, IG pastures had the most available forage differ among treatments (Table  5). Crude protein (1,450 kg/ha) while GG again had the least (250 kg/ Translate basic science to industry innovation Native warm-season grasses as equine forage Table 3. Biomarker change from baseline values in horses fed big bluestem, indiangrass, and orchardgrass hays Treatment Biomarker BB IG OG P-value Albumin, g/dL 0.0 ± 0.4 0.1 ± 0.2 −0.5 ± 0.5 0.0667 ALP , U/L −0.6 ± 23.8 48.0 ± 54.4 −8.0 ± 39.9 0.1061 AST , U/L −101.1 ± 42.8 −54.5 ± 26.2 −57.6 ± 34.7 0.3466 Bile acid, µmol/L −0.3 ± 1.2 −0.4 ± 1.2 −2.5 ± 2.1 0.1245 Direct bilirubin, mg/dL 0.0 ± 0.1 −0.1 ± 0.1 0.0 ± 0.1 0.6948 d ab a b GGT , U/L 1.0 ± 3.1 5.7 ± 4.2 0.0 ± 1.9 0.0166 e ab a b SDH , U/L −5.1 ± 2.5 −3.5 ± 6.6 −12.9 ± 8.5 0.0377 Total bilirubin, mg/dL 0.7 ± 0.4 0.7 ± 0.4 0.5 ± 0.3 0.7592 b b a Triglycerides, mg/dL −3.5 ± 10.2 −3.0 ± 7.7 16.0 ± 11.9 0.0062 Treatment means with different letters are statistically different at P < 0.05. Data presented are least squares means ± SE. Treatment: BB = big bluestem hay; IG = indiangrass hay; OG = orchardgrass hay. Alkaline phosphatase. Aspartate aminotransferase. Gamma glutamyl transferase. Sorbitol dehydrogenase. Table 4. Mean dry matter intake and change in bodyweight by treatment Treatment Variable BB IG OG P-value b b a DMI, kg/d 6.6 ± 0.8 7.0 ± 0.9 8.9 ± 1.8 0.0042 b b a DMI, % of BW 1.1 ± 0.1 1.3 ± 0.1 1.7 ± 0.3 0.0020 ab b a Change in BW −16.3 ± 19.0 −21.2 ± 26.2 3.8 ± 25.6 0.0357 Treatment means with different letters are statistically different at P < 0.05. Data presented are least squares means ± SE. Treatment: BB = big bluestem hay; IG = indiangrass hay; OG = orchardgrass hay. Table 5. Apparent digestibility of the three treatment hays fed Treatment Digestibility, % BB IG OG P-value Dry matter 38.8 ± 10.1 40.6 ± 8.4 43.2 ± 5.3 0.2636 b b b a CP 18.3 ± 13.5 30.1 ± 14.4 52.4 ± 3.9 <0.0001 ADF 51.3 ± 9.2 46.7 ± 7.7 48.5 ± 5.6 0.0979 NDF 47.6 ± 9.0 45.7 ± 8.0 46.5 ± 5.5 0.6415 b a b Ca −24.0 ± 26.1 15.6 ± 17.4 −13.7 ± 22.7 0.0002 b a ab P −55.4 ± 21.5 −30.3 ± 13.1 −47.6 ± 30.4 0.0076 Mg −20.6 ± 21.7 −23.2 ± 22.9 −13.0 ± 9.6 0.5823 K 58.7 ± 12.7 57.7 ± 11.7 57.3 ± 9.0 0.8599 Fe −213.2 ± 145.2 −144.6 ± 107.0 −89.3 ± 87.7 0.0561 Zn −34.0 ± 27.9 −29.3 ± 20.6 −45.1 ± 28.3 0.3722 b a b Cu −14.6 ± 18.0 13.3 ± 14.6 −8.6 ± 10.0 0.0033 b a b Mn −77.0 ± 50.8 −18.8 ± 23.2 −60.0 ± 22.5 0.0065 b a a Starch −27.0 ± 86.6 42.7 ± 19.8 39.5 ± 17.9 0.0096 WSC 55.7 ± 17.9 66.5 ± 7.0 59.4 ± 5.5 0.1806 ESC 66.1 ± 24.9 66.0 ± 19.1 81.3 ± 9.0 0.0588 Treatment means with different letters are statistically different at P < 0.05. Data presented are least squares means ± SE. Treatment: BB = big bluestem hay; IG = indiangrass hay; OG = orchardgrass hay. Crude protein. Acid detergent fiber. Neutral detergent fiber. Water-soluble carbohydrates. Ethanol-soluble carbohydrates. Translate basic science to industry innovation Ghajar et al. ha) (P  =  0.0354). BB pasture yields were inter- on GG, which had net negative removal over the mediate (620 kg/ha). The GG treatment again had course of the grazing trial, indicating an increase in the greater weedy species biomass, increasing to biomass of about 260 kg/ha (P = 0.0070). 780  kg/ha compared with 330  kg/ha and 170  kg/ Percent cover of IG and BB did not differ ha in the IG and BB treatments, respectively in July or September at the start of each grazing (P  =  0.0002). After the September grazing bout, bout (Figure  1). BW had higher percent cover in there were no differences in available forage among GG plots than IG or BB in July (P < 0.0001) and treatments. September (P  <  0.0001). GW had higher percent Total seasonal forage availability differed cover in GG than in IG or BB in July (P < 0.0001), among treatments, with the highest mean total and were higher in GG than in BB in September available forage in the IG treatment, at 4,340  kg/ (P = 0.0160). ha, and the lowest in the GG at 650 kg/ha (Table 6) After being grazed in July, over half the cover (P = 0.0140). Overall forage removed also differed of BB and IG plots was comprised of tram- between treatments, with the highest mean removal pled NWSG biomass, while no GG was trampled on IG plots at 3,880 kg/ha, and the lowest removal (P < 0.0001) (Table 7). After the September grazing Table 6. Mean available and removed NWSG forage and weed biomass (kg/ha) of treatments in the grazing trial NWSG Weeds Treatment Available Removed Biomass Removed July BB 2970 2400 130 70 GG 400 20 900 310 IG 2900 2460 510 210 September BB 620 440 170 80 GG 250 −280 780 240 IG 1460 1410 330 180 Total BB 3590 2890 300 140 GG 650 −260 1680 550 IG 4340 3880 830 390 P-value 0.014 0.007 0.0025 0.188 Removed forage was the difference between standing NWSG biomass at the start and finish of each grazing bout. A negative value indicates a gain rather than loss in biomass. P values reported are for repeated measures ANOVA comparing treatments across the course of the experiment. Treatment: BB = big bluestem; IG = indiangrass; OG = orchardgrass. Figure 1. Percent cover of native warm-season grasses (NWSG), grassy weeds (GW), and broadleaf weeds (BW) in July and September at the start of each grazing bout for the big bluestem (BB), gamagrass (GG), and indiangrass (IG) plots. Means with different letters are significantly different (P < 0.05). Translate basic science to industry innovation Native warm-season grasses as equine forage bout, about a third of the cover in BB plots was exceed acceptable limits after 2  wk. Voluntary in- trampled NWSG, and IG plots once again were take of these forages was lower than orchardgrass comprised of over 50% trampled forage from the intake in our study and lower than reported values most recent grazing bout. Again, GG plots had no for other species. However, this may be ideal for measurable trampling of native grass (P < 0.0001). horse owners struggling to find optimal forages for Biomass samples of NWSG from each plot horses prone to obesity and laminitis. Apparent di- were subsampled and composited for nutritional gestibility of some nutrients was low or negative for analysis at the beginning of the first grazing trial. the NWSG hays and orchardgrass on our study, Due to the sparseness of GG, nutritional analysis indicating a need to supplement these forages with could not be carried out on the harvested samples a ration balancer to ensure adequate intake of vita- of two of the three plots. Consequently, only BB mins, minerals, and protein. and IG were compared statistically. They did not differ in CP, ADF, NDF, DE, or NSC (Table 8). Toxicological Response Animal weight. Horses lost weight on all treat- The two NWSG hays tested did not cause bio- ments during the grazing bouts (Figure 2). Weight markers of toxicity to exceed levels deemed safe by changes in horses did not differ among treatments; the veterinarian monitoring horses on this study however, there was a trend for horses on the IG based on prior toxicological research conducted treatment to lose more weight than those on the BB with Thoroughbred geldings on this site (Mercer or GG treatments, at 1.5  kg/day lost on IG plots et al., 2020). As discussed above, one horse on the and about 0.5  kg/day in the other two treatments BB treatment had elevated biomarkers, but the lack (P = 0.0977). of any similar response in the other horses fed the hay as well as the continued elevated biomarkers DISCUSSION when that horse consumed cool-season pasture during a washout and monitoring period suggest a Experiment I.  The objectives of experiment I  cause other than diet. Additionally, we can find no were to determine if BB and IG hays induce a toxic case reports in the literature of BB hay causing a response in horses after 2  wk of feeding and to toxic response in horses. evaluate the potential use of these species as a hay Among biomarkers measured, only GGT in- source for equines. Biomarkers of toxicity did not creased in an NWSG hay (IG) relative to the OG treatment. GGT is an indicator of hepatic func- Table 7.  Mean percent cover of trampled NWSG tion, and changes were slight. To determine if this by treatment following each grazing bout response indicates the potential for hepatic insult Grazing bout caused by IG hay, a longer feeding trial may be ne- Species July September cessary for future research. Additionally, our results a a BB 67.7 ± 32.0 36.3 ± 34.1 cannot rule out the possibility of cystitis-ataxia for b b GG 0.0 ± 0.0 0.0 ± 0.0 horses consuming a diet of IG long-term, as cysti- a a IG 52.5 ± 34.4 54.5 ± 38.7 tis-ataxia does not affect the liver and is diagnosed P-value <0.0001 <0.0001 clinically rather than by blood samples (Adams et  al, 1969). None of our horses exhibited clinical Means not connected by the same letter are significantly different at P < 0.05. signs of cystitis-ataxia after 2 wk of consuming IG; Species: BB = big bluestem; IG = indiangrass; OG = orchardgrass. however, longer feeding trials should be conducted Table 8.  Forage nutritive value indices for big bluestem (BB), eastern gamagrass (GG), and indiangrass (IG) treatments in the grazing trial. No SD is reported for gamagrass because only one sample had suffi- cient biomass for chemical analysis Species CP, % ADF, % NDF, % DE, Mcal/kg NSC (%) BB 7.0 ± 0.3 45.5 ± 2.6 74.5 ± 0.5 1.5 ± 0.8 9.9 ± 1.3 GG 6.2 41.7 74.7 1.1 12.4 IG 6.0 ± 0.8 45.0 ± 0.9 74.7 ± 1.5 1.1 ± 0.6 8.3 ± 1.4 P-value 0.1419 0.7495 0.8117 0.4854 0.2090 Only BB and IG were compared statistically. Species: BB = big bluestem; IG = indiangrass; OG = orchardgrass. Translate basic science to industry innovation Ghajar et al. harvest. Nutritive values from a mixed hayfield of IG and BB from 2010 to 2012 in Tennessee (Keyser et al., 2012) averaged 66.8% NDF, 40.2% ADF, and 9.3% CP—substantially lower fiber concentrations and moderately higher CP than the NWSG hay in our study. Voluntary DMI of the NWSG hays was lower than values reported for mature warm-season grasses in past studies, such as 2.1% for “Coastal” bermudagrass (Cynodon dactylon) or 2.3% for Caucasian bluestem (Bothriochloa bladhii) (Crozier Figure 2. Average changes in equine bodyweight on the big bluestem et al., 1997; LaCasha et al., 1999). Voluntary DMI (BB), gamagrass (GG), and indiangrass (IG) treatments during the for horses consuming OG was within the normal grazing experiment expressed as kilograms per day with standard error. ranges of 1.5–3.1% BW described in the National Research Council’s guidelines for equine nutri- to determine if IG poses a risk for horses in tion, with IG falling slightly below the normal that regard. range at 1.3% and BB well below at 1.1% (National The declines in AST and SDH for all treatments Research Council, 2007). However, for obese horses during the feeding trial were likely a result of the needing reduced digestible energy intake, Virginia horses switching from a species-rich pasture dur- Cooperative Extension recommends reducing hay ing the washout periods to monospecific hay. On intake to 1–1.5% of the target BW while maintain- pasture, horses had exposure to greater varieties of ing constant forage availability to minimize risk of forbs and grasses, some of which can cause hepatic gastric ulcers (Porr and Crandell, 2008). insult. When fed a diet exclusively consisting of one Differences in NDF paralleled differences in in- grass species, this exposure is eliminated. A similar take among treatments, with OG having the lowest decline was measured in a study of acetaminophen NDF and highest intake, and the NWSG having pharmacokinetics in Thoroughbred geldings at the high NDF values and lower intake by horses. This same facility when horses were removed from pas- aligns with past research demonstrating the value ture and fed only hay (Mercer et al., 2020). of NDF as a predictor of intake, with lower NDF Triglycerides increased in the OG treatment but levels predicting higher voluntary intake (LaCasha decreased slightly in the NWSG treatments, which et al., 1999). can be explained by the loss of BW on the NWSG treatments and increase in BW on OG. Moderate weight loss in humans results in lower serum tri- Digestibility glyceride levels (Andersen et  al., 1995). Similarly, Dry matter digestibility (DMD) measured in Suagee et al. (2013) reported a positive relationship our study ranged from 38.8% to 43.2%, similar to between equine body condition and plasma trigly- values reported for lower-quality hay in previous cerides, as well as insulin concentration and plasma research, such as 43% in coastal bermudagrass triglycerides. These results align with our measure (Aiken et  al., 1989), 44% in Caucasian bluestem of limited increases in triglycerides for horses ex- (Bothriochloa bladhii) (Crozier et al., 1997), 38.5% periencing a minor increase in BW. in reed canarygrass (Phalaris arundinacea), and 42.1% in crested wheatgrass (Agropyron cristatum) Voluntary Dry Matter Intake and Nutritive Values (Cymbaluk, 1990). Apparent DMD of the hay treatments in our study was lower than those re- BB matures earlier than IG, and as the fields ported for another warm-season grass, teff, at three for both treatments were cut the same week, BB different stages of maturity, which ranged from hay had reached a more mature stage with a higher 51.5% to 60.6% when fed to horses (Staniar et al., proportion of reproductive tillers, while IG was 2010). The DMD of NWSG hays in our study were still vegetative. Orchardgrass hay was also in a also lower than those reported for alfalfa, popular vegetative state when cut in late spring. As hay in- hay for horses, which ranges from 58% to 64% creases in maturity, nutritional quality decreases (Crozier et  al., 1997; Sturgeon et  al., 2000). The and voluntary intake by horses declines (Staniar NWSG hays, while not highly digestible, are ad- et  al., 2010). The NWSG hays in our study were equate as roughage for equines susceptible to meta- cut later than is optimal, as prolonged rain-delayed bolic disorders. opportunities for cutting and curing at the site of Translate basic science to industry innovation Native warm-season grasses as equine forage Integrating NWSG Hays into the Equine Diet survival of the stand. Gamagrass plots were not sufficiently established for effective evaluation of Based on published nutrient requirements response to grazing. (National Research Council, 2007), the NWSG hays fed in our study do not meet DE requirements Forage Productivity for a mature horse at maintenance at the intake rates we observed. However, this may be advan- Forage biomass and regrowth in IG and BB tageous for horses prone to obesity and laminitis, were adequate to sustain each of the horses in this as reducing dietary energy and NSC in the diet study on 0.1 ha for about 3 wk, with total available of horses prone to obesity and laminitis is recom- forage of 4,344  kg/ha for IG and 3,587  kg/ha for mended (Geor, 2009). Additionally, the low con- BB. Had the plots experienced another year of es- centration of NSC in the NWSG hays is ideal for tablishment prior to the grazing trial, their forage horses susceptible to carbohydrate-induced lamin- production would have been higher, as NWSG do itis, as they were about half the recommended max- not reach full productivity until their second or imum range of 10–12% (Geor, 2009). The loss of third year of establishment (Harper et  al., 2011). BW observed on NWSG hays in our study was Mature, fertilized stands of BB and IG produced acceptable, especially given that the horses in this 6,290 and 5,590 kg of dry matter/ha, respectively, in study were hard-keeping off-track Thoroughbreds yield trials in Iowa (Hall et al., 1982). Forage yields and the trial took place in midwinter, when horses of these two species in Tennessee typically range require more energy to maintain thermal homeo- from 5,600 to 9,000 kg/ha (Harper et al., 2011). As stasis. As such, the energy levels in these hays may these plots were unfertilized monocultures grazed a be ideal for horses in need of a “diet” hay, resulting year after establishment, the lower yields relative to in optimal weight loss rather than either extreme other studies are to be expected. loss or weight gain. Differences in the seasonal timing of growth However, trace minerals and vitamins may be between species in our study may have impacted deficient in a diet of only BB or IG hay based on the quantity of forage regrowth between grazing nutrient values in our hay and observed intake bouts. The IG plots had higher regrowth than BB rates, and the ratio of Ca:P was approximately 1:1 (940 kg/ha vs. 50 kg/ha, respectively). The BB plots for the BB and 3:2 for IG rather than the optimal had produced reproductive tillers prior to the onset 2:1 recommended for horses (National Research of the grazing trial, and may have used more of Council, 2007). Additionally, CP was low in the their carbohydrate reserves producing reproductive NWSG hays, indicating a necessity for protein sup- tillers than IG as a result before being grazed, lim- plementation if fed long term. As such, it would iting available reserves for regrowth. IG plots did be advisable for equine managers to supplement not produce reproductive tillers until after the first the horse’s diet with a ration balancer formulated grazing trial. BB generally reaches maturity earlier to compensate for these deficiencies. This is recom- in the season than IG (Keyser et al., 2012). Mixing mended for all horses on pasture in Virginia, how- these two grasses in a pasture could provide more ever, and as such should not pose an additional uniform seasonal forage distribution than manag- challenge in nutritional management (Porr and ing as monocultures; however, binary mixtures of Greiwe-Crandell, 2009). these species may not appreciably increase overall Impaction colic should also be considered when yields compared to their monocultures, as a bio- weighing the risks and benefits of NWSG hay. fuel trial in the northern Great Plains found (Hong While impaction colic did not occur in the horses et al., 2013). on this study, high fiber content in hay is a contrib- uting factor to potential impactions. Impactions Animal Weight have been reported on other warm-season grasses high in fiber, such as bermudagrass and teff (Little The trend for greater weight loss in horses on and Blikslager, 2002; Staniar et al., 2010). IG plots was unlikely to be a result of differences Experiment II.  This study evaluated three in forage nutritive value, as DE and CP were NWSG species to determine their potential suit- similar between forage species. Higher forage ability as pasture grasses for horses in Virginia. Our intake on BB compared to IG may account for results suggest both IG and BB could be used as the trend; however, DMI was not measured dur- pasture grasses for horses in summertime if grazed ing the grazing trial. While no studies have com- at a suitable stocking rate to ensure the long-term pared the palatability of these NWSG species Translate basic science to industry innovation Ghajar et al. for horses, Dwyer and Sims (1964) conducted a CONCLUSIONS palatability trial of common prairie species with This study evaluated horse responses in a hay steers and found BB to be among the most pal- feeding trial and grazing experiment using three atable forages, with IG intermediate among the NWSG species in Virginia. BB and IG hay elicited 18 species tested. Though biomass of GG was no toxic effects in horses for the biomarkers we lower than either of the other two species, the measured. Voluntary intake on NWSG hays was abundance of palatable weeds in those plots such lower than on more common cool-season grass as white clover and crabgrass could in part ex- hays, which may be ideal for horses susceptible to plain the lower rate of weight loss by horses on metabolic disorders such as obesity and laminitis. those plots relative to IG, as many weedy species Horse owners with animals prone to these disorders common to Virginia pastures have high nutritive should consider integrating these two species into value (Abaye et al., 2009). their equine ration in combination with a ration balancer or mineral block. In pasture, BB and IG are productive species that can provide useful alternative summer forages Trampling for equine grazing systems. Gamagrass establish- Our study did not directly measure DMI ment was poor in our grazing experiment but GG from grazing, making it impossible to determine increased in productivity over the course of the how much of the forage removed during each grazing trial. For horse owners or land managers grazing bout was through ingestion rather than interested in forage options other than high-carbo- trampling. However, both BB and IG plots had hydrate cool-season pastures, IG and BB provide an more than 50% trampled NWSG cover after the opportunity to optimize equine summer grazing for first grazing bout, and about 35–55% trampled horses vulnerable to obesity and pasture-associated NWSG cover after the second bout, suggesting laminitis. If grazing NWSGs with horses, stocking the loss of forage to trampling was considerable. rate determinations should take the likelihood of As the grazing trial did not begin until July, both high forage losses to trampling damage into account. BB and IG were tall and more susceptible to lodging than they would have been earlier in the ACKNOWLEDGMENTS growing season. This activity was funded, in part, by the College Trampling damage is an important aspect of Agriculture and Life Sciences Pratt Endowment of equine impacts on pasture (Bott et  al., 2013). at Virginia Tech. While trampling simulations have been conducted Conflict of interest statement. None declared. to measure traffic tolerance in turf species for horses (Jaqueth, 2019), or to determine the effect LITERATURE CITED of trampling by feral horses in coastal ecosystems Abaye,  A.  O., G.  Scaglia  Alonso, C.  D.  Teutsch, and (Turner, 1987), we found no studies quantifying P.  M.  Raines. 2009. The nutritive value of common pas- pasture forage losses for actual, rather than simu- ture weeds and their relation to livestock nutrient require- lated, trampling by equines. A  study measuring ments. Blacksburg, VA: Va. Coop. Ext. cattle trampling impacts on pasture found that a Adams, L. G., J. W. Dollahite, W. M. Romane, T. L. Bullard, single trampling event reduced daily growth from and C.  H.  Bridges. 1969. Cystitis and ataxia associated 18 to 11kg/ha/day on a pasture in New Zealand, with sorghum ingestion by horses. J. Am. Vet. Med. Assoc. 155:518–524. and the effect persisted for more than 7 wk (Pande Aiken, G. E., G. D. Potter, B. E. Conrad, and J. W. Evans. 1989. et  al., 2000). Equines can move faster than cattle, Voluntary intake and digestion of Coastal Bermuda grass and their different hoof morphology may lend it- hay by yearling and mature horses. J. Equine Vet. Sci. self to greater sward damage. Reductions in sward 9:262–264. doi:10.1016/S0737-0806(89)80084-X yield and vigor by equine trampling may be more Andersen, R. E., T. A. Wadden, S. J. Bartlett, R. A. Vogt, and R. S. Weinstock. 1995. Relation of weight loss to changes pronounced. Future studies of equine grazing dy- in serum lipids and lipoproteins in obese women. Am. namics on NWSG swards would benefit from J. Clin. Nutr. 62:350–357. doi:10.1093/ajcn/62.2.350 incorporating sward height measurements, as well Bott,  R.  C., E.  A.  Greene, K.  Koch, K.  L.  Martinson, as determinations of equine DMI to provide more P. D. Siciliano, C. Williams, N. L. Trottier, A. Burk, and definitive assessments of the relative importance of A. Swinker. 2013. Production and environmental implica- herbivory and trampling on reductions of available tions of equine grazing. J. Equine Vet. Sci. 33:1031–1043. doi:10.1016/j.jevs.2013.05.004 forage. Translate basic science to industry innovation Native warm-season grasses as equine forage Crozier,  J.  A., V.  G.  Allen, N.  E.  Jack, J.  P.  Fontenot, and Acta Hortic. 93–98. doi:10.17660/ActaHortic.2008.782.8. M. A. Cochran. 1997. Digestibility, apparent mineral ab- LaCasha, P. A., H. A. Brady, V.  G. Allen, C. R. Richardson, sorption, and voluntary intake by horses fed alfalfa, tall and K. R. Pond. 1999. Voluntary intake, digestibility, and fescue, and caucasian bluestem. J. Anim. Sci. 75:1651– subsequent selection of Matua bromegrass, coastal ber- 1658. doi:10.2527/1997.7561651x mudagrass, and alfalfa hays by yearling horses. J. Anim. Curran, J. M., R. J. Sutherland, and R. L. Peet. 1996. A screen- Sci. 77:2766–2773. doi:10.2527/1999.77102766x ing test for subclinical liver disease in horses affected by Lee, S. T., B. L. Stegelmeier, D. R. Gardner, and K. P. Vogel. pyrrolizidine alkaloid toxicosis. Aust. Vet. J. 74:236–240. 2001. The isolation and identification of steroidal sapo- doi:10.1111/j.1751-0813.1996.tb15411.x genins in switchgrass. J. Nat. Toxins 10:273–281. Cymbaluk,  N. F. 1990. Comparison of forage digestion Little, D., and A. T. Blikslager. 2002. Factors associated with by cattle and horses. Can. J. Anim. Sci. 70:601–610. development of ileal impaction in horses with surgical doi:10.4141/cjas90-072 colic: 78 cases (1986–2000). Equine Vet. J. 34:464–468. Dwyer,  D.  D., and L.  S.  Sims. 1964. Preferences of steers for doi:10.2746/042516402776117773 certain native and introduced forage plants. Rangel. Ecol. Martinson, K., H. Jung, M. Hathaway, and C. Sheaffer. 2012. Manag. 17:83–85. The effect of soaking on carbohydrate removal and dry Geor,  R.  J. 2009. Pasture-associated laminitis. Vet. Clin. matter loss in orchardgrass and alfalfa hays. J. Equine Vet. North Am. Equine Pract. 25:39–50, v. doi:10.1016/j. Sci. 32:332–338. doi:10.1016/j.jevs.2011.11.009 cveq.2009.01.004 McIntosh, B. J. 2006. Circadian and seasonal variation in pas- Gorz,  H. J., F. A.  Haskins, R.  Dam, and K. P.  Vogel. 1979. ture nonstructural carbohydrates and the physiological Dhurrin in sorghastrum nutans. Phytochemistry. 18:2024. response of grazing horses [PhD thesis]. Blacksburg, VA: doi:10.1016/S0031-9422(00)82728-4 Virginia Polytechnic Institute and State University. Geor, R. J. 2010. Current concepts on the pathophysiology of Mercer,  M.  A., H.  C.  McKenzie, J.  L.  Davis, K.  E.  Wilson, pasture-associated laminitis. Vet. Clin. North Am. Equine D.  R.  Hodgson, T.  E.  Cecere, and B.  J.  McIntosh. 2020. Pract. 26:265–276. doi:10.1016/j.cveq.2010.06.001 Pharmacokinetics and safety of repeated oral dosing of Hall, K. E., J. R. George, and R. R. Riedl. 1982. Herbage dry acetaminophen in adult horses. Equine Vet. J. 52:120–125. matter yields of switchgrass, big bluestem, and indian- doi:10.1111/evj.13112 grass with N fertilization. Agron. J. 74:47–51. doi:10.2134/ Morgan,  S.  E., B.  Johnson, B.  Brewer, and J.  Walker. 1990. agronj1982.00021962007400010014x Sorghum cystitis ataxia syndrome in horses. Vet. Hum. Harper, C. A., G. E. Bates, M. P. Hansbrough, M. J. Gudlin, Toxicol. 32:582. J.  P.  Gruchy, and P.  D.  Keyser. 2011. PB1752 Native National Research Council. 2007. Nutrient requirements of warm-season grasses: identification, establishment and horses. 7th rev. ed. Washington, DC: Natl. Acad. Press. management for wildlife and forage production in the Pande,  T.  N., I.  Valentine, K.  Betteridge, A.  Mackay, and mid-south. Knoxville, TN: Univ. Tn. Coop. Ext. D. Horne. 2000. Pasture damage and regrowth from cattle Hong,  C.  O., V.  N.  Owens, D.  K.  Lee, and A.  Boe. 2013. treading. Proc. N. Z. Grassland Assoc. 62:155–160. Switchgrass, big bluestem, and indiangrass monocul- Porr, C. A. S., and K. Crandell. 2008. Easy Keepers: Managing tures and their two- and three-way mixtures for bioenergy Horses Prone to Obesity. Blacksburg (VA): Va. Coop. Ext. in the northern great plains. BioEnergy Res. 6:229–239. Staniar, W. B., J. R. Bussard, N. M. Repard, M. H. Hall, and doi:10.1007/s12155-012-9252-9 A. O. Burk. 2010. Voluntary intake and digestibility of teff Jaqueth, A. L., T. R. Turner, M. E. Iwaniuk, B. J. McIntosh, and hay fed to horses. J. Anim. Sci. 88:3296–3303. doi:10.2527/ A. O. Burk. 2019. Relative traffic tolerance of cool-season jas.2009-2668 turfgrasses and suitability for grazing by equine. J. Equine Sturgeon,  L.  S., L.  A.  Baker, J.  L.  Pipkin, J.  C.  Haliburton, Vet. Sci. 78:79–88. doi:10.1016/j.jevs.2019.04.005 and N.  K.  Chirase. 2000. The digestibility and mineral Johnson, A. L., T. J. Divers, M. L. Freckleton, H. C. McKenzie, availability of Matua, Bermuda grass, and alfalfa hay in E.  Mitchell, J.  M.  Cullen, and S.  P.  McDonough. 2006. mature horses. J. Equine Vet. Sci. 20:45–48. doi:10.1016/ Fall panicum (Panicum dichotomiflorum) hepatotoxicosis S0737-0806(00)80320–2 in horses and sheep. J. Vet. Intern. Med. 20:1414–1421. Suagee,  J.  K., B.  A.  Corl, M.  V.  Crisman, R.  S.  Pleasant, doi:10.1892/0891-6640(2006)20[1414:fppdhi]2.0.co;2 C.  D.  Thatcher, and R.  J.  Geor. 2013. Relationships be- Kagan,  I.  A., B.  H.  Kirch, C.  D.  Thatcher, C.  D.  Teutsch, tween body condition score and plasma inflammatory F.  Elvinger, D.  M.  Shepherd, and S.  Pleasant. 2011. cytokines, insulin, and lipids in a mixed population Seasonal and diurnal changes in starch content and sugar of light-breed horses. J. Vet. Intern. Med. 27:157–163. profiles of bermudagrass in the piedmont region of the doi:10.1111/jvim.12021 united states. J. Equine Vet. Sci. 31:521–529. doi:10.1016/j. Thatcher,  C. D., R. S.  Pleasant, R. J.  Geor, F.  Elvinger, jevs.2011.08.010 K. A. Negrin, J. Franklin, L. Gay, and S. R. Werre. 2008. Kephart, K. D., and D. R. Buxton. 1996. Nonstructural carbo- Prevalence of obesity in mature horses: an equine body hydrates in cool-and warm-season perennial grasses condition study. J. Anim. Physiol. Anim. Nutr. 92:222– adapted to shaded conditions. Res. Summ. 85–86. 222. doi:10.1111/j.1439-0396.2007.00789_8.x Keyser,  P., G.  Bates, J.  Waller, C.  Harper, F.  Allen, and Tompkins,  R.  D., C.  M.  Luckenbaugh, W.  C.  Stringer, E.  Doxon. 2012. SP731-D Producing hay from native K. H. Richardson, E. A. Mikhailova, and W. C. Bridges. warm-season grasses in the mid-South. Knoxville, TN: 2010. Suther prairie: vascular flora, species richness, and ed- Univ. Tn. Coop. Ext. aphic factors. Castanea. 75:232–244. doi:10.2179/09-041.1 Klein, J. D., L. A. Wood, and R. L. Geneve. 2008. Hydrogen Turner,  M.  G. 1987. Effects of grazing by feral horses, clip- peroxide and color sorting improves germination and ping, trampling, and burning on a Georgia salt marsh. vigor of eastern gamagrass (Tripsacum dactyloides) seeds. Estuaries. 10:54–60. doi:10.2307/1352025 Translate basic science to industry innovation http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Translational Animal Science Oxford University Press

Evaluating digestibility and toxicity of native warm-season grasses for equines

Download PDF
13 pages

Loading next page...
 
/lp/oxford-university-press/evaluating-digestibility-and-toxicity-of-native-warm-season-grasses-WGnFH9vOEn

References (39)

Copyright
© The Author(s) 2020. Published by Oxford University Press on behalf of the American Society of Animal Science.
eISSN
2573-2102
DOI
10.1093/tas/txaa224
Publisher site
See Article on Publisher Site

Abstract

†,1, ‡ † || † S.M. Ghajar , H. McKenzie , J. Fike , B. McIntosh , B.F. Tracy School of Plant and Environmental Sciences, Virginia Polytechnic Institute & State University, Blacksburg, VA 24061; Department of Large Animal Clinical Sciences, Virginia Polytechnic Institute & State University, || Blacksburg, VA 24061; and Mars Equestrian™, McLean, VA 22101 ABSTRACT:  Introduced cool-season grasses are Biomarkers for hepatotoxicity remained within dominant in Virginia’s grasslands, but their high acceptable ranges for all treatments. Apparent dry digestible energy and nonstructural carbohy- matter digestibility (DMD) did not differ among drate (NSC) levels pose a risk for horses prone to hays, ranging from 39% to 43%. NSC levels ranged obesity and laminitis. Native warm-season grasses from 4.4% to 5.4%, below maximum recom- (NWSGs) have lower digestible energy and NSC mended concentrations for horses susceptible to levels that may be more suitable for horses sus- laminitis. For the second objective, a grazing trial ceptible to laminitis. Although NWSGs have de- was conducted comparing IG, BB, and eastern sirable characteristics, they are novel forages for gamagrass (EG) (Tripsacum dactyloides) yields, horses. Little is known about NWSG intake or po- forage losses, changes in vegetative composition, tential toxicity to horses or how grazing by horses and effects on equine bodyweight. Nine, 0.1-ha may affect NWSG swards. The overall objectives plots were seeded with one of the three native of this research were to 1) assess voluntary intake, grass treatments, and each plot was grazed by one toxicological response, and apparent digestibility Thoroughbred gelding in two grazing bouts, one of NWSG hays fed to horses; and 2)  evaluate in July and another in September 2019. IG had the characteristics of three NWSG species under the greatest available forage, at 4,340 kg/ha, com- equine grazing. For the first objective, a hay feed- pared with 3,590 kg/ha from BB (P < 0.0001). EG ing trial using indiangrass (IG) (Sorghastrum plots established poorly, and had only 650  kg/ha nutans) and big bluestem (BB) (Andropogon gerar- available forage during the experiment. Grazing dii) was conducted with nine Thoroughbred geld- reduced standing cover of native grasses in IG ings in a replicated 3  × 3 Latin square design. and BB treatments by about 30%. Horses lost 0.5– Voluntary dry matter intake of IG and BB hays by 1.5  kg BW/d on all treatments. Findings suggest horses were 1.3% and 1.1% of BW/d, lower than IG and BB merit further consideration as forages orchardgrass (Dactylis glomerata), an introduced for horses susceptible to obesity and pasture-asso- cool-season grass, at 1.7% of BW/d (P = 0.0020). ciated laminitis. Key words: equine, forage, native grasses, pasture © The Author(s) 2020. Published by Oxford University Press on behalf of the American Society of Animal Science. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribu- tion, and reproduction in any medium, provided the original work is properly cited. Transl. Anim. Sci. 2021.5:1-13 doi: 10.1093/tas/txaa224 INTRODUCTION 1 More than half of horses in Virginia are over- Corresponding author: sghajar@vt.edu weight or obese (Thatcher et  al., 2008). Obesity Received September 18, 2020. puts a horse at risk for serious health issues such Accepted November 30, 2020. 1 Ghajar et al. as insulin resistance and laminitis (Geor, 2009). game species (Tompkins et  al., 2010). Common Laminitis is a disease characterized by an inflam- species native to much of the eastern United States matory response damaging the lamellar layer of include big bluestem (BB) (Andropogon gerar- the horse’s hooves, allowing the coffin bone to ro- dii), indiangrass (IG) (Sorghastrum nutans), and tate (Geor, 2010). It is painful, costly to treat, and switchgrass (Panicum virgatum). Though research may necessitate humane euthanasia of the horse. is lacking on the use of BB or IG as equine for- Although the mechanisms that precipitate ages, one study examining NSC in forages under laminitis are not fully understood, this inflamma- different light conditions found that NSC levels in tion in the hoof usually follows the consumption of BB never exceeded 12%, suggesting some NWSG large quantities of readily-fermented carbohydrates may have ideal NSC levels for horses susceptible to (Geor, 2010). When a horse with a predisposition to laminitis (Kephart and Buxton, 1996). obesity or insulin resistance consumes a large quan- Though the lower NSC concentration in tity of nonstructural carbohydrates (NSC)—carbo- NWSGs may be optimal for horses prone to meta- hydrates not forming the walls and membranes bolic disorders, questions remain regarding the of the plant, but rather starch and water-soluble safety of these grasses for horses. A  number of carbohydrates—it may prove more than the intes- studies have determined that the Panicum genus tine can readily digest, and the bolus of carbohy- causes hepatotoxicity in horses. Several species of drates can end up being fermented rapidly in the grass in that genus are common and widespread hindgut (McIntosh, 2006). This rapid fermentation throughout the eastern United States, including lowers hindgut pH, altering the microbiome, and switchgrass, fall panicum (Panicum dichotomi- causing the release of endotoxins into the blood- orum fl ), and many species commonly referred to as stream (Geor, 2010). For horses susceptible to “panic grasses” due to their large panicle seedheads. obesity and laminitis, it is recommended not to ex- After 14 horses at a boarding facility in Virginia fell ceed 10–12% NSC concentration in the diet (Geor, severely ill in 2004, the cause was determined to be 2010). Cool-season grass pastures can easily exceed their hay, which was largely comprised of fall pan- this limit throughout much of the year (McIntosh, icum (Johnson et al., 2006). A subsequent feeding 2006), and even grass hays commonly fed to horses trial of fall panicum to two research horses for 12 such as orchardgrass (Dactylis glomerata) can have d resulted in highly elevated biomarkers for hep- NSC concentrations above 12% (Martinson et al., atotoxicity in blood samples taken from the horses, 2012). As a result of the high NSC levels in many and histology revealed bile duct hyperplasia and common forage species, pasture-associated lamin- hepatocyte swelling (Johnson et al., 2006). itis may account for nearly half of all cases in the Other research found that switchgrass ingestion United States (Geor, 2009). by horses demonstrated hepatotoxicity as well, and Warm-season grasses, which are grasses with determined diosgenin, a steroidal sapogenin, to be higher optimal growth temperatures and often a the primary toxin (Lee et al., 2001). While the sap- C4 photosynthetic pathway, generally have lower ogenin is believed to be metabolized in a form that levels of NSC than cool-season forages, and no crystallizes in the liver of sheep affected by Panicum fructan, a simple carbohydrate thought to play a toxicity, the mechanism by which the chemical role in laminitis (Kagan et al., 2011). Staniar et al. damages the equine liver is not established, though (2010) found teff hay (Eragrostis tef) presented ad- it is thought to involve apoptosis rather than crys- equate nutrition and low carbohydrate levels for tallization (Johnson et al., 2006). horses. Kagan et al (2011) measured carbohydrate Aside from the documented toxicity of the levels of bermudagrass (Cynodon dactylon) at dif- Panicum genus for horses, we could find no stud- ferent stages of maturity, and at different times of ies linking NWSG to toxicity in horses. However, day, and found it to have suitably low carbohydrate one of the common native grasses mentioned levels to be considered safe for horses prone to earlier, IG, is related to the Sorghum genus. laminitis. While these studies on introduced forage Sorghum species, whether annual or perennial, species have provided alternatives to high-carbo- have been linked to cystitis ataxia in horses, a hydrate cool-season forages for horses, no stud- condition wherein hydrocyanic acid causes de- ies have examined the use of native warm-season generation of the nervous system (Morgan et al., grasses (NWSG) in equine forage systems. 1990). This results in loss of bladder control and NWSG were once abundant in the Piedmont of hind leg coordination, and is irreversible and Virginia as a result of Native American use of pre- often fatal (Adams et  al., 1969). The primary scribed fire to promote savannahs with abundant chemical that is hydrolyzed into hydrocyanic acid Translate basic science to industry innovation Native warm-season grasses as equine forage is dhurrin, which is also found in IG seedlings, medium, and low relative BW groups, and then one though in IG the concentration of dhurrin de- horse from each group was randomly assigned to clines as the plant matures (Gorz et  al., 1979). each of the three squares such that each square had Nevertheless, no cases of toxicity from IG have a similar mean BW. been documented in horses, and even-aged sor- Three hay types were used in the feeding trials. ghum hay has been found to be safe for horses, as NWSG hays were IG and BB donated in July 2018 the toxic compound does not survive prolonged by Ernst Conservation Seeds (Meadville, PA) from storage (Adams et al., 1969). pure stands normally used for seed production. In addition to the questions of safety and nu- A  common cool-season grass hay (orchardgrass tritional value of NWSGs for horses, we found cv ‘HLR’, Barenbrug) was used for comparison. no prior research on these grasses examining their Orchardgrass hay was produced on site in May response to grazing. As equine grazing can have 2018 at Virginia Tech’s Middleburg Agricultural different impacts on a pasture than cattle grazing Research & Extension Center. (Bott et al., 2013), it is necessary to conduct grazing The study consisted of a 10-d acclimation phase trials with horses rather than drawing conclusions and a 4-d digestibility trial. During the first 8 d of from research on cattle grazing and NWSGs to the acclimation phase (day 1 to day 8), horses were understand the potential impacts of horse grazing housed by a treatment group in three adjacent dry on NWSG swards. lots with access to run-in sheds for shelter, ad lib- This research evaluated IG and BB for use as itum white salt and water, and were fed their treat- equine forages. Specific objectives were to: ment hay ad libitum from round bales. On day 9, horses were moved to individual stalls (3.5 m × 3.5 1. Determine if feeding IG and BB hay causes hep- m). for the remainder of the period where they were atic insult to horses. fed their treatment hay and again had ad libitum ac- 2. Determine the voluntary intake of horses fed cess to clean water and white salt. On day 10, horses BB and IG hay as compared to a common hay were fitted with fecal collection harnesses (Equisan species, orchardgrass. Ltd, Australia) to ensure comfort and familiarity 3. Compare the nutritive value and apparent di- with the harness. The harness was also designed gestibility of these NWSG species to a common to collect urine; however urinary analyses were not cool-season grass species, orchardgrass, when conducted in this trial. From day 9 onwards, horses fed as hay. had group access to a dry lot for an hour per day 4. Compare forage productivity, nutritive value, for exercise and social time. The digestibility trial and short-term trampling effects in eastern began on day 11 and concluded on day 14. On day gamagrass (EG), BB, and IG swards grazed by 15, horses were turned out together into a mixed horses, and whether bodyweight gain differed cool-season pasture for a 2-wk washout between among forage types. experimental periods. Horses were weighed on a livestock platform scale on day 1, day 8, and day 15 of each period of the experiment. MATERIALS AND METHODS Blood samples were collected three times per Two experiments were conducted at the period per horse—once on day 1 (baseline), day Virginia Tech Middleburg Agricultural Research 8, and day 15 between 0700 and 0900 hours each & Extension Center in Middleburg, Virginia to day. Horses were not fasted prior to sampling. evaluate NWSG for use as hay and pasture spe- Samples were collected via jugular venipuncture cies for horses. Protocols for both experiments were into 10-mL vacutainer tubes, placed on ice, and approved by Virginia Tech’s Institutional Animal driven directly to Virginia Tech’s Marion DuPont Care and Use Committee. Scott Equine Medical Center in Leesburg, VA, for analysis. Plasma was analyzed for nine different markers of toxicity. The markers were selected Experiment I—Hay Feeding Trial based on past studies of Panicum toxicity and other A hay feeding trial was conducted in November common pasture-associated toxicities that caused and December of 2018 and January of 2019. A rep- elevated marker profiles in horse serum (Table  1) licated Latin square design with three treatments, (Curran et al., 1996; Johnson et al., 2006). Results three periods, and nine horses was used. The nine were forwarded to a veterinarian the same day to horses were Thoroughbred geldings (9–13 yr) and confirm that they were within acceptable ranges. 569  ± 38  kg BW. Horses were divided into high, Horses were also monitored daily for any changes Translate basic science to industry innovation Ghajar et al. Table 1.  Biomarkers assessed to detect potential block design with three replicates of three treat- hepatic insult to horses fed novel NWSG hays in ments: IG, BB, or gamagrass (GG). Plots were the study established on Fauquier-Eubanks and Purcellville- Tankerville soil series (fine, mixed, active mesic Albumin Typic Hapludults). Slopes at the study site ranged Alkaline phosphatase (ALP) from 7% to 15%. Aspartate aminotransferase (AST) Prior to the study, the site was managed as Bile acid Direct bilirubin cool-season pasture with tall fescue (Schedonorus Gamma glutamyl transferase (GGT) arundinaceus) as the dominant species. The site Sorbitol dehydrogenase (SDH) was sprayed with 4.7  L/ha glyphosate the third Total bilirubin week of April 2018. Two weeks later, prescribed Triglycerides fire was used to prepare a clean seedbed and en- sure fescue mortality. On June 1, glyphosate was in behavior that might have indicated an adverse re- applied again at 2.3  L/ha to kill a flush of weedy sponse to the novel hays being tested. species following the fire. The same week, IG and A digestibility trial was conducted in the last BB plots were seeded using a Truax FLEX-II no-till four days of each period. Bedding was removed drill at a depth of 6  mm and subsequently rolled from stalls and fecal collection harnesses were put with a water-filled roller to ensure adequate seed on each horse. Each horse was offered its treatment to soil contact. Gamagrass seeds were soaked in a hay at 2.5% BW dry matter based on the BW meas- 15% hydrogen peroxide (H O ) solution for 18 h to 2 2 ured on day 8 of the experimental period. Hay DM break seed dormancy (Klein et al., 2008), then were concentration was determined by taking approxi- drained, rinsed, and transferred to a Great Plains mately 20 cored samples per round bale being fed, 706NT seed drill. After a 2-d delay due to inclement drying the samples at 135 ºC for 2 h, and dividing the weather, the GG was seeded at a depth of 2 cm. dried weight by the original weight. Hay was split IG and BB plots were sprayed with imazapic into two daily feedings at 0800 and 2000 hours, and herbicide at a rate of 0.15 L/ha the week after seed- fed using hay nets. Orts were collected and weighed ing. GG plots were not sprayed with imazapic, as twice daily at 0700 and 1900 hours. Fecal collec- imazapic causes stunting and mortality in GG. In tion harnesses were emptied at least three times the second week of July, GrazonNextHL (active in- daily to ensure they did not become uncomfortable gredients: 2,4-D and aminopyralid) was applied to for the horse, at 0600, 1400, 2000, and if needed, all plots at a rate of 2.3 L/ha to control broadleaf 0000 hours. Feces were collected in tubes lined with weeds (BW). In mid-August the same year, another plastic bags which were kept shut to preserve mois- application of 0.3  L/ha imazapic was conducted ture, and total fecal output was weighed for each in IG and BB plots to control crabgrass (Digitaria 24-h period starting at 2000 hours the day prior to sanguinalis). No further herbicide applications were 2000 hours on the day of weighing. Two, 1-kg sub- applied to the plots. samples were collected after weighing and compos- The grazing trial began in July 2019. Horses iting each horse’s fecal output each day, and one of were first turned out into their plots on July 10, and these samples was dried at 55 °C until it reached a removed on July 24 when the majority of plots were constant weight to determine dry matter. The re- reduced to 20- to 30-cm stubble height. Nine (n = 9) maining sample was placed in a −20 °C freezer for Thoroughbred geldings aged 10–14 years (median: storage, and later thawed at room temperature for a 13)  and weighing an average of 550  ± 31  kg were day, then dried at 55 °C and sent to Equi-Analytical grouped by weight to reduce variation in grazing (Ithaca, NY) for chemical analyses. Grab samples and trampling pressure among treatments. Each of approximately 50  g were collected from each group was assigned to one treatment per grazing hay bale daily as they were fed to horses, and these bout, and then reassigned randomly to another samples were composited and submitted to Equi- treatment on the following bout. Each plot was Analytical for chemical analyses as well. grazed by one horse for the duration of the grazing bouts. Horses were each provided a shade structure for shelter from the weather, ad libitum access to Experiment II—Equine Grazing Trial water and white salt, and daily applications of fly The grazing trial was conducted from July repellant. through September 2019. Nine, 0.1-ha plots were Horses were then turned out into fescue-dom- established in May 2018 in a randomized complete inated pastures to allow NWSG plots to regrow Translate basic science to industry innovation Native warm-season grasses as equine forage until all plots had reached at least 46  cm height. 1 (baseline) and day 15. Changes in biomarkers of On September 4, horses were placed back on the toxicity were analyzed using mixed models with NWSG plots and removed on September 12, once treatment, period, and treatment × period as fixed some plots were estimated to have inadequate effects, and horse as a random effect. Mixed models forage to meet daily dry matter requirements of the with treatment and period as fixed effects and horse horses grazing them. as a random effect were used to analyze differences Plots were sampled for biomass and percent in apparent digestibility. Intake, BW changes, ap- cover using haphazardly-placed 20-cm × 50-cm parent digestibility, and biomarkers of toxicity are quadrats at the start of each grazing bout and presented as least squares means. shortly after the removal of horses. At the start Experiment II.  Biomass data were analyzed of the first grazing bout, nutritive samples were both separately by date, and in total using repeat- also collected from plots by compositing biomass ed-measures ANOVA. Available forage and forage subsamples and sending them to a commercial la- removed variables were analyzed with repeated boratory for chemical analyses (Equi-Analytical, measures ANOVA to determine differences among Ithaca, NY). Biomass samples (n = 5 per plot) were treatments. hand-clipped at ground level and separated into Percent cover was analyzed separately by date standing NWSG and weedy species, then dried in with ANOVA and pairwise comparisons made with a forced-air oven at 55 °C. NWSG biomass at the Tukey’s HSD. Cover variables analyzed include beginning of each grazing bout was categorized as NWSG standing cover, NWSG trampled cover, “available forage.” Trampled biomass was also har- GW, and BW. vested from within each quadrat; however, because Changes in equine weight for the two grazing of the difficulty in clipping fallen NWSG, these bouts were calculated on a per day basis to account data were discarded as unreliable. The difference for the differing lengths of each bout in the ana- between standing NWSG biomass at the start and lysis. Weight changes were compared with ANOVA. finish of each grazing bout was classified as “forage For all analyses, differences were considered signifi- removed” from the plot. Forage removal included cant when P < 0.05. both forage consumed by horses and forage that was trampled into the ground and thus “removed” RESULTS from the available forage pool. Percent cover was assessed visually as standing NWSG, grassy weeds Experiment I (GW), and BW (n = 10 quadrats per plot). Weeds Forage nutrient value. The three hays used in the were defined as any species not seeded in the plots. grazing trial were evaluated for 18 variables to es- In assessments made after the end of grazing bouts, timate nutritive value for horses. Digestible energy the percent cover of newly trampled NWSG was was greater in the IG hay at 1.9 Mcal/kg compared also assessed. to 1.8 for BB (P = 0.0345) (Table 2). However, NDF Horses were weighed on a digital livestock scale was also over 70 in both NWSG hays, while it was immediately before being turned out into the plots, 66.3 in the OG hay (P < 0.0001). Several nutrients and immediately upon removal. were also greater in the OG hay than in the NWSG hays, including Ca (P < 0.0001), P (P = 0.0009), Mg Statistical Analysis (P < 0.0001), K (P < 0.0001), and Fe (P = 0.0069). Experiment I.  Nutrient compositions for the Starch was greatest in the IG hay (P = 0.0049). three species of hay fed were compared by one-way Toxicological markers.  Biomarkers for toxicity ANOVA. If a difference was found, Tukey’s HSD stayed within parameters deemed acceptable by our was used for pairwise comparisons. Apparent di- veterinarian based on accepted normal ranges and gestibility was calculated by dividing the difference prior toxicological research on our research farm between average daily total nutrient intake and (Mercer et  al., 2020), with one exception. For one average daily nutrient excretion and dividing by horse, ALP, AST, bile acid, GGT, SDH, and total average daily total nutrient intake. Voluntary dry bilirubin were elevated past acceptable ranges on the matter intake (DMI) was compared using a mixed final sampling of period 1. The horse was immediately model with treatment, period, and treatment × removed from the study and was sampled frequently period as fixed effects, and horse as a random effect. to monitor biomarkers. The horse did not exhibit Biomarkers of toxicity were calculated as the overall clinical symptoms at any time. As no other horses change between values from samples taken on day were affected, and the horse’s biomarkers remained Translate basic science to industry innovation Ghajar et al. Table 2.  Mean nutrient composition values (±SE) for big bluestem, indiangrass, and orchardgrass hay treatments Treatment Variable BB IG OG P-value Dry matter, % 94.5 ± 0.5 95.0 ± 0.4 93.4 ± 4.8 0.6613 b a ab DE, Mcal/kg 1.8 ± 0.1 1.9 ± 0.0 1.8 ± 0.1 0.0345 b b b a CP , % 8.8 ± 1.7 7.4 ± 0.9 13.9 ± 1.0 <0.0001 c a b b ADF , % 49.4 ± 3.8 44.6 ± 2.7 41.3 ± 2.3 0.0013 d a a b NDF , % 76.1 ± 2.5 74.6 ± 2.2 66.3 ± 2.0 <0.0001 b a a Ca, % 0.2 ± 0.1 0.3 ± 0.0 0.4 ± 0.1 <0.0001 b b a P, % 0.2 ± 0.0 0.2 ± 0.0 0.3 ± 0.0 0.0009 b c a Mg, % 0.1 ± 0.0 0.1 ± 0.0 0.2 ± 0.0 <0.0001 b c a K, % 2.1 ± 0.3 1.6 ± 0.3 2.9 ± 0.1 <0.0001 b b a Fe, PPM 80.7 ± 27.3 97.6 ± 10.4 197.3 ± 90.7 0.0069 Zn, PPM 21.5 ± 4.5 19.2 ± 1.3 16.8 ± 3.1 0.0854 Cu, PPM 5.8 ± 1.2 6.0 ± 0.7 7.2 ± 0.8 0.0503 Mn, PPM 66.0 ± 21.1 84.4 ± 24.3 75.2 ± 16.5 0.3631 b a a Starch, % 0.4 ± 0.1 0.7 ± 0.1 0.6 ± 0.2 0.0049 WSC , % 4.0 ± 1.3 4.7 ± 0.7 4.4 ± 1.5 0.6642 ESC , % 3.2 ± 1.4 3.2 ± 0.9 2.2 ± 1.5 0.3223 NSC , % 4.4 ± 1.3 5.4 ± 0.8 5.0 ± 1.4 0.4275 Values with differing letters are significantly different at P < 0.05. Treatment: BB = big bluestem hay; IG = indiangrass hay; OG = orchardgrass hay. Crude protein. Acid detergent fiber. Neutral detergent fiber. Water-soluble carbohydrates. Ethanol-soluble carbohydrates. Nonstructural carbohydrates. elevated after removal from the hay and turnout on was more digestible for OG than in BB or IG cool-season mixed pastures, our veterinarian deter- (P < 0.0001). Starch was more digestible in OG and mined the elevated biomarkers most likely indicated IG than BB (P  =  0.0096). The IG treatment had either a response to an unknown insult or a recur- higher apparent digestibility of Ca (P  =  0.0002), rence of a past medical issue. The horse was on the Mn (P = 0.0065), and Cu (P = 0.0033). BB treatment; no other horse had a similar response to consuming the BB hay. Consequently, we omitted Experiment II that horse from toxicological analyses. In periods 2 and 3, an alternative horse was used. Forage characteristics.  Available forage differed Most biomarkers did not differ among treat- at the beginning of the grazing trial. BB and IG ments (Table 3). Of those that did, GGT increased standing biomass did not differ (about 3000 kg/ha), the most in the IG treatment (P = 0.0166) and SDH but the GG treatment had much lower available decreased the most (P = 0.0377). Triglycerides were forage (410 kg/ha) (P < 0.0001). Weedy species bio- lower in BB and IG treatments than in the OG mass also differed, with BB having the lowest weedy treatment (P = 0.0062). biomass at 130  kg/ha and GG having the most, Intake.  Voluntary DMI was greater for OG at 506  kg/ha (P  =  0.0002). At the end of the July than either of the NWSG hays, both in terms grazing bout, available forage did not differ among of mass-consumed (P  =  0.0042) or percent BW treatments, ranging from 386 kg/ha in the GG plots (P  =  0.002) (Table  4). Horses lost weight on both to 565  kg/ha in the BB plots. Weedy biomass dif- NWSG species tested, but gained weight on OG fered again, with GG having greater weedy biomass (P = 0.0357). For all measures of intake or change at 580 kg/ha than both the BB and IG plots, at 62 in BW, there was a treatment effect, but no period ef- and 295 kg/ha, respectively (P = 0.0003). fect or period × treatment interaction was detected. At the beginning of the September grazing Digestibility. Apparent DM digestibility did not bout, IG pastures had the most available forage differ among treatments (Table  5). Crude protein (1,450 kg/ha) while GG again had the least (250 kg/ Translate basic science to industry innovation Native warm-season grasses as equine forage Table 3. Biomarker change from baseline values in horses fed big bluestem, indiangrass, and orchardgrass hays Treatment Biomarker BB IG OG P-value Albumin, g/dL 0.0 ± 0.4 0.1 ± 0.2 −0.5 ± 0.5 0.0667 ALP , U/L −0.6 ± 23.8 48.0 ± 54.4 −8.0 ± 39.9 0.1061 AST , U/L −101.1 ± 42.8 −54.5 ± 26.2 −57.6 ± 34.7 0.3466 Bile acid, µmol/L −0.3 ± 1.2 −0.4 ± 1.2 −2.5 ± 2.1 0.1245 Direct bilirubin, mg/dL 0.0 ± 0.1 −0.1 ± 0.1 0.0 ± 0.1 0.6948 d ab a b GGT , U/L 1.0 ± 3.1 5.7 ± 4.2 0.0 ± 1.9 0.0166 e ab a b SDH , U/L −5.1 ± 2.5 −3.5 ± 6.6 −12.9 ± 8.5 0.0377 Total bilirubin, mg/dL 0.7 ± 0.4 0.7 ± 0.4 0.5 ± 0.3 0.7592 b b a Triglycerides, mg/dL −3.5 ± 10.2 −3.0 ± 7.7 16.0 ± 11.9 0.0062 Treatment means with different letters are statistically different at P < 0.05. Data presented are least squares means ± SE. Treatment: BB = big bluestem hay; IG = indiangrass hay; OG = orchardgrass hay. Alkaline phosphatase. Aspartate aminotransferase. Gamma glutamyl transferase. Sorbitol dehydrogenase. Table 4. Mean dry matter intake and change in bodyweight by treatment Treatment Variable BB IG OG P-value b b a DMI, kg/d 6.6 ± 0.8 7.0 ± 0.9 8.9 ± 1.8 0.0042 b b a DMI, % of BW 1.1 ± 0.1 1.3 ± 0.1 1.7 ± 0.3 0.0020 ab b a Change in BW −16.3 ± 19.0 −21.2 ± 26.2 3.8 ± 25.6 0.0357 Treatment means with different letters are statistically different at P < 0.05. Data presented are least squares means ± SE. Treatment: BB = big bluestem hay; IG = indiangrass hay; OG = orchardgrass hay. Table 5. Apparent digestibility of the three treatment hays fed Treatment Digestibility, % BB IG OG P-value Dry matter 38.8 ± 10.1 40.6 ± 8.4 43.2 ± 5.3 0.2636 b b b a CP 18.3 ± 13.5 30.1 ± 14.4 52.4 ± 3.9 <0.0001 ADF 51.3 ± 9.2 46.7 ± 7.7 48.5 ± 5.6 0.0979 NDF 47.6 ± 9.0 45.7 ± 8.0 46.5 ± 5.5 0.6415 b a b Ca −24.0 ± 26.1 15.6 ± 17.4 −13.7 ± 22.7 0.0002 b a ab P −55.4 ± 21.5 −30.3 ± 13.1 −47.6 ± 30.4 0.0076 Mg −20.6 ± 21.7 −23.2 ± 22.9 −13.0 ± 9.6 0.5823 K 58.7 ± 12.7 57.7 ± 11.7 57.3 ± 9.0 0.8599 Fe −213.2 ± 145.2 −144.6 ± 107.0 −89.3 ± 87.7 0.0561 Zn −34.0 ± 27.9 −29.3 ± 20.6 −45.1 ± 28.3 0.3722 b a b Cu −14.6 ± 18.0 13.3 ± 14.6 −8.6 ± 10.0 0.0033 b a b Mn −77.0 ± 50.8 −18.8 ± 23.2 −60.0 ± 22.5 0.0065 b a a Starch −27.0 ± 86.6 42.7 ± 19.8 39.5 ± 17.9 0.0096 WSC 55.7 ± 17.9 66.5 ± 7.0 59.4 ± 5.5 0.1806 ESC 66.1 ± 24.9 66.0 ± 19.1 81.3 ± 9.0 0.0588 Treatment means with different letters are statistically different at P < 0.05. Data presented are least squares means ± SE. Treatment: BB = big bluestem hay; IG = indiangrass hay; OG = orchardgrass hay. Crude protein. Acid detergent fiber. Neutral detergent fiber. Water-soluble carbohydrates. Ethanol-soluble carbohydrates. Translate basic science to industry innovation Ghajar et al. ha) (P  =  0.0354). BB pasture yields were inter- on GG, which had net negative removal over the mediate (620 kg/ha). The GG treatment again had course of the grazing trial, indicating an increase in the greater weedy species biomass, increasing to biomass of about 260 kg/ha (P = 0.0070). 780  kg/ha compared with 330  kg/ha and 170  kg/ Percent cover of IG and BB did not differ ha in the IG and BB treatments, respectively in July or September at the start of each grazing (P  =  0.0002). After the September grazing bout, bout (Figure  1). BW had higher percent cover in there were no differences in available forage among GG plots than IG or BB in July (P < 0.0001) and treatments. September (P  <  0.0001). GW had higher percent Total seasonal forage availability differed cover in GG than in IG or BB in July (P < 0.0001), among treatments, with the highest mean total and were higher in GG than in BB in September available forage in the IG treatment, at 4,340  kg/ (P = 0.0160). ha, and the lowest in the GG at 650 kg/ha (Table 6) After being grazed in July, over half the cover (P = 0.0140). Overall forage removed also differed of BB and IG plots was comprised of tram- between treatments, with the highest mean removal pled NWSG biomass, while no GG was trampled on IG plots at 3,880 kg/ha, and the lowest removal (P < 0.0001) (Table 7). After the September grazing Table 6. Mean available and removed NWSG forage and weed biomass (kg/ha) of treatments in the grazing trial NWSG Weeds Treatment Available Removed Biomass Removed July BB 2970 2400 130 70 GG 400 20 900 310 IG 2900 2460 510 210 September BB 620 440 170 80 GG 250 −280 780 240 IG 1460 1410 330 180 Total BB 3590 2890 300 140 GG 650 −260 1680 550 IG 4340 3880 830 390 P-value 0.014 0.007 0.0025 0.188 Removed forage was the difference between standing NWSG biomass at the start and finish of each grazing bout. A negative value indicates a gain rather than loss in biomass. P values reported are for repeated measures ANOVA comparing treatments across the course of the experiment. Treatment: BB = big bluestem; IG = indiangrass; OG = orchardgrass. Figure 1. Percent cover of native warm-season grasses (NWSG), grassy weeds (GW), and broadleaf weeds (BW) in July and September at the start of each grazing bout for the big bluestem (BB), gamagrass (GG), and indiangrass (IG) plots. Means with different letters are significantly different (P < 0.05). Translate basic science to industry innovation Native warm-season grasses as equine forage bout, about a third of the cover in BB plots was exceed acceptable limits after 2  wk. Voluntary in- trampled NWSG, and IG plots once again were take of these forages was lower than orchardgrass comprised of over 50% trampled forage from the intake in our study and lower than reported values most recent grazing bout. Again, GG plots had no for other species. However, this may be ideal for measurable trampling of native grass (P < 0.0001). horse owners struggling to find optimal forages for Biomass samples of NWSG from each plot horses prone to obesity and laminitis. Apparent di- were subsampled and composited for nutritional gestibility of some nutrients was low or negative for analysis at the beginning of the first grazing trial. the NWSG hays and orchardgrass on our study, Due to the sparseness of GG, nutritional analysis indicating a need to supplement these forages with could not be carried out on the harvested samples a ration balancer to ensure adequate intake of vita- of two of the three plots. Consequently, only BB mins, minerals, and protein. and IG were compared statistically. They did not differ in CP, ADF, NDF, DE, or NSC (Table 8). Toxicological Response Animal weight. Horses lost weight on all treat- The two NWSG hays tested did not cause bio- ments during the grazing bouts (Figure 2). Weight markers of toxicity to exceed levels deemed safe by changes in horses did not differ among treatments; the veterinarian monitoring horses on this study however, there was a trend for horses on the IG based on prior toxicological research conducted treatment to lose more weight than those on the BB with Thoroughbred geldings on this site (Mercer or GG treatments, at 1.5  kg/day lost on IG plots et al., 2020). As discussed above, one horse on the and about 0.5  kg/day in the other two treatments BB treatment had elevated biomarkers, but the lack (P = 0.0977). of any similar response in the other horses fed the hay as well as the continued elevated biomarkers DISCUSSION when that horse consumed cool-season pasture during a washout and monitoring period suggest a Experiment I.  The objectives of experiment I  cause other than diet. Additionally, we can find no were to determine if BB and IG hays induce a toxic case reports in the literature of BB hay causing a response in horses after 2  wk of feeding and to toxic response in horses. evaluate the potential use of these species as a hay Among biomarkers measured, only GGT in- source for equines. Biomarkers of toxicity did not creased in an NWSG hay (IG) relative to the OG treatment. GGT is an indicator of hepatic func- Table 7.  Mean percent cover of trampled NWSG tion, and changes were slight. To determine if this by treatment following each grazing bout response indicates the potential for hepatic insult Grazing bout caused by IG hay, a longer feeding trial may be ne- Species July September cessary for future research. Additionally, our results a a BB 67.7 ± 32.0 36.3 ± 34.1 cannot rule out the possibility of cystitis-ataxia for b b GG 0.0 ± 0.0 0.0 ± 0.0 horses consuming a diet of IG long-term, as cysti- a a IG 52.5 ± 34.4 54.5 ± 38.7 tis-ataxia does not affect the liver and is diagnosed P-value <0.0001 <0.0001 clinically rather than by blood samples (Adams et  al, 1969). None of our horses exhibited clinical Means not connected by the same letter are significantly different at P < 0.05. signs of cystitis-ataxia after 2 wk of consuming IG; Species: BB = big bluestem; IG = indiangrass; OG = orchardgrass. however, longer feeding trials should be conducted Table 8.  Forage nutritive value indices for big bluestem (BB), eastern gamagrass (GG), and indiangrass (IG) treatments in the grazing trial. No SD is reported for gamagrass because only one sample had suffi- cient biomass for chemical analysis Species CP, % ADF, % NDF, % DE, Mcal/kg NSC (%) BB 7.0 ± 0.3 45.5 ± 2.6 74.5 ± 0.5 1.5 ± 0.8 9.9 ± 1.3 GG 6.2 41.7 74.7 1.1 12.4 IG 6.0 ± 0.8 45.0 ± 0.9 74.7 ± 1.5 1.1 ± 0.6 8.3 ± 1.4 P-value 0.1419 0.7495 0.8117 0.4854 0.2090 Only BB and IG were compared statistically. Species: BB = big bluestem; IG = indiangrass; OG = orchardgrass. Translate basic science to industry innovation Ghajar et al. harvest. Nutritive values from a mixed hayfield of IG and BB from 2010 to 2012 in Tennessee (Keyser et al., 2012) averaged 66.8% NDF, 40.2% ADF, and 9.3% CP—substantially lower fiber concentrations and moderately higher CP than the NWSG hay in our study. Voluntary DMI of the NWSG hays was lower than values reported for mature warm-season grasses in past studies, such as 2.1% for “Coastal” bermudagrass (Cynodon dactylon) or 2.3% for Caucasian bluestem (Bothriochloa bladhii) (Crozier Figure 2. Average changes in equine bodyweight on the big bluestem et al., 1997; LaCasha et al., 1999). Voluntary DMI (BB), gamagrass (GG), and indiangrass (IG) treatments during the for horses consuming OG was within the normal grazing experiment expressed as kilograms per day with standard error. ranges of 1.5–3.1% BW described in the National Research Council’s guidelines for equine nutri- to determine if IG poses a risk for horses in tion, with IG falling slightly below the normal that regard. range at 1.3% and BB well below at 1.1% (National The declines in AST and SDH for all treatments Research Council, 2007). However, for obese horses during the feeding trial were likely a result of the needing reduced digestible energy intake, Virginia horses switching from a species-rich pasture dur- Cooperative Extension recommends reducing hay ing the washout periods to monospecific hay. On intake to 1–1.5% of the target BW while maintain- pasture, horses had exposure to greater varieties of ing constant forage availability to minimize risk of forbs and grasses, some of which can cause hepatic gastric ulcers (Porr and Crandell, 2008). insult. When fed a diet exclusively consisting of one Differences in NDF paralleled differences in in- grass species, this exposure is eliminated. A similar take among treatments, with OG having the lowest decline was measured in a study of acetaminophen NDF and highest intake, and the NWSG having pharmacokinetics in Thoroughbred geldings at the high NDF values and lower intake by horses. This same facility when horses were removed from pas- aligns with past research demonstrating the value ture and fed only hay (Mercer et al., 2020). of NDF as a predictor of intake, with lower NDF Triglycerides increased in the OG treatment but levels predicting higher voluntary intake (LaCasha decreased slightly in the NWSG treatments, which et al., 1999). can be explained by the loss of BW on the NWSG treatments and increase in BW on OG. Moderate weight loss in humans results in lower serum tri- Digestibility glyceride levels (Andersen et  al., 1995). Similarly, Dry matter digestibility (DMD) measured in Suagee et al. (2013) reported a positive relationship our study ranged from 38.8% to 43.2%, similar to between equine body condition and plasma trigly- values reported for lower-quality hay in previous cerides, as well as insulin concentration and plasma research, such as 43% in coastal bermudagrass triglycerides. These results align with our measure (Aiken et  al., 1989), 44% in Caucasian bluestem of limited increases in triglycerides for horses ex- (Bothriochloa bladhii) (Crozier et al., 1997), 38.5% periencing a minor increase in BW. in reed canarygrass (Phalaris arundinacea), and 42.1% in crested wheatgrass (Agropyron cristatum) Voluntary Dry Matter Intake and Nutritive Values (Cymbaluk, 1990). Apparent DMD of the hay treatments in our study was lower than those re- BB matures earlier than IG, and as the fields ported for another warm-season grass, teff, at three for both treatments were cut the same week, BB different stages of maturity, which ranged from hay had reached a more mature stage with a higher 51.5% to 60.6% when fed to horses (Staniar et al., proportion of reproductive tillers, while IG was 2010). The DMD of NWSG hays in our study were still vegetative. Orchardgrass hay was also in a also lower than those reported for alfalfa, popular vegetative state when cut in late spring. As hay in- hay for horses, which ranges from 58% to 64% creases in maturity, nutritional quality decreases (Crozier et  al., 1997; Sturgeon et  al., 2000). The and voluntary intake by horses declines (Staniar NWSG hays, while not highly digestible, are ad- et  al., 2010). The NWSG hays in our study were equate as roughage for equines susceptible to meta- cut later than is optimal, as prolonged rain-delayed bolic disorders. opportunities for cutting and curing at the site of Translate basic science to industry innovation Native warm-season grasses as equine forage Integrating NWSG Hays into the Equine Diet survival of the stand. Gamagrass plots were not sufficiently established for effective evaluation of Based on published nutrient requirements response to grazing. (National Research Council, 2007), the NWSG hays fed in our study do not meet DE requirements Forage Productivity for a mature horse at maintenance at the intake rates we observed. However, this may be advan- Forage biomass and regrowth in IG and BB tageous for horses prone to obesity and laminitis, were adequate to sustain each of the horses in this as reducing dietary energy and NSC in the diet study on 0.1 ha for about 3 wk, with total available of horses prone to obesity and laminitis is recom- forage of 4,344  kg/ha for IG and 3,587  kg/ha for mended (Geor, 2009). Additionally, the low con- BB. Had the plots experienced another year of es- centration of NSC in the NWSG hays is ideal for tablishment prior to the grazing trial, their forage horses susceptible to carbohydrate-induced lamin- production would have been higher, as NWSG do itis, as they were about half the recommended max- not reach full productivity until their second or imum range of 10–12% (Geor, 2009). The loss of third year of establishment (Harper et  al., 2011). BW observed on NWSG hays in our study was Mature, fertilized stands of BB and IG produced acceptable, especially given that the horses in this 6,290 and 5,590 kg of dry matter/ha, respectively, in study were hard-keeping off-track Thoroughbreds yield trials in Iowa (Hall et al., 1982). Forage yields and the trial took place in midwinter, when horses of these two species in Tennessee typically range require more energy to maintain thermal homeo- from 5,600 to 9,000 kg/ha (Harper et al., 2011). As stasis. As such, the energy levels in these hays may these plots were unfertilized monocultures grazed a be ideal for horses in need of a “diet” hay, resulting year after establishment, the lower yields relative to in optimal weight loss rather than either extreme other studies are to be expected. loss or weight gain. Differences in the seasonal timing of growth However, trace minerals and vitamins may be between species in our study may have impacted deficient in a diet of only BB or IG hay based on the quantity of forage regrowth between grazing nutrient values in our hay and observed intake bouts. The IG plots had higher regrowth than BB rates, and the ratio of Ca:P was approximately 1:1 (940 kg/ha vs. 50 kg/ha, respectively). The BB plots for the BB and 3:2 for IG rather than the optimal had produced reproductive tillers prior to the onset 2:1 recommended for horses (National Research of the grazing trial, and may have used more of Council, 2007). Additionally, CP was low in the their carbohydrate reserves producing reproductive NWSG hays, indicating a necessity for protein sup- tillers than IG as a result before being grazed, lim- plementation if fed long term. As such, it would iting available reserves for regrowth. IG plots did be advisable for equine managers to supplement not produce reproductive tillers until after the first the horse’s diet with a ration balancer formulated grazing trial. BB generally reaches maturity earlier to compensate for these deficiencies. This is recom- in the season than IG (Keyser et al., 2012). Mixing mended for all horses on pasture in Virginia, how- these two grasses in a pasture could provide more ever, and as such should not pose an additional uniform seasonal forage distribution than manag- challenge in nutritional management (Porr and ing as monocultures; however, binary mixtures of Greiwe-Crandell, 2009). these species may not appreciably increase overall Impaction colic should also be considered when yields compared to their monocultures, as a bio- weighing the risks and benefits of NWSG hay. fuel trial in the northern Great Plains found (Hong While impaction colic did not occur in the horses et al., 2013). on this study, high fiber content in hay is a contrib- uting factor to potential impactions. Impactions Animal Weight have been reported on other warm-season grasses high in fiber, such as bermudagrass and teff (Little The trend for greater weight loss in horses on and Blikslager, 2002; Staniar et al., 2010). IG plots was unlikely to be a result of differences Experiment II.  This study evaluated three in forage nutritive value, as DE and CP were NWSG species to determine their potential suit- similar between forage species. Higher forage ability as pasture grasses for horses in Virginia. Our intake on BB compared to IG may account for results suggest both IG and BB could be used as the trend; however, DMI was not measured dur- pasture grasses for horses in summertime if grazed ing the grazing trial. While no studies have com- at a suitable stocking rate to ensure the long-term pared the palatability of these NWSG species Translate basic science to industry innovation Ghajar et al. for horses, Dwyer and Sims (1964) conducted a CONCLUSIONS palatability trial of common prairie species with This study evaluated horse responses in a hay steers and found BB to be among the most pal- feeding trial and grazing experiment using three atable forages, with IG intermediate among the NWSG species in Virginia. BB and IG hay elicited 18 species tested. Though biomass of GG was no toxic effects in horses for the biomarkers we lower than either of the other two species, the measured. Voluntary intake on NWSG hays was abundance of palatable weeds in those plots such lower than on more common cool-season grass as white clover and crabgrass could in part ex- hays, which may be ideal for horses susceptible to plain the lower rate of weight loss by horses on metabolic disorders such as obesity and laminitis. those plots relative to IG, as many weedy species Horse owners with animals prone to these disorders common to Virginia pastures have high nutritive should consider integrating these two species into value (Abaye et al., 2009). their equine ration in combination with a ration balancer or mineral block. In pasture, BB and IG are productive species that can provide useful alternative summer forages Trampling for equine grazing systems. Gamagrass establish- Our study did not directly measure DMI ment was poor in our grazing experiment but GG from grazing, making it impossible to determine increased in productivity over the course of the how much of the forage removed during each grazing trial. For horse owners or land managers grazing bout was through ingestion rather than interested in forage options other than high-carbo- trampling. However, both BB and IG plots had hydrate cool-season pastures, IG and BB provide an more than 50% trampled NWSG cover after the opportunity to optimize equine summer grazing for first grazing bout, and about 35–55% trampled horses vulnerable to obesity and pasture-associated NWSG cover after the second bout, suggesting laminitis. If grazing NWSGs with horses, stocking the loss of forage to trampling was considerable. rate determinations should take the likelihood of As the grazing trial did not begin until July, both high forage losses to trampling damage into account. BB and IG were tall and more susceptible to lodging than they would have been earlier in the ACKNOWLEDGMENTS growing season. This activity was funded, in part, by the College Trampling damage is an important aspect of Agriculture and Life Sciences Pratt Endowment of equine impacts on pasture (Bott et  al., 2013). at Virginia Tech. While trampling simulations have been conducted Conflict of interest statement. None declared. to measure traffic tolerance in turf species for horses (Jaqueth, 2019), or to determine the effect LITERATURE CITED of trampling by feral horses in coastal ecosystems Abaye,  A.  O., G.  Scaglia  Alonso, C.  D.  Teutsch, and (Turner, 1987), we found no studies quantifying P.  M.  Raines. 2009. The nutritive value of common pas- pasture forage losses for actual, rather than simu- ture weeds and their relation to livestock nutrient require- lated, trampling by equines. A  study measuring ments. Blacksburg, VA: Va. Coop. Ext. cattle trampling impacts on pasture found that a Adams, L. G., J. W. Dollahite, W. M. Romane, T. L. Bullard, single trampling event reduced daily growth from and C.  H.  Bridges. 1969. Cystitis and ataxia associated 18 to 11kg/ha/day on a pasture in New Zealand, with sorghum ingestion by horses. J. Am. Vet. Med. Assoc. 155:518–524. and the effect persisted for more than 7 wk (Pande Aiken, G. E., G. D. Potter, B. E. Conrad, and J. W. Evans. 1989. et  al., 2000). Equines can move faster than cattle, Voluntary intake and digestion of Coastal Bermuda grass and their different hoof morphology may lend it- hay by yearling and mature horses. J. Equine Vet. Sci. self to greater sward damage. Reductions in sward 9:262–264. doi:10.1016/S0737-0806(89)80084-X yield and vigor by equine trampling may be more Andersen, R. E., T. A. Wadden, S. J. Bartlett, R. A. Vogt, and R. S. Weinstock. 1995. Relation of weight loss to changes pronounced. Future studies of equine grazing dy- in serum lipids and lipoproteins in obese women. Am. namics on NWSG swards would benefit from J. Clin. Nutr. 62:350–357. doi:10.1093/ajcn/62.2.350 incorporating sward height measurements, as well Bott,  R.  C., E.  A.  Greene, K.  Koch, K.  L.  Martinson, as determinations of equine DMI to provide more P. D. Siciliano, C. Williams, N. L. Trottier, A. Burk, and definitive assessments of the relative importance of A. Swinker. 2013. Production and environmental implica- herbivory and trampling on reductions of available tions of equine grazing. J. Equine Vet. Sci. 33:1031–1043. doi:10.1016/j.jevs.2013.05.004 forage. Translate basic science to industry innovation Native warm-season grasses as equine forage Crozier,  J.  A., V.  G.  Allen, N.  E.  Jack, J.  P.  Fontenot, and Acta Hortic. 93–98. doi:10.17660/ActaHortic.2008.782.8. M. A. Cochran. 1997. Digestibility, apparent mineral ab- LaCasha, P. A., H. A. Brady, V.  G. Allen, C. R. Richardson, sorption, and voluntary intake by horses fed alfalfa, tall and K. R. Pond. 1999. Voluntary intake, digestibility, and fescue, and caucasian bluestem. J. Anim. Sci. 75:1651– subsequent selection of Matua bromegrass, coastal ber- 1658. doi:10.2527/1997.7561651x mudagrass, and alfalfa hays by yearling horses. J. Anim. Curran, J. M., R. J. Sutherland, and R. L. Peet. 1996. A screen- Sci. 77:2766–2773. doi:10.2527/1999.77102766x ing test for subclinical liver disease in horses affected by Lee, S. T., B. L. Stegelmeier, D. R. Gardner, and K. P. Vogel. pyrrolizidine alkaloid toxicosis. Aust. Vet. J. 74:236–240. 2001. The isolation and identification of steroidal sapo- doi:10.1111/j.1751-0813.1996.tb15411.x genins in switchgrass. J. Nat. Toxins 10:273–281. Cymbaluk,  N. F. 1990. Comparison of forage digestion Little, D., and A. T. Blikslager. 2002. Factors associated with by cattle and horses. Can. J. Anim. Sci. 70:601–610. development of ileal impaction in horses with surgical doi:10.4141/cjas90-072 colic: 78 cases (1986–2000). Equine Vet. J. 34:464–468. Dwyer,  D.  D., and L.  S.  Sims. 1964. Preferences of steers for doi:10.2746/042516402776117773 certain native and introduced forage plants. Rangel. Ecol. Martinson, K., H. Jung, M. Hathaway, and C. Sheaffer. 2012. Manag. 17:83–85. The effect of soaking on carbohydrate removal and dry Geor,  R.  J. 2009. Pasture-associated laminitis. Vet. Clin. matter loss in orchardgrass and alfalfa hays. J. Equine Vet. North Am. Equine Pract. 25:39–50, v. doi:10.1016/j. Sci. 32:332–338. doi:10.1016/j.jevs.2011.11.009 cveq.2009.01.004 McIntosh, B. J. 2006. Circadian and seasonal variation in pas- Gorz,  H. J., F. A.  Haskins, R.  Dam, and K. P.  Vogel. 1979. ture nonstructural carbohydrates and the physiological Dhurrin in sorghastrum nutans. Phytochemistry. 18:2024. response of grazing horses [PhD thesis]. Blacksburg, VA: doi:10.1016/S0031-9422(00)82728-4 Virginia Polytechnic Institute and State University. Geor, R. J. 2010. Current concepts on the pathophysiology of Mercer,  M.  A., H.  C.  McKenzie, J.  L.  Davis, K.  E.  Wilson, pasture-associated laminitis. Vet. Clin. North Am. Equine D.  R.  Hodgson, T.  E.  Cecere, and B.  J.  McIntosh. 2020. Pract. 26:265–276. doi:10.1016/j.cveq.2010.06.001 Pharmacokinetics and safety of repeated oral dosing of Hall, K. E., J. R. George, and R. R. Riedl. 1982. Herbage dry acetaminophen in adult horses. Equine Vet. J. 52:120–125. matter yields of switchgrass, big bluestem, and indian- doi:10.1111/evj.13112 grass with N fertilization. Agron. J. 74:47–51. doi:10.2134/ Morgan,  S.  E., B.  Johnson, B.  Brewer, and J.  Walker. 1990. agronj1982.00021962007400010014x Sorghum cystitis ataxia syndrome in horses. Vet. Hum. Harper, C. A., G. E. Bates, M. P. Hansbrough, M. J. Gudlin, Toxicol. 32:582. J.  P.  Gruchy, and P.  D.  Keyser. 2011. PB1752 Native National Research Council. 2007. Nutrient requirements of warm-season grasses: identification, establishment and horses. 7th rev. ed. Washington, DC: Natl. Acad. Press. management for wildlife and forage production in the Pande,  T.  N., I.  Valentine, K.  Betteridge, A.  Mackay, and mid-south. Knoxville, TN: Univ. Tn. Coop. Ext. D. Horne. 2000. Pasture damage and regrowth from cattle Hong,  C.  O., V.  N.  Owens, D.  K.  Lee, and A.  Boe. 2013. treading. Proc. N. Z. Grassland Assoc. 62:155–160. Switchgrass, big bluestem, and indiangrass monocul- Porr, C. A. S., and K. Crandell. 2008. Easy Keepers: Managing tures and their two- and three-way mixtures for bioenergy Horses Prone to Obesity. Blacksburg (VA): Va. Coop. Ext. in the northern great plains. BioEnergy Res. 6:229–239. Staniar, W. B., J. R. Bussard, N. M. Repard, M. H. Hall, and doi:10.1007/s12155-012-9252-9 A. O. Burk. 2010. Voluntary intake and digestibility of teff Jaqueth, A. L., T. R. Turner, M. E. Iwaniuk, B. J. McIntosh, and hay fed to horses. J. Anim. Sci. 88:3296–3303. doi:10.2527/ A. O. Burk. 2019. Relative traffic tolerance of cool-season jas.2009-2668 turfgrasses and suitability for grazing by equine. J. Equine Sturgeon,  L.  S., L.  A.  Baker, J.  L.  Pipkin, J.  C.  Haliburton, Vet. Sci. 78:79–88. doi:10.1016/j.jevs.2019.04.005 and N.  K.  Chirase. 2000. The digestibility and mineral Johnson, A. L., T. J. Divers, M. L. Freckleton, H. C. McKenzie, availability of Matua, Bermuda grass, and alfalfa hay in E.  Mitchell, J.  M.  Cullen, and S.  P.  McDonough. 2006. mature horses. J. Equine Vet. Sci. 20:45–48. doi:10.1016/ Fall panicum (Panicum dichotomiflorum) hepatotoxicosis S0737-0806(00)80320–2 in horses and sheep. J. Vet. Intern. Med. 20:1414–1421. Suagee,  J.  K., B.  A.  Corl, M.  V.  Crisman, R.  S.  Pleasant, doi:10.1892/0891-6640(2006)20[1414:fppdhi]2.0.co;2 C.  D.  Thatcher, and R.  J.  Geor. 2013. Relationships be- Kagan,  I.  A., B.  H.  Kirch, C.  D.  Thatcher, C.  D.  Teutsch, tween body condition score and plasma inflammatory F.  Elvinger, D.  M.  Shepherd, and S.  Pleasant. 2011. cytokines, insulin, and lipids in a mixed population Seasonal and diurnal changes in starch content and sugar of light-breed horses. J. Vet. Intern. Med. 27:157–163. profiles of bermudagrass in the piedmont region of the doi:10.1111/jvim.12021 united states. J. Equine Vet. Sci. 31:521–529. doi:10.1016/j. Thatcher,  C. D., R. S.  Pleasant, R. J.  Geor, F.  Elvinger, jevs.2011.08.010 K. A. Negrin, J. Franklin, L. Gay, and S. R. Werre. 2008. Kephart, K. D., and D. R. Buxton. 1996. Nonstructural carbo- Prevalence of obesity in mature horses: an equine body hydrates in cool-and warm-season perennial grasses condition study. J. Anim. Physiol. Anim. Nutr. 92:222– adapted to shaded conditions. Res. Summ. 85–86. 222. doi:10.1111/j.1439-0396.2007.00789_8.x Keyser,  P., G.  Bates, J.  Waller, C.  Harper, F.  Allen, and Tompkins,  R.  D., C.  M.  Luckenbaugh, W.  C.  Stringer, E.  Doxon. 2012. SP731-D Producing hay from native K. H. Richardson, E. A. Mikhailova, and W. C. Bridges. warm-season grasses in the mid-South. Knoxville, TN: 2010. Suther prairie: vascular flora, species richness, and ed- Univ. Tn. Coop. Ext. aphic factors. Castanea. 75:232–244. doi:10.2179/09-041.1 Klein, J. D., L. A. Wood, and R. L. Geneve. 2008. Hydrogen Turner,  M.  G. 1987. Effects of grazing by feral horses, clip- peroxide and color sorting improves germination and ping, trampling, and burning on a Georgia salt marsh. vigor of eastern gamagrass (Tripsacum dactyloides) seeds. Estuaries. 10:54–60. doi:10.2307/1352025 Translate basic science to industry innovation

Journal

Translational Animal Science – Oxford University Press

Published: Dec 3, 2020

Keywords: equine; forage; native grasses; pasture

There are no references for this article.