Apparent digestibility, fecal particle size, and mean retention time of reduced lignin alfalfa hay fed to horses

Abstract

Reduced lignin alfalfa (Medicago sativa L.) has the potential to provide a higher-quality forage source for livestock by improving forage digestibility. This study was conducted to evaluate apparent digestibility when feeding reduced lignin and nonreduced lignin alfalfa hay to adult horses, and to examine mean fecal particle size (MFPS) and mean retention time (MRT) between alfalfa forage types. In 2017, reduced lignin (“54HVX41”) and nonreduced lignin (“WL355.RR”) alfalfa hay was harvested in Minnesota at the late-bud stage. Alfalfa hays were similar in crude protein (CP; 199 g/kg), neutral detergent fiber (NDF; 433 g/kg), and digestible energy (2.4 Mcal/kg). Acid detergent lignin concentrations were lower for reduced lignin alfalfa hay (74 g/kg) compared to nonreduced lignin alfalfa hay (81 g/kg). Dietary treatments were fed to six adult, stock-type horses in a crossover study. Experimental periods consisted of a 9-d dietary adaptation phase followed by a 5-d total fecal collection phase, during which horses were housed in individual boxstalls and manure was removed on a continuous 24-h basis. At 12-h intervals, feces were thoroughly mixed, subsampled in duplicate, and used for apparent digestibility and MFPS analysis. On day 2 of the fecal collection phase, horses were fed two indigestible markers, cobalt (Co) and ytterbium (Yb), which were fed as Co-ethylenediaminetetraacetic acid and Yb-labeled NDF residue, respectively. Additional fecal samples were taken at 2-h intervals following marker dosing until 96-h post-dosing to evaluate digesta MRT. Data were analyzed using the MIXED procedure of SAS, with statistical significance set at P ≤0.05. Dietary treatment (i.e., alfalfa hay type) was included as a fixed effect, while experimental period and horse were considered random effects. Dietary treatments were similar in dry matter intake (1.6% bodyweight) and time to consumption (7.6 h). Apparent dry matter digestibility (DMD) was greater for reduced lignin alfalfa (64.4%) compared to nonreduced lignin alfalfa (61.7%). Apparent CP and NDF digestibility did not differ between dietary treatments, averaging 78% and 45%, respectively. Dietary treatments were similar in MFPS (0.89 mm) and MRT for both liquid (23.7 h) and solid (27.4 h) phase material. These results indicate an improvement in DMD for reduced lignin alfalfa hay when fed to adult horses, with no change in forage consumption, fecal particle size, or digesta retention time.

Introduction

Alfalfa (Medicago sativa L.) is widely used as a forage source for horses due to its high nutrient content. Compared to grasses, legumes such as alfalfa are typically lower in neutral detergent fiber (NDF) and contain greater concentrations of protein, energy, and calcium (Sturgeon et al., 2000; Earing et al., 2010; Potts et al., 2010; Woodward et al., 2011). As a result, alfalfa is generally more digestible compared to grass forages when harvested at a similar stage of maturity. Researchers comparing forage digestibility for horses between alfalfa and grasses have reported average dry matter digestibility (DMD) ranging from 44% to 59% for grass hays and 58% to 73% for alfalfa hays (Crozier et al., 1997; LaCasha et al., 1999; Sturgeon et al., 2000; Edouard et al., 2008; Potts et al., 2010).

Although alfalfa is a preferred forage source for horses with higher dietary requirements, the digestibility and utilization of alfalfa by livestock can be hampered by its lignin content (Sewalt et al., 1997; Casler et al., 2002). Lignin is a complex phenolic polymer and the second most abundant component of secondary plant cell walls (Li et al., 2015). While it is essential for providing the strength and rigidity necessary for normal plant growth, the deposition of lignin into plant cell walls can reduce the feeding value of alfalfa by negatively affecting microbial degradation and the digestion of feed by intestinal enzymes (Buxton and Hornstein, 1986; Liu and Yu, 2011). Lignification has been reported to be the major factor limiting the in vitro DMD of whole plant forage (Reddy et al., 2005; Jung et al., 2012), and numerous studies have reported a strong inverse relationship between lignin concentration and forage digestibility (Albrecht et al., 1987; Buxton and Russell, 1988; Jung et al., 1997; Reddy et al., 2005).

Reduced lignin alfalfa cultivars have the potential to increase the digestibility of alfalfa compared to nonreduced lignin cultivars (Guo et al., 2001; Marita et al., 2003; Reddy et al., 2005; Getachew et al., 2011). Field research evaluating reduced lignin alfalfa under different harvest frequencies has demonstrated a reduction in total herbage acid detergent lignin (ADL) and an increase in estimated neutral detergent fiber digestibility (NDFD) and relative forage quality for reduced lignin alfalfa compared to nonreduced lignin cultivars (Grev et al., 2017, 2020; Getachew et al., 2018; Jungers et al., 2020). However, it remains to be seen if the reduction in ADL and increase in forage NDFD will translate into greater in vivo digestibility when fed to the animal directly. Results with experimental populations of reduced lignin alfalfa found that when reduced lignin alfalfa hay was fed to lambs, DMD and NDFD were greater compared to nonreduced lignin alfalfa hay (Mertens and McCaslin, 2008). Similarly, when reduced lignin alfalfa hay was included as 50% of the ration for lactating dairy cows, NDFD was increased and the additional forage digestibility resulted in 1.3 kg more milk production per head per day compared to the control diet (Weakley et al., 2008). While this information is promising, information on forage digestibility for commercially available reduced lignin alfalfa cultivars is lacking. In addition, alfalfa continues to be an important forage source for the equine industry, yet the effects of reduced lignification on forage digestibility in a hindgut fermenter has not yet been evaluated. Therefore, the objectives for this study were to evaluate apparent digestibility, mean fecal particle size (MFPS), and mean retention time (MRT) when feeding reduced lignin alfalfa hay to adult horses. The hypothesis was that reduced lignin and nonreduced lignin alfalfa hay would be similar in MFPS and MRT but that reduced lignin alfalfa hay would have increased apparent digestibility.

Materials and Methods

All experimental procedures were conducted according to those approved by the University of Minnesota Institutional Animal Care and Use Committee (1710-35228A).

Dietary treatments

Dietary treatments were two commercially available alfalfa cultivars, including one reduced lignin alfalfa cultivar (RL; “54HVX41”) and one nonreduced lignin alfalfa cultivar (NRL; “WL355.RR”). The “54HVX41” cultivar (Pioneer Hi-Bred International Inc., Johnston, IA) was produced via downregulation of the lignin biosynthetic genes (Barros et al., 2019). The “WL355.RR” cultivar (W-L Alfalfa, Ozark, MO) is a high-yielding, adapted cultivar with a similar fall dormancy but is not marketed as reduced lignin and served as a control cultivar for this experiment.

In April 2016, both alfalfa cultivars were seeded into a prepared seedbed in Minnesota at a rate of 18.7 kg/ha. Soil was a combination of an Angus-Le Sueur complex (1% to 6% slopes) and a Cordova loam (0% to 2% slopes), and soil fertility was amended to meet recommendations for alfalfa hay production according to University of Minnesota soil fertility guidelines (Kaiser et al., 2011). Hay for the study was harvested at the late-bud stage (Kalu and Fick, 1981) from the third regrowth on September 2, 2017. All forage was cut, raked, and baled using best management practices designed to minimize leaf loss and optimize forage quality and yield (Digman et al., 2011). Hay was baled into small-square bales, and alfalfa cultivars (i.e., dietary treatments) were individually identified using color-coded zip ties which were manually attached to each bale.

Immediately following baling, 20 hay bales of each alfalfa cultivar were randomly selected and cored (Penn State Forage Sampler, University Park, PA) to determine forage nutrient composition (Sheaffer et al., 2000a). Hay samples were dried in a forced-air oven at 60 °C for 48 h and ground to pass through a 1-mm screen in a Cyclotec (Foss, Hillerod, Denmark). Ground samples were analyzed for forage nutrient composition by a commercial forage testing laboratory (Equi-Analytical, Ithaca, NY) using the following methods: crude protein (CP) was calculated as the percentage of nitrogen multiplied by 6.25 (AOAC, 2010); NDF, acid detergent fiber (ADF), and ADL were measured using filter bag techniques (Ankom Technology, 2017a,b, 2020); starch, water-soluble carbohydrates, and ethanol-soluble carbohydrates were measured using techniques described by Hall et al. (1999); mineral concentrations were determined (Thermo Jarrell Ash IRIS Advantage HX Inductively Coupled Plasma Radial Spectrometer; Thermo Instrument Systems Inc., Waltham, MA) after microwave digestion (Microwave Accelerated Reaction System; CEM, Mathews, NC); and equine digestible energy (DE) was calculated using an equation developed by Pagan (1998). This initial analysis was completed to confirm forage nutrient composition for each of the alfalfa cultivars (data not reported).

In addition to the forage nutrient composition analysis, stem length measurements were completed to compare forage characteristics between RL and NRL alfalfa hay. Within each alfalfa hay type, 12 randomly selected samples consisting of one hay flake (approximately 1.0 kg) were used for stem length measurements. For each sample, stem length was measured on 100 randomly selected stems, which were averaged to estimate the mean stem length. Hay bales were then consolidated by alfalfa hay type and stored indoors throughout the duration of the study.

Experimental design and sample collection

The experiment was completed using a crossover design with two treatments (alfalfa hay type) and two periods during October and November 2017. Six adult (19.5 ± 4.8 yr; five mares and one gelding), stock-type horses with an average bodyweight (BW) of 544 ± 36 kg and body condition score (BCS; Henneke et al., 1983) of 5.7 ± 1.0 were blocked by weight and divided into two similar herds with three horses each. All horses had routine veterinary medical care performed and were current on vaccinations, deworming, and dental care. Horses had been acclimated to their herd and dry lot for approximately 5 mo, and herds remained together for the duration of the study. Each experimental period consisted of a 9-d dietary adaptation phase (day 1 to 9) followed by a 5-d total fecal collection phase (day 10 to 14; Earing et al., 2010; Potts et al., 2010; Staniar et al., 2010). For the 7-d prior to the start of the first experimental period, horses grazed alfalfa pastures containing both RL and NRL alfalfa during the day and were housed in a dry lot overnight with free-choice access to legume-grass mixed hay and water.

At the beginning of each adaptation phase (day 1), horses were weighed using a livestock scale (Weigh-Tronix, Fairmount, MN, PS2000) and BCS was determined (Henneke et al., 1983). For the duration of the adaptation phase, horse herds were housed in individual dry lots with ad libitum access to shelter and water. Horses received their experimental diet as long-stem hay fed from a hay net (Large Bale Net, Hay Chix, Taylor Falls, MN). Hay was provided at 2.5% of the herd BW split into two equal meals at 0800 and 1700 h (1.25% of herd BW at each feeding). Dry matter intakes (DMIs) were not measured during the adaptation phase; however, visual observations confirmed that all horses were readily consuming the alfalfa hay in a consistent manner and there were no changes in intake or feeding behavior observed between the adaptation and fecal collection phase. Manure and waste hay, or hay located outside of the feeder, was removed daily. With each morning feeding, horses were also given 0.9 kg of a commercially prepared ration balancer (Enrich Plus Ration Balancer, Purina, St. Louis, MO) to ensure that all nutritional requirements were met for adult horses at maintenance (NRC, 2007).

At the beginning of each fecal collection phase (day 10), horses were moved to individual rubber-matted boxstalls (3.6 × 3.6 m), where they were housed for the duration of the fecal collection phase. During this phase, the boxstalls remained unbedded to aid in fecal sample collection and prevent sample contamination. Each morning of the fecal collection phase, representative forage samples were obtained by randomly sampling hay bales from each dietary treatment using a core-sampler (Penn State Forage Sampler, University Park, PA). Hay cores were stored at −20 °C for later analysis; this analysis was done to obtain nutrient composition data for each alfalfa hay type and was used for apparent digestibility calculations. Hay was offered in long-stem form at 2% BW from hay nets (Half Bale Net, Hay Chix) and was fed in two equal portions (i.e., 1% BW at each feeding) at 0800 and 2000 h. Hay was fed at 2% BW rather than ad libitum so that each horse consumed a similar amount based on BW and to ensure consumption of the full hay meal. The time when horses began eating their hay meal and the time when horses finished the same hay meal was recorded to allow for calculation of total time to consumption (TTC). Any hay remaining in the stalls, including hay remaining in the hay net or on the stall floor, was considered orts and was collected and removed prior to each feeding. The amount of orts collected was minimal (3.3% ± 5.3% of hay offered), likely due to the reduction in the amount of hay offered (2% BW rather than ad libitum). However, all collected orts were weighed and subtracted from the daily amount of hay offered for determination of total daily hay intake. Any contaminated hay was collected by hand, air-dried (if wet), weighed, and subtracted from the daily amount of hay offered. Horses did not receive a ration balancer during the fecal collection phase, with the exception of the second day to aid in the consumption of the marked feed (described below). Horses had free-choice access to water throughout the fecal collection phase. The amount of water provided to each horse was recorded to allow for calculation of total daily water intake.

For the duration of the fecal collection phase, horses were hand walked twice daily for 15 min immediately prior to receiving their hay meal. Manure was removed from the stalls continuously on a 24-h basis to allow for determination of total daily fecal output and to reduce any possible contamination with hay or urine. Feces for each horse were collected individually into large plastic containers kept outside each horse’s stall and lined with plastic bags that remained closed throughout the day to retain moisture. Manure was removed from the stall immediately following defecation to minimize contamination; however, if contamination did occur the contaminated feces was collected by hand and kept separate so it could be included in the final fecal weight but excluded from the fecal subsamples used for analysis. Cumulative feces weight was recorded every 12 h (0800 and 2000 h), at which time the collected feces was thoroughly mixed using a spiral paint mixer attached to an electric drill and subsampled in duplicate. Each subsample contained approximately 10% of the total fecal mass collected over the 12 h and was placed in a sealed collection bag which was immediately stored at −20 °C for subsequent apparent digestibility and fecal particle size analysis. Across the 5-d fecal collection period, this resulted in a total of 10 sampling time points (2 per day collected at 0800 and 2000 h for 5 d) for a total of 20 subsamples collected per horse (2 subsamples at each time point). Total daily fecal output was calculated as the summed fecal weight from each 12-h period (0800 to 2000 h and 2000 to 0800 h).

Marker preparation

Indigestible markers were used to measure MRT of solute and particulate phase digesta through the entire digestive tract. Solute MRT was measured using cobalt (Co) in the form of Co-ethylenediaminetetraacetic acid (EDTA) as a solute-phase marker, and particulate MRT was measured using ytterbium (Yb) in the form of Yb-labeled NDF residue as a particulate phase marker. The Co-EDTA marker was prepared according to the methods of Udén et al. (1980). Briefly, a solution containing 25 g of Co(II) acetate 4H₂O, 29.2 g EDTA, 4.3 g LiOH H₂O, and 200 mL distilled water was prepared and heated until solutes were dissolved. The mixture was cooled and 20 mL of 30% hydrogen peroxide was added. After standing for 2 to 3 h at room temperature, 300 mL 95% ethanol was added and the mixture was stored overnight under refrigeration. The following day, the mixture was filtered, washed with 80% ethanol, and dried to a constant weight in a 60 °C forced-air oven.

The Yb-labeled NDF residue markers were prepared using the immersion method described by Earing (2011). Markers were prepared separately with both RL and NRL alfalfa hay so that horses were given Yb-labeled NDF residue that was consistent with their dietary treatment. Briefly, hay was chopped by passing through a Wiley Mill (Thomas Scientific, Swedesboro, NJ) with no screen four times. To prevent particle loss during the marker preparation process, hay used for marker attachment was kept in cloth bags throughout the labeling procedure. Hay was first prepared for marker attachment through the removal of soluble particles. Hay was soaked in boiling NDF solution for 1 h at a rate of 60 g chopped forage per 1 L NDF solution and then thoroughly rinsed with hot water to ensure soluble particle removal. Marker attachment to the resulting NDF residue was accomplished by soaking the residue at a rate of 100 g of NDF residue per 1 L of 0.007 M Yb solution for 24 h. The Yb solution was prepared by dissolving 2.96 g of Yb (III) acetate tetrahydrate in 1 L of distilled water. Following the 24-h soak, the NDF residue was soaked in tap water for 1 h and thoroughly rinsed. To ensure removal of any loosely bound Yb, the NDF residue was washed in 0.01 M acetic acid for 1.5 h and rinsed thoroughly. The final product was dried in a forced-air oven at 60 °C for 24 h or until a constant weight was reached.

Marker administration and sample collection

The morning of the second day of each total fecal collection period (day 11), horses received the prepared Co-EDTA and Yb-labeled NDF residue markers in a single dose immediately prior to their morning hay meal. Marker dosage was provided according to horse BW, with each horse receiving 9 mg of Co and Yb per kg BW^0.75. To encourage complete consumption, the marked feed was mixed with 0.9 kg of commercially available ration balancer (Enrich Plus Ration Balancer, Purina) and top dressed with a diluted molasses-water mixture. All horses readily consumed the marked feed in its entirety, after which the morning hay meal was provided.

In addition to the fecal collection procedures outlined previously, additional fecal samples were collected for marker concentration analysis. Following the marker ingestion, excreted feces were collected from each horse immediately following defecation every 2 h from 0 to 96 h post-marker ingestion. At each time point, the collected fecal sample was weighed, the time of excretion and fecal weight were recorded, and the sample was thoroughly mixed by hand. A subsample consisting of at least 400 g of wet feces was placed in a sealed collection bag and stored at −20 °C for subsequent marker concentration measurement. Following subsampling, any remaining fecal material was added to the ongoing 12-h cumulative fecal collection described previously. Weights for the collected marker subsamples were added to the summed total daily fecal output amount.

Sample analysis and calculations

Upon completion of the experiment, all hay and fecal samples were thawed and dried in a forced-air oven at 60 °C for 48 h. Dry matter (DM) concentration was determined by dividing the weight of the hay or feces after drying by the wet weight of the hay or feces as sampled. DMI was calculated by multiplying the hay DM by the total daily hay intake (hay fed minus orts). Hay DMI was expressed as a percentage of BW by dividing hay DMI by horse BW at the beginning of each period. Dry matter intake rate (DMIR) was determined by dividing the total amount of hay consumed (kg DM) by the TTC (h). Dry matter output (DMO) was calculated by multiplying the fecal DM by the total daily fecal output (summation of 12-h cumulative collection plus marker subsamples).

Hay and fecal samples for apparent digestibility analysis were ground to pass through a 1-mm screen in a Cyclotec (Foss) and analyzed for nutrient composition via wet chemistry by a commercial forage testing laboratory (Equi-Analytical). Samples were analyzed for CP, ADF, NDF, and ADL using the methods described previously. Apparent DMD was calculated based on the mean daily DMI and the mean daily fecal DMO using the following equation:

$$\text{DMD=} \frac{\text{DMI}-\text{Fecal} \text{DMO}}{\text{DMI}}$$

Calculations for individual nutrient digestibility followed the same format and were calculated using the following equation:

$$\text{Digestibility} \text{=} \frac{\text{Nutrient} \text{intake}-\text{Nutrient} \text{excreted} \text{in} \text{feces}}{\text{Nutrient} \text{intake}}$$

A particle size distribution analysis was performed on dried fecal particle size samples using a sieve shaker with 4, 2, 1, 0.5, and 0.25 mm stainless steel sieves (Gilson Co. Inc., Lewis Center, OH). A subsample of 50 g of dried feces was placed on the top screen and the stack of sieves was shook at 300 RPM on an agitator (New Brunswick Scientific, Edison, NJ) for 4 min. At this point, any fecal lumps remaining in the top sieve were broken up via manual processing and the sample was sieved for an additional 8 min. The material retained within each sieve was weighed, and MFPS was calculated based on a weighted average of the particle distribution within each sieve.

Fecal samples for marker concentration determination were ground to pass through a 1-mm screen in a Cyclotec (Foss) and sent to a commercial laboratory (University of Kentucky ERTL, Lexington, KY) for analysis. Briefly, samples were digested in concentrated nitric acid and analyzed for Co and Yb concentrations using inductively-coupled plasma spectrophotometry with wavelengths set at 236.379 and 222.447 nm for Co and Yb, respectively. Total tract MRT for both fluid (Co) and particulate (Yb) phases were calculated algebraically according to Blaxter et al. (1956) as:

$$\text{MRT}= \frac{\sum m_i{t}_i}{\sum m_i}$$

where m_i = the amount (ug/g) of marker at the ith sample and t_i = the time (h) elapsed between marker ingestion and the time the ith sample was collected. Total tract MRT was also calculated according to Thielemans et al. (1978) as:

$$\text{MRT}= \frac{\sum t_i{C}_i\Delta t_i}{\sum C_i\Delta t_i}$$

where t_i = time (h) elapsed between marker ingestion and the time the ith sample was collected, C_i = concentration (ug/g) of marker in the ith sample, and ∆t_i = time interval (h) between two consecutive samples. Both equations were utilized for MRT calculations to allow for comparison with the majority of previous literature.

Statistical analysis

Data was checked using a Q-Q plot to ensure normality and then all response variables were analyzed using the MIXED procedure of SAS (version 9.4; SAS Institute Inc., Cary, NC). Residuals were also checked for normality following model fitting to confirm that the assumption of normally distributed residuals had been met. Mean comparisons are reported as least square means ± SE, with statistical significance set at P ≤0.05. Means separations were performed on significant effects using Tukey’s honestly significant difference (HSD) test. For hay nutrient composition and forage characteristic response variables (i.e., stem length), individual forage samples comprised the experimental unit. The model included dietary treatment (i.e., alfalfa hay type) as a fixed effect. Experimental period and replicate were included as random effects.

For intake, apparent digestibility, and fecal particle size response variables, individual horse within dietary treatment was the experimental unit. The model included dietary treatment and treatment by time as fixed effects. Experimental period, horse, and the period by horse interaction were included as random effects. Time was included as a repeated measures according to the methods of Littell et al. (1998). For TTC and DMIR variables, data where horses did not finish their full hay meal prior to receiving the subsequent hay meal (i.e., within a 12-h time window) were excluded from the analysis.

For marker concentration and MRT response variables, individual horse within dietary treatment was the experimental unit. Area under the curve for Yb and Co markers was calculated using the trapezoidal method. The model included dietary treatment as a fixed effect. Experimental period, horse, and the period by horse interaction were included as random effects.

Results and Discussion

Nutrient composition and forage characteristics

Nutrient composition and forage characteristics for the RL and NRL alfalfa hays are shown in Table 1. CP, ADF, NDF, and equine DE were similar between RL and NRL alfalfa hays (P > 0.13), and concentrations for all nutrients fell within the normal range for alfalfa hay in the late-bud stage (Crozier et al., 1997; LaCasha et al., 1999; Sturgeon et al., 2000; Earing et al., 2010; Potts et al., 2010; Woodward et al., 2011). ADL was lower for RL alfalfa compared to NRL alfalfa (P < 0.01). Average ADL concentrations for alfalfa hay vary based on forage maturity, growing conditions, and harvest management but typically range between 70 and 120 g/kg (Crozier et al., 1997; LaCasha et al., 1999; Sturgeon et al., 2000). ADL concentrations for both RL and NRL alfalfa hay fell within this expected range, with RL alfalfa falling on the low end of normal. The nutrient profile for both alfalfa hays was sufficient to exceed the daily DE and CP requirements for adult horses at maintenance. At the intakes measured in this study, horses were consuming an average of 116 and 252% of their daily DE and CP requirements, respectively, with just the alfalfa hay alone.

Table 1.

Nutrient composition and forage characteristics of reduced lignin and nonreduced lignin alfalfa hay fed to adult horses at maintenance

Item	Reduced lignin	Nonreduced lignin	SE
	Nutrient composition1
DM, g/kg	892	892	1.68
CP, g/kg	196	202	3.53
ADF, g/kg	343	348	11.70
NDF, g/kg	430	435	8.06
ADL, g/kg	74b	81a	4.73
Equine DE, Mcal/kg	2.4	2.4	0.02
	Forage characteristics
Stem length, cm	25.3	24.6	0.63

¹All nutrient concentrations are reported on a DM basis.

^a,bWithin row, means without a common letter differ based on a Tukey’s HSD test (P ≤ 0.05).

ADL concentrations for RL alfalfa hay were 74 g/kg compared to 81 g/kg for NRL alfalfa hay, resulting in a 9% reduction in ADL for the RL alfalfa hay compared to the NRL alfalfa hay. Previous research comparing RL and NRL alfalfa hay also reported a reduction in ADL with little to no change in other nutrient components (Mertens and McCaslin, 2008). Similarly, RL alfalfa forage has shown a 6% to 24% reduction in total herbage ADL concentrations compared to NRL alfalfa cultivars (Getachew et al., 2011; Grev et al., 2017; Getachew et al., 2018; Barros et al., 2019). This reduction in total herbage lignin is likely a result of reductions in lignin within the stem fraction of the plant, as research has reported decreases in stem ADL concentrations with minimal differences in leaf ADL concentrations for RL alfalfa (Grev et al., 2020).

Stem length did not differ between alfalfa hay types (P = 0.45). This is the first study comparing forage characteristics such as stem length between RL and NRL alfalfa hay. Stem length was evaluated to ensure that forage particle length was similar across alfalfa hay types, as particle length has been shown to influence motility and transit time within the equine gastrointestinal tract (Drogoul et al., 2000). Stem to leaf ratios were not evaluated for the RL and NRL alfalfa hays; however, previous research has evaluated the stem-to-leaf ratio for the alfalfa cultivars used in this study and found no differences between cultivars, with an average stem-to-leaf ratio of 1.05 across the seeding and first production year (Grev et al., 2020). The lack of differences in forage CP between the two cultivars is also indicative of a similar level of leafiness, since leaves contain a greater proportion of protein compared to stems.

Intake and TTC

Intake for hay and water was similar across dietary treatments when expressed as kilograms per day, percentage of BW, and grams per kilogram of BW^0.75 (Table 2; P > 0.48). Across both treatments, DMI for alfalfa hay averaged 1.6% of BW and 78.0 g/kg BW^0.75. This intake level is within the expected range for horses and corresponds with other studies reporting DMI for horses fed legume or grass hay at maintenance (Ordakowski et al., 2001; Miyaji et al., 2008, 2014; Staniar et al., 2010; Woodward et al., 2011; Clauss et al., 2014). Although this is the first study reporting intake for RL alfalfa hay fed to horses, previous research with other species found no difference in DMI when RL and NRL alfalfa hay was fed to dairy cows as part of a total mixed ration (Weakley et al., 2008). Water intake in the present study averaged 34.3 kg/d across both dietary treatments and is comparable to previous studies reporting daily water intake for horses (Cuddeford et al., 1995; Pearson et al., 2001).

Table 2.

Forage DMI, water intake, total TTC, and DMI rate for adult horses fed reduced lignin and nonreduced lignin alfalfa hay

Item	Reduced lignin	Nonreduced lignin	SE
DMI, kg/d	8.9	8.8	0.54
DMI, % of BW	1.6	1.6	0.06
DMI, g/kg of BW0.75	78.3	77.8	3.34
Water, kg/d	34.1	34.5	2.45
Water, % of BW	6.2	6.3	0.34
Water, g/kg of BW0.75	301.2	306.1	17.53
TTC, h	7.3	7.9	0.93
DMIR, kg/h	0.72	0.67	0.11

TTC and DMIR averaged 7.6 h and 0.7 kg/h, respectively, and did not differ between dietary treatments (Table 2; P > 0.32). Hay nets used in the present study were comparable to the medium hay nets used by Glunk et al. (2014), who reported a mean TTC of 5.1 h and a DMIR of 0.99 kg/h when adult horses were fed a cool-season grass hay. The slightly longer TTC and lower DMIR observed in the present study could be due to a number of reasons, including changes in forage type (i.e., alfalfa vs. grass hay), differences in feeding time (i.e., PM meal fed at 2000 vs. 1600 h), or variation among individual horses (Crozier et al., 1997; LaCasha et al., 1999; Edouard et al., 2008; Hallam et al., 2012). Regardless, the lack of differences in DMI, TTC, and DMIR between RL and NRL alfalfa hays indicate that both hay types were equally accepted by the horses, and that rate of consumption was similar between alfalfa hay types.

Apparent digestibility

Apparent nutrient digestibility values for RL and NRL alfalfa hay are shown in Table 3. Forage DMD increased by 3% when horses consumed RL alfalfa hay compared to NRL alfalfa hay (P = 0.02). The DMD values observed in this study are comparable to results from previous studies, which have reported apparent DMD ranging from 57 to 73% for alfalfa hay fed to horses (Crozier et al., 1997; LaCasha et al., 1999; Sturgeon et al., 2000; Edouard et al., 2008; Earing et al., 2010; Potts et al., 2010). The variation in DMD seen among alfalfa hays is likely due to differences in forage nutrient composition resulting from several factors, including the cultivar grown, the stage of maturity at which the forage was harvested, the management practices implemented during harvest, and the weather conditions throughout the harvest process (Hall et al., 2000; Sheaffer et al., 2000b; Digman et al., 2011; Jungers et al., 2020). This is the first study comparing DMD between RL and NRL alfalfa hay fed to horses. However, the improvement in DMD for RL alfalfa hay is consistent with results from Mertens and McCaslin (2008), who reported a 3% to 5% increase in DMD when RL alfalfa hay was fed to lambs. These improvements in DMD for RL alfalfa hay are likely a result of decreased forage ADL concentrations for the RL alfalfa cultivar. It has been well established that the deposition of lignin in plant cell walls negatively affects forage digestibility (Albrecht et al., 1987; Jung et al., 1997; Reddy et al., 2005; Getachew et al., 2011). Therefore, a reduction in herbage lignin content should result in improved forage digestibility.

Table 3.

Apparent nutrient digestibility values for reduced lignin and nonreduced lignin alfalfa hay fed to adult horses at maintenance

Digestibility	Reduced lignin	Nonreduced lignin	SE
DMD, %	64.4a	61.7b	1.00
CPD, %	78.4	78.3	1.20
ADFD, %	48.1	44.5	2.39
NDFD, %	46.8	43.5	1.70
ADLD, %	30.8	23.3	5.64

^a,bWithin row, means without a common letter differ based on a Tukey’s HSD test (P ≤ 0.05).

Crude protein digestibility (CPD) did not differ between RL and NRL alfalfa hays (P = 0.82). Past studies evaluating CPD for alfalfa hay fed to horses found similar results, reporting CPD ranging from 73% to 83% (Crozier et al., 1997; LaCasha et al., 1999; Sturgeon et al., 2000; Potts et al., 2010; Woodward et al., 2011). Previous research has not compared CPD between RL and NRL alfalfa hay. The similarity in CPD between alfalfa hay types in the present study was likely due to a combination of factors, including the similar CP content between cultivars, the high availability of readily digested CP in alfalfa forage, and the ability of the microbial populations in the hindgut to utilize CP following pre-cecal digestion (Santos et al., 2011; Woodward et al., 2011).

Alfalfa hay treatments were statistically similar in ADF digestibility (ADFD; P = 0.12), NDFD (P = 0.16), and ADL digestibility (ADLD; P = 0.06). The observed fiber digestibility values are consistent with previous research, which has reported ADFD ranging from 21% to 55%, NDFD ranging from 24% to 57%, and ADLD ranging from 18% to 32% for alfalfa hay fed to horses (Crozier et al., 1997; LaCasha et al., 1999; Sturgeon et al., 2000; Earing et al., 2010; Potts et al., 2010). This is the first report comparing apparent ADFD, NDFD, and ADLD between RL and NRL alfalfa hay fed to horses. Previous work in other species has reported improvements in NDFD when RL alfalfa hay was either fed to lambs (Mertens and McCaslin, 2008) or included in the ration for lactating dairy cows (Weakley et al., 2008). Similarly, studies have documented improvements in in vitro stem and total herbage NDFD for RL alfalfa (Guo et al., 2001; Grev et al., 2017, 2020; Getachew et al., 2018). The lack of differences in fiber digestibility between RL and NRL alfalfa hay in the present study could be related to individual variability in fiber digestion among horses. Research has suggested that individual variability may be greater for horses compared to ruminants like cattle, and that individual horses may vary in their ability to ingest and digest their feed (Edouard et al., 2008). Similarly, microbiome work in horses has shown that the equine microbiome responds to small dietary changes in a very individualized manner (Gomez et al., 2021). The authors also recognize that orts are typically higher in cell-wall components compared to offered hay due to feed selection. Although the amount was minimal, there were some orts collected. Due to the limited quantity the nutrient composition of the orts was not analyzed, but if the orts did contain a higher percentage of cell-wall components compared to the offered hay this could have decreased the percentage of the fiber components ingested relative to the total DMI and subsequently affected fiber digestibility calculations. Coupled with the individual variability among horses, this may explain the lack of stronger population trends observed in the present study. Although ADFD and NDFD were not statistically greater for RL alfalfa hay in the present study, numerically there was a 3% to 4% increase in both ADFD and NDFD for the RL alfalfa hay. This, when combined with the individual variability, the significant improvements in NDFD reported for other species, and the increased NDFD documented in several field studies, indicates that further investigation into fiber digestibility of RL alfalfa forage in the horse is warranted.

Fecal particle size

Fecal particle size distribution and MFPS did not differ between RL and NRL alfalfa hay (Table 4; P = 0.98). For both alfalfa hay types, MFPS was 0.89 mm, with the greatest proportion of fecal particles retained in the 1.0 to 2.0 mm sieve followed by the 0.5 to 1.0 mm sieve. MFPS results from the present study are comparable to those reported by Clauss et al. (2014), who reported MFPS ranging from 0.73 to 1.55 mm for ponies consuming grass hay at different intake levels. Similarly, Carmalt et al. (2005) and Lapinskas et al. (2017) reported MFPS ranging from 0.90 to 1.53 mm for horses consuming diets containing various combinations of hay and concentrates. Average MFPS in the present study is slightly lower than results reported by Miyaji et al. (2011), who found MFPS ranging from 1.17 to 2.97 mm for horses fed chopped or ground timothy (Phleum pretense L.) hay. Differences in MFPS among studies are likely due to a number of aspects, including differences in feed type (i.e., hay vs. chopped hay vs. concentrate), diet nutrient composition, feeding level (i.e., ad libitum vs. regulated), number and size of sieves used, sieving method (i.e., wet vs. dry), and variability between individual horses (Udén and Van Soest, 1982; Carmalt et al., 2005; Müller, 2009; Miyaji et al., 2011; Lapinskas et al., 2017). Regardless, the lack of differences in fecal particle size distribution and MFPS between dietary treatments in the present study indicates that mastication and breakdown of feed was similar for each alfalfa hay type.

Table 4.

Fecal particle size distribution (% of DM) and MFPS for adult horses fed reduced lignin and nonreduced lignin alfalfa hay

Particle size distribution	Reduced lignin	Nonreduced lignin	SE
>4.0 mm, %	0.4	0.3	0.06
2.0–4.0 mm, %	18.6	18.3	2.62
1.0–2.0 mm, %	31.4	31.0	2.85
0.5–1.0 mm, %	29.1	29.0	2.38
0.25–0.5 mm, %	16.5	16.4	2.72
<0.25 mm, %	3.8	4.3	1.50
MFPS, mm	0.89	0.89	0.07

Mean retention time

Marker concentrations for both liquid (Co) and particulate (Yb) phase matter were not above detectable concentrations at 72 h post-dose and in some instances, prior to that time point. Therefore, marker concentrations were not determined past 72 h post-dosing. Mean marker excretion by dietary treatment for Co and Yb is shown in Figures 1 and 2, respectively. There was no effect of dietary treatment on marker concentration (average, area under the curve [AUC], peak, and time to peak [TTP]) within the liquid phase (Co; Table 5; P > 0.45). For the particulate phase (Yb), average marker concentration was lower (P = 0.03) and AUC was smaller (P = 0.02) for the RL compared to NRL alfalfa hay; however, there were no differences in peak or TTP marker concentration (Table 5; P > 0.06). The cause behind these differences in average marker concentration and AUC for the particulate phase is somewhat unclear but could possibly be related to the relationship between lignin and other cell wall components and the effects of this relationship on the digestion and excretion of the different fiber components; additional research is needed to further explore this concept.

Figure 1.

Mean Co (liquid phase) excretion for adult horses consuming RL and NRL alfalfa hay.

Figure 2.

Mean Yb (particulate phase) excretion for adult horses consuming RL and NRL alfalfa hay.

Table 5.

Marker concentration values and MRT for liquid (Co) and particulate (Yb) phase matter in the digestive tract of adult horses fed reduced lignin and nonreduced lignin alfalfa hay

Item	Reduced lignin	Nonreduced lignin	SE
Liquid phase (Co)
Average, mg/kg	57.4	55.8	3.71
AUC	3968.3	3963.2	263.87
Peak, mg/kg	261.3	270.9	15.13
TTP, h	17.5	18.2	1.57
MRT1, h	22.9	25.8	2.34
MRT2, h	23.2	22.9	1.71
Particulate phase (Yb)
Average, mg/kg	38.5a	45.1b	8.45
AUC	2182.4a	2678.0b	254.16
Peak, mg/kg	201.0	226.9	36.17
TTP, h	24.9	25.7	2.57
MRT1, h	27.5	27.0	2.23
MRT2, h	27.7	27.3	2.25

¹MRT calculated from Blaxter et al. (1956).

²MRT calculated from Thielemans et al. (1978).

^a,bWithin row, means without a common letter differ based on a Tukey’s HSD test (P ≤ 0.05).

There was no effect of alfalfa hay type on MRT for either liquid (Co) or particulate (Yb) phase matter (Table 5; P > 0.39). MRTs were similar when calculated using both Blaxter et al. (1956) and Thielemans et al. (1978) algebraic methods and averaged 23.7 and 27.4 h for Co and Yb, respectively. There is a very wide range in reported MRT for horses and ponies consuming forage or mostly forage diets, with liquid MRT ranging from 17.4 to 47.7 h and particulate MRT ranging from 23.1 to 53.3 h (Drogoul et al., 2000, 2001; Goachet et al., 2009; Miyaji et al., 2011, 2014; Rodrigues et al., 2012; Clauss et al., 2014). This large variation in MRT in the literature is likely due to a number of different factors, including forage source, inclusion of concentrates in the diet, diet nutrient composition, feeding level, intake rate, management strategies (e.g., time of feeding, use of hay nets, etc.), and variability between individual horses (Cuddeford et al., 1995; Drogoul et al., 2001; Pearson et al., 2001; Goachet et al., 2009; Miyaji et al., 2011; Clauss et al., 2014). However, the majority of reported MRT are between 18 and 34 h for liquid phase digesta and between 22 and 40 h for particulate phase digesta, both of which encompass the values reported in the present study.

The longer MRT for Yb compared to Co indicates selective retention of particulate phase digesta within the equine gastrointestinal tract. Previous research using both liquid and particulate phase markers has also reported a longer MRT for particulate phase matter compared to liquid phase matter (Drogoul et al., 2000; Goachet et al., 2009; Miyaji et al., 2011; Clauss et al., 2014), highlighting the importance of utilizing markers specific to the dietary component of interest. In the present study, the lack of differences between dietary treatments for both liquid and particulate phase matter indicates that both hays traveled through the gastrointestinal tract at similar rates, regardless of alfalfa hay type. Therefore, any differences in apparent digestibility between dietary treatments in the present study were likely not a result of differences in retention time within the gastrointestinal tract.

Summary and Conclusions

Reduced lignin and NRL alfalfa hays were similar in DM, CP, ADF, NDF, and equine DE but differed in ADL. ADL concentrations for RL alfalfa hay were 74 g/kg compared to 81 g/kg for NRL alfalfa hay, resulting in a 9% reduction in ADL for the RL alfalfa hay compared to the NRL alfalfa hay. The nutrient profile for both hays exceeded the daily DE and CP requirements for adult horses at maintenance. Average stem length did not differ between RL and NRL alfalfa hay, indicating a similar forage particle length across alfalfa hay types.

Hay and water intakes, TTC, and DMIR were similar for horses consuming RL and NRL alfalfa hays, indicating that both hay types were equally accepted and consumed by horses. Forage DMD was 3% greater when horses consumed RL alfalfa hay compared to NRL alfalfa hay. Apparent CPD, ADFD, NDFD, and ADLD did not differ statistically between RL and NRL alfalfa hay; however, numerically there was a 3% to 4% increase in ADFD and NDFD for RL alfalfa hay. Fecal particle size distribution and MFPS did not differ between RL and NRL alfalfa hay. There was no effect of dietary treatment on marker concentration (average, AUC, peak, and TTP) or MRT for liquid phase digesta. For particulate phase digesta, average marker concentration was lower and AUC was smaller for RL alfalfa hay compared to NRL alfalfa hay; however, there were no differences in peak marker concentration, TTP, or MRT.

Collectively, these results indicate an improvement in DMD when RL alfalfa hay is fed to adult horses, with no change in forage consumption, fecal particle size, or retention time within the gastrointestinal tract. Future research should further explore changes in fiber digestibility between RL and NRL alfalfa hays.

Glossary

Abbreviations

ADF

acid detergent fiber

ADFD

acid detergent fiber digestibility

ADL

acid detergent lignin

ADLD

acid detergent lignin digestibility

AUC

area under the curve

BCS

body condition score

BW

bodyweight

CP

crude protein

CPD

crude protein digestibility

DE

digestible energy

DM

dry matter

DMD

dry matter digestibility

DMI

dry matter intake

DMIR

dry matter intake rate

DMO

dry matter output

EDTA

ethylenediaminetetraacetic acid

ESC

ethanol-soluble carbohydrates

MFPS

mean fecal particle size

NDF

neutral detergent fiber

NDFD

neutral detergent fiber digestibility

NRL

nonreduced lignin

RL

reduced lignin

TTP

time to peak

WSC

water-soluble carbohydrates