Use of 3-nitrooxypropanol in a commercial feedlot to decrease enteric methane emissions from cattle fed a corn-based finishing diet

Abstract

The present study evaluated enteric CH₄ production, dry matter (DM) intake (DMI), and rumen fermentation in feedlot cattle supplemented with increasing concentrations of 3-nitrooxypropanol (3-NOP). A total of 100 crossbred steers (body weight, 421 ± 11 kg) was randomly assigned to one of four treatments (n = 25/treatment): control (no 3-NOP) or low (100 mg/kg DM), medium (125 mg/kg DM), and high (150 mg/kg DM) doses of 3-NOP. The study was comprised of 28 d of adaptation followed by three 28-d periods, with CH₄ measured for 7 d per period and cattle remaining on their respective diets throughout the 112-d study. Each treatment group was assigned to a pen, with the cattle and diets rotated among pens weekly to allow the animals to access the GreenFeed emission monitoring (GEM) system stationed in one of the pens for CH₄ measurement. Measured concentration (mg/kg DM) of 3-NOP in the total diet consumed (basal diet + GEM pellet) was 85.6 for low, 107.6 for medium, and 124.5 for high doses of 3-NOP. There was a treatment × period interaction (P < 0.001) for DMI; compared with control, the DMI was less for the low and high doses in period 1, with no differences thereafter. Compared with control (10.78 g/kg DMI), CH₄ yield (g/kg DMI) was decreased (P < 0.001) by 52%, 76%, and 63% for low, medium, and high doses of 3-NOP, respectively. A treatment × period effect (P = 0.048) for CH₄ yield indicated that the low dose decreased in efficacy from 59% decrease in periods 1 and 2 to 37% decrease in period 3, while the efficacy of the medium and high doses remained consistent over time. Irrespective of dose, hydrogen emissions increased by 4.9-fold (P < 0.001), and acetate:propionate ratio in rumen fluid decreased (P = 0.045) with 3-NOP supplementation, confirming that other hydrogen-utilizing pathways become more important in the CH₄-inhibited rumen. The study indicates that supplementation of corn-based finishing diets with 3-NOP using a medium dose is an effective CH₄ mitigation strategy for commercial beef feedlots with a 76% decrease in CH₄ yield. Further research is needed to determine the effects of 3-NOP dose on weight gain, feed conversion efficiency, and carcass characteristics of feedlot cattle at a commercial scale.

Introduction

Many enteric methane (CH₄) mitigation strategies have been identified (Hristov et al., 2013; Beauchemin et al., 2020), including the investigational CH₄ inhibitor 3-nitrooxypropanol (3-NOP; DSM Nutritional Products Ltd., Kaiseraugst, Switzerland). This inhibitor reduces methanogenesis in the rumen by inactivating the enzyme methyl-coenzyme M reductase used by archaea (Duin et al., 2016). Some of the advantages of 3-NOP for CH₄ mitigation include: the quantity needed is small (1 to 2 g/d), lack of negative effects on digestibility (Romero-Perez et al., 2014) and ruminal fiber degradability (Zhang et al., 2020), rapid degradation in the rumen to compounds naturally occurring in the rumen (e.g., nitrate, nitrite, and 1,3-propanediol; Duin et al., 2016), sustained efficacy over time (Romero-Perez et al., 2015), and low safety risks (Thiel et al., 2019a, 2019b). However, 3-NOP is not yet approved for commercial use.

In previous studies, 3-NOP has been shown to decrease enteric CH₄ emissions from beef cattle by up to 80% (Romero-Perez et al., 2014; Vyas et al., 2016a, 2018; Kim et al., 2019; McGinn et al., 2019). A meta-analysis by Dijkstra et al. (2018) indicated that the extent of CH₄ decrease when using 3-NOP depended upon the type of animal (dairy vs. beef), diet composition (greater effects in high grain vs. high forage diets), dose (positively related), and method of supplementation (more effective when added to feed and fed continuously compared with giving a pulse ruminal dose directly via a fistula twice a day). Beef feedlots offer a unique opportunity to administer feed additives for CH₄ mitigation because the cattle receive a daily ration that is highly managed, and, thus, daily provision of an optimum dose of 3-NOP is feasible.

There is interest in evaluating the responses of feedlot cattle to diets supplemented with 3-NOP as a means of decreasing the carbon footprint of beef production. For dry-rolled barley grain-based diets fed over a 105-d period, Vyas et al. (2016a) reported a CH₄ yield (kg CH₄/kg dry matter [DM]) reduction of 9% and 81% for cattle fed 100 and 200 mg 3-NOP/kg DM, respectively, as compared with the control group. In a follow-up 238-d study (backgrounding and finishing phases) that used dry-rolled barley grain-based diets supplemented with 200 mg/kg DM in the backgrounding phase and 125 mg/kg DM in the finishing phase, Vyas et al. (2018) reported that CH₄ yield decreased by 42% and 37% during the backgrounding and finishing phases due to 3-NOP, respectively. Furthermore, in a study conducted at a commercial feedlot that used steam-flaked barley-based finishing diets, McGinn et al. (2019) reported a 70% (±18%) reduction in CH₄ yield for cattle fed 3-NOP (125 mg/kg DM). Yet, when 3-NOP (100 mg/kg DM) was added to a diet consisting of 70% whole shelled corn, Kim et al. (2019) reported that enteric CH₄ yield was not significantly decreased compared with control. The reason for the lack of response to 3-NOP in the study by Kim et al. (2019) is not clear but may relate to the type of grain fed (corn vs. barley) and the degree of processing (whole vs. processed). In addition, only a small number of animals (n = 9) were used in the study of Kim et al. (2019), which was a replicated 3 × 3 Latin Square design with limited emission measurements taken over 3 d.

Starch concentration is greater in corn grain than barley grain (72% vs. 57% of DM; NASEM, 2016), and CH₄ emissions from feedlot cattle are less for cattle fed corn compared with barley (Beauchemin and McGinn, 2005). While dry-rolled and temper-rolled barley grain is the mainstay of the western Canadian feedlot industry, feedlots in the United States and elsewhere primarily use steam-flaked, dry-rolled, or high moisture corn grain (Samuelson et al., 2016; Cowley et al., 2019). Thus, it is important to determine the effect of supplementing steam-flaked corn-based feedlot finishing diets with 3-NOP on CH₄ production. Furthermore, it is important to evaluate CH₄ mitigation strategies using commercially relevant conditions. Thus, the objective of this study was to determine the optimum concentration of 3-NOP to reduce enteric CH₄ emissions in a finishing diet containing steam-flaked corn grain fed to cattle in a commercial feedlot.

Materials and Methods

All experimental procedures were in accordance with the guidelines of the Canadian Council on Animal Care (2009), and the use of 3-NOP was approved by the Veterinary Drug Directorate of Health Canada (DSTS No. 207170).

Experimental design and site description

The research was conducted at a small pen facility within a commercial beef cattle feedlot located near Nanton, Alberta, Canada, with a one-time full capacity of ~25,000 cattle. The experiment was conducted between November 7, 2017, and February 27, 2018, for a total of 112 d. The study was structured as a completely randomized design with an adaption phase (28 d) followed by three consecutive 28-d measurement periods and four treatments. The cattle remained on their respective diets throughout the entire study. The adaptation phase allowed the animals to adapt to 3-NOP (DSM Nutritional Products Ltd., Kaiseraugst, Switzerland) in the ration and become familiar with the GreenFeed emission monitoring (GEM) system (C-Lock Inc., Rapid City, SD, USA) used for enteric CH₄ measurement, as described below.

Experimental animals and feeding

Body weight (BW) of individual animals was measured before feeding (nonfasted BW) on two consecutive days at the start of the experiment. A total of 100 crossbred steers (average BW ± SD: 421 ± 11 kg) were randomly assigned to one of the four dietary treatments (25 cattle per treatment): control basal diet (corn-based finishing diet without 3-NOP) or basal diet supplemented with different target levels of 3-NOP (mg/kg DM): low (100), medium (125), or high (150). In addition to the basal diet, all cattle received pellets (without 3-NOP) in the GEM system. Therefore, the levels of 3-NOP in the final diets consumed (basal diet + pellets) were expected to be less than the concentrations of 3-NOP in the basal diet alone.

During the adaptation phase, the concentration of 3-NOP in the 3 treatment diets was 75 mg/kg dietary DM for the first 7 d followed by the appropriate level (100, 125, or 150 mg/kg dietary DM) thereafter, such that the cattle within each treatment group were maintained on their target dose throughout the study. The basal diet contained (DM basis) 72.6% steam-flaked corn (flaked on site), 18% dry corn distiller’s grains with solubles, 7.5% barley silage, and minerals and vitamins added to exceed the nutrient requirements of animals gaining 2 kg/d (Table 1; NASEM, 2016). Feed additives included monensin ionophore (25 mg/kg DM; Rumensin, Elanco Canada Limited, Guelph, ON, Canada) and an antibiotic for liver abscess control (35 mg/kg DM; Chlor 100, Bio-Agri Mix, Mitchell, ON, Canada). Micro-ingredients (monensin, antibiotic, minerals, vitamins, and a portion of the 3-NOP) were added to the ration via a water-based micro-ingredient machine. Feed was prepared in two batches daily, with the control ration prepared before the other rations. Feed for the treatment groups was prepared by first dispensing the appropriate amount of 3-NOP for the 100 mg/kg group from the micro-ingredient machine. Once the cattle were fed the low level of 3-NOP, an additional 3-NOP was added by hand to the remaining feed in the mixer and fed to the cattle in the 125 mg/kg group. The process was repeated for the 150 mg/kg group. The rations were mixed thoroughly before delivery into the feed bunks, and prior to starting the experiment, the mixer underwent testing to validate the weigh scale and mixing consistency.

Table 1.

Ingredient inclusion rate and composition (mean ± SD) of basal diet

Item	Basal diet1
Ingredients, % DM2
Steam-flaked corn grain	72.6
Dry distillers grains with solubles	18.0
Barley silage	7.5
Micro-ingredients3	1.9
Composition, % DM
DM, % as-fed	73.2 ± 0.88
OM	94.6 ± 0.29
CP	14.7 ± 1.17
Starch	49.7 ± 2.06
NDF	16.9 ± 1.00
ADF	5.3 ± 0.31
Crude fat	4.3 ± 0.20

¹Combined analysis of control diet and diets containing 100, 125, and 150 mg 3-NOP/kg DM.

²On a DM basis, steam-flaked corn grain contained: 10.1 ± 0.84% NDF, 2.6 ± 0.47% ADF, 8.2 ± 0.33% CP, and 75.0 ± 7.72% starch; dry distillers grains contained: 35.5 ± 0.75% NDF, 10.2 ± 0.01% ADF, 34.1 ± 1.07% CP, and 3.4 ± 0.20% starch; and barley silage contained: 48.6 ± 0.32% NDF, 25.8 ± 0.07% ADF, 12.2 ± 0.23% CP, and 14.8 ± 0.40% starch.

³Includes: limestone, vitamin-trace mineral premix formulated to meet or exceed vitamin and trace mineral requirements (NASEM, 2016), monensin sodium (to provide 25 mg/kg DM; Rumensin, Elanco Canada Limited, Guelph, ON, Canada), chlortetracycline (to provide 35 mg/kg DM; Chlor 100, Bio-Agri Mix, Mitchell, ON, Canada). Micro-ingredients were included at 1.90% and 1.89% for control and 3-NOP diets, respectively.

Animals were housed by treatment in four adjacent pens surrounded by a 3-m high low-porosity wooden slatted fence to provide protection from the prevailing winds. Pen size per treatment was 16.5 × 31 m (20.5 m² per animal), which is within the recommended range by the Canadian Council on Animal Care (2009) for feedlot operations (19 to 28 m² per animal). Each pen was equipped with four electronic feed bunks (GrowSafe, Calgary, AB, Canada) to measure DM intake (DMI) of individual animals. Animals were fed twice a day (0900 and 1500 hours) for ad libitum intake (5% orts). Orts were removed and weighed weekly. Animal health was monitored daily. For the high-dose treatment, one animal died on day 56 (on January 2, 2018), and another animal was removed on day 106 (February 21, 2018) due to health issues unrelated to treatment.

Emission measurements

Methane and hydrogen (H₂) gases were measured using the GEM system. The GEM system was placed in one of the pens, with fencing and posts positioned to ensure that only a single animal could access the system at a time. Each pen of animals and their treatment diet was moved rotationally once a week, such that a different group of animals accessed the GEM system each week. Thus, each treatment group had access to the GEM system for 7 d within the adaptation phase and in each subsequent 21-d period. Using this approach allowed us to eliminate any possible pen effect because all animals spent the same amount of time in each pen. Details on the use of the GEM system in this manner were reported by Alemu et al. (2019).

The GEM system allows free movement of animals (in and out of the system) and gasses are measured only when the animal’s head is in the “head chamber” unit as determined by the proximity sensor. The system is equipped with an radio-frequency identification reader to recognize individual animals via an electronic ear tag. Upon visiting the system, animals are provided with pellets from an overhead hopper (as bait) to keep them in the unit for sufficient time to achieve a representative measurement during eructation. The pellet was composed of (DM basis) ground corn (75.0%), dried molasses (18.05%), dry corn distiller’s grains with solubles (3.0%), canola meal and oil (3.0%), and urea (0.95%).

Maximum daily pellet drops per animal from the GEM system was set to 36 per 24 h to restrict the amount of pellets consumed. Animals could visit the system anytime during the day but they were eligible for pellet drops only during six visits per 24 h with a maximum of six drops per visit such that animals were required to wait for 4 h between GEM visits. With an average measured pellet drop weight of 35.2 g DM (SD = 2.82, n = 90) and pellet drop interval of 35 s, about 211 g of pellet was delivered during each visit. The system measured CH₄ and H₂ continuously together with airflow, temperature, atmospheric pressure, and relative humidity. Each gas was analyzed by a separate nondispersive infrared analyzer, which was calibrated weekly using a zero (semipure nitrogen) and span CH₄ (319 mg/m³) and carbon dioxide (CO₂; 9,253 mg/m³) gases, with nitrogen as the balance gas. The purpose of the calibration was to define sensor responses to a known concentration of gasses. Four times during the experiment, the air flux sensor was calibrated by releasing a gravimetrically determined quantity of CO₂ into the system using a 90-g prefilled CO₂ cylinder for 3 min (at least three times). The amount released was compared with the calculated capture (97.4% CO₂ recovery, SD = 4.8, n = 4). To avoid interference, animals were kept at a distant from the GEM system during calibration. In order to maintain a consistent airflow rate between 27 and 40 L/s in the collection tube (Velazco et al., 2016), the air filter was cleaned and changed regularly (every 3 to 5 d).

Sample collection

Feed ingredients and diets offered were sampled every day and composited by week for analyses of chemical composition and 3-NOP concentration. To examine the possible time effects on 3-NOP concentration of the diet, samples of the diets were taken daily for 2 wk within each period from the feed bunks immediately after feed delivery in the morning (0900 hours), 6 h after the morning feeding, and before the next morning feeding (24 h after morning feeding). Samples were frozen at −20 °C until sent for 3-NOP analysis. Samples of pellets offered in the GEM system were collected every 2 wk for chemical analysis.

Within treatment, 4 animals (a total of 16 animals) were selected for rumen fluid collection via oral lavage on the last day of the study (Lodge-Ivey et al., 2009). The samples were taken before the morning feeding using a tube connected to a vacuum pump. Care was taken to avoid saliva contamination by discarding the first ~200 mL of sample and any subsequent sample with signs of saliva. The samples were filtered through two layers of polyester monofilament fabric (355 μm mesh opening) to separate the liquid and solid fractions, with the liquid retained for analysis of volatile fatty acids (VFA) and ammonia nitrogen (NH₃-N). For VFA and NH₃-N analyses, 5-mL aliquot rumen fluid samples were acidified with 1 mL 25% metaphosphoric acid and with 1 mL 0.2 M sulfuric acid, respectively, and frozen at −20 °C until analysis.

Laboratory analysis

Composited samples of ingredients, diets, and GEM pellets were analyzed for DM content by drying at 55 °C for 72 h in a forced air oven. Dried samples were ground through a 1-mm screen using a Wiley mill (Thomas Scientific, Swedesboro, NJ) for chemical analyses. Analytical DM concentration of the ground samples was determined by drying at 135 °C for 2 h (method 930.15; AOAC, 2016). The ash concentration of feed was determined to calculate organic matter (OM) concentration (method 942.05; AOAC, 2016). Subsamples were further ground with a ball grinder (mixer mill MM200, Retsch, Haan, Germany) and analyzed for nitrogen using flash combustion (AOAC, 2016, method no. 990.03; Carlo Erba Instruments, Milan, Italy). The crude protein (CP) concentration of ingredients was calculated by multiplying the nitrogen concentration by 6.25. The neutral detergent fiber (NDF) concentration was determined using the procedure of AOAC (2016, method 2002.04) with heat-stable α-amylase and sodium sulfite, and acid detergent fiber (ADF) concentration was determined sequentially (AOAC, 2016, method 973.18) with both procedures modified for the ANKOM Fiber Analyzer (ANKOM Technology Corp., Fairport, NY, USA). Crude fat concentration was determined by ether extraction for 6 h (AOAC, 2016, method no 920.39; E-816 Hot Extraction Unit, Büchi Labortechnik AG, Flawil, Switzerland). Frozen samples of the diets were shipped to DSM Nutritional Products (Basel, Switzerland) for the measurement of 3-NOP concentration using high-performance liquid chromatography (method AP.227089.01; DSM Nutritional Products) and propanediol mononitrate as standard.

Concentrations of VFA were measured using an automated gas-liquid chromatograph (Model 6890, Hewlett-Packard; Palo Alto, CA, USA) with a capillary column (30 m × 0.25 mm × 1 μm; Supelco Nukol 24207, Sigma-Aldrich Canada Co., Oakville, ON, Canada) and flame ionization detector. For VFA, the oven temperature was 150 °C for 1 min, which was then increased by 5 °C/min to 195 °C and held at this temperature for 5 min. The injector temperature was 225 °C, the detector temperature was 250 °C, and the carrier gas was helium. Crotonic acid (Sigma-Aldrich Canada Co., Oakville, ON, Canada) was used as internal standard for the determination of VFA (Ottenstein and Bartley, 1971). Ruminal NH₃-N concentration was determined by the salicylate–nitroprusside–hypochlorite method using a flow injection analyzer (Rhine et al., 1998).

Calculations and statistical analysis

The DMI of each animal was calculated as the sum of dietary DM consumed from the basal diets offered in the GrowSafe feed bunks and the DM consumed as GEM pellet. To calculate gaseous emissions from the GEM system, only visits in which the animal’s head was in close proximity to the sensor for at least 3 min were used in the analysis (described as useful/good visits) as 3-min samples are the minimum to capture at least three eructations per sampling event (Arthur et al., 2017; Beck et al., 2018). To calculate an average daily emission rate (production, g/d) of CH₄ for each animal for the period (7 d of measurement/treatment), the fluxes measured for each animal during useful/good visits within the period were compiled into six 4-h blocks corresponding to time of day (i.e., time bins). Only cattle with ≥10 useful/good visits spread in at least five of the six 4-h time blocks within a period of measurement (7 d) were used in the final analysis to ensure that the full diurnal cycle of emission was represented, as CH₄ emissions fluctuate diurnally (Gunter and Bradford, 2015; Manafiazar et al., 2016). The average flux for each 4-h block was calculated for each animal within period and averaged to obtain a daily emission rate per animal for the period. Hydrogen production was calculated using an arithmetic averaging method with a straight averaging of the visit fluxes (Manafiazar et al., 2016). Methane and H₂ were expressed as yields (g/kg DM) by dividing the daily fluxes by total DMI (basal diet + GEM pellet) of the animal, and CH₄ energy was expressed as a proportion of gross energy intake (GEI) according to IPCC (2019; GE content of feed = 4.409 Mcal/kg DM and energy content of CH₄ = 13.30116 Mcal/kg).

The data were analyzed using the mixed procedure of SAS (SAS Inst. Inc., Cary, NC) with animal as experimental/observational unit (Bello et al., 2016). DMI (kg/d) and CH₄ and H₂ emissions (g/d) and yields (g/kg DMI) for the animals that had adequate GEM visitation data were analyzed using a model that included treatment (control, low, medium, and high), period (1, 2, and 3), and their interaction as fixed effects, with period considered as a repeated measure in the model. To capture potential variation due to individual animals, initial weight was used as a random effect in all analyses. VFA and NH₃-N concentrations were analyzed using a similar model excluding the repeated effect of period and the treatment by period interaction. Kenward–Roger`s option was used in the model statement to estimate denominator degrees of freedom. Time-series covariance structures were assessed and the one with the lowest Akaike and Bayesian information criteria (autoregressive of order-one) was selected. Residual plots were used to check the validity of the underlying statistical assumptions of homogeneity of variances and normality. Treatment means were compared using Tukey’s multiple comparison procedure. Statistical significance was declared at P ≤ 0.05 and trends are discussed at P ≤ 0.10.

Results

The nutrient composition of the basal diet is shown in Table 1. Analyzed chemical composition of the pellet used in the GEM system was (DM basis): 15.3 ± 0.55% CP, 43.6 ± 0.60% starch, 15.0 ± 2.27% NDF, and 7.3 ± 0.65% ADF.

DMI and emissions

The effect of 3-NOP supplementation on total DMI of the cattle, comprised of basal diet consumed in the GrowSafe feed bunks and pellets offered in the GEM system, is reported in Table 2. Overall, relative to the control treatment (11.46 kg/d), total DMI was 6.1% and 6.4% less (P < 0.001) for the low and high doses of 3-NOP, respectively. However, there was a treatment × period interaction (P < 0.001), indicating that the differences in DMI among the treatments changed over time. Figure 1 shows that the DMI of cattle fed the medium dose of 3-NOP was similar to that of control cattle in all periods, while the DMI of the cattle fed the low and high doses was less than that of control cattle in period 1, with no differences thereafter. The observed variation in total DMI of cattle was due to the basal diet intake from the GrowSafe bunks, which also had a treatment × period interaction (P < 0.001; Table 2). Although the total DMI for the low and high doses was less than that of control, the intake of GEM pellet was actually 22% to 25% greater (P = 0.045) for these treatments relative to the control treatment (0.49 kg/d). When only the cattle used to measure CH₄ emissions were considered, the effect of 3-NOP on intakes of total DM and basal diet was similar to that of all cattle, but with a tendency (P = 0.06) for treatment × period interaction for basal diet. Pellet intake for the CH₄ group animals was not affected by treatment (P = 0.16).

Table 2.

Initial BW and DMI for all steers (n = 25) used in the experiment as well as DMI and visits to the GEM system for steers that met the criteria for measuring gaseous emissions. Steers fed a corn-based finishing diet without (control) or supplemented with low, medium, and high doses of 3-NOP

	Treatment1						P-value2
Item	Control	Low	Medium	High	SEM	3-NOP	Period	Int
All animals (n = 25)3
Starting BW, kg	420	421	421	421	3.25	0.99	…	…
Total DMI, kg/d	11.46	10.77	11.48	10.73	0.29	<0.001	0.43	<0.001
Period 1	11.69a	10.48b	10.99ab	10.39b
Period 2	11.07	11.20	11.23	10.58
Period 3	11.21ab	10.56b	11.61a	10.31b
Basal dietary DMI (GrowSafe bunks)	10.97	10.24	10.86	10.12	0.21	<0.001	0.24	<0.001
Period 1	11.38a	10.07a	10.63ab	10.22b
Period 2	10.72	10.66	10.82	10.27
Period 3	10.82ab	9.98bc	11.12a	9.85c
Pellet DMI (GEM system)	0.49b	0.53ab	0.63a	0.61ab	0.06	0.045	<0.001	0.08
Animals visiting the GEM system4
No. of animals5	9–14	12–18	12–17	10–15	…	…	…	…
Total DMI, kg/d	11.22	10.49	11.57	10.77	0.37	0.003	0.50	0.04
Basal diet DMI (GrowSafe bunks)	10.66a	9.86b	10.89a	10.14b	0.36	0.004	0.58	0.06
Pellet DMI (GEM system)	0.57	0.64	0.67	0.61a	0.05	0.16	0.04	0.48
Visits to the GEM system/animal
No. of 4-h blocks in which visits occurred	5.76	5.84	5.91	5.86	0.09	0.43	0.33	0.80
Weekly visits per animal	22.40	27.31	28.10	29.29	2.96	0.12	0.09	0.48
Visit duration (min:s)	4:51b	4:53b	5:10a	5:06ab	0.24	0.01	<0.001	0.21

¹The measured 3-NOP concentration was 0, 85.6, 107.6, and 124.5 mg/kg dietary DM consumed for the control, low, medium, and high treatment, respectively.

²Int, interaction between 3-NOP and period.

³For the high-dose treatment, one animal died on day 56, and another animal was removed on day 106 due to health issues unrelated to treatment.

⁴Visits were compiled into six 4-h blocks corresponding to the time of day. Only animals with ≥10 “good” weekly visits in at least five of the six 4-h time blocks were selected for final analysis within period.

⁵The number of animals in periods 1, 2, and 3 was 9, 10, and 14 for control; 12, 16, and 18 for low; 12, 15, and 17 for medium; 10, 12, and 15 for high treatments, respectively.

^a–cValues within a row with different letters differ at P ≤ 0.05.

Figure 1.

Total DMI (diet offered in the GrowSafe feed bunks plus pellet offered in the GEM system) of all cattle during adaptation (day 1 to 28) and the three experimental periods (day 29 to 112, n = 25/treatment). Values within periods with different letters differ (P ≤ 0.05), and error bars indicate the standard error of the mean.

Of a possible 25 cattle per treatment, 48% to 68% of animals in period 1, 60% to 80% in period 2, and 68% to 84% in period 3 visited the GEM system, depending upon the treatment group. After the visitation criteria were applied (i.e., ≥10 good weekly visits in at least five of the six 4-h time blocks), only the data for 9, 10, and 14 animals for control; 12, 16, and 18 animals for low; and 12, 15, and 17 animals for medium as well as 10, 12, and 15 animals for high treatments were used in periods 1, 2, and 3, respectively (Table 2). The number of average weekly visits to the GEM system was 26.8 and was not affected by treatment. On average, each visit lasted for about 5 min with only the visits for the medium dose being slightly longer (5:10 min; P < 0.01) relative to control (4:51 min). The selected animals used in the emission analysis visited the GEM system throughout the 24-h period, with activity in an average of 5.84 out of 6.0 possible 4-h blocks. However, when examined over the entire group of cattle, many animals failed to visit the GEM system between 0300 and 0600 hours and 1900 and 2000 hours (Figure 2), which may have partly contributed to some animals being excluded in the final analysis.

Figure 2.

Diurnal pattern of visits to the GreenFeed system for all steers (n = 25/treatment). The steers were fed a corn grain-based finishing diet without (control) or supplemented with low, medium, and high doses of 3-NOP. The arrows indicate the time of feeding at 0900 and 1500 hours, and 0000 hours indicates midnight. Error bars indicate standard deviation.

The measured concentrations of 3-NOP in the basal diets offered in the GrowSafe bunks and calculated concentrations in the total DMI (diet plus pellet offered in the GEM system) for the animals used for the emission measurements are presented in Table 3. The calculated concentration of 3-NOP in the total DMI, which accounts for the intake of basal diet and pellet, was 85.6, 107.6, and 124.5 mg/kg DM for the low, medium, and high doses, respectively. Recovery of 3-NOP ranged between 88.8% and 91.2% of the target concentration. The concentration of 3-NOP in the basal diet was consistent throughout the day for all treatments.

Table 3.

The targeted and measured concentration of 3-NOP in the diets (mean ± SD) and estimated daily intake of 3-NOP for the subset of feedlot cattle that met the criteria for measuring gaseous emissions

	Treatment
Item	Low	Medium	High
Target 3-NOP concentration in basal diet, mg/kg DM	100	125	150
Measured 3-NOP concentration basal diet (n = 9)1, mg/kg DM	91.2 ± 2.87	114.0 ± 3.23	133.1 ± 4.68
Recovery, %	91.2	91.2	88.8
Calculated 3-NOP concentration in total diet consumed2, mg/kg total DM	85.6	107.6	124.5
Measured 3-NOP concentration in basal diet (n = 2)3, mg/kg DM
At morning feeding (0900 hours)	92.5 ± 5.72	106.6 ± 8.78	137.9 ± 0.52
6 h after morning feeding (1500 hours)	94.0 ± 3.88	107.5 ± 3.04	135.3 ± 6.65
Before next day morning feeding (24 h after morning feeding)	91.6 ± 8.90	109.7 ± 5.25	129.4 ± 4.90

¹Basal diet was provided in GrowSafe feed bunks, sampled daily, and composited by week for analyses of 3-NOP concentration. 3-NOP concentration in the control diet was zero.

²Calculated based on total DMI (basal diet DMI + pellet DMI in the GEM system) and measured concentration of 3-NOP in the basal diet.

³Based on diet samples collected from the GrowSafe feed bunks at different sampling points of the day for 2 consecutive weeks per period.

Feeding 3-NOP decreased (P < 0.001) CH₄ yield (g/kg DMI), CH₄ production (g/d), and percentage of GEI lost as CH₄ (Table 4). Compared with the CH₄ yield (11.32 g/kg DMI) of the control treatment, the greatest decrease (76.0%) occurred for the medium dose followed by high (63.3%) and low (52.2%) doses. However, there was a significant (P = 0.048) treatment × period interaction for CH₄ yield and production, because the low dose decreased in efficacy in period 3 compared with periods 1 and 2, while the efficacy of the medium and high doses remained consistent over time. The observed decrease in CH₄ yield as affected by dose in the present study was assessed relative to the literature for beef cattle fed high-grain diets supplemented with 3-NOP (Figure 3).

Table 4.

Methane (CH₄) and hydrogen (H₂) emissions of steers fed a corn-based finishing diet without (control) or supplemented with low, medium, and high doses 3-NOP that met the criteria for measuring gaseous emissions

	Treatment1					P-value2
Item	Control	Low	Medium	High	SEM	3-NOP	Period	Int
CH43
g/d	126.4	54.8	28.6	41.2	14.94	<0.001	0.62	0.020
Period 1	139.6a	48.4b	22.7b	28.9b
Period 2	124.1a	50.4b	32.9b	56.4b
Period 3	115.4a	65.6b	30.2c	38.3c
g/kg total DMI	11.32	5.41	2.72	4.16	1.77	<0.001	0.60	0.048
Period 1	12.13a	4.95b	2.32b	2.65b
Period 2	11.35a	4.64b	3.06b	5.88b
Period 3	10.47a	6.65b	2.79c	3.94c
% GE intake	3.42	1.63	0.82	1.26	0.53	<0.001	0.60	0.048
Period 1	3.66a	1.49b	0.70b	0.80b
Period 2	3.43a	1.40b	0.92bc	1.78b
Period 3	3.16a	2.01b	0.84c	1.19c
H2
g/d	0.53	2.42	2.62	2.61	0.22	<0.001	<0.001	0.004
Period 1	0.60c	2.81b	3.05ab	3.26a
Period 2	0.56b	2.66a	2.62a	2.48a
Period 3	0.41c	1.80b	2.18a	2.08ab
g/kg total DMI	0.049	0.231	0.229	0.243	0.01	<0.001	<0.001	<0.001
Period 1	0.054b	0.274a	0.272a	0.308a
Period 2	0.053b	0.242a	0.231a	0.226a
Period 3	0.039b	0.176a	0.185a	0.194a

¹The measured 3-NOP concentration was 0, 85.6, 107.6, and 124.5 mg/kg dietary DM consumed for the control, low, medium, and high treatment, respectively.

²Int, interaction between 3-NOP and period.

³Only animals with ≥10 “good” weekly visits in at least five of the six 4-h time blocks were selected for final analysis. Useful/good visits were selected based on the distance of the animal’s head from the proximity sensor and the duration that the animal’s head in the “head chamber.”

^a–cValues within a row with different letters differ at P ≤ 0.05.

Figure 3.

Decrease in methane (CH₄) yield (g/kg DMI) observed in the current study (red circles) compared with published literature (black circles) for feedlot cattle fed high-grain-based finishing diets. Literature studies were: Kim et al. (2019), McGinn et al. (2019), and Vyas et al. (2016a, 2016b, 2018).

For all treatments, there was a large variation in CH₄ production (g/d) among animals (Figure 4), with an average coefficient of variation (cv) of 19.9%, 78.1%, 39.6%, and 85.4% for the control, low, medium, and high treatments, respectively. The average hourly CH₄ emissions after feeding were substantially less for cattle fed 3-NOP compared with control cattle (Figure 5). The diurnal pattern of CH₄ emissions that occurred post-morning feeding was not evident when cattle were fed 3-NOP.

Figure 4.

Average methane emission (g/d) by experimental period for individual finishing steers fed high-grain diets without (control) or supplemented with low, medium, and high doses of 3-NOP. Only animals with ≥10 “good” weekly visits in at least five of the six 4-h time blocks were included. Animals that visited the GEM system only in period 2 or 3 and met the selection criteria are shown in red. The cv was 19.9%, 78.1%, 39.6%, and 85.4% for the control, low, medium, and high doses, respectively.

Figure 5.

Diurnal pattern of methane emissions from feedlot steers fed a high-grain diet without (control) or supplemented with low, medium, and high doses of 3-NOP. Animals were fed twice a day at 0900 hours (0:00) in the morning and 1500 hours (6:00) in the afternoon (indicated by arrows). Error bars indicate the standard error of the mean. Only animals with ≥10 “good” weekly visits in at least five of the six 4-h time blocks were included, and means are reported in Table 4.

Hydrogen production and yield increased (P < 0.001) with the level of 3-NOP inclusion in the diet, with period (P < 0.001) and treatment × period interaction (P ≤ 0.004) effects (Table 4). Hydrogen emissions for the control cattle remained very low throughout the study, whereas H₂ emissions for the 3-NOP cattle were about five times greater than control in period 1 for all doses. However, H₂ emissions from the cattle fed 3-NOP decreased each period with a significant decrease in period 3 for the cattle fed the low dose compared with the other 3-NOP treatments. The average diurnal pattern of H₂ production by treatment (Figure 6) generally followed a pattern similar to that of CH₄ production (Figure 5).

Figure 6.

Diurnal pattern of hydrogen production from steers fed a high-grain diet without (control) or supplemented with low, medium, and high doses of 3-NOP. Animals were fed twice a day at 0900 hours (0:00) in the morning and 1500 hours (6:00) in the afternoon (indicated by the arrows). Error bars indicate the standard error of the mean. Only animals with ≥10 “good” weekly visits in at least five of the six 4-h time blocks were included, and means are reported in Table 4.

Rumen fermentation

Rumen fermentation was altered by 3-NOP supplementation. Total VFA concentration tended (P = 0.060) to be affected by 3-NOP supplementation, with a slight decrease for the medium dose (Table 5). The molar proportion of acetate was less (P < 0.001) for cattle fed 3-NOP compared with control, while the molar proportion of propionate only tended (P = 0.059) to be greater for cattle receiving 3-NOP supplementation. Isobutyrate was greater (P = 0.001) for medium and high doses of 3-NOP, but there was no treatment effect on butyrate, valerate, isovalerate, or caproate (P ≥ 0.115). Acetate:propionate ratio was 30.5% less (P = 0.045) for low and medium doses compared with control. In addition, NH₃-N was not affected by treatment (P = 0.119).

Table 5.

Ruminal VFA and ammonia-nitrogen (NH₃-N) concentrations for a subset of steers (n = 4) fed a corn-based finishing diet without (control) or supplemented with low, medium, and high doses of 3-NOP

	Treatment1					P-value
Item	Control	Low	Medium	High	SEM	3-NOP
Total VFA, mM	133.0	144.1	104.8	139.7	9.79	0.06
VFA, mol/100 mol
Acetate (Ac)	47.11a	39.99b	42.24b	41.28b	0.850	<0.001
Propionate (Pr)	36.57	45.11	46.16	40.84	2.803	0.06
Butyrate	10.46	8.91	5.89	11.48	2.140	0.12
Valerate	2.45	3.01	2.31	2.75	0.391	0.61
Isobutyrate	1.01b	1.02b	1.38a	1.55a	0.131	0.001
Isovalerate	2.08	1.53	1.83	1.79	0.375	0.29
Caproate	0.32	0.42	0.19	0.31	0.072	0.21
Ac:Pr ratio	1.31a	0.90b	0.92b	1.06a	0.100	0.05
NH3-N, mM	7.49	6.15	3.34	6.04	1.795	0.12

¹The measured 3-NOP concentration was 0, 85.6, 107.6, and 124.5 mg/kg dietary DM consumed for the control, low, medium, and high treatment, respectively.

^a,bValues within a row with different letters differ at P ≤ 0.05.

Discussion

Numerous enteric CH₄ mitigation strategies have been proposed for ruminant production systems (Hristov et al., 2013; Beauchemin et al., 2020), but few have been evaluated in commercial operations and over entire feeding periods lasting several months. The study design allowed us to evaluate the effects of increasing levels of 3-NOP supplementation over a 112-d period on DMI and CH₄ production in an environment relevant to commercial feedlots, albeit with smaller pens (25 animals/pen) than used by most of the large commercial operations (>100 animals/pen).

The concentrations of 3-NOP in the basal diet delivered to the feed bunks were analyzed to ensure the target concentrations were achieved, which is particularly important in a dose titration study. The 8.8% to 11.2% lower than targeted concentrations of 3-NOP in the basal diet were consistent for all three levels of 3-NOP and may be attributed to mixing of the diet, sampling at the feed bunk, dispersion in the feed bunk, or extraction and measurement in the lab. Recovery of most feed additives typically ranges from 80% to 120%, thus the recovery of 89% to 91% of added 3-NOP would be acceptable against the label claim.

The 52.2% to 76.0% decrease in CH₄ yield, depending upon the 3-NOP dose, compared with the control diet indicates that 3-NOP is a highly effective mitigation agent in corn grain-based feedlot finishing diets. This is an important finding because the only other published study that used corn grain-based feedlot cattle diets reported no effect of 3-NOP (100 mg/kg DM) on CH₄ yield (Kim et al., 2019). The reasons for the lack of response to 3-NOP in that study relative to the observed decrease in CH₄ yield in the present study are unclear but may be related to the conditions used: whole shelled corn rather than steam-flaked corn, Latin square vs. completely randomized design, short duration of feeding period (21 vs. 112 d), and protocol for measuring CH₄ using the GEM system. In a longer-term study, Vyas et al. (2016a) fed cattle a finishing diet containing dry-rolled barley grain supplemented with 3-NOP and reported a 9% and 81% decrease in CH₄ yield for low (100 mg 3-NOP/kg DM) and high (200 mg 3-NOP/kg DM) doses, respectively. Vyas et al. (2018) also fed dry-rolled barley-based finishing diets and reported a 37% decrease in CH₄ yield for diets containing 3-NOP (125 mg/kg DM). In a study conducted at a commercial feedlot that used steam-flaked barley-based finishing diets, McGinn et al. (2019) reported a 70 ± 18% reduction in CH₄ yield for cattle fed 3-NOP (125 mg/kg DM). Combining the results of the present study with those from previously published studies for beef cattle fed finishing diets supplemented with 3-NOP revealed a linear response to targeted dose level (treatment means [n] = 14, P = 0.004, R² = 0.510; Figure 3), with considerable variation due to the conditions of the studies. Factors that may have contributed to this variability include: method used to measure CH₄ emissions (chambers, GEM, and micrometeorology), duration that cattle were fed 3-NOP (weeks vs. months), target dose of 3-NOP differing from measured dose, fiber content of the diet (greater fiber content reduces 3-NOP effectiveness; Dijkstra et al., 2018), source of grain (corn vs. barley), and grain processing. These factors require further study to improve the ability to predict the CH₄ decrease when using 3-NOP in beef cattle diets.

While 3-NOP clearly decreased CH₄ yield in the present study, the dose–response was surprisingly not linear. The medium dose was considered as optimum due to its significantly greater reduction in CH₄ yield and lack of effect on DMI. In a meta-analysis of data from beef and dairy studies, Jayanegara et al. (2018) reported a linear decrease in CH₄ (% of GEI) with an increasing target dose of 3-NOP added to feed (n = 29, P < 0.001, R² = 0.49). This linear relationship was confirmed in a meta-analysis of beef and dairy literature by Dijkstra et al. (2018) and is apparent in Figure 3 for beef cattle. For feedlot finishing diets, Vyas et al. (2016b) also reported a linear effect of 3-NOP dose between 100 and 200 mg/kg DM on CH₄ yield (maximum decrease of 45.2%). However, at lower doses (50 and 75 mg/kg), CH₄ yield actually increased slightly, indicating that there may be a minimum threshold dose needed in high-grain diets before a decrease in CH₄ yield occurs. The greater decrease in CH₄ yield over the experimental periods for the medium dose compared with low and high doses was not expected. Upon examination of the data for individual animals (Figure 4), there was considerably less variability among animals in the medium-dose group (cv = 39.6%), compared with the other treatment groups (cv ranging between 78.1% and 85.4%). With the low- and high-dose groups, there were some animals with relatively high emissions that did not appear responsive to 3-NOP, which accounts for the greater mean CH₄ yields for those treatment groups compared with the medium dose. 3-Nitrooxypropanol inhibits methanogens by binding to the active site of the methyl-coenzyme M reductase enzyme in the last step of methanogenesis (Duin et al., 2016), thus effectiveness may depend upon the resident methanogen community within individual animals. However, there is limited information regarding the changes in microbial community composition of animals fed 3-NOP (Martinez-Fernandez et al., 2018; Zhang et al., 2020), particularly in relation to individual animal responses to 3-NOP, which is an area that needs further study.

The efficacy of 3-NOP for CH₄ reduction declined for the low dose in period 3 indicating a possible rumen adaptation at low levels of supplementation, which is an important finding of the study. To ensure the long-term effects of 3-NOP were adequately assessed, we adopted stringent criteria in the data analysis of CH₄. Few beef cattle studies have examined temporal changes in CH₄ responses to 3-NOP. Romero-Perez et al. (2015) reported a 59% decrease in CH₄ yield for beef cattle fed restrictively a high forage diet supplemented with 3-NOP (280 mg/kg DM), with no decline in effectiveness when emissions were measured monthly over 112 d using respiratory chambers. However, McGinn et al. (2019) used micrometeorological techniques to measure pen-level emissions at a commercial feedlot using steam-flaked barley-based finishing diets supplemented with 3-NOP (125 mg/kg DM) and reported an average decrease in CH₄ production of 72% at the start of the study declining linearly to 61% after 90 d. No other long-term study with beef cattle has performed repeated measurements of CH₄ emissions over time to our knowledge, but a dairy cow study reported no decline in mitigation effectiveness of 3-NOP over 12 wk (Hristov et al., 2015; 40 to 80 mg 3-NOP/kg DM; 30% decrease in CH₄ emissions).

The lack of postprandial fluctuation in CH₄ emissions over the day for the cattle fed 3-NOP diets contrasts with the fluctuating emissions of control cattle (Figure 5) and has been reported previously for feedlot cattle using respiratory chambers (Vyas et al., 2018). With the GEM system, achieving an accurate assessment of diurnal pattern of emissions can be challenging because of the sporadic nature of the visits to the GEM (Gunter and Bradford, 2015). We overcame that limitation by using a strict criterion for the data analysis requiring that the animals visited in at least five of the six 4-h blocks corresponding to the time of day. The resulting average visitation of 5.84 (±0.06 SD) blocks in the study ensured an accurate assessment of the diurnal pattern of CH₄ production.

Methane represents the largest sink of H₂ in the rumen. Most of the H₂ produced in the rumen by the microbiota is transferred to methanogens and used in methanogenesis (Janssen, 2010); however, other H₂ using pathways such as formation of formate, propionate, valerate, caproate, and heptanoate, saturation of fatty acids, reduction of nitrate and sulfate, and microbial protein synthesis become more important when methanogenesis is inhibited (Guyader et al., 2017). Hydrogen gas can also accumulate in the rumen. For example, when methanogenesis was suppressed by feeding 3-NOP, H₂ emissions increased in dairy cows (Hristov et al., 2015) and beef cattle (Vyas et al., 2018). In the present study, the H₂ emissions of cattle fed 3-NOP were increased approximately 4.9-fold compared with control. 3-Nitrooxypropanol spared 85 to 117 g/d of CH₄, depending upon the dose of 3-NOP, which is equivalent to approximately 21 to 29 g/d of H₂ (4 mol of hydrogen atoms required for 1 mol of CH₄). The observed H₂ emission of 2.42 to 2.62 g/d for the cattle fed 3-NOP indicates that most (> 85%) of the spared H₂ from CH₄ mitigation was redirected to other sinks or toward dissolved H₂. The results of a meta-analysis indicated that the energy losses as H₂ resulting from CH₄ inhibition varied widely across experiments but were consistently much lower than the energy saved in CH₄ (Ungerfeld, 2019). In the present study, H₂ emissions due to 3-NOP supplementation declined over time, especially for the high concentration of 3-NOP (Table 4), indicating a possible adaptation of the ability of the rumen microbes to redirect H₂ to other alternate sinks. A similar decline in H₂ gas was observed after 6 wk of feeding 3-NOP to dairy cows (Hristov et al., 2015).

A decrease in DMI has been observed previously when feeding beef cattle diets containing 3-NOP (Romero-Perez, 2014; Vyas et al., 2016a, 2018). The 0% to 6.4% average decrease in total DMI (all animals) depending upon the level of 3-NOP in the present study is in agreement with those previous reports. The treatment × period interaction (Figure 1) revealed that the effects of 3-NOP dose on DMI varied over the study, with cattle fed low and high doses adapting their intakes after the first period. Lee et al. (2019) proposed that the decrease in DMI when feeding 3-NOP to beef cattle is caused by: 1) improved energy status as a result of decreased energy losses in the form of CH₄, 2) metabolic effects resulting from an increase in rumen propionate concentration, which when oxidized in the liver produces a hypophagic response (Allen, 2000), and 3) possible changes in organoleptic properties of the diet. Although we confirmed that 3-NOP decreased energy lost as CH₄ (% of GEI; 3.42, 1.63, 0.82, and 1.26 for control, low, medium, and high treatments, respectively), animal performance (growth and carcass characteristics) was not measured in the study. The observed decrease in acetate proportion and a tendency for increased propionate proportion in rumen fluid of cattle fed 3-NOP supports the theory that DMI may have been decreased by feeding 3-NOP due to a shift in metabolic precursors. Lee et al. (2019) reported that cattle rapidly acclimatize to 3-NOP (within 7 days), and that organoleptic factors likely only contribute to short-term decreases in DMI when feeding 3-NOP.

Conclusions

This 112-d study demonstrates that supplementing a steam-flaked corn-based finishing diet without and with low (85.6 mg/kg dietary DM), medium (107.6 mg/kg dietary DM), and high (124.5 mg/kg dietary DM) doses of 3-NOP under commercial feedlot conditions decreased CH₄ yield by 52% to 76%, depending upon the dose. The dose–response was quadratic, with a maximum decrease (76%) in CH₄ yield with the medium dose. Furthermore, the efficacy of the low dose of 3-NOP for CH₄ reduction declined in the final period of the study. Supplementing the diet with a low or high dose of 3-NOP decreased DMI by 6.1% or 6.4% overall, respectively, with the effect observed mainly in the initial period of the study. The observed increase in H₂ gas emissions and decrease in acetate molar proportion in rumen fluid confirm that other H₂-utilizing pathways become more important in the CH₄-inhibited rumen. Further studies conducted under commercially relevant conditions are needed to determine the effects of the optimum dose of 3-NOP on performance and carcass characteristics of feedlot cattle.

Glossary

Abbreviations

ADF

acid detergent fiber

BW

body weight

CP

crude protein

CV

coefficient of variation

DM

dry matter

DMI

dry matter intake

GEM

GreenFeed emission monitoring

GEI

gross energy intake

3-NOP

3-nitrooxypropanol

NDF

neutral detergent fiber

OM

organic matter

VFA

volatile fatty acids

Conflict of interest statement

A.W.A., A.S., C.W.B., L.K.D.P., S.M.M., and K.A.B. declare no conflict of interest. M.K. is an employee of DSM Nutritional Products and the company supplied the 3-NOP product used in the study.