Odd-chain fatty acids as an alternative method to predict ruminal microbial nitrogen flow of feedlot Nellore steers fed grain-based diets supplemented with different nitrogen sources

Abstract

This study aimed to evaluate the use of total odd-chain fatty acids (OCFA) as a marker to estimate microbial nitrogen flow (MicN) and calculate the efficiency of microbial nitrogen synthesis (EMNS) in Nellore steers fed high-concentrate diets supplemented with different nitrogen supplements (NS). Ruminally and duodenally cannulated Nellore steers (n = 6; 354 ± 12 kg) were used in a 6 × 6 repeated switchback design balanced for residual effects. Treatments were arranged in a 3 × 3 factorial of three nitrogen (N) supplements (urea plus soybean meal; corn gluten meal; dried distillers’ grains plus solubles) and three microbial markers (OCFA; double-labeled urea, 15N; microbial nucleic acid bases, MNAB). The total mixed ration was composed of fresh chopped sugarcane as the forage source in an 83:17 concentrate: forage ratio (dry matter basis). Linear regression was used to develop predictions of MicN from OCFA using 15N and MNAB as response variables. Microbial N flow was underestimated by the MNAB marker compared to 15N. Neither NS nor their respective interactions with the marker methods (MM) affected MicN or EMNS (P > 0.05). However, MicN was different for 15N and MNAB (P > 0.001 for both treatments). Marker methods affected EMNS in all energetic bases (total digestible carbohydrates P < 0.001; rumen-fermentable carbohydrates P < 0.001; organic matter truly degradable in the rumen P < 0.001). Equations that utilized OCFA as a regressor to predict MicN under different MM resulted in good fits of the data as observed by the coefficient of determination (R²; 15N = 0.78; MNAB = 0.69). Microbial N flow estimated from OCFA was overpredicted (15N by 7.46%; MNAB by 4.30%) compared with observed values. The OCFA model presented a small slope bias when methodological validation was applied (15N = 0.96%; MNAB = 3.90%), ensuring reliability of the proposed alternative method. Based on the conditions of this experiment, OCFA may be a suitable alternative to other methods that quantify MicN under different dietary conditions.

Lay Summary

Nutritional strategies that maximize microbial nitrogen supply to the small intestine may improve cattle performance. Nevertheless, in vivo quantification generally requires sensitive or expensive methods and often yields highly variable results. In the present work, we investigated the use of duodenal odd-chain fatty acids (OCFA) as an alternative method to predict microbial nitrogen flow (MicN) and calculated its efficiency on different energetic bases under different dietary nitrogen supplements. We utilized total OCFA flow (TOCFAf) to predict MicN by two well-established conventional methods: 1) 15N, considered the gold standard and 2) microbial nucleic acid bases. Models presented a positive relationship between TOCFDf and response variables, and under validation, both demonstrated low estimation bias. Under the conditions of this experiment, OCFA appeared to serve as an alternative marker to quantify ruminal MicN for beef cattle.

Odd-chain fatty acids may be a suitable alternative marker method to quantify ruminal microbial nitrogen flow in beef cattle.

Introduction

Microbial nitrogen (MicN) synthesized in the rumen can supply 50% to 90% of the metabolizable protein required by beef cattle (Poppi and McLennan, 1995; NASEM, 2016). Ruminal MicN provides the animal with a highly metabolizable amino acid profile necessary for biological processes like tissue development, lactation, and fetal growth (Mariz et al., 2018a). Diet composition, particularly rumen degradable protein (RDP) content, ruminal energy availability, and dilution rate are considered essential factors to increase the efficiency of MicN synthesis (Clark et al., 1992; Dijkstra et al., 1998; Firkins et al., 2007). Further, nitrogen (N) recycling positively influences MicN efficiency when dietary RDP is limited (Poppi and McLennan, 1995; Silva et al., 2019).

Postruminal MicN flow and efficiency may be assessed using external markers like double-labeled urea (15N) or internal markers like microbial nucleic acid bases (MNAB), which are structural base components of DNA (Broderick and Merchen, 1992; Valadares et al., 2010; Liu et al., 2017). According to Rotta et al. (2014), 15N and MNAB are adequate marker methods (MM) to estimate MicN. However, analysis across a wide range of dietary nitrogen supplements (NS) is necessary to determine the most suitable MicN marker for a particular study. Despite the increasing use of MNAB, this marker still presents great variability of results (Del Valle et al., 2019). While 15N is considered a more reliable MM compared with purine-centric procedures (Reynal et al., 2005), its methodology requires constant rumen or jugular infusion plus an isotope ratio mass spectrometry analyzer (Reynal et al., 2005; Titgemeyer et al., 2012), rendering it a time-consuming and expensive method.

Odd-chain fatty acids (OCFA) are stable compounds of the bacterial phospholipid membrane. They have potential utility as internal markers because they are derived from microbes (Castro-Montoya et al., 2016), they are only found in trace amounts in most concentrates and forages (Diedrich and Henschel, 1990). However, if propionate is not directly supplemented or, in the case of dairy animals, administered via mammary gland perfusion (Massart-Leen et al., 1983, Berthelot et al., 2001, Vaz et al., 2006), OCFA may not be optimal for predicting microbial N synthesis. Castro-Montoya et al. (2016) evaluated milk OCFA as a predictor of purine derivative excretion in lactating cows. However, milk OCFA poorly explained variation in purine derivative excretion, potentially due to differences in individual microbial OCFA absorption rates, blood transport, and de novo synthesis in the mammary glands. Liu et al. (2017) observed a strong correlation between MNAB and OCFA in the rumens of dry and lactating dairy cows. They suggested that OCFA and C15:0 isomers could estimate MicN and the efficiency of microbial nitrogen synthesis (EMNS).

Keeney et al. (1962) suggested that OCFA sampled from the proximal duodenum may be a suitable alternative MM to predict MicN synthesis in cattle. To our knowledge, no studies evaluated OCFA as an internal MM to estimate MicN and EMNS in feedlot beef cattle fed high-concentrate diets with different dietary NS. Thus, this study aimed to evaluate OCFA as an alternative method to 15N and MNAB to estimate MicN and EMNS in feedlot Nellore steers fed high-concentrate diets supplemented with different NS. We hypothesized that OCFA may be a suitable internal MM to predict MicN and EMNS in feedlot Nellore steers. In addition, the effects of different NS on MicN and EMNS were evaluated.

Materials and Methods

The study was conducted at UNESP, Jaboticabal, São Paulo, Brazil. Animal care and handling procedures followed the Brazilian College of Animal Experimentation Guidelines (COBEA) and were approved by the Ethics, Bioethics, and Animal Welfare Committee of São Paulo State University – UNESP, Jaboticabal under Protocol 16.668/16.

Animals, experimental design, and diets

The experimental trial utilized six Nellore steers fitted with rumen (silicone; oval-shaped), proximal duodenum, and distal ileum cannulas (both silicone; T-shaped). Steers were approximately 18 mo of age with average body weights (BWs) of 354 ± 12 kg. Animals were randomly assigned to one of three dietary treatments in a repeated switchback design that spanned across six experimental periods. Each period consisted of 14 d of diet adaptation (Machado et al., 2016) followed by 9 d of infusions and sample collections (23 d per period). Animals were individually housed in partially covered metabolism pens (12 m² each) with concrete floors, feed bunks, and ad libitum water access. The total experiment length was 138 d.

Before starting the experiment, steers were vaccinated with a long-lasting antiparasitic agent (Ivermectin, 200 μg/kg BW; Boehringer Ingelheim Animal Health, Paulinia, SP, Brazil) and initial BWs were recorded after 16 h of total fasting. Steers also underwent a 4-wk adaptation period to a high-concentrate diet prior to the experimental period. Total mixed rations consisted of fresh chopped sugarcane plus concentrate in an 83:17 concentrate:forage ratio (dry matter basis; DM). Fresh sugarcane was chopped in a stationary chopper (TRF 400, TRAPP, Jaraguá do Sul, SC, Brazil) to a length of 8 mm and used as the primary roughage source. The base concentrate mixes were comprised of ground corn and a mineral mix (Table 1). Varying NS were then blended into the base concentrate mix to develop three treatment diets: 1) a urea and soybean meal blend (UR+SBM); 2) corn gluten meal (CGM); and 3) dried distillers’ grains plus solubles (DDGS). Nitrogen sources were chosen due to their differences in rumen undegradable protein (RUP) composition, which impacts microbial N synthesis. Diets were developed to be isoenergetic with contrasting RUP to investigate MM sensitivity across different NS. CGM and DDGS were chosen because they are relatively high in RUP, and UR+SBM was chosen because they are high in RDP. The concentrates were mixed according to the proposed composition of each diet in a mixer (MRO plus 500, MFW, São Paulo, SP, Brazil). The concentrates were mixed by hand with the chopped sugarcane in the feed bunks.

Table 1.

Ingredients and chemical compositions (% DM) of diets fed to feedlot steers in the study

Item	Treatments1
	UR+SBM	CGM	DDGS
Ingredient composition
Sugarcane, chopped	15.8	16.7	17.4
Ground corn	76.4	72.0	63.0
Soybean meal	4.12	-	-
DDGS2	-		17.1
Corn gluten meal	-	8.81	-
Urea	1.12	-	-
Mineral mix3	2.55	2.49	2.45
Chemical composition4
DM, %	78.0	78.6	78.9
OM, % DM	95.5	95.7	95.8
CP, % DM	13.3	12.8	12.5
NDF, % DM	23.1	25.9	30.0
EE, % DM	4.57	4.38	4.76
RDP, % CP	66.8	45.5	49.3
RUP, % CP	33.1	54.4	50.7

UR+SBM, urea + soybean meal; CGM, corn gluten meal; DDGS, dried distillers’ grains plus solubles.

Dried distiller’s grains plus solubles.

Mineral mix composition (per kg of DM): 220 g Ca, 20 g P, 60 g Na, 25 g S, 10 g Mg, 100 mg Co, 500 mg Cu, 50 mg I, 1,500 mg Zn, 9 mg Se, 1,500 mg Mn, 100,000 IU vitamin A, and 50 g sodium bicarbonate.

DM, dry matter; OM, organic matter; CP, crude protein; NDF, neutral detergent fiber; EE, ether extract; RDP, rumen degradable protein; RUP, rumen undegradable protein.

Dietary treatments were formulated to meet an average daily gain (ADG) of 1.25 kg/d according to requirements outlined by BR-Corte (Valadares et al., 2016). Animals were fed twice a day at 0600 and 1600 hours. Individual refusals were removed and weighed the following morning, and animals were fed ad libitum so that approximately 10% refusals (as-fed basis) remained each day. The proportion of ingredients and the chemical composition of the diets are shown in Table 1.

Digesta duodenal flow estimation

From days 18 to 20 of each experimental period, 800 mL samples of duodenal digesta were collected in plastic bags tied with elastic to the duodenal cannula, allowing free digesta flow into the bags. Duodenal samples were collected at 9 h intervals at 1030 and 1930 hours on day 18, at 0430, 1330, and 2230 hours on day 19, and at 0730, 1630, and 0130 hours on day 20 (Allen and Linton, 2007). After collection, total duodenal digesta (TDD) samples were divided into two subsamples: the first was used to estimate digesta flow (375 mL) and the second was used for isolation of the microbial fraction (425 mL). Samples used to estimate the flow of nutrients were immediately frozen after each collection at −20 °C. At the end of each experimental period, frozen samples were thawed, mixed, and pooled for each animal. Samples used to isolate microbial fractions were refrigerated (4 °C), pooled by animal and day, and subsequently centrifuged.

Isolation of the rumen and duodenal microbial fraction was carried out following methodology described by Reynal et al. (2005) with adaptations suggested by Krizsan et al. (2010) to estimate MicN. Pooled digesta samples (1 L) were filtered through a 100 μm nylon filter (44% surface pore area; Sefar Nitex 100/44, Sefar, Thal, Switzerland). The retained material was washed with saline solution (800 mL of 0.9% wt/vol NaCl) in a separate container, then stored for further isolation of solid-associated bacteria (PAB). The filtered liquid sample was processed to obtain the liquid-associated bacteria (LAB). After centrifugation (1,000 × g for 10 min at 5 °C), the pellet was stored for PAB isolation (as described later). The supernatant was centrifuged at 11,250 × g for 30 min at 5 °C. The supernatant was discarded and 200 mL of McDougall’s solution was added to the pellet. Then, samples were centrifuged at 16,500 × g for 20 min at 5 °C, and the resulting pellet composed of LAB was frozen (−80 °C), freeze–dried for 72 h, and ground through a 1-mm screen (Thomas Scientific, Swedesboro, NJ) for further analysis. For PAB isolation, 700 mL of saline solution containing 1% Tween-80 (v/v) was added to the vessels containing the pellet from the first centrifugation of LAB isolation. The solid phase was retained in the nylon filter, homogenized for 30 s, and refrigerated (4 °C) overnight to detach any bacteria attached to feed particles. The samples were then filtered through a nylon filter, the obtained liquid was centrifuged (1,000 × g for 10 min at 5 °C), and the obtained supernatant was centrifuged (11,250 × g; 30 min at 5 °C). McDougall’s buffer (200 mL) was added to the resulting pellet then centrifuged again at 16,250 × g for 20 min at 5 °C. The resulting pellet (PAB) was frozen (−80 °C) and freeze–dried for 72 h.

Chemical analysis

At the end of each experimental period, samples of TDD were freeze–dried. Samples of feed, refusals, and feces were dried at 55 °C for 72 h. Samples of refusals and feces were pooled proportionately to total fecal output or total refusals. All samples described above were ground in a Wiley mill (Thomas Scientific, Swedesboro, NJ) to pass through a 1- and 2-mm screen. Samples ground through a 1-mm screen were analyzed for DM (INCT-CA, number G-003/1), organic matter (OM; INCT-CA, number G-004/1), and ether extract (EE; Randall procedure; method INCT-CA number G-005/1) according to the standardized analytical practices described by INCT (INCT-CA; Detmann et al., 2012). Neutral detergent fiber (NDF) was analyzed following the methodology described by Van Soest et al. (1991) and adapted for the ANKOM 200 fiber analyzer (ANKOM Technology, Fairport, NY) using α-amylase, without sodium sulfite and expressed inclusive of residual ash. Total N was analyzed by the DUMAS method using Leco-FP 528 LC equipment (2013 LECO Corporation, St. Joseph, MI) according to the protocol described by Etheridge et al. (1998) and converted to crude protein (CP) using a factor of 6.25. Samples ground through a 2-mm screen were used for indigestible neutral detergent fiber (iNDF) determination as recommended by Valente et al. (2011) after in situ incubation for 288 h.

Microbial markers

Microbial nucleic acid bases

Analysis of total purines to determine MicN from MNAB was performed in PAB, LAB, and TDD according to Ushida et al. (1985) and adapted for perchloric acid hydrolysis of nucleotides (Makkar and Becker, 1999) and for the precipitation of free purines over silver nitrate (Obispo and Dehority, 1999).

¹⁵N¹⁵N-urea

Prior to marker infusion, 410 mL of a solution containing 2.6 g of double-labeled urea (¹⁵N¹⁵N-urea per L), required for 82 h was prepared per animal. The solution was prepared under sterile techniques in a laminar flow chamber, filtered on a bacteriological filter (0.22 μm; Millipore Corporation, Billeric, MA, USA) in a sterilized glass vessel, sealed with a sterile rubber septum, and stored at 4 °C until infusion. On day 17 of each period, a jugular catheter was placed in the steers for isotope infusion (Cambridge Isotope Laboratories, Andover, MA, 98% purity), which was used as an external marker to assess urea kinetics (Souza et al., 2021) and to estimate duodenal MicN flow. The catheter (BD Angiotech, 14-gauge, 133 mm; Becton Dickinson, Sandy, UT) was inserted into the vein by percutaneous venipuncture after skin disinfection using a 10% iodine solution. After catheter placement, 10 mL saline solution (0.9% NaCl and 10 IU heparin per mL) was infused every 6 h until 0600 hours on day 19 to avoid patency loss, at which time infusion of ¹⁵N¹⁵N-urea started. From days 19 (0600 hours) to 22 (1600 hours), the ¹⁵N¹⁵N-urea solution was continuously infused at a rate of 5.00 mL/h, allowing the delivery of 0.610 mmol labeled N-urea per h through an infusion pump (BS-9000 Multi-Phaser, Braintree Scientific Inc., Braintree, MA). On day 22 of each experimental period, ruminal and duodenal digesta samples (each 200 mL) were collected at 0, 2, 4, 6, 8, and 10 h after morning feeding (0600 hours) for determination of 15N enrichment of natural matter. The schedule used for sampling was developed according to Titgemeyer et al. (2012) and corresponded to 72 to 82 h of infusion, the period during which 15N isotopic enrichment theoretically reaches a plateau (Wickersham et al., 2009). Samples were frozen at −80 °C and subsequently freeze–dried for 72 h.

Fecal samples from days 18 and 21, duodenal digesta from day 22, and pooled ruminal bacteria from day 22 were analyzed for 15N isotopic enrichment as described by Rotta et al. (2014). Briefly, 15N atom excess was analyzed in an isotope ratio mass spectrometer (Delta S; Finnigan MAT, Bremen, Germany). Approximately 4 mg of sample was weighed and placed in 5 × 8 mm capsules. The ratio of stable N isotopes (15N:14N) was analyzed according to international standards as Δ per thousand and was converted to percentages of atoms in excess (APE, %).

Odd-chain fatty acids

OCFA were analyzed using a gas chromatography-mass spectrometer (GCMS; CG Model Shimatzu 20-10, with automatic injection; Shimatzu Corporation, Kyoto, Japan) with an SP-2560 column (100 m × 0.25 mm in diameter, 0.02 mm thickness; Supelco, Bellefonte, PA, USA). The analysis was performed on PAB, LAB, and TDD samples according to the methodology described by Palmquist and Jenkins (2003). Machine settings are described as follows: column temperature: stationary at 240 °C for 15 min; injector and detector temperature: 250 °C; column pressure: 318.6 kPa; internal standard: nonadecanoic acid methyl ester (C:19:0; Sigma-Aldrich); carrier gas: hydrogen. The OCFA compounds were identified by standards, co-injection, and relative retention time of the internal standard C:19:0.

The MicN was directly quantified using 15N and MNAB as MM, described earlier, due to their direct association with CP. However, OCFA is associated with the EE fraction due to its lipid extraction (Palmquist and Jenkins, 2003). As a result, MicN quantification directly from OCFA was not possible. To the authors’ knowledge, no correction factor exists in the literature that enables the conversion between EE and CP for MicN estimates. As such, linear regression was used to develop predictions of MicN from total OCFA flow (TOCFAf) using 15N and MNAB as response variables.

Calculations

Nutrient flow

Duodenal dry matter flow (DMf) was estimated using iNDF as a single marker calculated according to Harvatine and Allen (2006) equation 1:

$${\displaystyle \begin{array}{ll}{\displaystyle \mathrm{D}\mathrm{M}\mathrm{f}(\mathrm{k}\mathrm{g}/\mathrm{d})=}& \mathrm{F}\mathrm{e}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{N}\mathrm{D}\mathrm{F}(\mathrm{g}/\mathrm{k}\mathrm{g}\mathrm{D}\mathrm{M})\\ & /\mathrm{D}\mathrm{u}\mathrm{o}\mathrm{d}\mathrm{e}\mathrm{n}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{N}\mathrm{D}\mathrm{F}(\mathrm{g}/\mathrm{k}\mathrm{g}\mathrm{D}\mathrm{M}).\end{array}}$$ (1)

Duodenal flow of total carbohydrates (TCf; kg/d) was calculated according to the following equation 2.

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{T}\mathrm{C}\mathrm{f}(\mathrm{k}\mathrm{g}/\mathrm{d})=\mathrm{O}\mathrm{M}\mathrm{f}-(\mathrm{E}\mathrm{E}\mathrm{f}+\mathrm{C}\mathrm{P}\mathrm{f}),}\end{array}}$$ (2)

where $\mathrm{O}\mathrm{M}\mathrm{f},\mathrm{E}\mathrm{E}\mathrm{f},\mathrm{C}\mathrm{P}\mathrm{f}$ refer, respectively, to dietary OM, EE, and CP flow, calculated by the multiplication of DMf and concentration of each nutrient on the true digesta (kg/d).

Bacterial and duodenal 15N enrichments were corrected from fecal background samples collected before 15N infusion. Microbial N flow of 15N (MicN_15N; g/d) was calculated by multiplying duodenal N flow and 15N APE ratio according to equations 3 and 4:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{P}\mathrm{E}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}15\mathrm{N}\mathrm{\%}=[(\mathrm{P}\mathrm{A}\mathrm{B}15\mathrm{N}+\mathrm{L}\mathrm{A}\mathrm{B}15\mathrm{N})/2]/\mathrm{T}\mathrm{D}\mathrm{D}15\mathrm{N},}\end{array}}$$ (3)

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{M}\mathrm{i}\mathrm{c}{\mathrm{N}}_{15\mathrm{N}}\left(\mathrm{g}/\mathrm{d}\right)=\mathrm{N}\mathrm{f}\left(\mathrm{k}\mathrm{g}/\mathrm{d}\right)\ast 15\mathrm{N}\mathrm{A}\mathrm{P}\mathrm{E}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\left(\mathrm{\%}\right),}\end{array}}$$ (4)

where PAB15N is the 15N enrichment on PAB; LAB15N is the 15N enrichment on LAB (APE); TDD15N is the 15N enrichment on duodenal digesta (APE); and Nf is the duodenal nitrogen flow.

Nitrogen flow utilizing MNAB (MicN_MNAB; g/d) was calculated from the ratio between the N-RNA flow in the duodenal digesta and the average N content in the bacterial pellet of PAB and LAB, using equations 5 and 6:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{N}\mathrm{R}\mathrm{N}\mathrm{A}\mathrm{f}\left(\mathrm{g}/\mathrm{d}\right)=\mathrm{D}\mathrm{M}\mathrm{f}\left(\mathrm{k}\mathrm{g}/\mathrm{d}\right)\ast \mathrm{N}\mathrm{R}\mathrm{N}\mathrm{A}(\mathrm{g}/\mathrm{d}),}\end{array}}$$ (5)

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{M}\mathrm{i}\mathrm{c}{\mathrm{N}}_{\mathrm{M}\mathrm{N}\mathrm{A}\mathrm{B}}\left(\mathrm{g}/\mathrm{d}\right)=\mathrm{N}\mathrm{R}\mathrm{N}\mathrm{A}\mathrm{f}/[(\mathrm{N}\mathrm{P}\mathrm{A}\mathrm{B}+\mathrm{N}\mathrm{L}\mathrm{A}\mathrm{B})/2],}\end{array}}$$ (6)

where DMf is the duodenal dry matter flow; NRNA is the nitrogen content of true RNA; NPAB is the nitrogen content on PAB; NLAB is the nitrogen content on LAB.

Linear regressions were developed for each OCFA (C11, C13, C15, C15:1, C17, C17:1, and C21) to create the calibration curve. Concentrations of OCFA were corrected by duodenal DMf followed by the sum across each respective OCFA. Then, total OCFA flow (TOCFAf; g/d) was calculated by adjusting each OCFA’s concentration by the average of its respective LAB and PAB. Total OFCA flow (TOCFAf; g/d) of LAB was obtained according to equations 7 and 8:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{O}\mathrm{C}\mathrm{F}\mathrm{A}\mathrm{f}(\mathrm{g}/\mathrm{d})=\mathrm{D}\mathrm{M}\mathrm{f}(\mathrm{k}\mathrm{g}/\mathrm{d})\ast \mathrm{\%}\mathrm{A}\mathrm{G}\mathrm{C}\mathrm{I}\mathrm{D}\mathrm{D}\mathrm{T},}\end{array}}$$ (7)

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{O}\mathrm{C}\mathrm{F}\mathrm{A}\mathrm{f}\left(\mathrm{g}/\mathrm{d}\right)=\mathrm{O}\mathrm{C}\mathrm{F}\mathrm{A}\mathrm{f}/[(\mathrm{O}\mathrm{C}\mathrm{F}\mathrm{A}\mathrm{P}\mathrm{A}\mathrm{B}+\mathrm{O}\mathrm{C}\mathrm{F}\mathrm{A}\mathrm{L}\mathrm{A}\mathrm{B})/2].}\end{array}}$$ (8)

Energetic bases

Total digestible nutrients (TDN) were calculated according to Weiss (1999) equation 9:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{T}\mathrm{D}\mathrm{N}(\mathrm{k}\mathrm{g}/\mathrm{d})=\mathrm{C}\mathrm{P}\mathrm{d}+\mathrm{N}\mathrm{D}\mathrm{F}\mathrm{d}+\mathrm{N}\mathrm{F}\mathrm{C}\mathrm{d}+(2.25\ast \mathrm{E}\mathrm{E}\mathrm{d}),}\end{array}}$$ (9)

where CPd, NDFd, NFCd, and EEd refer to the digestible fraction of dietary CP, NDF, non-fiber carbohydrates, and EE, respectively.

Rumen fermentable carbohydrates (RFC) were calculated by multiplying the total carbohydrate intake by ruminal digestibility of total carbohydrates according to equations 10–12:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{I}\mathrm{T}\mathrm{C}(\mathrm{k}\mathrm{g}/\mathrm{d})=\mathrm{O}\mathrm{M}\mathrm{c}-(\mathrm{C}\mathrm{P}\mathrm{c}+\mathrm{E}\mathrm{E}\mathrm{c}),}\end{array}}$$ (10)

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{R}\mathrm{D}\mathrm{T}\mathrm{C}\mathrm{\%}=\mathrm{I}\mathrm{T}\mathrm{C}-\mathrm{T}\mathrm{C}\mathrm{f}/\mathrm{I}\mathrm{T}\mathrm{C}\ast 100,}\end{array}}$$ (11)

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{R}\mathrm{F}\mathrm{C}(\mathrm{k}\mathrm{g}/\mathrm{d})=\mathrm{I}\mathrm{T}\mathrm{C}\ast \mathrm{R}\mathrm{D}\mathrm{T}\mathrm{C}/100,}\end{array}}$$ (12)

where OMc, CPc, and EEc refer to OM, CP, and EE consumed; TCf is the duodenal total carbohydrate flow.

Organic matter truly degradable in the rumen (OMTD) was calculated according to equation 13:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{O}\mathrm{M}\mathrm{T}\mathrm{D}(\mathrm{k}\mathrm{g}/\mathrm{d})=\mathrm{O}\mathrm{M}\mathrm{c}-(\mathrm{O}\mathrm{M}\mathrm{f}+\mathrm{O}\mathrm{M}\mathrm{m}\mathrm{i}\mathrm{c}),}\end{array}}$$ (13)

where OMc, OMf and OMmic represent OM consumed, flow, and microbial flow, respectively.

Efficiency of microbial nitrogen synthesis

EMNS was calculated by the ratio between MicN and each energetic base (i.e., TDN, RFC, and OMTD), according to equation 14:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{E}\mathrm{M}\mathrm{N}\mathrm{S}\left(\mathrm{g}/\mathrm{k}\mathrm{g}\right)=\mathrm{M}\mathrm{i}\mathrm{c}\mathrm{N}(\mathrm{g}/\mathrm{d})/\mathrm{T}\mathrm{D}\mathrm{N},\mathrm{R}\mathrm{F}\mathrm{C},\mathrm{o}\mathrm{r}\mathrm{O}\mathrm{M}\mathrm{T}\mathrm{D}(\mathrm{k}\mathrm{g}/\mathrm{d}).}\end{array}}$$ (14)

Statistical analysis

Statistical analyses were performed using R Studio version 3.5.1 (R Core Team, 2018). Data were tested using mathematical assumptions for normality and homoscedasticity utilizing the Shapiro–Wilk and Bartlett’s tests, respectively.

The lme4 (Bates et al., 2014) and Nlme (Pinheiro et al., 2018) libraries in R Studio were used to fit linear, quadratic, and cubic regressions to model relationships of MicN from 15N and MNAB on TOCFAf (g/d). Based upon the coefficient of determination (R²) and residual standard deviation (Sy.x), the statistical method that best fit the data was a simple linear regression, according to the model described later (equation 15):

$${\displaystyle \begin{array}{l}{\displaystyle Y_i={\beta }_0+{\beta }_1{x}_i+e_i.}\end{array}}$$ (15)

where $Y_i$ represents the MicN (15N and MNAB; g/d), of the i^th level of tested fixed effects; ${\beta }_0$ and ${\beta }_1$ are the parameters to be estimated for each of the i^th level of tested fixed effects (i.e., NS and MM); x_i regressor refers to TOCFAf (independent variable); $e_i$ is the residual error ~ N (0, o²).

To quantify how well the regression lines from the candidate models fit the data, the standard deviation of the residuals (Sy.x) was calculated, according to equation 16:

$${\displaystyle \begin{array}{l}{\displaystyle Sy.x\surd [(residua{l}^2)/(n-K)],}\end{array}}$$ (16)

where $K$ is the number of parameters fit in the regression. The value $n-K$ is the number of degrees of freedom in the regression. Reliability tests verifying the model’s accuracy and precision were further evaluated based on Bland–Altman Analysis (blandr library; Datta, 2017).

Nitrogen sources and MM effects were evaluated considering a repeated switchback design (Tempelman, 2004) in a factorial arrangement (A × B). The fixed effects of factor A corresponded to the NS in the diet (UR+SBM, CGM, and DDGS) and factor B corresponded to the MM (15N, MNAB, and OCFA), factors interactions (A × B), and treatment error. Experimental period, animal, and treatment × animal interaction were considered random effects, as well as the candidate model residuals. Tukey’s post hoc test (agricolae library; de Mendiburu, 2021) was applied when analysis of variance indicated a significant difference between means. Select animal performance measures were analyzed by multiple comparison of means (emmeans library; Lenth al., 2019) between frequency of dependent variables of interest using Tukey’s method (multcomp library; Hothorn et al., 2008). Statistical significance was declared at P ≤ 0.05.

Results

Models predicting MicN

Linear regression equations were developed to predict the relationship between MicN15N (R² = 0.78) and MicNMNAB (R² = 0.69) against TOCFAf (equations 17 and 18):

$${\displaystyle \begin{array}{l}{\displaystyle {\mathrm{M}\mathrm{i}\mathrm{c}\mathrm{N}}_{15\mathrm{N}}\left(\mathrm{g}/\mathrm{d}\right)=1.328\ast \mathrm{T}\mathrm{O}\mathrm{C}\mathrm{F}\mathrm{A}\mathrm{f}\left(\mathrm{g}/\mathrm{d}\right)+0.07619,}\end{array}}$$ (17)

$${\displaystyle \begin{array}{l}{\displaystyle {\mathrm{M}\mathrm{i}\mathrm{c}\mathrm{N}}_{\mathrm{M}\mathrm{N}\mathrm{A}\mathrm{B}}\left(\mathrm{g}/\mathrm{d}\right)=0.8909\ast \mathrm{T}\mathrm{O}\mathrm{C}\mathrm{F}\mathrm{A}\mathrm{f}\left(\mathrm{g}/\mathrm{d}\right)-8.858.}\end{array}}$$ (18)

Figure 1 shows the positive linear relationship between predicted MicN_15N (Graph A; R² = 0.78), MicN_MNAB, and (Graph B; R² = 0.69) TOCFAf. Figure 2 represents the method validation of the proposed models, where a positive linear relationship between observed and predicted MicN_15N and MicN_MNAB was detected.

Figure 1.

Relationship between total odd-chain fatty acid flow (TOCFAf; g/d) and microbial nitrogen flow (MicN; g/d) calculated under distinct conventional marker methods. The positive linear relationship between TOCFAf and MicN under the double-labeled urea (15N) method (A; R² = 0.78). The positive linear relationship between TOCFAf and microbial nucleic acid bases (B; R² = 0.69). Solid points, solid lines, and dashed lines represent observed MicN, the regression line for predicted MicN, and 95% confidence interval, respectively.

Figure 2.

Relationship between observed double-labeled urea microbial nitrogen flow (MicN15N) and microbial nitrogen flow (MicN) predicted by equation 1 (A; slope bias = 0.96) and relationship between observed microbial nucleic acid bases (MNAB) microbial nitrogen flow (MicNMNAB) and MicN predicted by equation 2 (B; slope bias = 3.90). Figure 2 compares both predicted and observed values for MicN and outlines the positive linear relationship between both methods. Solid points, solid lines, and dashed lines represent observed MicN, the regression line for predicted MicN, and standard error, respectively.

Slope bias or tested methodology error for equation 17 was 0.96% and 3.90% for equation 18. After the odd-chain fatty acids flow adjustments, results were compared between the proposed treatments (i.e., NS and MM; Table 3).

Table 3.

Observed vs. predicted means of microbial nitrogen flow and efficiency using observed MNAB or 15N compared with MNAB or 15N predicted from odd-chain fatty acids in feedlot steers supplemented with different nitrogen sources¹

Item1	15N2						MNAB3
	Observed			Predicted			Observed			Predicted			SE4	P-value
	UR+SBM5	CGM6	DDGS7	UR+SBM5	CGM6	DDGS7	UR+SBM5	CGM6	DDGS7	UR+SBM5	CGM6	DDGS7		MM8	NS9	MM×NS10
MicN	89.8	93.6	99.5	108.4	98.2	99.2	66.8	51.5	52.5	63.8	57.0	57.7	4.49	<0.001	0.66	0.69
EMNS, g/kg
TDN	22.7	24.5	26.2	24.3	26.0	23.1	15.2	13.4	13.0	14.1	15.5	13.3	0.99	<0.001	0.89	0.87
RFC	32.9	34.2	35.1	37.1	36.3	30.8	23.5	18.5	17.2	21.8	20.8	17.8	1.73	<0.001	0.87	0.97
OMTD	22.6	25.9	26.1	25.2	26.1	22.9	15.4	14.2	12.4	14.7	14.8	13.2	1.20	<0.001	0.99	0.97

MicN, microbial nitrogen flow; EMNS, efficiency of microbial nitrogen synthesis; TDN, total digestible nutrients; RFC, rumen-fermentable carbohydrates; OMTD, organic matter truly degradable in the rumen.

15N, double-labeled urea.

MNAB, microbial nucleic acid bases.

SE, standard error.

UR+SBM, urea + soybean meal.

CGM, corn gluten meal.

DDGS, dried distillers’ grains plus solubles.

MM, microbial markers: observed 15N and observed MNAB.

NS, dietary nitrogen sources.

MM × NS, interaction between microbial markers and dietary nitrogen source.

Observed vs. predicted MicN and efficiency

There were no differences between any treatment for any performance measure (Table 2; P > 0.05 for DMI, BW, DMI% BW, and ADG). There were no interactions between NS and MM (P > 0.05) for any evaluated parameter. Additionally, NS affected neither MicN (P = 0.66) nor EMNS in any energetic base (TDN, RFC, OMTD P = 0.89; P = 0.87; P = 0.99, respectively; Table 3).

Table 2.

Select performance measures of feedlot steers supplemented with different nitrogen sources (n = 6)

Item	Treatments1
	UR+SBM	CGM	DDGS	SEM	P-value
DMI2, kg/d	5.85	5.34	4.87	0.404	0.25
BW3, kg	366	365	364	9.90	0.98
DMI4, % BW	1.61	1.47	1.34	0.116	0.28
ADG5, kg	0.511	0.381	0.743	0.443	0.66

UR+SBM, urea + soybean meal; CGM, corn gluten meal; DDGS, dried distillers’ grains plus solubles.

Dry matter intake.

Body weight.

Dry matter intake as a percentage of BW.

Average daily gain.

MM differed in both MicN and EMNS (P < 0.01) in all energetic bases. MicN predicted by OCFA was overestimated by 4.3% compared with observed MNAB (observed $\overline{X}=56.9$; predicted $\overline{X}=59.5$) and overestimated by 7.46% compared with observed 15N (observed $\overline{X}=94.3$; predicted $\overline{X}=101.9$). Microbial N was not affected (P > 0.05) by dietary NS and results were consistent in both MM (Table 3). EMNS predicted by OCFA was slightly overestimated compared with observed MNAB based upon the overall averages for EMNS in all calculated energetic bases, such as TDN (observed $\overline{X}=13.9$; predicted $\overline{X}=14.3$), RFC (observed $\overline{X}=19.7$; predicted $\overline{X}=20.1$), and OMTD (observed $\overline{X}=14.0$; predicted $\overline{X}=14.2$). Additionally, EMNS predicted by OCFA had similar predictions across different energetic bases such as TDN (observed $\overline{X}=24.5$; predicted $\overline{X}=24.5$) and OMTD (observed $\overline{X}=24.9$; predicted $\overline{X}=24.7$) when compared with observed 15N. Analysis by each ingredient demonstrated EMNS predictions by OCFA were overestimated for both 15N and MNAB MM, except for treatments containing DDGS on 15N and UR+SBM on MNAB, where the model underestimated EMNS (Table 3).

Post hoc analyses comparing MM are described in Table 4. There is a challenge when performing post hoc tests on predicted MicN_15N and MicN_MNAB. Since both originated from developed equations, it is only possible to compare them once the orthogonal contrast is applied. As such, the predicted MicN must be analyzed only according to the respective observed MicN to allow a direct comparison. This means predicted MicN_15N and MicN_MNAB must be compared with observed MicN_N15 and MicN_MNAB, respectively, since 15N and MNAB were calculated in distinct ways. Markers did not differ when the experimental values of MicN and EMNS (all tested energy bases) were compared with model predictions (P > 0.05; Table 4).

Table 4.

Tukey’s post hoc test on marker method means of feedlot steers supplemented with different nitrogen sources

Item1		15N		MNAB
		Obs2	Pred3	Obs2
MicN, g/d, P-value
MNAB	Obs2	<0.01	<0.01	-
	Pred3	<0.01	<0.01	0.99
15N	Obs2	-	0.85	-
EMNS, g/kg, P-value
TDN
MNAB	Obs2	<0.01	<0.01	-
	Pred3	<0.01	<0.01	0.99
15N	Obs2	-	0.99	-
RFC
MNAB	Obs2	<0.01	<0.01	-
	Pred3	<0.01	<0.01	0.97
15N	Obs2	-	0.98
OMTD
MNAB	Obs2	<0.01	<0.01	-
	Pred3	<0.01	<0.01	0.99
15N	Obs2	-	0.99	-

MicN, microbial nitrogen flow; MNAB, microbial nucleic acid bases; 15N, double-labeled urea; EMNS, efficiency of microbial nitrogen synthesis; TDN, total digestible nutrients; RFC, rumen-fermentable carbohydrates; OMTD, organic matter truly degradable in the rumen.

Obs, observed data.

Pred, predicted data.

Discussion

Overview and limitations

Information presented in this study sheds light on the possibility of using OCFA as an alternative microbial MM to estimate MicN and EMNS in cattle, which is notable due to the lack of information regarding OCFA use in the existing literature. Limitations of our dataset included low feed intakes and ADG (Table 2). These variables were lower than expected for animals under normal feedlot conditions. However, it is important to note that the animals had three cannulas (rumen, duodenum, and ileum) necessary for digesta collections. Although we did not perform collections of ileal digesta in this study, the cannula is still considered a stressor to the animals. Additionally, animals were receiving jugular and ruminal infusions and were submitted to multiple collections throughout the experimental periods. Therefore, a reduction in intake and ADG was expected. Indeed, our DM intake results are similar to the average reported in two meta-analyses that evaluated urea kinetics in beef cattle with similar BW. Souza and White (2021) reported an average DM intake of 5.72 kg/d, and Eisemann and Tedeschi (2016) reported 6.0 kg/d. However, we believe treatment responses with the observed intakes can still validate whether OCFA is a useful MM within the confines of our study.

Nitrogen supplements

Sugarcane and ground corn comprised the largest fraction of each treatment because they are common ingredients used in Brazilian feedlots (Table 1; Ferraz and Felicio, 2010). Millen et al. (2009) described fresh chopped sugarcane as the primary roughage source used by Brazilian feedlot nutritionists. Nutritionally, sugarcane slows the passage rate in the gastrointestinal tract and provides necessary structural carbohydrates (Freitas et al., 2006). A stable rumen environment was essential for NS to be compared. Both sugarcane and corn afford satisfactory amounts of fermentable carbohydrates (Freitas et al., 2006; Nunes et al., 2020) that provide energy substrate for rumen microorganisms, and contribute to an ideal rumen environment (i.e., desirable pH for proteolytic bacteria). Similarly, MicN and EMNS levels among the sources of N signaled a balanced rumen environment in our study. This may be due to recycled excess N, which could have kept N supply constant in the case of NS with lesser ruminal degradability such as CGM and DDGS (Wickersham et al., 2008). The results of MicN in this study suggested that tested NS provided ruminal NH₃-N concentrations above 5 mg/dL, which has been described as the minimum value requirement for optimal ruminal MicN synthesis (Firkins et al., 2007; Souza et al., 2021).

MicN and efficiency

Roman-Garcia et al. (2016) reported that DM intake was the predominant variable that demonstrated a strong, direct, and positive relationship with MicN as well as a moderate, indirect, and positive relationship with EMNS. A high correlation was found between MicN and EMNS (i.e., heavier cattle, greater intake, greater MicN, and a potentially greater EMNS). According to our findings, both observed and predicted MicN from our tested MM (15N and MNAB) align with the literature when adjusted for animal MicN, DM intake, and BW (Mariz et al., 2018b).

Although the EMNS followed a positive and linear response with MicN in all MM and energetic bases tested, an underestimation was observed in our models when compared with those predicted by BR-Corte (Valadares et al., 2016) and observed by Rotta et al. (2014). This may explain lower DM intake from steers in this study. Further, De Tonissi et al. (2003) mentioned changes in intake behavior could be observed when utilizing beef cattle as the main experimental unit in metabolism studies. Due to the experimental invasiveness of this study, DM intake may have been compromised, which consequently affected EMNS (De Tonissi et al., 2003).

Microbial marker comparison

The statistical differences observed between 15N and MNAB on MicN and EMNS disagree with previous literature (Carro and Miller, 2002; Mariz et al., 2018a). This was an expected finding once calculations to obtain observed MicN_15N and MicN_MNAB rates were done according to respective sampling sites. Our findings suggest that MNAB is an established conventional MM but susceptible to high variation, which directly interferes with the interpretation of results. Del Valle et al., 2019 observed MicN overestimation utilizing MNAB relative to 15N, while Askar et al. (2005) observed underestimations of MicN using MNAB, further suggesting the variability of MNAB as an MM.

Total OCFA appeared to be a good predictor of MicN flow, and consequently of EMNS, mainly when 15N was used as an MM predictor (MicN_15N R² = 0.78) compared with MNAB (MicN_MNAB R² = 0.69). The R² observed by our predictions are similar to Vlaeminck et al. (2005), who tested OCFA and MNAB as MM in regression models in lactating dairy cows. The latter developed two main equations for the respective prediction with a similar coefficient of determination as presented in this study (equation 1_R2 = 0.72; equation 2_R2 = 0.78). Beyond the R², our methodologic validation demonstrated low slope bias from both MicN_15N ($e$ = 0.96%) and MicN_MNAB ($e$ = 3.90%), which was even smaller when compared with Vlaeminck et al. (2005) ($e$ = 5.39%). OCFA appeared to be a reliable marker within the conditions of our study, but postruminal modifications of OCFA such as hindgut de novo synthesis of OCFA, or fatty acid chain elongation in duodenal epithelial or liver cells profile may occur. As such, endogenous OCFA production could limit the utility of OFCA when estimating microbial N. However, postruminal changes in OCFA appear to be partially due to dietary characteristics (Vlaeminck et al., 2015). As such, corrections may be applied to biomarker models when dietary factors are anticipated to impact endogenous OCFA production (such as the depression of de novo fatty acid synthesis in cattle consuming high concentrate diets). In order to develop any such correction factors, the magnitude of diet-induced endogenous OCFA modification warrants further research.

Double-labeled urea is considered the “gold standard” for MM evaluation because of its low error yield (Broderick and Merchen, 1992). However, unlike the 15N “gold standard,” OCFA are internal markers already present in the ruminant gastrointestinal tract. Therefore, jugular or ruminal infusions are not necessary with OCFA. Gas chromatography analysis, utilized by the OCFA method, is a commonly performed analysis in nutrition laboratories compared with isotope ratio mass spectrometry analysis needed for the 15N method. Few laboratories can analyze isotopes in Brazil. Further, 15N analyses are costly and laborious. To the authors’ knowledge, no studies have been published comparing the “gold standard” method to OCFA.

One benefit of OCFA compared with MNAB is the more precise execution of the OCFA experimental method. By carefully following the steps described by Palmquist and Jenkins (2003), the fatty acid extraction can be performed with no significant stability concerns. On the other hand, MNAB methodology contains several steps that can be flagged as unstable, especially because RNA can be easily damaged (Popova et al., 2010). Nucleotide hydrolysis, such as purine precipitation, requires close attention while executing. A simple mistake during sample preparation will directly interfere with sample absorbance analysis, resulting in bias on the reading values. Lastly, the MNAB procedure requires more extensive labor, and needs a vast selection of lab glassware and reagents compared with OCFA analysis.

Conclusion

Under the conditions of this experiment, OCFA demonstrated to be a reliable method to quantify MicN and EMNS and may be capable of replacing other conventional MM. Models developed in the present study may be utilized in future research studies under similar experimental conditions. Further work to explore the robustness of the models should include validation using external data.

Glossary

Abbreviations

BW

body weight

CGM

corn gluten meal

CP

crude protein

DDGS

dried distillers’ grains plus solubles

DM

dry matter

DMf

duodenal dry matter flow

EE

ether extract

EMNS

efficiency of microbial nitrogen synthesis

iNDF

indigestible neutral detergent fiber

LAB

liquid-associated bacteria

MicN

microbial nitrogen

MM

marker methods

MNAB

microbial nucleic acid bases

MP

metabolizable protein

NDF

neutral detergent fiber

NS

nitrogen supplements

OCFA

odd-chain fatty acids

OCFAf

odd-chain fatty acid flow

OMTD

organic matter truly degradable in the rumen

RDP

ruminal degradable protein

RFC

rumen-fermentable carbohydrates

RUP

rumen undegradable protein

PAB

solid-associated bacteria

SBM

soybean meal

TDD

total duodenal digesta

TDN

total digestible nutrients

TOCFAf

total OFCA flow

Acknowledgments

We are thankful to São Paulo Research Foundation (FAPESP) for financial support (grant 2017/08854-0; 2016/22022-4; 2016/16796-7; 2017/02034-0).

Conflict of interest statement

The authors have no conflicts of interest to disclose.