Influence of nutritional management prior to adaptation to a feedlot diet on ruminal microbiota of Nellore cattle

ABSTRACT The objective of this study was to evaluate the effect of either a limited forage intake or concentrate supplementation prior to the adaptation to high-concentrate diets on dry matter intake, ruminal pH, bacteria, and protozoa of Nellore cattle. The experiment was designed as a two 3×3 Latin square, and six cannulated Nellore steers were used. Each experimental period was composed by three feeding phases: pre-adaptation (14 days), adaptation (12 days), and finishing (seven days) diet, in a total of 33 days per period. The steers were assigned to one of three pre-adaptation dietary treatments: control (Tifton hay fed ad libitum + mineral supplement), restriction (Tifton hay fed at 1.4% of BW + mineral supplement), and concentrate (Tifton hay fed ad libitum + 0.5% of BW of a mix of concentrate feedstuffs and mineral supplement). The adaptation period consisted of two adaptation diets, which contained 72 and 79% concentrate for six days each. The finishing diet contained 86% concentrate. During the pre-adaptation phase, restricted cattle had higher pH than concentrate-fed cattle. There was a reduction in M. elsdenii relative population in cattle from either restriction or concentrate groups. During adaptation and finishing phases, cattle from concentrate group had smaller F. succinogenes populations compared with the control group. The previous nutritional backgrounds impact ruminal microbiota during adaptation and finishing phases without causing any negative effect on ruminal pH. Feeding concentrate prior to the adaptation positively impacted the transition to high-concentrate diets and promoted increased dry matter intake.


Introduction
The Brazilian cattle industry is characterized by animals finished in pastures, but to attend the consumer market demand in recent years, the beef industry has changed by increasing the number of cattle finished in feedlots. However, transitioning cattle from pasture to a feedlot requires adjustments, because the diets consumed before feedlot arrival are typically forage-based. Therefore, some processes of adapting ruminal microorganisms to effective use of readily fermentable carbohydrate are necessary to avoid metabolic disorders (Millen et al., 2016;Rigueiro et al., 2021). Furthermore, this transition results in ruminal fermentation and microbial changes, as well as the enlargement of the ruminal epithelium to accommodate the increase of short-chain fatty acid production (Bevans et al., 2005).
Several studies were conducted previously to determine the most appropriate adaptation period for Nellore cattle receiving high-concentrate diets in Brazilian feedlots Watanabe et al., 2022), and authors reported that Nellore cattle should not be adapted in less than 14 days. However, in all of these studies, the cattle went through a ten-day receiving period, where they consumed a forage-based diet ad libitum to standardize the ruminal microbiota. However, Brazilian feedlots receive cattle whose previous nutritional history is unknown. Feedlot operations in Brazil commonly receive cattle from either grazing system supplemented with concentrate feedstuffs or grazing systems that typically are unable to support maintenance requirements because of the poor quality of tropical grasses during the dry season . Besides, Silvestre and Millen (2021), in a survey with Brazilian feedlot nutritionists, reported that 2.78% of the interviewed nutritionists did not adopt any reception program, and that cattle start on adaptation diet without a period to suppress possible carryover effects of a previous nutritional background.
Thus, it was hypothesized that cattle maintained under nutritional restriction or grazing with concentrate feedstuffs during the pre-adaptation period would have different ruminal microbiota, affecting the rumen fermentation patterns and animal performance. The present study was conducted to evaluate the effect of either a limited forage intake or concentrate supplementation prior to the adaptation to high-concentrate diets on dry matter intake (DMI), ruminal microbiota, and pH.

Material and Methods
All procedures involving the use of animals in this study were in agreement with the guidelines of Nacional Council of Animal Control and Experimentation (CONCEA) and were approved by the local Ethical Committee for Animal Research (protocol number 20/2016 -06/15/2016). The experiment was carried out in Dracena, São Paulo, Brazil (21°29' S, 51°52' W, 421 m).

Animals, treatments, and management
Six 20-month-old yearling cannulated Nellore bulls (236±20 kg) were randomly assigned to a replicated 3×3 Latin square design. Each experimental period was composed by three feeding phases: pre-adaptation (14 days), adaptation (12 days), and finishing (seven days) diet, in a total of 33 days per period. Animals were randomly distributed into Latin squares according to the type of diet (Table 1) provided in the pre-adaptation period, which represented the treatments: control (Tifton hay fed ad libitum plus a supplement), restriction (Tifton hay fed at 1.4% of body weight plus a supplement), and concentrate (Tifton hay fed ad libitum plus 0.5% of body weight of a mix of concentrate feedstuffs and supplement). After the pre-adaptation phase, the diets were the same for all animals (adaptation and finishing diets). Likewise, cattle were submitted to a seven-day washout between periods.
The adaptation phase consisted of two adaptation diets, which contained 72 and 79% concentrate offered ad libitum for six days each. The finishing diet, contained 86% concentrate and was offered for seven days. All the diets were composed of sugarcane bagasse, Tifton hay, cracked corn grain, cottonseed meal, urea, limestone, and mineral supplement ( Table 2). The diets were formulated according to the Large Ruminant Nutrition System (Fox et al., 2004) (Tables 1 and 2).
Influence of nutritional management prior to adaptation to a feedlot diet on ruminal microbiota of Nellore cattle Pinto et al. 3 On the first and last day of each period, cattle were weighed for body weight assessment. The Nellore cattle were housed in individual pens (6 m of linear bunk space and 72 m 2 of pen space per animal) with free access to water. Cattle were fed ad libitum once a day at 08:00 h, and leftovers were weighed in the next day at 07:00 h. For the restriction treatment, the animals were fed Tifton hay at 1.4% of BW plus supplement. The amount of feed offered was adjusted daily based on orts left before morning feed delivery (target leftover rate of 5% relative to the quantity of feed offered).
Dry matter intake was calculated every day by weighing and determining the dry matter (DM) of feed and the leftover feed.  Ingredients (

Ruminal pH measurements
Ruminal pH was continuously measured using a pH data logger (Model T7-1 LRCpH, Dascor, Escondido, CA, USA; Penner et al., 2006) on days 5 (pre-adaptation phase), 16 (adaptation phase), and 27 (finishing phase). The data logger was inserted before feeding each day and was removed 24 hours later. The systems were initialized to record data at ten-minute intervals. Before the insertion in the rumen and after removal from the rumen, each electrode was standardized at pH 7.0 and 4.0. The pH data were recorded at 0, 4, 8, and 12 h after feeding (8, 12, 16, and 20 h).

Ruminal protozoa counting
For ruminal ciliated protozoa counting, 10 mL of ruminal contents were collected through the ruminal cannula with a vacuum pump. Samples were stored in vials containing 20 mL of 50% formaldehyde.
The sampling was carried out on days 8 (pre-adaptation), 18 (adaptation), and 30 (finishing) at 4, 8, and 12 h after feeding for each period. Protozoa were identified (genera Isotricha, Dasytricha, Entodinium, and Diplodiniinae subfamily) and counted using a Neubauer Improved Bright-Line counting chamber (Hausser Scientific Partnership, Horsham, PA, United States) by optical microscopy (Olympus CH-2 R, Japan; Dehority, 1993). Samples for protozoa counting were not collected at 0 h to avoid opening the rumen canula before collecting samples for ruminal bacteria 4 h after feeding.
The ruminal samples (solid + liquid) were collected by manually evacuating the rumen through the cannula 4 h after feeding. The ruminal content was weighed (solid and liquid phase, separately), and the proportion of solid and liquid in the rumen of each animal was calculated. For each sample 50 g of ruminal content were used according to the proportion calculated in the rumen evacuation for each animal (e.g., 30% liquid and 70% solid, so 15 g of liquid and 35 g of solid composed a 50 g of ruminal content sample). Samples were processed immediately after collection as described by Yu and Morrison (2004)   Influence of nutritional management prior to adaptation to a feedlot diet on ruminal microbiota of Nellore cattle Pinto et al. 5 in a final volume of 20 μL per reaction. The extracted DNA was used as a template in a real-time PCR reaction using specific primers for the desired rumen bacteria (Table 3), herewith a universal primer for eubacteria (universal).
The PCR amplification protocol was as follows: an initial denaturation step at 95 °C for 10 min, then 44 cycles of heating and cooling at 95 °C for 15 s, followed by annealing step at 60 °C for 30 s, and extension at 72 °C for 30 s. The samples were run in duplicate, and a negative control was included in each assay to assess the specificity of PCR reaction. The melting curves were analyzed at the end of the reactions to verify the specificity of each amplification.

Statistical analysis
Data were analyzed in a replicated Latin square design by SAS software (Statistical Analysis System, version 9.1), and tests for normality (Shapiro-Wilk's and Kolmogorov-Smirnov's) and heterogeneity of treatment variances (GROUP option of SAS) were performed before analyzing the data. The effects of period, square, period × square, square × treatments, animal nested within square, and period animal nested within square were considered random factors. The qPCR of ruminal bacteria was analyzed by Mixed procedure of SAS. The model accounted for the same effects as described above. Results were considered significant at P≤0.05 level. All means presented are least squares means, and effects were separated by PDIFF option of SAS. The mathematical model used was: in which y ijkl = observed value of the dependent variable, μ = overall mean, τ i = treatment effect, ρ j = period effect, σ k = Latin square repetition effect, a l (σ k ) = animal within Latin square repetition, and e ijkl = random residual error.
The variables involving DMI, rumen protozoa, and ruminal pH were analyzed by MIXED procedure of SAS with repeated measures (Littell et al., 1998). The model accounted for the same effects as described above plus time and its interactions with treatments. Results were considered significant at P≤0.05 level. All means presented are least squares means, and effects were separated by PDIFF option of SAS. The mathematical model used was: in which y ijklm = observed value of the dependent variable, μ = overall mean, τ i = treatment effect, ρ j = period effect, σ k = Latin square repetition effect, ο l = time effect, Θ il = interaction between treatments and time, a m (σ k ) = animal within Latin square repetition, and e ijklm = random residual error.

Results
For DMI, an interaction was observed between treatments and day on feed, in which cattle from concentrate group had greater intakes during most part of the feeding period (Figure 1).
During the pre-adaptation phase, restricted cattle had a higher pH than concentrate-fed cattle (P = 0.05). However, there were no differences between cattle from control group and those from other treatments (Table 4). Moreover, a quadratic response was observed between time and ruminal pH (P = 0.01), in which the lowest pH was measured 12 h after feeding (0 h: 6.59, 4 h: 6.66, 8 h: 6.52, 12 h: 6.39; SE = 0.07; data not shown). Regarding the adaptation and finishing phases, no effect of treatments was observed (P>0.05). Nevertheless, a quadratic relationship was observed (P<0.01) between rumen pH and time for the adaptation phase ( In the pre-adaptation phase, cattle fed concentrate had a larger Entodinium population than cattle from other treatments (P = 0.02), resulting in a larger total protozoa counting for steers fed concentrate as well (P<0.01, Table 5). In addition, there was an interaction (P = 0.04) between treatments and time after feeding for Isotricha counts, in which only at 8 h after feeding cattle either fed concentrate or restricted presented smaller Isotricha populations (Figure 2). No further differences for protozoal Dasytricha and Diplodinium populations were observed during the pre-adaptation phase (P>0.05).
During the adaptation phase, cattle fed concentrate had the lowest counting of Dasytricha (P<0.01) and Isotricha (P = 0.05) when compared with cattle from control or restricted groups. Furthermore, cattle from restricted group showed higher number of Dasytricha (P<0.01) and a smaller number of Isotricha (P = 0.05) than the control animals. No effect of treatments was observed (P>0.05) on Entodinium and Diplodinium populations and total protozoa.
Regarding the finishing phase, steers either restricted or fed concentrate had larger Diplodinium (P<0.01) and Dasytricha (P<0.01) relative populations than control animals. Furthermore, restricted cattle had lower (P = 0.02) numbers of Isotricha when compared with either control or concentrate-fed steers. Finally, a time effect was observed (P = 0.03) for Isotricha, in which its counts decreased linearly (4 h: 2.33; 8 h: 1.20; 12 h: 2.20). No significant effect (P>0.05) was detected for Entodinium relative populations and total protozoa. Regarding the real-time PCR of ruminal bacteria, there was no effect of treatments on F. succinogenes population (P>0.05, Table 6) during the pre-adaptation phase. However, there was a reduction in M. elsdenii relative population (P<0.01) in cattle from either restriction or concentrate groups. Likewise, restricted cattle had smaller S. bovis relative population when compared with animals receiving concentrate (P = 0.05).   Regarding the adaptation and finishing phases, no effect (P>0.05) of treatments were observed on M. elsdenii and S. bovis relative populations. However, cattle from concentrate group had smaller F. succinogenes populations (P = 0.05) when compared with animals from the control group.

Discussion
The present study was part of a larger research performed by this research group, a compendium of studies assessing the effect of either nutritional restriction or intake of concentrate feedstuffs prior to the adaptation to high-concentrate diets on animal performance, DMI, ruminal fermentation, and microbiota. Pereira et al. (2020) aimed to test the hypothesis that cattle from pasture, coming either from nutritional restriction or from intake of concentrate feedstuffs prior to the adaptation period, require a different adaptation length and present different overall feedlot performance. The authors reported that either restriction or concentrate supplementation before beginning the adaptation period to high-concentrate diets did not impact adaptation length, and both may be used as nutritional strategies to improve performance and carcass characteristics of feedlot Nellore cattle. In this context, it was hypothesized that restriction or intake of concentrate feedstuffs prior to the adaptation could affect rumen fermentation patterns and, consequently, ruminal microbiota. Thus, Pinto et al. (2020) reported that cattle previously exposed to concentrate exhibited decreased bacterial richness during the pre-adaptation phase and increased bacterial diversity during the adaptation phase. Moreover, restricted animals had lower DMI during the adaptation phase, as well as lower DM digestibility, starch, and total digestible nutrients when compared with cattle consuming concentrate.
In the present study, greater ruminal fermentation during pre-adaptation phase decreased ruminal pH, which may have negatively affected M. elsdenii, a pH-sensitive microorganism (Nocek, 1997). However, we observed an increase in Entodinium populations, which is considered the most pH-tolerant species when compared with other genera of rumen protozoa (Mackie et al., 1978;Lyle et al., 1981). Furthermore, Entodinium populations present high amylase activity to digest engulfed starch granules (Nagaraja, 2016). Entodinum ferment cell wall carbohydrates, as well as starch and soluble sugars, but in general, these microorganisms use starch as main source to growth, which may have favored Entodinium population in the rumen of concentrate-fed cattle, since the number of protozoa can be relatively low in animals receiving exclusive forage diets and higher in forage and grain mixtures (Veira, 1986).
During the pre-adaptation, restricted steers presented higher rumen pH than concentrate-fed cattle, due to the lack of substrate available for fermentation based on the DMI (Figure 1). In addition, the lack of substrate may have played a role in reducing M. elsdenii and S. bovis populations during preadaptation without negative effects on protozoa. When concentrate diets were introduced during adaptation phase, bacterial populations were reestablished, and no differences were observed when restricted animals were compared with cattle from control or concentrate groups during adaptation and finishing.
The lower pH during pre-adaptation phase may have negatively affected the F. succinogenes relative populations in both adaptation and finishing phases, as well as Isotricha and Dasytricha populations during adaptation. Fibrobacter succinogenes is considered the major ruminal cellulolytic bacterium in the rumen, which is sensitive to low ruminal pH; almost none of ruminal cellulolytic bacteria grow significantly at pH values below 6.0 (Weimer, 1996). Likewise, Isotricha and Dasytricha populations were negatively affected when the level of concentrate in diets became higher (Dehority, 1995). However, since they do not ferment structural carbohydrates (Nagaraja, 2016), the lower ruminal pH during adaptation and finishing phase when compared with pre-adaptation may have had more impact on their populations. On the other hand, the lower inclusion of roughage sources in the diet in the adaptation and finishing phases may have played a more significant role in reducing F. succinogenes population than the ruminal pH itself, since there would be smaller amounts of structural carbohydrates available for fermentation entering the rumen. Pinto et al. (2020) reported that the lower relative abundances of F. succinogenes could be related to the larger area of pH below 6.2, resulting in a reduction in total tract digestibility of neutral and acid detergent fibers.
In the finishing phase, steers receiving concentrate reestablished protozoa populations that were negatively affected in previous phases, such as Diplodinium, which may be related to the increased DMI presented by cattle fed concentrate without negatively impacting ruminal pH. Moreover, Dasytrichia and Diplodinium populations were higher in restricted cattle than in the control cattle, which may be related to the lower intake presented by these animals most of the feeding period. So, the intake of concentrate feedstuffs during pre-adaptation phase may have promoted positive effects in the process of cattle adaptation to high-concentrate diets. In this context, early exposure to concentrate feedstuffs is thought to prepare the ruminal bacterial community for higher levels of non-fibrous carbohydrates Pinto et al., 2020). Thus, Pinto et al. (2020) reported an increase in ruminal starch degradability in animals exposed to concentrate feedstuffs prior to the adaptation phase.

Conclusions
The previous nutritional background impacts dry matter intake and ruminal microbiota during adaptation and finishing phases without causing any negative effect on rumen pH. Furthermore, feeding concentrate prior to the adaptation positively impacts the transition to high-concentrate diets, since cattle receiving concentrate partially reestablishes the ruminal microbiota during the finishing phase, even presenting greater dry matter intake.