The current research was carried out to determine the associations between the rumen microbiota and traits related with feed efficiency in a Holstein cattle population (n = 30) using whole metagenome sequencing. Improving feed efficiency (FE) is important for a more sustainable livestock production. The variability for the efficiency of feed utilization in ruminants is partially controlled by the gastrointestinal microbiota. Modulating the microbiota composition can promote a more sustainable and efficient livestock. This study revealed that most efficient cows had larger relative abundance of Bacteroidetes (P = 0.041) and Prevotella (P = 0.003), while lower, but non-significant (P = 0.119), relative abundance of Firmicutes. Methanobacteria (P = 0.004) and Methanobrevibacter (P = 0.003) were also less abundant in the high-efficiency cows. A de novo metagenome assembly was carried out using de Bruijn graphs in MEGAHIT resulting in 496,375 contigs. An agnostic pre-selection of microbial contigs allowed high classification accuracy for FE and intake levels using hierarchical classification. These microbial contigs were also able to predict FE and intake levels with accuracy of 0.19 and 0.39, respectively, in an independent population (n = 31). Nonetheless, a larger potential accuracy up to 0.69 was foreseen in this study for datasets that allowed a larger statistical power. Enrichment analyses showed that genes within these contigs were mainly involved in fatty acids and cellulose degradation pathways. The findings indicated that there are differences between the microbiota compositions of high and low-efficiency animals both at the taxonomical and gene levels. These differences are even more evident in terms of intake levels. Some of these differences remain even between populations under different diets and environments, and can provide information on the feed utilization performance without information on the individual intake level.

Lameness is considered one of the most common welfare and productive problems in dairy cattle. The objective of this study was to evaluate differences in lying behavior between moderately lame and nonlame lactating cows under commercial conditions. Data were collected from 10 free-stall commercial herds, which were feeding on exactly the same ration once daily. All lactating cows were scored for lameness according to a 1 to 5 locomotion scoring system. Only cows with a lameness score between 1 and 4 were considered in the study. In each herd, between 10 and 15 lame cows (scored as 3 or 4) were chosen, and for each lame cow, a nonlame cow (scored as 1) within the same parity and similar days in milk was also selected. Pendant data loggers were then placed on the right hind leg of each cow for 10 d to record lying behavior at 1-min intervals. In addition, the time of feed delivery was recorded in each herd on a daily basis. Total daily lying time, daily number of lying bouts, lying bout duration, laterality (side of recumbence), and lying behavior around feed delivery time were evaluated using a mixed-effects model that accounted for the fixed effects of lameness, days in milk, parity, and the interaction between parity and lameness, plus the random effects of herd. Total daily lying time (721±24.2 min/d) tended to increase with days in milk, but it was not affected by lameness or parity. Likewise, no differences were found in the number of lying bouts (9.6±0.49/d) or laterality (47±2.6% of time lying on the right side). However, the mean bout duration was longer in lame (89.3±3.89 min) compared with nonlame (80.7±3.90 min) cows. It is interesting that lame cows stood up 13 min later than nonlame cows relative to the time when the ration was delivered. In addition, lame cows lay down 19 min earlier than nonlame ones after the feed was delivered, which implies that nonlame cows spent more time standing, and probably eating, than did lame cows. It was concluded that lame cows have longer lying bouts than nonlame animals, and that lying behavior around feed delivery time may be an effective proxy to identify moderately lame cows.

The objectives of this study were (1) to use endoscopy to collect biopsies from the rumen and colon epithelia to describe changes in gene expression in these 2 tissues as cows move from a dry to a lactation ration and (2) to evaluate the potential influence that supplementation of live yeast could exert on these 2 epithelia. Twenty-one Holstein cows were split into 2 treatments and received either 300 g/d of corn containing 1 × 10 cfu/d of live yeast (LY; n = 10) or 300 g/d of corn with no supplementation (control; n = 11) starting 21 ± 2.6 d (average ± SD) before until 21 d after calving. At 14 ± 2.6 d before the expected calving date, and exactly at 7 and 21 d after calving, rumen and colon biopsies were obtained from each cow using an endoscope. Total RNA was extracted from rumen and colon tissues, and the expression of IL10, TNFA, TLR4, IL1B, PCNA, MKI67, SGLT1, BAX, CASP3, OCLN, CLDN4, HSPA1A, HSPB1, DEFB1, and MCT1 (the latter only in rumen samples) was quantified by quantitative PCR. Overall, fluctuations in expression of the selected genes in the colon between the 2 stages of production and the 2 treatments were smaller than those found in the rumen. In the rumen epithelium, expression of TLR4 and DEFB1 was greatest before calving, with LY cows having a greater expression of TLR4 than control cows. Similarly, expression of IL10 was greatest in LY cows before calving. Expression of TNFA in the rumen epithelium of control cows was lowest at 21 DIM but in LY cows was kept steady among production stages. The expression of PCNA and MKI67 in the rumen epithelium was greatest at 7 DIM, indicating a high proliferation rate of this epithelium after calving. In the colon mucosa, expression of TLR4 and DEFB1 was greater than in the rumen, and DEFB1 expression was greater in LY cows than in control cows. The use of an endoscope allowed us to study the dynamics of rumen epithelium adaptation to increased supply of concentrate after calving, consisting of increased epithelia remodeling, reduction of the TLR4, and increased IL10 expression. Furthermore, the rumen epithelium of dry cows responded rapidly to live yeast, with changes in the expression of genes involved in the immune response becoming evident after 7 d of exposure to yeast. The expression of genes related to the immune response (mainly TLR4 and DEFB1) in the colon mucosa was greater than in the rumen, and the expression of DEFB1 was further stimulated by live yeast. It is concluded that the use of an endoscope allows the study of gene expression patterns in the rumen and hindgut epithelia. We report marked changes in the rumen wall and more modest changes in the colon when transitioning from a dry to a lactation ration. Furthermore, supplementation of live yeast fostered and increased expression of genes regulating inflammation and epithelial barrier in the rumen, and in the colon it increased the expression of DFEB1 coding for an antimicrobial peptide.

With the objective of evaluating the potential effects of sodium bicarbonate or a magnesium-based product on rumen pH and milk performance of dairy cattle exposed to a dietary challenge, 30 lactating Holstein cows (648 ± 67 kg of body weight; 44.4 ± 9.9 kg/d of milk yield; 155 ± 75 d in milk) were blocked by parity (9 primiparous and 21 multiparous) and randomly distributed to 3 treatment groups. One group received a total mixed ration (TMR) that acted as a control (CTR), a second group (SB) received the same TMR but with an additional supplementation of 0.8% of sodium bicarbonate, and a third group (MG) received the same TMR as CTR but an additional supplementation of 0.4% of a magnesium-based product (pHix-Up, Timab, Dinard, France). After 1 wk of exposure to this TMR, all 3 rations were supplemented with 1 kg/d of barley, which was then increased 1 kg/wk until reaching 3 kg/d of barley during wk 4 of the study. Every kilogram of barley replaced 1 kg of forage in the diet. Individual feed intake and behavior were monitored using electronic feed bins. Seven cows per treatment were equipped with an intraruminal bolus that recorded pH every 15 min. As the severity of the barley challenge increased, dry matter intake decreased, but this decrease was more pronounced in SB cows than in MG cows, with an intermediate response for CTR cows. The MG cows produced more milk when challenged with 2 or 3 kg/d of additional barley than when challenged with 1 kg/d, whereas CTR cows produced less milk with the 3 kg/d challenge compared with 1 or 2 kg/d, and the SB cows maintained milk production. Milk fat content decreased with barley challenges, with CTR cows experiencing a more severe decrease than SB cows, which maintained stable butterfat values throughout the study, and MG cows showed a decline in milk fat content only with the 3 kg/d of additional barley. Meal size was also reduced as the severity of barley challenge increased, and this reduction was more modest in MG cows than in SB cows. The number of daily meals consumed by SB and MG cows was more constant than that recorded in CTR cows. Cows on the CTR and SB treatments showed a marked decrease in rumen pH with the 3 kg/d of additional barley, whereas MG cows maintained stable rumen pH during the barley challenges and had greater average rumen pH (5.93 ± 0.04) than CTR cows (5.83 ± 0.04) with the 3 kg/d of additional barley; SB cows showed intermediate values (5.85 ± 0.04). Last, MG cows spent less time (32.3 ± 6.1%) with rumen pH ≤5.8 when exposed to the 3 kg/d of barley challenge than CTR and SB cows (50.7 ± 5.02%). In conclusion, supplementation with MG prevents the decline in dry matter intake and milk production induced by a rumen challenge, whereas supplementation with SB prevents the decay in milk production but does not prevent the decrease in feed intake. These changes were probably due to the ability of the MG treatment to prevent a reduction in rumen pH when challenging cows with 3 kg/d of additional barley in the ration.

The aim of this study was to evaluate the potential effects of methyl donor supplementation of pregnant animals in the presence or absence of a concomitant lactation on the methylome of the offspring. Twenty Holstein cows, 10 nulliparous (non-lactating while pregnant) and 10 multiparous (lactating while pregnant) were blocked by parity and randomly assigned to an i.m. weekly injections of a placebo (CTRL) or a solution containing methyl donors (MET). After calving, 5 calves randomly selected from each treatment (two born to non-lactating and three to lactating dams) were blood-sampled to determine their full methylome. There were more than 2,000 CpG differentially methylated between calves born to CTRL and those born to MET, and also between calves born to lactating and non-lactating dams. Most of the differences affected genes involved in immune function, cell growth regulation and differentiation, kinase activity, and ion channeling. We conclude that the coexistence of pregnancy and lactation affects the methylome of the offspring, and that supplementation of methyl donors early in gestation has also consequences on the methylome.