Effects of a short 40-d dry period with only a close-up diet (SHORT) and a conventional 60-d dry period with a 39-d far-off and a 21-d close-up diet (CONV) on milk production, feed intake, blood and rumen parameters, liver gene expression and rumen microbiota profile were compared in 11 second-parity and 15 third and later parity cows. Milk production was recorded daily during the first 16 wks of lactation. Differential liver gene expression was assessed by affymetrix microarray analysis and DNA extracted from rumen samples was subjected to Illumina sequencing for exploring the microbiome profile. The SHORT treatment reduced milk yield and DMI after calving in third and later parity cows, but not in second-parity cows when compared to the CONV treatment. Cows on the SHORT treatment had higher concentrations of NEFA in blood plasma and tended to have higher liver TAG immediately after calving. These effects tended to be greater in third- and later parity cows compared to second-parity cows. Expression patterns of genes involved in β-oxidation at the first week of lactation compared to those at three weeks before calving showed lower hepatic β-oxidation capacity in cows on the SHORT treatment compared to those on the CONV treatment. During this period, the expression of DGAT, a key gene in the triglyceride synthesis, increased in SHORT-treatment cows while it remained unchanged in CONV-treatment cows. The expression patterns of genes involved in gluconeogenesis showed a higher capacity at first week after calving in cows on the SHORT compared to those on the CONV treatment. Our study also showed that the SHORT treatment increased the relative abundance of Firmicutes and reduced the relative abundance of Bacteroidetes compared to the CONV treatment and reduced the shifting of rumen microbiota from before to after calving. Results also demonstrated that the rumen microbiota was more stable in the SHORT treatment during the transition period. Based on these results, a 40-d dry period management with only a close-up diet might be beneficial for second parity cows. However, this treatment may be detrimental for older cows as excessive energy intake and fat deposition during the dry period in these animals result in lower milk production and higher mobilization of NEFA and accumulation of fat in the liver. / February 2016

Este trabalho foi desenvolvido com a finalidade de testar alterações nas metodologias in vitro de produção de gases (diferentes proporções de inóculo) e incorporação de 32P (velocidade de centrifugação, número de lavagens e diferentes tipos de inóculos) para adequá-las às condições climáticas e às especificidades da microbiota dos ruminantes mantidos no clima tropical brasileiro. Além dessas técnicas utilizadas, foram feitas determinações de análise bromatológica com posterior análise de nutrientes digestíveis totais, degradabilidade ruminal in situ e digestibilidade in vitro a partir da técnica dos dois estágios. Os substratos testados foram divididos em, fibrosos (feno de alfafa e capim colonião), concentrados (milho grão e farelo de soja) e fermentado (silagem de milho). O delineamento experimental utilizado foi inteiramente casualizado, todos os substratos foram avaliados para cada uma das variáveis com pelo menos três repetições no tempo. Cada variável foi comparada por análise de variância considerando a hipótese de nulidade dentro dos substratos. As correlações entre as metodologias foram testadas quando biologicamente significantes e compatíveis. A técnica de incorporação in vitro de 32P, obteve resultados satisfatórios (P<0,05) em relação à velocidade de centrifugação e número de lavagens. Os diferentes tipos de inóculos testados (100% fase líquida do rúmen, 50% fase líquida e 50% fase sólida) não apresentaram diferença significativa (P>0,05). As diferentes proporções do inóculo testado na técnica in vitro de produção de gases (A contendo 100% de fase líquida e 0% de fase sólida; inóculo B, 75% de fase líquida e 25% de fase sólida; inóculo C, 67% de fase líquida e 33% de fase sólida; inóculo D, 50% de fase líquida e 50% de fase sólida) apresentaram diferença significativa (P<0,01); os fatores analisados permitiram a interpretação de que usando o inóculo com maior porcentagem de fase sólida, o processo fermentativo foi mais parecido ao potencial de produção de gases. Para a técnica in vitro de incorporação de 32P sugere-se usar uma velocidade de centrifugação de 39.907g, 4 lavagens e um inóculo com apenas a fase líquida do conteúdo ruminal. É sugerido que para a técnica in vitro de produção de gases pode ser usado um inóculo contendo 50% de fase sólida e 50% de fase líquida do conteúdo ruminal. / This study was developed to test the adaptations in in vitro methodologies of gas production (different proportion of solid phase in the inoculum) and 32P incorporation technique (centrifugation speed, number of washings and different kinds of inoculum) to adequate them to the climate condition and to the microbiota specifications of the ruminants kept in the Brazilian tropical climate. Besides these techniques, bromatological analysis determinations were made with further total digestible nutrients and also, in situ rumen degradability and in vitro digestibility from the two-stage technique. The tested substrata were divided in fibrous (Panicum grass and Lucerne hay), concentrated (soybean meal and corn grain) and fermented (corn silage). The experimental design was completely randomized, where all the substrata were evaluated for each of the variables with at least three repetitions. Each variable was compared for analysis of variance considering the hypothesis, within substrata. The correlations between methodologies were tested when biologically significant and compatible. The 32P in vitro incorporation technique obtained satisfactory results (P<0,05) in relation to the centrifugation speed and the number of washings. The two different kinds of tested inocula (100% rumen liquid phase, and 50% both liquid and solid phase) showed no significant difference (P>0,05). The different inoculum proportion tested for in vitro gas production (inoculum A containing 100% liquid phase and 0% solid phase; inoculum B, 75% liquid phase and 25% solid phase; inoculum C, 67% liquid phase and 33% solid phase; and inoculum D, 50% both liquid and solid phases) showed significant difference; the analyzed factors allowed the interpretation that using the inoculum with higher solid phase percentage, the fermentation process was more similar to the gas production potential. In the in vitro 32P incorporation technique it is suggested to use a centrifugation speed of 39,907g, 4 washings and an inoculum with only liquid phase from the rumen. It is suggested that for the in vitro gas production technique can be used an inoculum containing 50% of solid phase and 50% of liquid phase.

Quatro bubalinos adultos com fístulas ruminais foram utilizados com o objetivo de se estudar os efeitos da ingestão de níveis de fósforo sobre as características do metabolismo ruminal e digestão das rações. Foram avaliados os seguintes parâmetros: pH; concentração de amônia; volume e taxa de passagem do líquido ruminal; quantificação de minerais no conteúdo ruminal; digestibilidade com marcador (Cr2O3) e estimativa de excreção fecal de minerais. Os animais foram delineados em Quadrado Latino (4x4), com ingestão de quatro diferentes níveis de fósforo na dieta (8, 12, 15 e 18 gramas), mantendo-se a relação volumoso:concentrado foi de 85:15 e com o uso de cana de açúcar como único volumoso. Cada sub-período compreendeu 28 dias, sendo 14 de adaptação. Não houve influencia no pH,n a concentração de amônia, no volume e na taxa de passagem do líquido ruminal; nas concentrações de minerais no conteúdo ruminal (P, Ca, Mg, Na, K, S, Fe, Cu, Zn e Mn) e nos coeficientes de digestibilidade da matéria seca, fibra em detergente neutro, proteína bruta e energia bruta. Observou-se excreção linear de fósforo nas fezes em função ao fósforo ingerido da dieta, podendo ser estimada pela equação de regressão Y = 0,03 + 0,610 * X, onde Y é a quantidade de fósforo excretado nas fezes diariamente em g/kg0,75 e X é a quantidade de fósforo ingerido diariamente em g/kg0,75. / Four mature buffaloes fitted with rumen cannulas were used to study the effects of different levels of the phosphorus intake on the ruminal metabolism and the digestion in the diets. Ammonia concentration, pH, fluid and solid outflow rates, rumen volume; concentration of mineral in the ruminal content; digestibility and estimate of faecal output of minerals were determined. The animals were designed in Latin Square experiment (4x4) with each period of 28 d of duration. Treatments were applied four rations with different levels of P intake (8, 12, 15 e 18 grams). The proportion forage for concentrate was 85:15 with only using fresh chopped sugar cane. No differences were observed among the treatments in the rumen parameters evaluated and digestibility. Increasing in phosphorus intake resulted in simple linear regression equation Y = 0.03 + 0.610 * X, where P output daily (g (kg bodyweight)-0,75) was used as the dependent variable (Y) and P intake daily (g / kg-0,75) was used as the independent variable (X).

Quatro bubalinos adultos com fístulas ruminais foram utilizados com o objetivo de se estudar os efeitos da ingestão de níveis de fósforo sobre as características do metabolismo ruminal e digestão das rações. Foram avaliados os seguintes parâmetros: pH; concentração de amônia; volume e taxa de passagem do líquido ruminal; quantificação de minerais no conteúdo ruminal; digestibilidade com marcador (Cr2O3) e estimativa de excreção fecal de minerais. Os animais foram delineados em Quadrado Latino (4x4), com ingestão de quatro diferentes níveis de fósforo na dieta (8, 12, 15 e 18 gramas), mantendo-se a relação volumoso:concentrado foi de 85:15 e com o uso de cana de açúcar como único volumoso. Cada sub-período compreendeu 28 dias, sendo 14 de adaptação. Não houve influencia no pH,n a concentração de amônia, no volume e na taxa de passagem do líquido ruminal; nas concentrações de minerais no conteúdo ruminal (P, Ca, Mg, Na, K, S, Fe, Cu, Zn e Mn) e nos coeficientes de digestibilidade da matéria seca, fibra em detergente neutro, proteína bruta e energia bruta. Observou-se excreção linear de fósforo nas fezes em função ao fósforo ingerido da dieta, podendo ser estimada pela equação de regressão Y = 0,03 + 0,610 * X, onde Y é a quantidade de fósforo excretado nas fezes diariamente em g/kg0,75 e X é a quantidade de fósforo ingerido diariamente em g/kg0,75. / Four mature buffaloes fitted with rumen cannulas were used to study the effects of different levels of the phosphorus intake on the ruminal metabolism and the digestion in the diets. Ammonia concentration, pH, fluid and solid outflow rates, rumen volume; concentration of mineral in the ruminal content; digestibility and estimate of faecal output of minerals were determined. The animals were designed in Latin Square experiment (4x4) with each period of 28 d of duration. Treatments were applied four rations with different levels of P intake (8, 12, 15 e 18 grams). The proportion forage for concentrate was 85:15 with only using fresh chopped sugar cane. No differences were observed among the treatments in the rumen parameters evaluated and digestibility. Increasing in phosphorus intake resulted in simple linear regression equation Y = 0.03 + 0.610 * X, where P output daily (g (kg bodyweight)-0,75) was used as the dependent variable (Y) and P intake daily (g / kg-0,75) was used as the independent variable (X).

Three non-lactating Holstein cows fitted with rumen cannula were used to determine crude protein and dry matter rate of disappearance of two protein supplements: 1) soybean meal and 2) bypass protein blend by using the in situ bag technique. Rate disappearance (%/hr) was higher for soybean meal. Two collection periods were completed using 6 cows with a minimum of 21 d adaptation to the treatment top dressed on to the total mixed ration. Ruminal concentration of ammonia N, blood ammonia, and urea did not differ between treatments. Total concentration of volatile fatty acids was higher for bypass protein blend-fed cattle as well as percent molar concentration of propionate, butyrate, and valerate, while pH was lower. Total protozoa, and total and cellulolytic viable bacteria populations did not differ. Four of the cows received a dose of 5 ml of Prostaglandin F,a. Blood and cervical muc us samples obtained showed no difference in blood ammonia and urea concentration . Forty-six Holstein cows were assigned to one of the two treatments (top dressed on the t otal mixed ration), according to parity during the following 125 d postpartum. Daily dry matter intake and milk production were recorded. Feed, orts, and feces were sampled. Milk samples were collected weekly and analyzed for components. Percent lactose and solid non-fat showed higher for cattle fed the bypass prorein blend. Starting on day 10 postpartum, cows were observed for signs of estrus and bred at first estrus observed after 45 d postpartum. Cervical mucus and blood collected at first standing estrus, and first, seco nd, and third service, did not show a significant difference in urea concentration between rations. Twice-weekly collected blood samples showed similar monthly mean concentration of ammonia, urea, and progesterone profile for both treatment groups. Percent pregnancy, services per pregnancy, first service pregnancy and embryo mortality showed no significant difference. Motility and survivability of bull sperm were evaluated by incubating thawed semen in different levels of previously observed and/or ammonia. physiological There were no concentrations of urea significant treatment differences observed. Detrimental effects of treatment on sperm were not detected.

"January 2003" / Includes bibliographical references (leaves 172-204) / xx, 204 leaves : ill. ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Thesis (Ph.D.)--University of Adelaide, Dept. of Animal Science, 2003

The effect of untreated roughages on digestibility and rumen fill of the gut was reviewed as physical mechanism influencing the regulation of roughage intake. The review of literature also focused on identifying factors that affect the way in which urea-ammoniation alters the roughage intake, digestibility and performance of ruminant animals. Trials were carried out with fistulated cows to address to what extent concentrate proportion and urea-ammoniation affected microbial colonization and degradation of roughage diets in the rumen. One interest of this study was to develop a model that would help to predict the benefit associated with urea-treatment of roughages. The first trial (Chapter 3) investigated the effect of urea-ammoniation of roughage and concentrate proportion of the diet on degradation of roughages, and the benefit associated with the treatment of roughages. Four rumen-fistulated Jersey cows were fed on a basal diet composed of either urea treated (3 kg of urea per 100 kg of straw) or untreated Eragrostis curvula hay. These basal diets were supplemented with concentrate composed of maize meal (78%) and cotton seed cake (22%). The concentrate contributed 0, 25, 50 and 75% of the total ration and hay the rest. The experiment consisted of 6 periods. Each period lasted 19 days, comprising 12 days of adaptation to the experimental diet followed by 6 days degradability measurements and 1-day rumen fluid collection. During each period the 4 Jersey cows were randomly allocated to 4 of the 8 dietary treatments, ensuring that each diet was fed to 3 animals during the entire experimental period. The experimental roughages used in this trial were wheat (Triticum sativum) straw, barley (Hordeum Vulgare) straw, coastcross (k11) (Cynodon hybrid) hay, veld hay (natural grass), oat (Avena sativa) straw, oat (Avena sativa) hay, maize (Zea mays) stover, kikuyu (Pennisetum clandestinum) grass, weeping love grass (Eragrostsis curvula) and Italian rye (Lolium multiflorum) grass. Each roughage (sample) was subdivided into two equal portions, one of which was then treated with urea. The urea solution was prepared by dissolving 30 g of urea in 0.4 liter of water. The solution was fully distributed over I kg of roughage. Treated roughages were sealed tightly and stored at room temperature for 5 weeks in plastic bags. Immediately after opening, the different roughages, including the untreated ones, were sun dried, chopped fine by hand and ground through a 2-mm screen in a laboratory mill. About 3 g of each sample was weighed into labeled nylon bags. The bags were tied to a stainless steel disc with 10 evenly spaced small holes drilled through the periphery of the disc serving as anchor points. The bags were incubated (in duplicate per time interval) in the rumen for 120, 96, 72, 48, 24, 12, 6 and 3 h, sequentially. The treated samples were incubated in animals fed treated hay, while untreated samples were incubated in animals given untreated hay. Immediately after removal from the rumen, the bags, including the 0 hour ones, which had not been incubated but soaked in warm water for I hour, were washed in 6 cycles (each lasting 4 minutes) in a semi-automatic washing machine. The washed bags were then dried in a forced draught oven at 60 degrees C for 48 hours, cooled in a desicator and weighed. The pH of the rumen fluid ranged between 6.5 and 6.8 for all diets. Rumen ammonia concentration was higher (P<0.002) when the basal diet consisted of urea treated hay. Increasing the concentrate proportion in the diet had the desired effect of increasing rumen ammonia concentration without severely affecting pH. Urea-ammoniation increased (P<O.OOO1) the slowly degradable fraction (B), potential degradability (PD), effective degradability (ED) of dry matter and neutral detergent fiber (NDF), decreased (P>0.05) lag time (LT) but had no effect on the rate of degradation (c) of dry matter. Concentrate proportions affected (P<0.05) the slowly degradable fraction, potential degradability, lag time and effective degradability but had no effect (P>0.05) on the rate of degradation of dry matter (DM). Maximum and minimum values of the slowly degradable fraction, potential degradability and effective degradability of DM and NDF were obtained at the 25 and 75% concentrate levels, respectively. Within urea-ammoniation, roughage type affected (P<O.OO1) the B-fraction, PO and EO of OM and NDF degradation. Rate of degradation of DM of untreated roughages varied from 0.022 h(-1) in wheat straw to 0.087 h(-1) in rye grass, while for urea treated roughages it varied from 0.022 h(-1) in oat straw to 0.082 h(-1) in rye grass. Rye grass degraded almost three to four times faster than urea treated oat or untreated wheat straw. Urea-ammoniation was less effective in increasing DM and cell wall degradation rates (c) of rye grass compared to wheat straw. The results showed that low quality roughages such as wheat straw benefited relatively the most from urea-ammoniation. The effect of urea-ammoniation and dietary manipulation on microbial colonization (Chapter 4) of fiber particles in the rumen of animals was also investigated in two experiments. In Experiment 1, the cows were fed on rations comprising either urea-ammoniated or untreated Eragrostis curvula hay supplemented with concentrate at hay to concentrate ratio of 100:0, 75:25, 50:50, 25:75, resulting in eight different rumen environments. The experiment consisted of two periods. Each period lasted 12 days of adaptation to the experimental diet followed by one-day incubation of urea-ammoniated and untreated barley straw. Experiment 2 consisted of two urea-ammoniated (7.5 kg of urea per 100 kg of hay) hay levels (20 and 40% of the total ration) and concentrate levels (60 and 80%). Fistulated Jersey cows were adapted for 12 days after assigning to the dietary treatment. Feed was given at the rate of 9.0 kg day(-1) per animal portioned into equal meals of 4.50 kg each and offered at 08:00 and 16:00 every day. About 3 g of urea-ammoniated or untreated barley (Hordeum vulgare) straw, ground through a 2-mm screen, was weighed into a labelled nylon bag and incubated for 3, 6 or 12 h in the rumen of the fistulated cows. Microbes adhering to incubated fiber particles were examined under the Environmental Scanning Electron Microscopy (ESEM) and analysed on the image analyser. Depending on morphology , the microbes were divided into three groups: bacilli (rod), cocci (round) and others (spiral, fimbrea and cluster ; not specifically defined or undefined microbes). Urea-ammoniation of dietary roughage decreased (P<O.OO1) bacilli counts and total bacteria count but had no effect on count of the undefined group of microbes on fiber particles in the rumen of cows (Experiment 1). Concentrate proportions had no effect (P>0.05) on bacilli, cocci and total bacterial count on fiber particles. However, the results from electron micrograph observations revealed that the total bacterial count tended to decrease as the concentrate level increased in the diet of cows. Bacilli, cocci, undefined group of microbes and total count of microbes increased (P<0.05) as length of incubation increased. In Experiment 2, incubated feed, concentrate proportion and time of incubation had no effect (P>O.05) on bacilli , others (undefined group of microbes) and total count of fiber-adhering microbes in the rumen of cows. However, increasing concentrate in the diet of cows tended to decrease (P<O.07) the count of fiber-adhering cocci. The total count of microbes on fiber particles was higher in animals fed 80% concentrate as compared to 60% concentrate. The benefit derived from urea treatment in terms of B-fraction, effective degradability and potential degradability of DM and fiber of roughages increased with increasing the NDF content. Therefore, the important conclusions drawn from the results of the present study is that urea-ammoniation of roughages should be done strategically and that high quality roughages may give little return per unit of cost of ammoniation. This means that the benefit associated with urea-ammoniation would be justified for poor quality roughages only. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 2003.

Made available in DSpace on 2014-11-10T11:10:00Z (GMT). No. of bitstreams: 0 Previous issue date: 2014-04-22Bitstream added on 2014-11-10T11:57:23Z : No. of bitstreams: 1 000791240.pdf: 1034820 bytes, checksum: 4b0b273b843efc0acef3c24358ec0816 (MD5) / Objetivou-se com este estudo caracterizar as alterações na microbiota ruminal, parâmetros ruminais, consumo, digestibilidade das dietas e a eficiência de síntese microbiana, em novilhos Nelore confinados alimentados com diferentes relações volumoso:concentrado, utilizando como fontes de volumoso o feno de Tifton 85 ou cana-de-açúcar. Foram realizados dois experimentos, no experimento 1 utilizou a cana-de-açúcar como fonte de volumoso e no experimento 2 utilizou-se o feno de tifton 85 como volumoso. Em ambos os experimentos foram testadas diferentes relações V:C (70:30; 60:40; 40:60 e 20:80). Em cada experimento, utilizaram-se oito novilhos Nelore (331±8 kg PV), cânulados no rumen, distribuídos em duplo quadrado latino 4x4 balanceados para o controle do efeito residual. No experimento 1 (cana-de-açúcar), o aumento da proporção de concentrado na dieta reduziu a população de Fibrobacter succinogenes e Ruminococus flavefaciens, e aumentou a população de Selenomonas ruminantium e Megasphaera elsdenii, porém o CDFDN não foi alterado. O aumento da participação de carboidratos não estruturais na dieta favoreceu a síntese de proteína microbiana e reduziu a população de bactérias fibrolíticas. A cana-de-açúcar como fonte de volumoso em dieta com proporções crescente de concentrado pode otimizar a síntese de proteína microbiana sem alterar digestibilidade da fibra. No experimento 2 (feno de Tifton 85), o aumento da proporção de concentrado na dieta reduziu a população de Fibrobacter succinogenes, Ruminococus flavefaciens e Ruminococcus albus e aumentou a população de Selenomonas ruminantium e Megasphaera elsdenii e Streptococcus bovis e o CDFDN diminuiu com o aumento da proporção de concentrado na dieta. Feno de Tifton 85 em dietas com altas proporções de concentrado pode minimizar o risco de distúrbios ruminais em novilhos confinados / This trial aimed to characterize the changes in ruminal microbiota, ruminal fermentation, intake, diet digestibility and microbial efficiency in Nellore steers fed with different forage:concentrate proportions, using as sources of forage Tifton 85 hay or sugar cane. Two experiments were conducted: in experiment 1 the sugar cane was used as forage source and in Experiment 2 the Tifton 85 hay was used as forage source. In both experiments were tested different F:C proportions (70:30, 60:40, 40:60 and 20:80). On each experiment were usedeight Nellore steers ( 331 ± 8 kg BW) cannulated in the rumen, in a double latin square 4x4 balanced to control the residual effect. In experiment 1 (sugar cane), increasing the proportion of concentrate in the diet reduced the population of Fibrobacter succinogenes and Ruminococus flavefaciens, and increased the population of Selenomonas ruminantium and Megasphaera elsdenii, but the digestibility of NDF has not changed. The increased participation of non-structural carbohydrates in the diet favored microbial protein synthesis and reduced the population of fibrolytic bacteria. The use of sugar cane as forage source associated with the increases of concentrate proportions in the diet can optimize microbial protein synthesis without change the fiber digestibility. In experiment 2 (Tifton 85 hay), increasing the proportion of concentrate in the diet reduced the population of Fibrobacter succinogenes, Ruminococcus albus and Ruminococus flavefaciens and increased the population of Selenomonas ruminantium, Megasphaera elsdenii and Streptococcus bovis and the digestibility of NDF decreased when increases the proportion of concentrate in the diet. Tifton 85 hay in diets with high proportions of concentrate can minimize the risk of ruminal disorders in feedlot steers

L'acidose ruminale est l'une des préoccupations majeures des exploitations laitières actuelles. Les levures vivantes (LV) ont été largement étudiées et utilisées chez les vaches laitières pour stabiliser la fermentation ruminale. Récemment, la mesure du potentiel redox ruminal (Eh, en mV) a été considérée comme un outil intéressant pour indiquer le trouble de la fermentation ruminale. L'effet positif de LV sur Eh ruminal a été rapporté, mais il reste variable selon les conditions expérimentales. Les objectifs de ce travail étaient de fournir une meilleure compréhension du mode d'action de LV et de définir la condition optimale de l'utilisation de LV chez les vaches laitières. La première partie de ce travail a consisté en une analyse quantitative des résultats de 22 expériences avec des vaches laitières canulées. La deuxième partie de ce travail a consisté à vérifier certains des résultats de l'analyse quantitative par une expérience chez des vaches en lactation. En utilisant l'analyse quantitative de données existantes provenant d'expériences antérieures, nous avons clarifié la relation entre le Eh ruminal et d'autres paramètres ruminaux principaux tels que le pH et le profil VFA, et suggéré que les variations de Eh pourraient être liées au transfert d'électrons dans les réactions dans le rumen. En outre, la réponse du Eh après la supplémentation en LV était également liée à celle du profil AGV ruminal, suggérant que l'effet de LV sur le profil VFA était atteint par l'augmentation du pouvoir réducteur, reflétant un meilleur transfert d'électrons dans le rumen. L'analyse a en outre démontré que la régulation du Eh ruminal par LV serait particulièrement efficace lorsque le risque de troubles digestifs est élevé. Puisque l'influence des caractéristiques de la ration sur le Eh ruminal a été quantifiée, l'effet de LV dans un régime donné pourrait être estimé indirectement. En outre, l'analyse quantitative a également révélé que la réponse de Eh suite à la supplémentation en LV était associée à la quantité de sucres solubles ingérée. L'expérience in vivo chez des vaches en début de lactation a confirmé un effet plus important de LV sur Eh ruminal avec une ration riche en sucres solubles, et a démontré que la supplémentation en LV avait un impact sur la richesse des bactéries, et que les métabolites ont également été influencés par la supplémentation en LV, probablement associée à la diminution du Eh ruminal. / Ruminal acidosis is one of the major concerns of current dairy farms. Live yeasts (LY) have been extensively studied and used in dairy cows for stabilization of rumen fermentation. Recently, measurement of ruminal redox potential (Eh, in mV) has been considered as an interesting tool to indicate ruminal fermentation disorder. The positive effect of LY on ruminal Eh has been reported, but it remains variable according to the experimental conditions. The aims of this work was to provide better understanding of mode of actions of LY, and to define the optimal condition of LY utilization in dairy cows. The first part of this work consisted to quantitative analysis of existing results from 22 experiments with cannulated dairy cattle. The second part of this work consisted to verify some of the results from quantitative analysis by an in vivo experiment in lactating cows. By using quantitative analysis of existing data from previously conducted experiments, we clarified the relationship between ruminal redox and other main ruminal parameters such as pH and VFA profile, and suggested that Eh variations might be related to the transfer of electrons in the reactions producing VFAs in the rumen. Moreover, response of ruminal Eh following live yeast supplementation was also related to that of ruminal VFA profile, which suggested that the effect of LY on VFA profile was achieved via the increase of reducing power, possibly reflected improved electron transfer and use in the rumen. The analysis further demonstrated that the regulation of ruminal Eh by LY would be particularly effective when risk of digestive disorder is high. Since the influence of dietary characteristics on ruminal Eh was quantified, the effect of LY in a given diet could be indirectly estimated. In addition, quantitative analysis also associated the response of ruminal Eh following LY supplementation to the intake of soluble sugars. The in vivo experiment in early-lactating cows confirmed greater effect of LY on ruminal Eh in diet rich in soluble sugars, and further demonstrated that i) LY supplementation tended to impact the richness of ruminal bacteria, and ii) some unidentified metabolites were also influenced by LY supplementation, probably associated to the decrease of ruminal Eh.

L'objectif de ce travail est de caractériser l'effet des levures vivantes Sc47 sur le métabolisme ruminal et l'utilisation digestive de la ration en lien avec le statut oxydo-réducteur du rumen chez des vaches laitières recevant des rations qui diffèrent dans leur composition. En effet, le milieu ruminal est très anaérobie et réducteur : les micro-organismes qui s'y développent sont à l'origine de ces conditions physico-chimiques particulières caractérisées par des valeurs de potentiel rédox (Eh) ou indice de Clark (rH) basses : respectivement dans cette étude entre –213 et –147 mV et entre 6.04 et 7.48 unités rH. L'impact de trois constituants du régime des vaches laitières d'une part sur le statut réducteur du milieu ruminal et d'autre part sur son métabolisme et/ou l'utilisation digestive de la ration a été testé en interaction avec l'addition de levures vivantes : un fourrage sec chez la vache tarie, deux concentrés azotés différant par le niveau de solubilité ruminal des protéines qu'il contient et enfin la nature de l'amidon, rapidement ou lentement dégradable dans le rumen chez la vache en lactation. Il apparaît nettement que le régime alimentaire ainsi que le niveau d'ingestion des animaux influencent directement le statut réducteur du rumen : un niveau réducteur bas étant favorable à l'activité d'une flore cellulolytique alors que des niveaux plus élevés s'avèrent être un indicateur de l'apparition de troubles métaboliques du rumen. En effet, le pouvoir réducteur ruminal est corrélé avec l'activité fermentaire qui s'y développe ainsi qu'à la structure des communautés bactériennes en présence. D'après nos résultats, l'effet de la levure vivante sur les conditions réductrices du rumen semble fortement conditionné par le niveau du statut réducteur induit par le régime alimentaire : les conditions réductrices ruminales peuvent être renforcées dans la mesure où le Eh initial se trouve, dans notre étude, entre –174 et –152 mV. Ce renforcement serait favorable à l'activité de la flore fibrolytique si celle-ci n'est pas dominante, ce groupe fonctionnel participant alors également au maintien des conditons réductrices. En outre, la levure vivante aurait un impact direct sur les bactéries protéolytiques favorisant la quantité de protéines "by pass" si bien que la valeur azotée de la ration peut être améliorée en particulier en termes de PDIA. / The objective of this work is to characterize the effect of live yeast Sc47 on ruminal metabolism and ration digestibility in relation to ruminal redox status of dairy cows fed diets that differed in their composition. Indeed, the rumen environment is very anaerobic and very reducing : the inhabiting micro-organisms are the main sources of particular physic-chemical conditions characterized by low values of redox potential (Eh) or Clark’s exponent (rH): in this study, values were between –213 and –147 mV and between 6.04 and 7.48 units for Eh and rH, respectively. The impact of three constituting ingredients of the diet of dairy cows was investigated. On the one hand, the reducing status of rumen was observed and on the other hand, the metabolism and/or digestibility of the diet with or without addition of live yeast was explored. The tested constituents were: hay for dry dairy cows, two concentrates differing in nitrogen levels of ruminal solubility, and two energetic concentrates differing in rate of ruminal degradation of starch – quickly or slowly degradable in the rumen of lactating dairy cows. It clearly appeared that the diet and the level of dry matter intakes of animals directly influenced the reducing ruminal status: low levels being favorable to the activity of cellulolytic microflora whereas higher levels appeared to be indicators of metabolic disorders occuring in the rumen. Indeed, the reducing power was correlated to rumen fermentative activity and the structure of bacterial communities involved. According to our results, the effect of live yeast on rumen reducing conditions appeared strongly influenced by the level of the intrinsic reducing status induced by diet: reducing conditions in the rumen could be strengthened to the extent that the original values ranged between –174 and –152 mV. This reinforced ruminal conditions would uphold the activity of fibrolytic microflora if it is not dominant, this functional group then participating to the maintenance of reducing conditions. In addition, live yeast has a direct impact on proteolytic bacteria promoting by-pass proteins so that the protein value of the ration may be improved especially in terms of PDIA.