Synergistic fibrolysis in the rumen by cellulolytic Ruminococcus flavefaciens and non-cellulolytic Selenomonas ruminantium: Evidence in defined cultures

To investigate the rumen bacterial interaction between cellulolytic Ruminococcus flavefaciens and non‐cellulolytic Selenomonas ruminantium, fiber digestibility and fermentation products were determined in defined cultures consisting of these two species. Avicel, orchardgrass hay, rice straw and alfalfa hay were used as substrates for 72 h incubation to monitor digestibility, volatile fatty acids, succinate, lactate and bacterial number. In monoculture, R. flavefaciens digested the fiber sources at 21–32%, while S. ruminantium strains did not. When R. flavefaciens was cocultured with one of three different strains (GA192, S137 and S150) of S. ruminantium, fiber digestion exceeded the value recorded by R. flavefaciens alone. In particular, cocultures with S. ruminantium S137 showed significantly higher digestibility for all the fiber sources than R. flavefaciens alone (P < 0.05). Propionate production and growth of S. ruminantium was notable in all cocultures but not in monocultures. Succinate was accumulated in monoculture of R. flavefaciens, while the accumulation was not observed in cocultures. These results indicate that R. flavefaciens provides fiber hydrolysis products to S. ruminantium as growth substrates. In addition, S. ruminantium could activate R. flavefaciens by rapidly consuming the products. Such cross‐feeding between cellulolytic and non‐cellulolytic bacteria could enhance fiber digestion, although the extent of the enhancement may depend on strain combinations.     

Uptake of milk with and without solid feed during the monogastric phase: Effect on fibrolytic and methanogenic microorganisms in the gastrointestinal tract of calves

Microbial communities are affected by diet and play a role in the successful transition from milk to a solid diet. The response of microorganisms in the gastrointestinal tract of Holstein bull calves to the uptake of milk with solid feed (control treatment; CT), or milk without solid feed (milk‐only treatment; MT) during the first 3 weeks of life was investigated. Samples were collected from the rumen (fluid and tissue), abomasum (fluid), cecum (fluid and tissue) and feces at 7, 14 and 20 days of age. Calf weight was higher on days 14 and 20 in the MT than the CT. In the rumen at 14 days, the fibrolytic bacteria Fibrobacter succinogenes and Prevotella ruminicola increased in the CT and Ruminococcus flavefaciens increased in the MT. This suggests that R. flavefaciens is not strictly fibrolytic and that it might use milk as a substrate or other microbial species might supply a substrate. Diet affected methanogens, but this may have been due to an indirect effect via an association with Geobacter spp. or other syntrophic partners. The treatments also affected microorganisms in the abomasum, cecum and feces. Our results contribute to an understanding of diet, microbes in the gastrointestinal tract and weaning.     

Microbial population in the rumen of swamp buffalo (Bubalus bubalis) as influenced by coconut oil and mangosteen peel supplementation

Four, rumen fistulated swamp buffalo bulls were used to study microbial populations in the rumen when supplemented with coconut oil and mangosteen peel. Animals were randomly assigned to a 4 × 4 Latin square design. Four treatments were un‐supplemented (Control), supplementation with coconut oil at 50 g/kg (CO5), supplementation with mangosteen peel at 30 g/kg (MP3) and supplementation with CO5 and MP3 (COM), of total DM intake. Animals received concentrate at 10 g/kg of BW, and rice straw was given ad libitum. Abundance of total bacteria was increased by CO5 supplementation, whereas populations of protozoa and Fibrobacter succinogenes were reduced by CO5 and COM supplementation. Dietary supplementation did not affect methanogen, Ruminococcus flavefaciens or Ruminococcus albus abundances. Dietary treatments changed denaturing gradient gel electrophoresis (DGGE) band patterns of methanogens and protozoa when compared with the control group, especially when supplemented with MP3. Supplementation of COM resulted in the greatest difference in pattern of DGGE bands for total bacteria compared with the control. Coconut oil and mangosteen peel supplementation resulted in changing of rumen microbial abundances and communities; however, combination of them could be more benefit to improve rumen fermentation of swamp buffalo fed on rice straw.     

Use of bean husk as an easily digestible fiber source for activating the fibrolytic rumen bacterium Fibrobacter succinogenes and rice straw digestion

A series of in sacco and in vitro studies were carried out to evaluate bean husks for activation of fibrolytic rumen bacteria and rice straw digestion. First, lablab bean husk, chickpea husk and rice straw were suspended in the rumen of sheep to analyze the bacterial consortium developed on each fiber source. Known members of fiber‐associating bacteria were found on both lablab bean husk and rice straw, but some of these bacteria were lacking on chickpea husk. Second, a pure culture study was carried out using six strains of Fibrobacter succinogenes. Both husks stimulated the growth of all tested strains, including a strain that did not grow on rice straw. The strain OS128 that showed the highest growth on rice straw displayed even higher growth on lablab bean husk without a time lag. Finally, two‐step incubations were carried out to determine whether prior incubation of rumen fluid with husks stimulates subsequent rice straw digestion. Higher digestibility of rice straw was recorded in the second‐round incubation following the first incubation with bean husks. These results suggest that the tested bean husks improve the digestion of rice straw by activating fibrolytic F. succinogenes and other associated bacteria.     

Effect of non‐starch‐polysaccharide‐degrading enzymes as feed additive on the rumen bacterial population in non‐lactating cows quantified by real‐time PCR

The effects of non‐starch‐polysaccharide‐degrading enzymes, added to a maize silage‐ and grass silage‐based total mixed ration (TMR) at least 14 h before feeding, on the rumen bacterial population were investigated. Six non‐lactating Holstein Friesian cows were allocated to three treatment groups using a duplicate 3 × 3 Latin square design with three 31‐day periods (29 days of adaptation and 2 days of sampling). Treatments were control TMR [69% forage and 31% concentrates on a dry matter (DM) basis] or TMR with 13.8 or 27.7 ml/kg of feed DM of Roxazyme G2 liquid with activities (U/ml enzyme preparation) of xylanase 260 000, β‐glucanase 180 000 and cellulase 8000 (DSM Nutritional Products, Basel, Switzerland). The concentrations of 16S rDNA of Anaerovibrio lipolytica, Fibrobacter succinogenes, Prevotella ruminicola, Ruminococcus flavefaciens, Selenomonas ruminantium and Treponema bryantii, and their relative percentage of total bacteria in rumen samples obtained before feeding and 3 and 7 h after feeding and from two rumen fractions were determined using real‐time PCR. Sampling time had only little influence, but bacterial numbers and the composition of the population differed between the transition layer between rumen fluid and the fibre mat (fraction A) and the rumen fluid (fraction B) highlighting the importance to standardize sampling. The 16S rDNA copies of total bacteria and the six bacterial species as well as the population composition were mainly unaffected by the high levels of exogenous enzymes supplemented at all sampling times and in both rumen fractions. Occasionally, the percentages of the non‐fibrolytic species P. ruminicola and A. lipolytica changed in response to enzyme supplementation. Some increases in the potential degradability of the diet and decreases in lag time which occurred collaterally indicate that other factors than changes in numbers of non‐particle‐associated bacteria are mainly responsible for the effects of exogenous enzymes.     

Interaction of rumen bacteria as assumed by colonization patterns on untreated and alkali‐treated rice straw

Colonization patterns of representative rumen bacteria were compared between untreated rice straw (UTS) and sodium hydroxide‐treated rice straw (SHTS). UTS and SHTS were incubated in the rumen of sheep for 10 min, 1, 2, 6, 12, 24, 48 and 96 h using the nylon bag method. The population sizes of 13 representative bacterial species or groups were quantified by real‐time PCR. The total bacterial population size (abundance) was similar in both UTS and SHTS. Fibrobacter succinogenes showed a higher population size compared to other fibrolytic species and was detected at a higher level in SHTS (3.7%) than in UTS (2.6%). Ruminococcus albus and Ruminococcus flavefaciens were also detected at higher levels in SHTS (0.15% and 0.29%) than in UTS (0.03% and 0.18%). Population sizes of non‐fibrolytic species, such as Selenomonas ruminantium, Anaerovibrio lipolytica and Succinivibrio dextrinosolvens were higher in UTS than in SHTS. Coefficient of determination (r²) on population changes between bacterial species or groups were higher in UTS than in SHTS, suggesting the necessity of stronger bacterial interactions for UTS digestion. Therefore, not only colonization of fibrolytic species, but also synergistic interactions between different bacterial species may be key to the ruminal digestion of rice straw.     

Ruminal fermentation and microbial ecology of buffaloes and cattle fed the same diet

Although buffaloes and cattle are ruminants, their digestive capabilities and rumen microbial compositions are considered to be different. The purpose of this study was to compare the rumen microbial ecology of crossbred water buffaloes and cattle that were fed the same diet. Cattle exhibited a higher fermentation rate than buffaloes. Methane production and methanogen density were lower in buffaloes. Phylogenetic analysis of Fibrobacter succinogenes‐specific 16S ribosomal RNA gene clone library showed that the diversity of groups within a species was significantly different (P < 0.05) between buffalo and cattle and most of the clones were affiliated with group 2 of the species. Population densities of F. succinogenes, Ruminococcus albus and R. flavefaciens were higher until 6 h post‐feeding in cattle; however, buffaloes exhibited different traits. The population of anaerobic fungi decreased at 3 h in cattle compared to buffaloes and was similar at 0 h and 6 h. The diversity profiles of bacteria and fungi were similar in the two species. The present study showed that the profiles of the fermentation process, microbial population and diversity were similar in crossbred water buffaloes and crossbred cattle.     

In vitro gas production in rumen fluid of buffalo as affected by urea‐calcium mixture in high‐quality feed block

This study aimed to determine the effect of urea‐calcium sulphate mixture (U‐cas) levels in high‐quality feed block (HQFB) on ruminal digestibility, fermentation and gas kinetics in rumen fluid of swamp buffalo by using in vitro techniques. The treatments were seven levels of U‐cas incorporated in HQFB at 0, 3, 6, 9, 12, 15 and 18% and the experimental design was a completely randomized design. Gas production rate constants for the insoluble fraction, potential extent of gas and cumulative gas were linearly increased with increasing levels of U‐cas in HQFB. The in vitro dry matter digestibility, in vitro organic matter digestibility, true digestibility and microbial mass were altered by treatments and were greatest at 18% U‐cas supplementation. Concentrations of propionate were linearly increased with increasing levels of U‐cas and was highest with U‐cas supplementation at 18%. The NH₃–N concentration was highest when urea was added in the HQFB while NH₃–N concentration tended to be reduced with increasing level of U‐cas. The findings suggest supplementation of 18% U‐cas in HQFB improves kinetics of gas production, rumen fermentation, digestibility and microbial mass as well as controlling the rate of N degradation in the rumen of swamp buffalo.     

ORIGINAL ARTICLE: Description of development of rumen ecosystem by PCR assay in milk‐fed,weaned and finished lambs in an intensive fattening system

This study examined the reticulo‐rumen characteristics of the microbial community and its fermentative characteristics in milk‐fed, at weaning and finished lambs in a conventional fattening system. Five lambs were assigned to each of three groups: milk‐fed lambs slaughtered at 30 days (T30), weaned lambs slaughtered at 45 days (T45) and ‘finished lambs’ slaughtered at 90 days (T90). At slaughter, rumen size, fermentation parameters (pH, volatile fatty acids and microbial enzyme activity) and protozoal counts were recorded. Quantitative PCR was used to quantify the genes encoding 16S and 18S ribosomal DNA of the rumen bacterial and protozoal populations, respectively, and the sequential colonization of the rumen by cellulolytic (Ruminococcus albus, Ruminococcus flavefaciens) and amylolytic (Prevotella ruminicola, Streptococcus bovis) bacteria, and protozoa (Entodinium sp.). Denaturing gradient gel electrophoresis was used to study the development of rumen microbiota biodiversity. Intake of solid food before weaning caused a significant increase in rumen weight (p < 0.0001) and bacterial DNA (p < 0.05) and volatile fatty acid analysis concentration (p < 0.01), whereas pH declined. In milk‐fed lambs, cellulolytic bacteria were evident after 30 days. Thereafter, in the 45‐day and 90‐day groups, the proportions of R. flavefaciens decreased and R. albus increased. Amylolytic bacteria were present in milk‐fed lambs; the proportion of P. ruminicola increased in fattening lambs and S. bovis was the least abundant species. Protozoal concentrations were irregular; milk‐fed lambs had a significant number of protozoa species from Entodinium and subfamily Isotrichiidae, but they disappeared at weaning. Lamb rumen were refaunated in some individuals at 90 days (Entodinium and subfamily Diplodiniinae spp.), although individual concentrations were variable.     

The relationships between odd‐ and branched‐chain fatty acids to ruminal fermentation parameters and bacterial populations with different dietary ratios of forage and concentrate

The objectives of this study were to investigate the effect of different dietary ratios of forage and concentrate (F:C) on ruminal odd‐ and branched‐chain fatty acids (OBCFAs) contents and to evaluate the relationships between OBCFA and ruminal fermentation parameters as well as bacterial populations tested by real‐time PCR technique. The experimental design was a 3 × 3 Latin square. Three rumen‐fistulated dry Holstein cows were fed three rations with different dietary F:C ratios (F:C; 30:70, 50:50 and 70:30). The rumen samples were collected every two hours (0600, 0800, 1000, 1200, 1400, 1600, 1800, 2000, 2200, 2400, 0200 and 0400 h) over three consecutive days in each sampling period. The results showed that rumen OBCFA profiles are significantly (p < 0.05) affected by the dietary F:C ratios. The concentrations of C11:0, C13:0, iso‐C15:0, iso‐C16:0, iso‐C17:0 and C17:0 were higher in the cows fed dietary F:C ratio of 70:30 than those fed with other two rations. However, the concentrations of anteiso‐C15:0, C15:0 and total OBCFA were on the lowest level in the high forage diet. Correlation and regression analysis showed that ruminal OBCFAs had strong relationships with ruminal fermentation parameters and bacterial populations. In particular, the iso‐fatty acids had potential power to predict butyrate and isoacids metabolized in the rumen, whereas the fatty acids with 17 carbon atoms correlated with ruminal NH₃‐N content. The OBCFA contents have different relationships with fibrolytic and starch bacteria in the rumen. C17:0 and its isomers might be used to predict populations of fibrolytic bacteria.     

Bean husks as a supplemental fiber for ruminants: Potential use for activation of fibrolytic rumen bacteria to improve main forage digestion

This study evaluated the suitability of easily digested fiber sources as a supplemental fiber to improve overall fiber digestion in ruminants. First, the degradation of five fibrous feedstuffs and the stimulatory effects on rumen bacteria were examined in situ. Chickpea and lablab bean husks were selected for their potential use due to their large degradable fraction (> 94%), which had a stimulatory effect on fibrolytic rumen bacteria such as Fibrobacter succinogenes. Second, a possible improvement in the digestibility of rice straw diet by husk supplementation was monitored in vivo. Four dietary treatments comprising RS (rice straw and concentrate), CHM (RS supplemented with Myanmar chickpea husk), CHE (RS with Egyptian chickpea husk) and LH (RS with lablab bean husk) were allocated to four wethers. The digestibility of acid detergent fiber was 3.1–5.5% greater in CHM and LH than RS. Total volatile fatty acid concentration was higher in LH than other treatments. Acetate proportion was higher in LH than RS. Ruminal abundance of F. succinogenes was 1.3–1.5 times greater in CHM and LH than RS. These results suggest that bean husk supplementation, especially lablab bean husk, might improve the nutritive value of rice straw diet by stimulating fibrolytic bacteria.     

Ability to utilize lactate and related enzymes of a ruminal bacterium, Selenomonas ruminantium

The ability of a nitrate‐reducing and nitrite‐reducing strain of Selenomonas ruminantium ssp. lactilytica (TH1) to utilize lactate was examined at the cell and enzyme levels. The TH1 strain was found to possess NAD‐independent D‐lactate dehydrogenase (iD‐LDH), with little or no lactate racemase or L‐lactate dehydrogenase, implying that TH1 virtually utilizes only the D‐form of lactate. Therefore, the introduction of lactate racemase to TH1 may enhance its ability to utilize lactate in the rumen where both D‐lactate and L‐lactate are produced. Because lactate utilization by Megasphaera elsdenii in the rumen may increase methanogenesis, it is desirable to increase lactate utilization by S. ruminantium, which may decrease methanogenesis. However, the specific activity of iD‐LDH, which represents the amount of enzyme per cell, in TH1 was approximately threefold lower than M. elsdenii. Properties of iD‐LDH, such as optimal pH and temperature, affinity for D‐lactate, and effect of metal ions, did not differ greatly between TH1 and M. elsdenii. The specific activity of iD‐LDH in TH1 increased as the D‐lactate concentration in the medium increased, suggesting that iD‐LDH synthesis is regulated in response to D‐lactate. On the contrary, no iD‐LDH activity was detected when TH1 was grown in the presence of glucose, even when D‐lactate was present. This result suggests that iD‐LDH synthesis is strongly suppressed by glucose. In order to improve the ability of S. ruminantium to utilize lactate and reduce nitrate and nitrite, it is important to enhance iD‐LDH synthesis in addition to the introduction of lactate racemase.     

Genetic analysis of river,swamp and hybrid buffaloes of north‐east India throw new light on phylogeography of water buffalo (Bubalus bubalis)

This study analysed buffaloes from north‐east India and compared their nuclear and mitochondrial DNA variations with buffaloes of mainland India, China, Mediterranean and South‐East Asia. Microsatellite genotypes of 338 buffaloes including 210 from six north‐east Indian buffalo populations and three mainland Indian breeds were analysed to evaluate their genetic structure and evolutionary relationships. Phylogenetic analysis and multidimensional scaling plot of pairwise F_ST revealed the clustering of all swamp‐type buffaloes of north‐east India with Lower Assamese (significantly hybrid type) buffaloes in one plane and all the mainland river buffaloes in another plane while the upper Assamese buffaloes being distinct from both these clusters. Analysis of mtDNA D‐loop region of 530‐bp length was performed on 345 sequences belonging to 23 buffalo populations from various geographical regions to establish the phylogeography of Indian water buffalo. The swamp buffaloes of north‐east India clustered with both the lineages of Chinese swamp buffalo. Multidimensional scaling display of pairwise F_ST derived from mitochondrial DNA data showed clustering of upper Assamese, Chilika and Mediterranean buffaloes distinctly from all the other Indian buffalo populations. Median‐joining network analysis further confirmed the distinctness and ancestral nature of these buffaloes. The study revealed north‐east region of India forming part of the wider hybrid zone of water buffalo that may probably extend from north‐east India to South‐East Asia.     

Characteristics of various fibrolytic isozyme activities in the rumen microbial communities of Japanese Black and Holstein Friesian cattle under different conditions

Rumen microorganisms produce various fibrolytic enzymes and degrade lignocellulosic materials into nutrient sources for ruminants; therefore, the characterization of fibrolytic enzymes contributing to the polysaccharide degradation in the rumen microbiota is important for efficient animal production. This study characterized the fibrolytic isozyme activities of a rumen microbiota from four groups of housed cattle (1, breeding Japanese Black; 2, feedlot Japanese Black; 3, lactating Holstein Friesian; 4, dry Holstein Friesian). Rumen fluids in all cattle groups showed similar concentrations of total volatile fatty acids and reducing sugars, whereas acetic acid contents and pH were different among them. Predominant genera were commonly detected in all cattle, although the bacterial compositions were different among cattle groups. Zymograms of whole proteins in rumen fluids showed endoglucanase activities at 55 and 57 kDa and xylanase activity at 44 kDa in all cattle. Meanwhile, several fibrolytic isozyme activities differed among cattle groups and individuals. Treponema, Succinivibrio, Anaeroplasma, Succiniclasticum, Ruminococcus, and Butyrivibrio showed positive correlations with fibrolytic isozyme activities. Further, endoglucanase activity at 68 kDa was positively correlated with pH. This study suggests the characteristics of fibrolytic isozyme activities and their correlations with the rumen microbiota.     

DGAT1, GH, GHR, PRL and PRLR polymorphism in water buffalo (Bubalus bubalis)

The polymorphism of several genes has been shown to affect the milk composition traits in dairy cattle, including DGAT1‐exon8 K232A, GH‐intron3 MspI, GH‐exon5 AluI, GHR‐exon8 F279Y, PRL‐exon3 RsaI and PRLR‐exon3 S18N. However, the polymorphism and effects of these genes on the milk traits of water buffalo are still unclear. In this study, four DNA pooling samples from Murrah, Nili‐ravi, Murrah‐Nili‐Swamp crossbreed and Chinese swamp buffalo were constructed, respectively, and polymorphism of these sites was investigated using PCR–Single‐strand conformation polymorphism and sequencing. Twenty‐eight inter‐specific single‐nucleotide polymorphism (SNPs) were found in these six assayed gene fragments between buffalo and dairy cattle, including nine intra‐specific SNPs among buffalo groups. All buffalo fixed a K allele genotype in DGAT1‐exon8, MspI⁺ restriction site(c nucleotide) and AluI⁺ site(c nucleotide) at intron3 and exon5 of GH gene, F allele genotype of F279Y mutation in GHR gene, RsaI^? restriction site at PRL‐exon3/exon4 and N allele genotype of S18N mutation at PRLR‐exon3. It provides an indirect evidence that water buffalo have fixed alleles with genotypes reported in dairy cattle, which is thought to be responsible for high milk fat, high protein content and low milk yield. Moreover, three new intra‐specific SNPs were found including 275th bp (c/t) in DGAT1 of Murrah buffalo, 109th bp (t/a) in PRL‐exon3/exon4 and 43rd bp (c/t) in PRLR‐exon3 of Chinese swamp buffalo. Information provided in this study will be useful in further studies to improve buffalo breeding for better lactation performances.     

Seasonal differences in rumen bacterial flora of wild Hokkaido sika deer and partial characterization of an unknown bacterial group possibly involved in fiber digestion in winter

Rumen digesta was obtained from wild Hokkaido sika deer to compare bacterial flora between summer and winter. Bacterial flora was characterized with molecular‐based approaches and enrichment cultivation. Bacteroidetes was shown as a major phylum followed by Firmicutes, with similar proportions in both seasons. However, two phylogenetically unique groups in Bacteroidetes were found in each season: unknown group A in winter and unknown group B in summer. The ruminal abundance of unknown group A was the highest followed by Ruminococcus flavefaciens in winter. Moreover, the abundance of these two was higher in winter than in summer. In contrast, the abundance of unknown group B was higher in summer than in winter. In addition, this group showed the highest abundance in summer among the bacteria quantified. Unknown group A was successfully enriched by cultivating with oak bark and sterilized rumen fluid, particularly that from deer. Bacteria of this group were distributed in association with the solid rather than the liquid rumen fraction, and were detected as small cocci. Accordingly, unknown group A is assumed to be involved in degradation of fibrous materials. These results suggest that wild Hokkaido sika deer develop a rumen bacterial flora in response to changes in dietary conditions.     

Jersey steer ruminal papillae histology and nutrigenomics with diet changes

The transition from a high forage to a high concentrate diet is an important milestone for beef cattle moving from a stocker system to the feedlot. However, little is known about how this transition affects the rumen epithelial gene expression. This study assessed the effects of the transition from a high forage to a high concentrate diet as well as the transition from a high concentrate to a high forage diet on a variety of genes as well as ruminal papillae morphology in rumen fistulated Jersey steers. Jersey steers (n = 5) were fed either a high forage diet (80% forage and 20% grain) and transitioned to a high concentrate diet (20% forage and 80% grain) or a high concentrate diet (40% forage and 60% grain) and transitioned to a high forage diet (100% forage). Papillae from the rumen were collected for histology and RT‐qPCR analysis. Body weight had a tendency for significant difference (p = .08). Histological analysis did not show changes in papillae length or width in steers transitioning from a high forage to a high concentrate diet or vice versa (p > .05). Genes related to cell membrane structure (CLDN1, CLDN4, DSG1), fatty acid metabolism (CPT1A, ACADSB), glycolysis (PFKL), ketogenesis (HMGCL, HMGCS2, ACAT1), lactate/pyruvate (LDHA), oxidative stress (NQO1), tissue growth (AKT3, EGFR, EREG, IGFBP5, IRS1) and the urea cycle (SLC14A1) were considered in this study. Overall, genes related to fatty acid metabolism (ACADSB) and growth and development (AKT3 and IGFBP5) had a tendency for a treatment × day on trial interaction effect. These profiles may be indicators of rumen epithelial adaptations in response to changes in diet. In conclusion, these results indicate that changes in the composition of the diet can alter the expression of genes with specific functions in rumen epithelial metabolism.     

Digestibility, methane production and nitrogen balance in sheep fed ensiled or fresh mixtures of sorghum–soybean forage

The feeding value of a mixture of sorghum and soybeans plants, either fresh or ensiled, was evaluated with sheep. Sorghum and soybeans were harvested during the Cuban rainy season and ensiled in a ratio of 0.6/0.4 (w/w, as feed) with molasses and a bacterial inoculant. Silos were opened between 162 and 182 d post ensiling during the Cuban dry season and silages were fed to six pelibuey sheep (including two fistulated). Six other sheep (also including two fistulated) were fed sorghum and soybean in the same proportion, but freshly harvested during the dry season. The experiment lasted 21 d (14 d adaptation and 7 d data collection period). Silage quality parameters included pH, ammonia, lactate, and short chain fatty acids (SCFA). Further, both fresh and ensiled diets were offered to study the rumen fermentation characteristics (pH, ammonia, lactate, SCFA, in situ degradability and methane), duodenal flow of microbial protein (assessed through urinary purine derivatives secretion), fecal degradability of nutrients and urinary N excretion. From these measurements ME value and degradable CP supply at the small intestine (DCPSI) were estimated. Silage was of excellent quality. Compared to fresh forage feeding silage increased molar propionate proportion and rumen microbial protein synthesis and reduced methane emission. Fresh forage showed lower rumen degradability and total digestibility. Further, the higher ME concentration (11.2 vs. 10.3 MJ/kg DM) and DCPSI (84.7 vs. 56.1 g/kg DM) of the silage would allow a higher milk production or daily gain as compared with fresh forage available during the Cuban dry season.