Recipient Organization
UNIVERSITY OF CALIFORNIA, DAVIS
410 MRAK HALL
DAVIS,CA 95616-8671
Performing Department
Animal Science
Non Technical Summary
Ruminants (cows, sheep, goats) are unique because most of the protein they digest comes from microbes growing in the forestomach (rumen). Increasing this microbial protein is a perennial goal of ruminant nutrition, but progress has stalled because microbes grow with poor efficiency and for reasons that are unclear. In fact, we do not know how inefficient microbes really are; it is uncertain how much cellular energy (ATP) they generate during metabolism and have available for growth. Our overarching hypothesis is that microbes grow with poor efficiency because they direct ATP energy away from growth and towards energy sinks. We further hypothesize that microbes generate more ATP energy during metabolism than previously realized, and thus they grow with even poorer efficiency than thought. To test these hypotheses, we propose the specific aims of 1) determining how much ATP energy microbes expend on two energy sinks (maintenance and glycogen synthesis) 2) determining how much ATP energy cells yield during metabolism of carbohydrate (glucose). We will use experiments spanning from laboratory cultures of bacteria to dairy cows, and we expect our results to show both energy sinks (maintenance, glycogen synthesis) are large and important to decreasing efficiency. However, we expect the exact importance of each sink varies with the animal diet and microbial group. We further expect our results to show that microbes generate up to 50% more ATP energy during metabolism of glucose than previously recognized. We anticipate the impact of this project will be to reveal mechanisms behind poor growth efficiency, guide strategies to increase microbial efficiency, and reduce feeding of protein to ruminants. We anticipate that we could improve efficiency by at least 5%, reducing feed costs for the US dairy industry by $122 million/yr.
Animal Health Component
10%
Research Effort Categories
Basic
90%
Applied
10%
Developmental
(N/A)
Goals / Objectives
Objective 1. Determine how much ATP that mixed rumen bacteria expend on maintenance when provided differentcarbohydrates (glucose, starch, fiber).Objective 2. Determine how much ATP that mixed rumen microbes expend on glycogen synthesis in vivo.Objective 3. Determine the ATP yield from the major fermentation pathways of glucose (hexose).
Project Methods
Aim 1.Determine how much ATP that mixed rumen bacteria expend on maintenance when provided different carbohydrates (glucose, starch, fiber).Study 1 (Mixed bacteria)We will inoculate a system of continuous-culture fermenters with rumen contents from four lactating cows fed a corn silage based diet. To maintain bacteria at different growth rates, fermenters will be maintained at a range of dilution rates (2.0, 6.0, 8.0, 12.0/% h) by adjusting rate of buffer inflow. Carbohydrate (fiber, starch, or glucose) will be manually fed at 2-h intervals (to maintain near steady-state conditions). Buffer will be infused continuously and will include N sources (trypticase peptides, ammonia, urea), minerals, and clarified rumen fluid. After the 7-d adaptation period, we will sample fermenter contents and effluent once daily over a 3-d measurement period. We will measure concentration of bacterial protein (g L-1) in fermenter contents by using quantitative PCR against 16S ribosomal DNA as described in Aim 2 (below). The composition of bacterial taxa will be determined with sequencing as we have described previously.Study 2 (Pure culture of bacteria)In order to determine if maintenance is impact by carbohydrate source per se (not shifts in bacterial taxa), we will repeat experiments in study 1 but with a single pure culture. We will use Fibrobacter succinogenes S85, which is one of the few bacteria that can grow on either fiber (cellulose) or glucose.AnalysisFor study 1, we will calculate maintenance using regression and then determine if its value differs across carbohydrate sources. Expressing results on an ATP basis (mol ATP-1 g cell protein-1 h-1), we will determine maintenance as ATP production at zero growth rate. After calculating maintenance, we will formally determine if maintenance differs across carbohydrate sources (glucose, starch, fiber) using ANOVA. We will determine if composition of bacterial taxa, also, differs across energy sources. Using 16S ribosomal DNA sequence data, we will calculate within-sample (α) and between-sample (β) diversity using QIIME and procedures we have established, then compare values of these diversity metrics across energy sources using ANOVA. For study 2, experiments for the pure culture will be designed, analyzed, and interpreted similarly, but starch will not be included as a treatment, and no sequence data will be analyzed.Aim 2.Determine how much ATP that mixed rumen microbes expend on glycogen synthesis in vivo.Study 1 (Flows of microbial glycogen and protein)Ruminally-cannulated cows (n = 8) in late lactation will be fed a corn-silage based ration in two equal meals. Following a 14-d adaptation period, we will determine flow of microbial glycogen and protein over a 4-d measurement period. To determine these flows, we will measure 1) total digesta flow (kg/d) using the triple marker technique and omasal sampling, 2) concentration of microbes within digesta (kg/kg) using qPCR, and 3) concentration of glycogen and protein within microbes (kg/kg) using chemical analysis. To determine the concentration of protozoa and bacteria in omasal digesta, we will use small ribosomal subunit DNA (16S and 18S rDNA) as a microbial marker. To determine the concentration of glycogen in isolated bacteria and protozoa, we will use glucose released from glycogen after enzymatic hydrolysis.Study 2 (Dynamics and source carbohydrate for glycogen deposition)We will use n = 4 ruminally-cannulated cows managed as in study 1. Following 14 d of adaptation, we will pulse dose 13Clabeled carbohydrate (glucose or maize cellulose in amounts below) via the cannula during the morning feeding (n = 4 animals per carbohydrate). At 4-h intervals over a 48-h measurement period, we will sample rumen contents, isolate groups as described above, and measure 13C-enrichment of glycogen with isotope ratio mass spectrometry. For study 1, we will calculate ATP required to synthesize microbial glycogen and protein flowing through the omasum. We compare these values (ATP required for glycogen vs. protein) to determine if glycogen is indeed a major sink for ATP. First, we will calculate omasal flows of glycogen and protein, then calculate total ATP required for synthesis glycogen and protein. We will do calculations for bacteria and protozoa separately, and then formally compare values for these two microbial groups using ANOVA. For study 2, we will determine the fraction of carbohydrate deposited into glycogen (mol/mol) over time from 1) total glycogen, 2) 13C-enrichment of glycogen, and 3) 13C dosed. We will compare the fraction deposited across microbial group (protozoa, bacteria) and times after feeding using ANOVA.Aim 3.Determine the ATP yield from the major fermentation pathways of glucose (hexose)Study 1 (Ion pumps Rnf and Ech)We will determine if Ech and Rnf have activity in 1) pure cultures of bacteria and 2) mixed bacteria isolated from the rumen. We will include bacteria that encode and are expected to have activity for 1) both Rnf and Ech or 2) Rnf only. These bacteria are selected based on genome searches. In additional experiments, mixed bacteria will be isolated from rumen fluid as described for Aim 2. To assay for activity of Rnf, we will first prepare cell membranes by 1) disrupting cells in a French pressure cell press then 2) pelleting membranes by ultracentrifugation. All components for the assay (e.g., membranes, buffer, redox cofactors) will be added to a cuvette. Rnf activity will be monitored spectroscopically by reduction of the oxidized NAD (NADox) in the presence of reduced ferredoxin (Fdred). To assay for activity of Ech, we will follow a two-step approach. First, we will determine if bacteria produce hydrogen from reduced ferredoxin. Second, we will determine if bacteria pump protons in response to adding reduced ferredoxin.Study 2 (Glycolytic enzyme Pfp)Using similar approaches to study 1, we will determine if bacteria and protozoa have an unconventional glycolytic enzyme, Pfp, which spares ATP during glycolysis. For bacteria, we will also determine if bacteria have the proton pump HppA, which generates pyrophosphate needed by the enzyme Pfp. For pure cultures of bacteria, we will include species that encode and are expected to have activity of Pfp.As controls, we will include bacteria that do not encode Pfp.Mixed bacteria and protozoa will be isolated from rumen fluid as described for Aim 2. To assay for activity of Pfp, we will prepare cytoplasmic extract by a similar method as in study 1 for membranes, but we will use the supernatant (not pellet) resulting from ultracentrifugation. To determine the kinetics of Pfp vs. Pfk, we will use enzyme derived from Pr. brevis GA33, a bacterium known to have activity of these enzymes. We will produce the enzyme recombinantly in E. coli (using an Invitrogen pET TOPO cloning and expression kit). To assay for activity for the HppA proton pump, which forms pyrophosphate, we will prepare inverted membrane vesicles also using P. brevis GA33.We will determine if protons are pumped into the vesicles after adding pyrophosphate.AnalysisFor study 1, we will determine if activity of Rnf or Ech for each bacterium is different from 0 using ANOVA. For Ech, this will in fact involve examining if two different activities (hydrogen production, proton pumping) are different from 0. For study 2, we will analyze Pfp activity using methods for study 1. For recombinantly produced Pfp and Pfk, we will compare parameters Km, Vmax, and Ki using t-tests. Further, we will estimate the activity of Pfp and Pfk using these parameters, the Michaelis-Menten equation, and plugging in concentrations of substrates and products. As in study 1, all experiments will be repeated at least three times.