Item No. 1 of 1

INVESTIGATOR: JEWELL, K. A.; Jewell, K.

PERFORMING INSTITUTION:
UNIV OF WISCONSIN
21 N PARK ST STE 6401
MADISON, WISCONSIN 53715-1218

IDENTIFYING RUMINAL MICROBIAL IMPACT ON MILK PRODUCTION EFFICIENCY IN LACTATING DAIRY COWS

NON-TECHNICAL SUMMARY: The ruminant community, such as that present in the rumen of dairy cows, is responsible for the conversion of feedstock into volatile fatty acids and other compounds necessary for milk production and animal survival. The full impacts of the rumen microbe-host interaction, particularly the long-term impact of specific populations on milk production, remains incompletely understood. We hypothesize that the structure of the ruminal community (for example, presence or absence of specific microbes) will correlate with host data, inluding milk production efficiency. We will perform a long-term experiment to concurrently monitor microbial populations, rumen function, and milk production efficiency in fistulated, lactating dairy cows. Microbiome data for both bacteria and fungi will be collected using culture-independent next-generation, high-throuput pyrosequencing. Combining long-term microbial, dietary and host data will result in an unprecendented total picture of cow function not possible using other microbial characterization tecniques. The results of this study will provide an important framework for both improving our knowledge of host-microbe interactions, and for designing methods by which to maximize milk production. This information may be used in the future to guide dietary regimes or even probiotic treatments to encourage the growth of microbes associated with improving milk production efficiency and quality.

OBJECTIVES: The main objective of this study is to correlate ruminal microbial (bacterial and fungal) populations with milk production characteristics in dairy cows. The main objective will be split into three aims. Aim 1: Select high- and low-efficiency fistulated cows and maintain them on a feeding regime related to lactation for sampling periods covering the first two lactation cycles. Aim 2: Quantify milk production efficiency and host factors during two lactation cycles for high- and low-efficiency cows. Aim 3: Correlate milk efficiency and host factors to host microbiomes using next-generation pyrosequencing. During the first year we will create a preliminary model for host-microbe-production, and this model will be improved upon based upon data from the second year's samples. This final descriptive model for the interrelationship of dairy production qualities (milk production and feed consumption) with ruminal microbial populations and population structure will be useful in the future to guide dietary regimes or even probiotic treatments to encourage the growth of specific microbes associated with improving milk production efficiency and quality. Preliminary and final results from this project will be presented at the 2013 American Society for Microbiology General Meeting, the 2013 Congress on Gastrointestinal Function, and the 2014 International Symposium on Microbial Ecology. The data from this study will also be used to write manuscript(s) for publication in peer-reviewed journals, and will be made available for usage in University of Wisconsin Extension Services publications and lectures. Finally, this project will involve multiple undergraduates, with a recruitment emphasis on students interested in pursuing careers in dairy, animal, and veterinary sciences.

APPROACH: Aim 1: Twenty first-lactation, fistulated Holstein heifers will be selected and maintained over the course of two lactation cycles. These cows will be maintained on a single, herd-standard milking diet during the sampling periods, and it is expected that among these twenty animals we will be able to form two cohorts of high- and low-efficiency milk production cows. A successful cohort will be considered established by the identification and maintenance of healthy animals throughout the term of the study. Aim 2: Milk production efficiency will be calculated at three periods over the course of two lactation cycles for the cohorts selected in Aim 1. Early (~75 days in milk), middle (~150 days in milk) and late (~250 days in milk) lactation and feeding data will be collected to coincide with rumen sampling. Volatile fatty acid profiles will be determined on rumen liquids using HPLC, and rumen pH measured during sample collection. This data will be used to determine the changes over time for the cows as they progress through two lactation cycles. Data from this aim will be used to monitor the performance of each animal, with classification of "high" and "low" efficiency dependent upon the results of this aim. Aim 3: We will extract total genomic DNA from the rumen liquids and solids following a standard disruption/phenol-chloroform protocol and use this DNA as template for 16S rDNA (bacterial) and 18S rDNA (fungal) gene sequencing. Sequencing will be performed using next-generation 454 pyrosequencing technology on a Roche GS Junior machine, with data analysis performed using the bioinformatic software suite MOTHUR. This aim will be considered successful if we can establish complete sequence coverage for both bacterial and fungal ruminal populations (as determined by rarefaction curve analysis of the sequencing data). The entire study will be considered successful if we can build a model for the interrelationships of ruminal microbes with host factors, whether or not a significant pattern can be found linking these elements.

PROGRESS: 2012/08 TO 2014/07
Target Audience: Nothing Reported Changes/Problems: The initial goal of this project was to sequence both the bacterial and fungal communities present in the rumen. After much in-group debate and effort this goal was changed to sequencing only the bacterial community. The decision to focus on the bacteria was based on the following arguments: (1) other recently published research has shown that the fungal community is low in diversity, making the use of large-scale sequencing inappropriate; (2) the ruminal bacteria have been demonstrated to be responsible for the bulk of feed digestion and volatile fatty acid production; (3) sequencing the bacterial community to the desired coverage depth required the use of six sequencing runs on the 454 GS Junior, which was the entire initial sequencing budget. Because the fungal community is still of interest and may play in important role in determining the total efficiency of the lactating cow, we are discussing continuing this project after the term of the Fellowship ends with the inclusion of a targeted sequencing technique, such as quantitative PCR. All DNA from the ruminal samples is stored at -80C for potential future studies. What opportunities for training and professional development has the project provided? During the course of this Fellowship K. Jewell directly trained two high school students (Elizabeth Strassman and Caroline McCormick), two undergraduates (Jacob Breaker and Amy Speich), a rotating Ph.D. graduate student (Brent Anderson), and a visiting doctoral student from Nigeria (Lukman Omoniyi) in the methods and theory required for the Fellowship experiments. Elizabeth Strassman is currently enrolled at the University of Wisconsin-Milwaukee and is an active member of the Air Force, with the goal of pursuing a career in medicine as a surgeon. Caroline McCormick is matriculating at the University of Wisconsin-Madison this fall (2014), will be continuing to perform research in our lab, and intends to pursue a career as a medical examiner. Jacob Breaker is matriculating this fall (2014) at the University of Wisconsin-Madison Veterinary school and intends to become a large-animal veterinarian. Amy Speich is majoring in food science at UW-Madison and intends to join her family dairy business post-graduation. Brent Anderson is still studying for his degree at UW-Madison in the lab of Dr. Wang, and Lukman Omoniyi has graduated with his Ph.D in animal nutrition from the University of Nigeria. The experiences that they gained during their involvement with this Fellowship explicitly helped each of these students in their studies and determining their career paths, and gave K. Jewell continued education in mentee training and counseling. How have the results been disseminated to communities of interest? As reported under "Other Products," K. Jewell organized and led two two-hour educational sessions for high and middle school students on rumen function using a live, fistulated cow for the Science Olympiad Badger Invitational Tournament 2014 Workshop, UW-Madison. The major results were also presented at a public seminar at UW-Madison as a part of K. Jewell's dissertation defense. The following talks were presented by the advisor of K. Jewell, Dr. Garret Suen: 1. Suen, G. The Influence of the Ruminal Microbiome on Milk Production in Dairy Cows. Invited Speaker, Department of Microbiology, Cornell University, Ithaca, NY, USA, March 27, 2014. Invited by: Prof. Esther R. Angert. 2. Suen, G. The Role of Symbiotic Microbial Communities in Two Prolific North and South American Herbivores. Invited Speaker, Department of Microbiology, Universidade Federal de Viçosa, Viçosa, Minas Gerais, Brazil, July 24, 2013. Invited by: Prof. Hilario C. Montovani, Universidade Federal de Viçosa. 3. Suen, G. The Ruminal Microbial Ecosystem and its Influence on Milk Production in Dairy Cows. Invited Speaker, Department of Microbiology and Molecular Genetics, Michigan State University, East Lansing, MI, USA, March 12, 2013. Invited by: Prof. Lee Kroos, Michigan State University. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

IMPACT: 2012/08 TO 2014/07
What was accomplished under these goals? (1) From our starting group of 22 cows, we were able to maintain 14 fistulated Holstein dairy cows through their first two full lactations as a research herd. We collected milk, feed, and ruminal samples from all 14 animals at early, middle, and later periods of lactation for both cycles. We collected milk production and feed consumption data for all cows at all periods, and body weight fortnightly for all cows during their second lactation cycle. (2) Using the cows' milk and feed parameters, we calculated the gross feed efficiency (GFE) of each cow during each sampling period. We collected and processed a total of 504 ruminal samples for total DNA extraction and for volatile fatty acid (VFA) profile, 1176 milk samples for composition, and 52 feed samples with 84 feed refusal samples for compositional analysis. (3) We completed sequencing of the bacterial 16S rRNA gene from the liquid and solid ruminal fractions for all samples using an in-house Roche 454 GS Junior pyrosequencer. We analyzed the resulting sequence data in terms of GFE and other study-associated factors, and found significant correlations between the total ruminal bacterial community structure and GFE, as well as specific members within that community. The two-year data set also revealed consistent and surprising shifts in the total ruminal bacterial community as the animals aged. We wrote three manuscripts ready for submission for publication or already published: one on a collaboration project on West African Dwarf Sheep with a Nigerian research group performed by K. Jewell using the techniques learned through this Fellowship (published in 2014); one on a secondary question developed during the course of this Fellowship regarding methodological during the generation of 16S rRNA gene sequence libraries; and the third presenting the combined results of the two-year efficiency and ruminal community study.

PUBLICATIONS (not previously reported): 2012/08 TO 2014/07
1. Type: Journal Articles Status: Published Year Published: 2014 Citation: L.A. Omoniyi, K.A. Jewell, O.A. Isah, A.P. Neumann, C.F.I. Onwuka, O.M. Onagbesan, G. Suen. An analysis of the ruminal bacterial microbiota in West African Dwarf sheep fed grass- and tree-based diets. Journal of Applied Microbiology. 116(5), 1094-1105.
2. Type: Journal Articles Status: Other Year Published: 2014 Citation: Kelsea Jewell, Caroline McCormick, Christine Odt, Paul Weimer, Garret Suen. Long-term ruminal bacterial community dynamics in lactating dairy cows as they relate to host milk production. Applied and Environmental Microbiology. Manuscript in prep.
3. Type: Journal Articles Status: Other Year Published: 2014 Citation: Kelsea Jewell, Elizabeth Strassman, Caroline McCormick, Alissa Hanshew, Heidi Goodrich-Blair, Katherine McMahon, Nasia Safdar, Cameron Currie, Garret Suen. 16S rRNA-based community analyses are biased by initial PCR amplification. Journal of Microbiological Methods. Manuscript in prep.

PROGRESS: 2012/08/01 TO 2013/07/31
Target Audience: We aim to help bioinformatics researchers understand how their techniques can be used to answer applied research questions in fields not traditionally associated with their work, and how lab work can be used as a way to encourage younger students to have an interest in research science. As a part of answering this goal Kelsea Jewell presented at the Midwest Association of Core Directors Meeting on Oct. 25th, 2012 with a talk titled "454 Pyrosequencing in education and research: Undergraduates and academics." The experimental design and early sample collection for our NIFA grant project was shown as part of a presentation given at the International Society for Microbial Ecology (ISME) meeting Aug. 19-24th, 2012 titled "Determining the relationships between ruminal microbes and milk production efficiency in dairy cows." ISME conferences bring together microbiologists, bioinformaticists, environmental scientists, ecologists, and a wide variety of related researchers and professional scientists. Changes/Problems: The only significant change to our project has been a shifting of our timeline to presenting our results at later sets of conferences from originally anticipated. This was occasioned by the time needed to process the large number of samples collected during the first lactation cycle, especially in regards to compositional analysis of feed refusals. We have since adjusted our protocols to increase the efficiency of sample processing. What opportunities for training and professional development has the project provided? We recruited a high school junior, Caroline McCormick, as a mentee to Kelsea Jewell through the Youth Apprenticeship Program (YAP) in Wisconsin. This student has been trained in sampling and data analysis protocols, and has been active in collecting and processing samples for this research project. In addition, we have had the involvement of: a visiting doctoral student from Nigeria (Lukman Omoniyi), an undergraduate from a working Wisconsin dairy farm (Amy Speich), an undergraduate intending to become a large-animal veterinarian (Jacob Breaker), and a rotating graduate student (Brent Anderson). How have the results been disseminated to communities of interest? Study structure and early observations have been used in educational and scientific presentations as outlined previously in the section on "Target Audiences." Because of the length of time required for full sample processing and determination of which animals to use for microbiome analysis, we were unable to have enough preliminary data to justify attending the 2013 ASM general meeting. We still anticipate having preliminary, and then full, data available in time for later conferences and publications. What do you plan to do during the next reporting period to accomplish the goals? As a part of our goal to disseminate our information to a wide variety of audiences we have become involved in several local, public lectures, Kelsea Jewell has given presentations on this project and previous work for the seminar series "Wednesday Nite @ The Lab" (Aug. 28th, 2013) on the UW-Madison campus, "Science Happens in Taverns" (Sept 18th, 2013) at The Library Tavern in Madison, WI, and will be giving a seminar at the "PLATO Frontiers of Life Sciences" (Oct. 2, 2013) at the UW Space Place in Madison, WI. As a part of professional development for Kelsea Jewell, she attended the "26th ADSA Discover Conference on Food and Animal Agriculture: Dairy Feed Efficiency" (Sept. 23-26th, 2013). It is still our intention to have the final results of this study presented at the 2014 ISME conference, and to produce peer-reviewed publications from our results. We are currently preparing the data from the preliminary work done in support of this grant as a manuscript, with grant acknowledgement.

IMPACT: 2012/08/01 TO 2013/07/31
What was accomplished under these goals? Aim 1: We completed one full cycle of monitoring and sampling for 18 of our original 22 cohort animals. From our cohort, we have selected 14 cows of particular interest based upon their residual feed intake (RFI) values and fat corrected milk (FCM) to corrected feed consumed ratios (corrected based on refusal compositional analysis) as compared to the total herd average values. We have selected cows based on first lactation data, with an emphasis placed on those animals (a) consistently higher than herd average (with low RFI), (b) consistently lower than herd average (with high RFI), and (c) showing a marked increased or decrease in these values over the course of their entire first lactation. For these 14 cows we have begun rumen sample processing so as to created a library of microbial DNA from rumen solid and liquid fractions. The samples from all cohort cows have been retained, and all cohort cows are still being tracked through their second lactation. Aim 2: We collected milk samples and milk weights from all cows for periods of seven days overlapping their early, middle, and late lactation ruminal sample collection dates. The composition of these samples was determined and used to calculated FCM values that we used as a part of Aim 1. We also recorded and tracked mastitis (sub-clinical and clinical) and other health conditions (ruminal pH during sampling periods, incidences of morbidity as reported by farm staff between sampling periods). We have stored ruminal liquids for later processing for volatile fatty acid analysis by HPLC. Aim 3: We have not yet begun pyrosequencing.

PUBLICATIONS: 2012/08/01 TO 2013/07/31
No publications reported this period.