Elucidating the role of the rumen microbiome in cattle feed efficiency and its 1 potential as a reservoir for novel enzyme discovery 2 3 by 4 5 André Luis Alves Neves 6 7 8 9 10 11 12 A thesis submitted in partial fulfillment of the requirements for the degree of 13 14 15 Doctor of Philosophy 16 17 in 18 19 Animal Science 20 21 22 23 24 25 Department of Agricultural, Food and Nutritional Science 26 University of Alberta 27 28 29 30 31 32 33 34 35 36 37 © André Luis Alves Neves, 2019 38 39 40 Abstract 1 2 The rapid advances in omics technologies have led to a tremendous progress in our 3 understanding of the rumen microbiome and its influence on cattle feed efficiency. 4 However, significant gaps remain in the literature concerning the driving forces that 5 influence the relationship between the rumen microbiota and host individual variation, and 6 how their interactive effects on animal productivity contribute to the identification of cattle 7 with improved feed efficiency. Furthermore, little is known about the impact of mRNA- 8 based metatranscriptomics on the analysis of rumen taxonomic profiles, and a strategy 9 for the discovery of lignocellulolytic enzymes through the targeted functional profiling of 10 carbohydrate-active enzymes (CAZymes) remains to be developed. Study 1 investigated 11 the dynamics of rumen microorganisms in cattle raised under different feeding regimens 12 (forage vs. grain) and studied the relationship among the abundance of these 13 microorganisms, host individuality and the diet. To examine host individual variation in 14 the rumen microbial abundance following dietary switches, hosts were grouped based on 15 the magnitude of microbial population shift using log2-fold change (log2-fc) in the copy 16 numbers of bacteria, archaea, protozoa and fungi. Three groups of log2-fc in the bacterial 17 and fungal abundance (Low, log2-fc < -1; Stable, -1 < log2-fc < 1; and High, log2-fc > 1) 18 were identified from the magnitude of change in baseline rumen microbial populations. 19 By monitoring the microbial population shift within the same animal in response to the 20 diet, significant ecological features of rumen microorganisms were identified and shed 21 new light on their dynamic roles in animal feed utilization and individual variation. Study 22 2 compared the outcomes of two methods, Kraken (mRNA based) and a pipeline 23 developed in-house based on Mothur (16S rRNA based), concerning the taxonomic 24 ii profiles (bacteria and archaea) of rumen microbial communities using total RNA 1 sequencing of rumen fluid collected from cattle with different feed conversion ratios (FCR). 2 Both approaches revealed a similar phyla distribution of the most abundant taxa, with 3 Bacteroidetes, Firmicutes, and Proteobacteria accounting for approximately 80% of total 4 bacterial abundance. For bacterial taxa, although 69 genera were commonly detected by 5 both methods, an additional 159 genera were exclusively identified by Kraken. Kraken 6 detected 423 species, while Mothur was not able to assign bacterial sequences to the 7 species level. For archaea, both methods generated similar results only for the 8 abundance of Methanomassiliicoccaceae and Methanobrevibacter ruminantium. 9 Although Kraken enhanced the microbial classification at the species level, identification 10 of bacteria or archaea in the rumen was limited due to a lack of reference genomes for 11 the rumen microbiome. Study 3 investigated the effect of cattle breeds on specific ruminal 12 taxonomic microbial groups and functions associated with FCR, using two genetically 13 related Angus breeds as a model. Total RNA was extracted from rumen content samples 14 collected from purebred Black and Red Angus bulls fed the same forage diet and then 15 subjected to metatranscriptomic analysis. Multivariate discriminant analysis (sPLS-DA) 16 and analysis of composition of microbiomes (ANCOM) were conducted to identify 17 microbial signatures characterizing Black and Red Angus cattle. Although Black and Red 18 Angus are genetically similar, sPLS-DA detected 25 bacterial species and ten functions 19 that differentiated the rumen microbial signatures between those two breeds. ANCOM 20 revealed an association between FCR and breed with Chitinophaga pinensis and 21 Clostridium stercorarium, suggesting that these bacterial species may play a key role in 22 the feed conversion efficiency of forage-fed bulls. Study 4 combined selective pressure 23 iii to enrich the rumen for lignocellulolytic microbes with bioinformatic tools to guide the 1 discovery of unknown CAZymes in the microbiome. It was demonstrated that the rumen 2 microbiome increased the abundance of lignocellulolytic bacteria, such as Fibrobacter 3 succinogens, and a diverse set of CAZymes over time, including 18 uncharacterized 4 members of the family GH11 (xylanases) and three of the family GH45 (endoglucanases). 5 Further experiments confirmed the lignocellulolytic activity of xylanase using such 6 approach. In summary, the data presented in this thesis provide fundamental knowledge 7 on the role of the rumen microbiome in cattle feed efficiency and offers opportunities to 8 further explore the potential of the rumen as a source for novel enzyme discovery. 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 iv Preface 1 2 This thesis is an original work by Andre Luis Alves Neves and is part of a 3 collaborative project between Profs. Leluo Guan at the University of Alberta and Prof. Kim 4 H. Ominski of the University of Manitoba. The Veterinary Services and the Animal Care 5 Committee of the University of Manitoba granted ethical approval for all experimental 6 procedures described in Chapters 2, 3, 4 and 5. 7 Chapter 2 of this thesis has been submitted to FEMS Microbiology Ecology as 8 “Dynamics of microbial populations driven by interactions between diet and host shed 9 light on individualized rumen microbiota” by Andre L. A. Neves, Yanghong Chen, Eoin 10 O’Hara, Tim McAllister, Kim H. Ominski and Le Luo Guan. (2019). ALAN was responsible 11 for laboratory and data analysis, as well as manuscript writing. YC performed the 12 laboratory analysis, while LLG, KHO, TM, and EOH revised and contributed to the writing 13 of the manuscript. 14 Chapter 3 of the thesis has been published as Andre L. A. Neves, Fuyong Li, 15 Bibaswan Ghoshal, Tim McAllister and Le Luo (2017). “Enhancing the resolution of rumen 16 microbial classification from metatranscriptomic data using Kraken and Mothur”. Frontiers 17 Microbiology; 8: 2445. doi: 10.3389/fmicb.2017.02445. ALAN and BG executed Kraken, 18 and FL executed the pipeline based on Mothur. ALAN analyzed the data and performed 19 the statistical analysis. ALAN, FL and BG executed the experiment and wrote the 20 manuscript. LLG and TM contributed to the experiment and revised the manuscript. 21 Chapter 4 has been submitted to Animal as “Taxonomic and functional 22 assessment reveals the effect of Angus breed genetics on rumen microbial signatures” 23 by Andre L. A. Neves, Yanhong Chen, Kim-Anh L. Cao, Siddhartha Mandal, Thomas J. 24 Sharpton, Tim McAllister, and Le Luo Guan. ALAN run Kraken, analyzed the data, 25 performed the statistical analysis, and wrote the manuscript. KALC and SM helped ALAN 26 run the sPLS-DA and ANCOM, respectively. TJS helped ALAN run ShotMAP. YC 27 assisted with the laboratory analysis. LLG and TM contributed to the manuscript writing. 28 v Dedication 1 2 This thesis is dedicated to my Lord and Savior Jesus Christ, who said “I am the 3 resurrection, and the life: he that believeth in me, though he were dead, yet shall he live. 4 And whosoever liveth and believeth in me shall never die. Believe thou this?”, KJV 5 Bible, John 11.25,26. 6 7 8 I sincerely appreciate your support and guidance: I love you my Lord Jesus! 9 10 This thesis is also dedicated to my beautiful wife, Euzilene Neves, and our beloved son, 11 Daniel Neves! 12 13 Thanks Lene for the patience and encouragement during this time, I love you and 14 Daniel! 15 16 17 18 19 20 21 22 23 24 25 vi Acknowledgements 1 2 First, I would like to express my gratitude to my supervisor Dr. Le Luo Guan for her 3 dedication, mentoring, advice and patience during my Ph.D. program. I am proud to say 4 to everyone that I had an excellent supervisor, who provided me with the best 5 opportunities to learn and grow in various aspects of my professional career and personal 6 life. You may have no idea about the important role you played in my life, Dr. Guan! 7 Throughout these four years, you helped me get rid of all academic fears, and I am 8 passionate about science because you taught me so. Thanks for accepting me as your 9 student in May/2015 and for making me what I am now. I will take your academic and life 10 teachings for the rest of my days! Thank you very much! 11 I also want to thank the funding bodies AITF for the doctorate scholarship and 12 EMBRAPA for the financial support during the program. 13 To my committee members: Drs. Tim McAllister, Michael Steele, and Paul Stothard. 14 Thanks for providing me with constructive suggestions on my projects that contributed to 15 my professional development. Additionally, I want to thank Dr. McAllister for introducing 16 me to Dr. Guan in 2015 and for providing me with resources to improve my English skills 17 when I started writing my first paper in English in 2015. Since then, I have written in a 18 much faster pace! I also appreciated being a student of Drs. Steele and Stothard. I 19 enjoyed all the classes at the UofA, especially those at DRTC! 20 I thank Dr.