Omics techniques for rumen microbiology › wp-content › uploads › 2014 › 04 › Matthias-He… · SMB Cow rumen metagenome Library name Insert Size Sequencing Bases (Gb) Reads

SMB

Matthias Hess M.Sc., Ph.D.

Wiley Research Fellow – DOE Environmental Molecular Science Laboratory Assistant Professor – Washington State University

Omics techniques for rumen microbiology

Joint RuminOmics/Rumen Microbial Genomics Network (RO/RMGN) Workshop, Aberdeen, UK June 16th, 2014

SMB

Outline

•  Introduction •  The Hess Lab •  Rumen Systems Microbiology

•  Omics Techniques •  Challenges & solutions

•  Perspectives

SMB

The Hess Lab

Ecosystem-driven enzyme discovery

Cultivation-based techniques

Cultivation-independent techniques

Wet-lab experiments Bioinformatics

SMB

•  Impacts health and productivity of host animals •  Rumen microbes produce ~21% of anthropogenic methane •  A complex system

•  1010 bacteria/mL •  106 protozoa/mL •  105 fungi/mL

The Rumen

SMB

Systems Microbiology toolbox

Warnecke and Hugenholtz 2007; Genome Biol. 8; 231

SMB

•  Introduction •  The Hess Lab •  Rumen Systems Microbiology

•  Omics Techniques •  Challenges & solutions

•  Perspectives

Outline

SMB

Challenge: Dynamic & complex environment

Koropatkin; Nature Reviews Microbiology 10; 323

Spatial dynamics

Carbohydrate utilization along the length of the digestive tract: •  variable transit times •  different carbohydrates are digested in different gut regions •  complex microbe-microbe and microbe-host interactions

Interspecies-Interactions

SMB

High Resolution Imaging

Piao et al 2014; Frontiers in Microbiology

Visual atlas of the rumen microbiome: Request for pure cultures with FISH probe available

SMB

DNA, RNA and protein profiles

Direct sequencing of DNA, RNA and proteins extracted from an environmental sample can provide snapshots of the complete microbiome and subsequently insights into changes of metabolic processes. Ø  DNA sequencing (well established) Ø  RNA sequencing (moderately established for environmental samples) Ø  Protein profiling (little established for environmental samples)

SMB

DNA, RNA and protein profiles

Direct sequencing of DNA, RNA and proteins extracted from an environmental sample can provide snapshots of the complete microbiome and subsequently insights into changes of metabolic processes. Ø  DNA sequencing (well established)

•  Amplicon sequencing -  16S rRNA gene for prokaryotes (methods, databases and software well

established) -  Internal Transcribed Spacer (ITS) region for fungi (less established; methods

databases and software are being established) -  Functional genes

•  Shotgun sequencing Ø  RNA sequencing (moderately established for environmental samples) Ø  Protein profiling (little established for environmental samples)

SMB

Cost of nucleotide sequencing

SMB

28S 18S

5.8S

DNA amplicon sequencing

Ø  Phylogenetic marker genes

-  16S rRNA (& various hypervariable regions) -  Internal Transcribed Spacer (ITS) -  18S rRNA

Hypervariable regions of 16S rRNA gene (Pseudomonas)

Bodilis et al 2012; PLoS ONE 7(4): e35647

Inter Genic Spacer

Transcribed rRNA genes & spacers

ITS1 ITS2

Organization of ribosomal genes in eukaryotes

SMB

DNA amplicon sequencing (16S rRNA gene & ITS)

Fiber colonization during rumen-degradation

Piao et al 2014; Frontiers in Microbiology

SMB

DNA amplicon sequencing

Ø  Functional marker genes -  detection of genes with relevant function

-  mcrA (methanogenesis) -  nifH (nitrogen fixation) -  dsrAB (sulfate reduction)

Fraction of methanogens in human gut microbiome is age dependent

Mihajjlvski et al 2010; Env. Micro. Rep. 2:272

Compartmentalized microbial nitrogen fixation in the gut of Odontotaenius disjunctus

Ceja-Navarro et al 2014; ISME J 8:6

nifH expression level 1 5 27 2

SMB

Cow rumen metagenome

Library name Insert Size Sequencing Bases (Gb) Reads (M) GAII_200bp 200bp 2x125bp 17.3 138.3 GAII_3kb* 3kb 2x36bp 4.5 124.5 GAII_3kb 2x75bp 27.2 357.6 GAII_5kb* 5kb 2x36bp 4.0 110.8 GAII_5kb 2x75bp 22.8 300.6 HiSeq_300bp 300bp 2x36bp 3.9 108.9 2x100bp 188.2 1,863.6 Total 267.9 3,004.3

* titration data for calibration of full-length run

~3 billion reads

~268 Gb metagenomic data from rumen microbiome

~58,000 Escherichia coli genomes

Hess et al 2011; Science 331; 463

SMB

DNA shotgun sequencing

0.01

0.1

1

10

100

0.01 0.1 1 10 100 1000

~80  CPU  years  (completed  in  2090)  

~1  day  on  NERSC’s  30,000-­‐CPU-­‐cluster  

Cow  Rumen  I  (20  Gb)  

Cow  Rumen  HiSeq;  2011  (240  Gb)  

Termite  Hindgut;  2007  (0.04  Gb)  

Sargasso  Sea;  2004    (1  Gb)  

Global  Ocean  Survey;  2007  (6  Gb)  

Data [Gbp]

Run

ning

tim

e [C

PU-y

ears

]

BLAST time

SMB

DNA shotgun sequencing

Ø  Targeted gene discovery (e.g. carbohydrate active enzymes)

Hess et al 2011; Science 331; 463

CAZy database increased by ~30%

K-mer Filter (read selection /

trimming)

Velvet Assembly

Gene prediction

HMMSearch (GH & CBM

domains)

111 Gb

179,092 scaffolds >1kb (1.9 Gb)

2.5 M genes

27,755 CAZymes genes

Gene prediction 268 Gb

85,740 CAZymes in CAZy DB

SMB

DNA shotgun sequencing

Ø  Assembly of individual genomes from metagenome data

Hess et al 2011; Science 331; 463

SMB

Ø  DNA sequencing (well established)

•  Amplicon sequencing •  Shotgun sequencing

•  Provides overview of metabolic repertoire of microbial community •  Can be used to assemble genomes from individual community

members •  Targeted gene discovery

Ø  RNA sequencing (moderately established for environmental samples) Ø  Protein profiling (little established for environmental samples)

Community Profiling

SMB

RNA sequencing

He et al 2010; Nature Methods 7; 807

Ø  RNA sequencing (moderately established for environmental samples) •  identification of genes that are expressed •  rRNA needs to be removed for functional gene information •  t1/2 of mRNA is short and varies between different transcripts

SMB

Community Profiling

Ø  DNA sequencing (well established)

•  Amplicon sequencing •  Shotgun sequencing

•  Provides overview of metabolic repertoire of microbial community •  Can be used to assemble genomes from individual community

members •  Targeted gene discovery

Ø  RNA sequencing (moderately established for environmental samples) Ø  Protein profiling (little established for environmental samples)

SMB

Database search

Protein Profiling

Hettich et al 2013; Anal. Chem. 2013, 85, 4203

SMB

Challenge: Data Integration & verification

Warnecke and Hugenholtz 2007; Genome Biol. 8; 231

Microbial Ecology omics toolbox

SMB

Challenge: Data Integration Ø  Omics platforms generate non-standardized output Ø  No general SOP for sample prep and data analysis (even with same platform) Ø  Different scientific questions Ø  Different background of “users” Ø  Developing in-house solutions is currently not economical (even at the institution-

level) Some Integrative Data Analysis Tools

16S rRNA gene sequencing ITS sequencing

Metagenomics Metatranscriptomics

Metaproteomics Metametabolomics

Imaging

SMB

Hess et al 2011; Science 331; 463

Identification of cellulolytic proteins using high-throughput screening assays

Large scale in-vivo production of cellulolytic proteins

Expression of target sequences

In-vitro In-vivo

Synthesis of target nucleotide sequence

Direct amplification from metagenomic DNA Codon optimized sequence

Chemical synthesis

Challenge: Verification of assembly at the gene level

SMB

Warnecke & Hugenholtz, 2007

Beyond gene & genome inventories

SMB

Metagenome

Metatranscriptome

Metaproteome

Hess et al; unpublished

Reconstruction of metabolic pathways Glycolysis/Gluconeogenesis

For microbial dark matter: Piao et al (2014) Biotechnology & Bioengineering

SMB

Perspectives

•  BUILD REFERENCE GENE/GENOME DATABASE •  Define Standard Operational Procedures •  Take advantage of JGI & EMSL resources and

data generated through existing CSPs (Hungate1000; Fungi1000; Rumen MT, MP & MB)

•  Research Topic in Frontiers in Systems Microbiology

•  Visual atlas of rumen microbiome

[email protected]

SMB

Hailan Piao; WSU – Postdoc Erik Hawley; WSU – Graduate Student

UI Urbana-Champaign: Roderick Mackie

LBNL/DOE JGI: Eddy Rubin, Nikos Kyrpides, Igor Grigoriev, Susannah Green Tringe, Stefanie Malfatti, Erika Lindquist, Natalia Ivanova, Jeff Froula, Chongbin Du, Amrita Pati and Kristin Tennessen

PNNL/EMSL: Scott Baker, Jon Magnuson, Lili Pasa-Tolic, Dave Culley, Kenneth Bruno, Heather Brewer, Angela Norbeck and Nancy Isern

UC Berkeley: Doug Clark, Tae-Wan Kim, Chris Sommerville and Stefan Bauer

University of Bielefeld, GER: Alex Sczyrba

Acknowledgements

SMB

Q & A