Microbiome and Inflammation Core
Equipment & Facilities
Microbiome and Inflammation Lab
Under the direction of Drs. Huffnagle and Young, the following equipment (and training) for microbiome and inflammation research are located in 6200 MSRBIII and 4618 Med Sci II and available for use without cost to Center Members:
- Roche MagNA Pure LC 2.0 System - for nucleic acid isolation
- 2 Beckman spectrophotometers
- Nanodrop spectrophotometer
- Agilent 2100 Bioanalyzer
- 5 - 96 sample thermal-cyclers for RT-PCR
- Roche LightCycler ® 480 Real-Time PCR System
- Doc-It gel imaging system
- 4 iMAC computers for data analysis
- Electrophoresis units for either horizontal or vertical gel slabs
- Electrotransfer unit & other equipment
- X-ray film developer
- 5 Biosafety cabinets (for tissue culture and microbiology)
- 3 CO2 incubators
- 2 Ambient air heated incubators with shakers for growing cultures
- Two person gloved anaerobe chamber from Coy Labs
- Anaerobic culture supplies & incubators
- Coulter cell counter
- 2 Inverted phase microscopes, one equipped with a digital camera and connected to an iMac computer with image capture software
- 4 Light microscopes, one equipped with a digital camera and connected to an iMac computer with image capture software
- Fluorescent microscope with multiple filters, equipped with a digital camera and connected to a Dell computer with image capture software
- Biotek ELISA plate washer • Biotek reader for ELISA and enzyme kinetics
Germ-free & Gnotobiotic Mouse Facilities
The germ free mouse resource at the University of Michigan was established by Dr. Kathryn Eaton and consists of two barrier-isolated animal rooms (880 square feet) located near the Microbiome Lab in the Biomedical Sciences Research Building. It currently houses a total of 12 gnotobiotic isolators, as well as 200 square feet of office and storage space. Four of the isolators are large, capable of holding up to 50 mouse cages each. These are used to house breeding pairs of mice and are maintained completely germ free at all times. Eight isolators are smaller and capable of housing up to 12 mouse cages. These are used as quarantine, surgical, and experimental isolators and used to house germ-free or experimentally colonized mice depending on the experimental protocol. Sterility is maintained by routine methods and ensured by weekly aerobic and anaerobic bacterial culture. Routine monitoring of sterility will be performed by weekly aerobic and anaerobic culture and Gram-stains of feces and moistened chow and by serologic and molecular monitoring of sentinel mice. Germ-free mouse strains are derived by embryo transfer. A breeding colony of germ-free Swiss Webster mice (currently over 150 mice) is maintained for use as embryo transfer recipients, vasectomized male breeders, and breeding stock. Superovulated donor mice are mated, and blastocysts are collected aseptically and transferred to the germ-free hood. Pseudopregnant recipients are aseptically removed from the isolators to the hood, and blastocysts are transferred to the uterus. In some cases embryos are collected and frozen until pseudopregnant recipients are available. Frozen two cell embryos are transferred to the Fallopian tube. The recipient mice are placed into a quarantine isolator until the pups are weaned and sterility is assured, and then the pups are transferred to one of the breeding isolators. Surgeries and related procedures are performed by trained surgical technicians with the assistance of staff in the U of M transgenic core.
University of Michigan DNA Sequencing Core
Under the direction of Dr. Robert Lyons, the University of Michigan's DNA Sequencing Core is located near the Microbiome Lab in Medical Science Research Building II. It provides U-M investigators access to automated DNA sequencing technology on a recharge basis. The Core processes samples primarily on ABI Model 3730 sequencers. Two Next-Gen pyrosequencing platforms are available at the Sequencing Core: the Illumina Genome Analyzer IIx and the Roche 454FLX Genome Sequencer. Although available for microbial genome sequencing, the use of these two sequencers is primarily for mammalian genome sequencing.
The bioinformatics support for the Microbiome & Inflammation Core will be directed by Dr. Patrick Schloss and is located in 5713 Med Sci II and 6200 MSRBIII. Additional bioinformatics and computer support for the project will be provided by the University of Michigan Center for Computational Medicine (www.ccmb.med.umich.edu) including analytic tools on high-performance computers.
T-RFLP data will be analyzed using a program developed in our laboratory called K9 (this program runs under OSX 10.4.x or above and is freely available under the GNU public license (http://www-personal.umich.edu/~jre/Microbiome_Core/K9.html). This program has been designed to streamline the analysis T-RFLP data by pulling together functions from many different sources in to one discrete and easy-to-use package. Detection of specific TRFs is carried out using slight modifications of the R (http://www.r-project.org) and Perl scripts that use a statistical method to separate “true” peaks from noise and then bins TRFs of similar size. Once binned, the Bray-Curtis distance is determined for each of the groups, a hierarchical cluster analysis of the binned TRFs is performed and a dendrogram displaying the relatedness between communities generated.
All DNA sequences will be analyzed using a phylotype and OTU-based approach, using the algorithms implemented in mother. Phylotypes have the advantage of basing an analysis on previously defined taxonomic lineages for which we have some phenotypic knowledge. The challenge of this approach is that there is no standard taxonomy and all taxonomies are incomplete as demonstrated by the myriad new lineages identified in nearly every microbial ecology study. OTU-based approaches overcome these limitations by allowing the data to speak for themselves in forming groupings for any genetic distance threshold. All sequences will be assigned to a phylotype using the 16S rRNA classifier implemented within mothur, which is based on a naïve Bayesian approach. Because there is discordance in the taxonomies used by different sources, we will use multiple taxonomies including that of the RDP, greengenes, SILVA, and NCBI. Phylotype assignments will be summarized by generating pie charts, heat-maps, and bar graphs for each sample. Sequences will also be assigned to OTUs by calculating pairwise sequence distances that will be clustered using an unsupervised furthest neighbor (i.e. complete-linkage) algorithm using mothur. A representative sequence will be identified from each OTU and processed through the taxonomy classifier in order to associate each OTU with a phylogenetic anchor. Based on the number of sequences assigned to each phylotype or OTU, we will use mothur to calculate non-parametric estimators of richness and diversity within individual samples (e.g. Chao1, ACE, Shannon index). These frequency data will also be used to compare the membership and structure of communities by generating dendrograms based on commonly used b-diversity metrics (e.g. Jaccard, Morisita-Horn indices). We will make use of the datasets deposited in the Human Microbiome Project’s Data Analysis and Coordination Center (DACC; http://www.hmpdacc.org) to perform meta-analyses comparing the community structure of our samples to those from other investigators. These dendrograms will enable us to use indicator analysis, which can identify OTUs that are responsible for the structure of the tree. In addition, we will use t-tests and analysis of variance on each phylotype or OTU to identify those populations that are differentially represented under different conditions.
Core users will receive extensive support for developing their experimental design and carrying out their analyses. Schloss will work with users to implement and tailor analyses specific to users’ needs and as new algorithms are proposed in the literature. For example, we will advise users on how to design analyses using tools such as BLAST, MG-RAST, and SEED, which can be used to obtain phylogenetic and functional information from metagenomic shotgun sequencing projects. Annual 3-day workshops will be offered for core users and their staff that describe the theory and practice of analyzing microbial community data.