2 Metagenome Analysis
39
2005, Rusch et al. 2007) or library-independent pyrosequencing (Angly et al. 2006,
Edwards et al. 2006, Biddle et al. 2008, Dinsdale et al. 2008). If, however, gene
order and phylogenetic affiliation of gene fragments are of interest, end-sequencing
or screening of large insert libraries is the method of choice (DeLong et al. 2006).
This approach allows selection and further sequencing of clones that carry genes of
interest.
Studies aimed at studying the genetic capability of certain species within a highly
diverse microbial community should use an enrichment of target cells to reduce
screening and sequencing effort. If this is successful, a shotgun sequencing approach
or a pyrosequencing approach followed by sequence assembly can be applied. If it
is only possible to obtain a small number of target cells at a high level of purity,
a DNA amplification step has to be included (Mussmann et al. 2007, Podar et al.
2007). If pre-enrichment of target populations is not possible, the generation of large
insert libraries can be attempted as this allows a considerably reliable phylogenetic
affiliation of sequences (Beja et al. 2000, Krüger et al. 2003, Teeling et al. 2004,
Bryant et al. 2007).
An intelligent combination of the available tools can result in significant savings
in both time and money.
2.3.2 Enrichment Strategies
There are various methods available to enrich subpopulations of microbial communities prior to a metagenomic study.
Plankton samples have been fractionated by serial filtration through filters of
different pore sizes. Eukaryotic phytoplankton and particles were removed using
meshes and filters with pore sizes ranging from 20 to 0.8 μm (Schmidt et al. 1991,
Stein et al. 1996, Beja et al. 2000, de la Torre et al. 2003, Lopez-Garcia et al. 2004,
Venter et al. 2004, Angly et al. 2006, Culley et al. 2006, DeLong et al. 2006, MartínCuadrado et al. 2007, Rusch et al. 2007). The filtrate was either further subdivided
(Rusch et al. 2007) or prokaryotic cells were concentrated using filters of generally
0.22 μm pore size, or by centrifugation (Schmidt et al. 1991, Beja et al. 2000).
For metagenomic studies of viruses, prokaryotic cells were removed using filters
of 0.22 μm pore size (Angly et al. 2006, Culley et al. 2006, Rusch et al. 2007).
Prokaryotes inhabiting marine invertebrates have been successfully separated from
host tissue cells by density gradient centrifugation (Hughes et al. 1997, Schirmer
et al. 2005, Woyke et al. 2006, Robidart et al. 2008). Benthic microbial consortia
were detached from sediment particles by sonication, separated from those particles
by density gradient centrifugation, and concentrated and separated from single cells
by filtration (Schleper et al. 1998, Hallam et al. 2003). In other cases, it may be
even possible to manually isolate cell filaments. This was applied in a metagenome
analysis of multicellular Beggiatoa filaments (Mussmann et al. 2007). Enrichment
of genomic DNA based on G+C content is an alternative to the enrichment of intact
cells, if the G+C content of the target cells is significantly different from that of the
39
2005, Rusch et al. 2007) or library-independent pyrosequencing (Angly et al. 2006,
Edwards et al. 2006, Biddle et al. 2008, Dinsdale et al. 2008). If, however, gene
order and phylogenetic affiliation of gene fragments are of interest, end-sequencing
or screening of large insert libraries is the method of choice (DeLong et al. 2006).
This approach allows selection and further sequencing of clones that carry genes of
interest.
Studies aimed at studying the genetic capability of certain species within a highly
diverse microbial community should use an enrichment of target cells to reduce
screening and sequencing effort. If this is successful, a shotgun sequencing approach
or a pyrosequencing approach followed by sequence assembly can be applied. If it
is only possible to obtain a small number of target cells at a high level of purity,
a DNA amplification step has to be included (Mussmann et al. 2007, Podar et al.
2007). If pre-enrichment of target populations is not possible, the generation of large
insert libraries can be attempted as this allows a considerably reliable phylogenetic
affiliation of sequences (Beja et al. 2000, Krüger et al. 2003, Teeling et al. 2004,
Bryant et al. 2007).
An intelligent combination of the available tools can result in significant savings
in both time and money.
2.3.2 Enrichment Strategies
There are various methods available to enrich subpopulations of microbial communities prior to a metagenomic study.
Plankton samples have been fractionated by serial filtration through filters of
different pore sizes. Eukaryotic phytoplankton and particles were removed using
meshes and filters with pore sizes ranging from 20 to 0.8 μm (Schmidt et al. 1991,
Stein et al. 1996, Beja et al. 2000, de la Torre et al. 2003, Lopez-Garcia et al. 2004,
Venter et al. 2004, Angly et al. 2006, Culley et al. 2006, DeLong et al. 2006, MartínCuadrado et al. 2007, Rusch et al. 2007). The filtrate was either further subdivided
(Rusch et al. 2007) or prokaryotic cells were concentrated using filters of generally
0.22 μm pore size, or by centrifugation (Schmidt et al. 1991, Beja et al. 2000).
For metagenomic studies of viruses, prokaryotic cells were removed using filters
of 0.22 μm pore size (Angly et al. 2006, Culley et al. 2006, Rusch et al. 2007).
Prokaryotes inhabiting marine invertebrates have been successfully separated from
host tissue cells by density gradient centrifugation (Hughes et al. 1997, Schirmer
et al. 2005, Woyke et al. 2006, Robidart et al. 2008). Benthic microbial consortia
were detached from sediment particles by sonication, separated from those particles
by density gradient centrifugation, and concentrated and separated from single cells
by filtration (Schleper et al. 1998, Hallam et al. 2003). In other cases, it may be
even possible to manually isolate cell filaments. This was applied in a metagenome
analysis of multicellular Beggiatoa filaments (Mussmann et al. 2007). Enrichment
of genomic DNA based on G+C content is an alternative to the enrichment of intact
cells, if the G+C content of the target cells is significantly different from that of the