multiple platforms including metagenomics, metabolomics, and
metaproteomics is key for a full functional assessment on the role
of the microbiota in the intestine [3].
Metaproteomic analysis to determine the protein composition
of the gut microbiome can be performed on fecal samples [4, 5] or
colonic lavages [6, 7]. Fecal samples can be collected noninvasively
and in large quantities; however, the samples lack intestinal regionspecific information and the sample preparation often involves steps
that deplete the host cells (e.g., centrifugation, filtration) and
thereby the interplay between the microbiota and its host cannot
be addressed. Colonic lavages can be collected during colonoscopy
by injecting small amounts of sterile water onto the mucosal surface, which is then aspirated and analyzed [6, 7]. This method of
sample collection allows for the site-specific microbiota assessment
and the characterization of the direct effect on protein expression
between the various species and its effect on host response [7].
Mass spectrometry analysis enables to simultaneously characterize both the host and microbial components in contrast to
DNA/RNA-based sequencing techniques. This does not limit the
analysis to the bacterial component of the microbiome but in
parallel identifies archaea, fungi, and potential virus species and
proteins [6]. Microbial protein identification by mass spectrometry
in intestinal samples is hampered by various challenges, such as the
enormous species variety and a high protein sequence homology
between closely related microbes, which makes species-level assignment challenging and the lack of appropriate databases required for
protein identification [8]. The species complexity and homology
can be overcome by combing results at higher taxonomic level such
as genus and family as is a common procedure in 16S microbial
hypervariable region sequencing, with the added benefit of assessing protein functional groups. Establishing a suitable protein database, including all proteomes for each potential microbial species in
the intestine, has been more challenging due to the highly variable
composition. This is a first requirement as conventional methods
for protein identification require prior knowledge of all the species
in a given sample while adding too many entries will limit the search
sensitivity [9]. Different methods have been established to create
these protein databases, either by including all available proteins for
the species identified by genomics sequencing or by analyzing a
limited number of samples to establish a reference protein database
[4, 6]. Alternatively, de novo spectral sequencing can be applied to
identify the peptides in a sample independent of a protein database.
This method, although less frequently applied, has gained recent
interest mainly due to the increased availability of high-resolution
mass spectrometers allowing for the highly accurate detection of
peptide fragments [10].
168
Sjoerd van der Post and Liisa Arike
metaproteomics is key for a full functional assessment on the role
of the microbiota in the intestine [3].
Metaproteomic analysis to determine the protein composition
of the gut microbiome can be performed on fecal samples [4, 5] or
colonic lavages [6, 7]. Fecal samples can be collected noninvasively
and in large quantities; however, the samples lack intestinal regionspecific information and the sample preparation often involves steps
that deplete the host cells (e.g., centrifugation, filtration) and
thereby the interplay between the microbiota and its host cannot
be addressed. Colonic lavages can be collected during colonoscopy
by injecting small amounts of sterile water onto the mucosal surface, which is then aspirated and analyzed [6, 7]. This method of
sample collection allows for the site-specific microbiota assessment
and the characterization of the direct effect on protein expression
between the various species and its effect on host response [7].
Mass spectrometry analysis enables to simultaneously characterize both the host and microbial components in contrast to
DNA/RNA-based sequencing techniques. This does not limit the
analysis to the bacterial component of the microbiome but in
parallel identifies archaea, fungi, and potential virus species and
proteins [6]. Microbial protein identification by mass spectrometry
in intestinal samples is hampered by various challenges, such as the
enormous species variety and a high protein sequence homology
between closely related microbes, which makes species-level assignment challenging and the lack of appropriate databases required for
protein identification [8]. The species complexity and homology
can be overcome by combing results at higher taxonomic level such
as genus and family as is a common procedure in 16S microbial
hypervariable region sequencing, with the added benefit of assessing protein functional groups. Establishing a suitable protein database, including all proteomes for each potential microbial species in
the intestine, has been more challenging due to the highly variable
composition. This is a first requirement as conventional methods
for protein identification require prior knowledge of all the species
in a given sample while adding too many entries will limit the search
sensitivity [9]. Different methods have been established to create
these protein databases, either by including all available proteins for
the species identified by genomics sequencing or by analyzing a
limited number of samples to establish a reference protein database
[4, 6]. Alternatively, de novo spectral sequencing can be applied to
identify the peptides in a sample independent of a protein database.
This method, although less frequently applied, has gained recent
interest mainly due to the increased availability of high-resolution
mass spectrometers allowing for the highly accurate detection of
peptide fragments [10].
168
Sjoerd van der Post and Liisa Arike