D. Metabolomics
In line with the other omics methods, metabolomics provides a temporally limited snapshot
of an organism’s chemical potential, pooling
all metabolites available in the sample, or
enriching for secondary metabolome fingerprint, using an adequate extraction method.
Experiments aiming at the discovery of novel
NPs thus usually involve an untargeted screening of organic extracts of whole cultures, followed by a chromatographic separation and
mass spectrometry for compound analysis.
Lately, a shift has been observed toward in
situ analysis strategies, including mass spectrometry imaging (MSI), circumventing the
need for sample preparation. As the analytical
methods are beyond the scope of the present
work, the reader is directed to the excellent
reviews by Cox et al. (2014), Wolfender et al.
(2015), and Netzker et al. (2018).
Dereplication (the process of identification
of known compounds in a mixture) of experimental data represents a key step in metabolomics—indeed, rediscovery of SMs has
hampered research in this field for years (see
above).
For illustration, an OSMAC-coupled de-replication
approach was used to identify three new penitremone
derivatives from a marine-derived strain of Penicillium
canescens amidst a large number of already known
compounds (Vansteelandt et al. 2012).
Multiple public data repositories are available, including METLIN (Zhu et al. 2013),
GNPS (Wang et al. 2016c), or MassBank
(Horai et al. 2010), where researchers are
encouraged to deposit their spectrometric data
to aid dereplication efforts and acceleration of
the process of novel compound discovery. The
software tools used in database interrogation
usually also provide the option of performing
multivariate statistical analysis, e.g., principal
component analysis and partial least squares
(Cox et al. 2014; Covington et al. 2017), as well
as hierarchical clustering and pathway enrichment analysis for further exploration of the
data [for an overview of software, see Liu and
Locasale (2017)].
Molecular networking based on metabolomic data, i.e., creating clusters of molecule
families according to correlations between fragment ions was used to identify new analogs of
fungisporin and nidulanin A in A. nidulans
(Klitgaard et al. 2015). Metabolomics research
targeting secondary metabolites has often seen
appropriation of methods originally designed
for different research purposes [e.g., MS imaging, discussed in Netzker et al. (2018)]. This
also holds true for computational strategies
assisting with the analysis of complex data. A
commonly used tool for mapping of compounds into metabolic pathways and networks
is Cytoscape, with the metabolism plugin Metscape, originally devised for human metabolome
experiments (Gao et al. 2010). Other platforms
include the KEGG Atlas (Okuda et al. 2008) or
MetaMapR (Grapov et al. 2015), a software
which can use information from KEGG and
PubChem databases to find associations
between metabolites which have been detected
in a metabolomics approach, but whose biochemistry and structure are not yet known,
and integrate this information with genomic
or proteomic data into metabolic networks.
E. Integration of Omics Data
Due to inherent limitations and properties of
each of the omics methods, it is possible to
introduce bias and/or obtain an imperfect, partial understanding of a biological system at
hand (Yeger-Lotem et al. 2009). Following the
trend evident in other areas, fungal SM research
has also started moving toward the adoption of
an integrative approach to novel compound
discovery, as well as mechanistic characterization of biosynthetic processes and the roles of
SMs (Macheleidt et al. 2016; Hautbergue et al.
2018).
An excellent example of the power of a
multi-omics strategy is the study of the rice
pathogen Fusarium fujikuroi, which amalgamated genome mining, transcriptome and proteome studies, chromatin immunoprecipitation,
and metabolic profiling to reveal the complexities in the regulation of the secondary metabo11 New Avenues Toward Drug Discovery in Fungi
285
In line with the other omics methods, metabolomics provides a temporally limited snapshot
of an organism’s chemical potential, pooling
all metabolites available in the sample, or
enriching for secondary metabolome fingerprint, using an adequate extraction method.
Experiments aiming at the discovery of novel
NPs thus usually involve an untargeted screening of organic extracts of whole cultures, followed by a chromatographic separation and
mass spectrometry for compound analysis.
Lately, a shift has been observed toward in
situ analysis strategies, including mass spectrometry imaging (MSI), circumventing the
need for sample preparation. As the analytical
methods are beyond the scope of the present
work, the reader is directed to the excellent
reviews by Cox et al. (2014), Wolfender et al.
(2015), and Netzker et al. (2018).
Dereplication (the process of identification
of known compounds in a mixture) of experimental data represents a key step in metabolomics—indeed, rediscovery of SMs has
hampered research in this field for years (see
above).
For illustration, an OSMAC-coupled de-replication
approach was used to identify three new penitremone
derivatives from a marine-derived strain of Penicillium
canescens amidst a large number of already known
compounds (Vansteelandt et al. 2012).
Multiple public data repositories are available, including METLIN (Zhu et al. 2013),
GNPS (Wang et al. 2016c), or MassBank
(Horai et al. 2010), where researchers are
encouraged to deposit their spectrometric data
to aid dereplication efforts and acceleration of
the process of novel compound discovery. The
software tools used in database interrogation
usually also provide the option of performing
multivariate statistical analysis, e.g., principal
component analysis and partial least squares
(Cox et al. 2014; Covington et al. 2017), as well
as hierarchical clustering and pathway enrichment analysis for further exploration of the
data [for an overview of software, see Liu and
Locasale (2017)].
Molecular networking based on metabolomic data, i.e., creating clusters of molecule
families according to correlations between fragment ions was used to identify new analogs of
fungisporin and nidulanin A in A. nidulans
(Klitgaard et al. 2015). Metabolomics research
targeting secondary metabolites has often seen
appropriation of methods originally designed
for different research purposes [e.g., MS imaging, discussed in Netzker et al. (2018)]. This
also holds true for computational strategies
assisting with the analysis of complex data. A
commonly used tool for mapping of compounds into metabolic pathways and networks
is Cytoscape, with the metabolism plugin Metscape, originally devised for human metabolome
experiments (Gao et al. 2010). Other platforms
include the KEGG Atlas (Okuda et al. 2008) or
MetaMapR (Grapov et al. 2015), a software
which can use information from KEGG and
PubChem databases to find associations
between metabolites which have been detected
in a metabolomics approach, but whose biochemistry and structure are not yet known,
and integrate this information with genomic
or proteomic data into metabolic networks.
E. Integration of Omics Data
Due to inherent limitations and properties of
each of the omics methods, it is possible to
introduce bias and/or obtain an imperfect, partial understanding of a biological system at
hand (Yeger-Lotem et al. 2009). Following the
trend evident in other areas, fungal SM research
has also started moving toward the adoption of
an integrative approach to novel compound
discovery, as well as mechanistic characterization of biosynthetic processes and the roles of
SMs (Macheleidt et al. 2016; Hautbergue et al.
2018).
An excellent example of the power of a
multi-omics strategy is the study of the rice
pathogen Fusarium fujikuroi, which amalgamated genome mining, transcriptome and proteome studies, chromatin immunoprecipitation,
and metabolic profiling to reveal the complexities in the regulation of the secondary metabo11 New Avenues Toward Drug Discovery in Fungi
285