Finally, we note that synthesis of some scaffolds requires the action of additional enzymes,
such as a trans-enoyl reductase, a trans-acyltransferase, a trans-thioesterase, or another
synthase. For example, tenellin synthetase
from Beauveria bassiana requires trans-acting
enoyl reductase for correct polyketide scaffold
assembly (Halo et al. 2008); during lovastatin
synthesis, both acyltransferase and thioesterase
are necessary in order to release the polyketide
products from the synthases LovF (Xie et al.
2009) and LovB (Xu et al. 2013), respectively,
whereas for the synthesis of the first intermediate toward asperfuranone, two polyketide
synthases are needed (Chiang et al. 2009). In
these cases, a multi-gene insertion strategy is
required; see Sect. II.C.
2. Reconstitution of Secondary Metabolite
Pathways in a Heterologous Host
Establishing entire pathways is more challenging as several genes need to be functionally
expressed. In the following section, we will
describe various strategies that have been used
to ensure expression of some or all genes from a
BGC.
In an early study by Sakai et al., a 20 kb
fragment of DNA containing the citrinin BGC
from Monascus purpureus was inserted into an
E. coli-Aspergillus shuttle cosmid and ectopically integrated into the genome of A. oryzae
to produce 4 mg/L of citrinin (Sakai et al. 2008).
This yield was increased approximately 400fold to 1.48 mg/L when the cluster-specific
transcriptional regulator encoding gene
(ctnA) was constitutively expressed under control of the A. nidulans trpC promoter. Pathways
may also be transferred using a vector set developed by Pahirulzaman et al., which allows up to
12 different genes to be ectopically integrated
into the genome via three vectors containing
different selectable markers (Pahirulzaman
et al. 2012). Hence, pathways may be partly or
entirely established in a fungus to produce key
intermediates for pathway elucidation as well as
to identify the final product. Using this system
and A. oryzae as a host, He and Cox produced a
detailed model of the citrinin pathway and generated a cell factory producing citrinin to final
titers of 19.1 mg/L (He and Cox 2016).
Defined integration of entire BGCs into
fungal chromosomal expression sites has also
been used to establish functional SM pathways.
One advantage of this method is that reverse
genetics can be applied to dissect the pathway
once the pathway has been functionally established in a new host. This principle has been
exploited to successfully transfer all 12 biosynthetic genes required for geodin production
from A. terreus to A. nidulans. Additionally, to
activate the biosynthetic genes, the TF gene of
the cluster (gedR) was equipped with a constitutive A. nidulans promoter prior to genomic
integration. Subsequently, the efficient gene
deletion toolbox of A. nidulans was used to
identify gedL as the gene encoding the halogenase necessary for geodin production (Nielsen
et al. 2013).
A recent screening-friendly method for
linking BGC to metabolites uses FACs as vectors for gene transfer (see Sect. II.B.1) as they
can accommodate DNA fragments containing
even large BGCs. FAC-based BGC libraries from
A. terreus, Aspergillus wentii, and Aspergillus
aculeatus have been transferred and analyzed
in A. nidulans resulting in the assignment of 17
compounds to BGCs (Bok et al. 2015; Clevenger
et al. 2017).
3. Synthetic Pathway Setups for Secondary
Metabolite Production
For some compounds of interest, it is not possible to establish a cell factory based on the
natural pathway. This is the case if the biosynthetic pathway is unknown or it has only been
partly elucidated or if enzymes are compartmentalized in the native producer in a manner
that cannot be implemented in the new host. In
these cases, it may be possible to make a cell
factory based on a semi-synthetic or synthetic
pathway with enzymes of known (or predicted)
activities from various species. In this way, it
may be possible to synthesize the desired compound, but in a manner, which is different
from the pathway in the native host. For example, production of the pharmacologically rele10 Filamentous Fungi as Hosts for Heterologous Production of Proteins and Secondary. . .
253
such as a trans-enoyl reductase, a trans-acyltransferase, a trans-thioesterase, or another
synthase. For example, tenellin synthetase
from Beauveria bassiana requires trans-acting
enoyl reductase for correct polyketide scaffold
assembly (Halo et al. 2008); during lovastatin
synthesis, both acyltransferase and thioesterase
are necessary in order to release the polyketide
products from the synthases LovF (Xie et al.
2009) and LovB (Xu et al. 2013), respectively,
whereas for the synthesis of the first intermediate toward asperfuranone, two polyketide
synthases are needed (Chiang et al. 2009). In
these cases, a multi-gene insertion strategy is
required; see Sect. II.C.
2. Reconstitution of Secondary Metabolite
Pathways in a Heterologous Host
Establishing entire pathways is more challenging as several genes need to be functionally
expressed. In the following section, we will
describe various strategies that have been used
to ensure expression of some or all genes from a
BGC.
In an early study by Sakai et al., a 20 kb
fragment of DNA containing the citrinin BGC
from Monascus purpureus was inserted into an
E. coli-Aspergillus shuttle cosmid and ectopically integrated into the genome of A. oryzae
to produce 4 mg/L of citrinin (Sakai et al. 2008).
This yield was increased approximately 400fold to 1.48 mg/L when the cluster-specific
transcriptional regulator encoding gene
(ctnA) was constitutively expressed under control of the A. nidulans trpC promoter. Pathways
may also be transferred using a vector set developed by Pahirulzaman et al., which allows up to
12 different genes to be ectopically integrated
into the genome via three vectors containing
different selectable markers (Pahirulzaman
et al. 2012). Hence, pathways may be partly or
entirely established in a fungus to produce key
intermediates for pathway elucidation as well as
to identify the final product. Using this system
and A. oryzae as a host, He and Cox produced a
detailed model of the citrinin pathway and generated a cell factory producing citrinin to final
titers of 19.1 mg/L (He and Cox 2016).
Defined integration of entire BGCs into
fungal chromosomal expression sites has also
been used to establish functional SM pathways.
One advantage of this method is that reverse
genetics can be applied to dissect the pathway
once the pathway has been functionally established in a new host. This principle has been
exploited to successfully transfer all 12 biosynthetic genes required for geodin production
from A. terreus to A. nidulans. Additionally, to
activate the biosynthetic genes, the TF gene of
the cluster (gedR) was equipped with a constitutive A. nidulans promoter prior to genomic
integration. Subsequently, the efficient gene
deletion toolbox of A. nidulans was used to
identify gedL as the gene encoding the halogenase necessary for geodin production (Nielsen
et al. 2013).
A recent screening-friendly method for
linking BGC to metabolites uses FACs as vectors for gene transfer (see Sect. II.B.1) as they
can accommodate DNA fragments containing
even large BGCs. FAC-based BGC libraries from
A. terreus, Aspergillus wentii, and Aspergillus
aculeatus have been transferred and analyzed
in A. nidulans resulting in the assignment of 17
compounds to BGCs (Bok et al. 2015; Clevenger
et al. 2017).
3. Synthetic Pathway Setups for Secondary
Metabolite Production
For some compounds of interest, it is not possible to establish a cell factory based on the
natural pathway. This is the case if the biosynthetic pathway is unknown or it has only been
partly elucidated or if enzymes are compartmentalized in the native producer in a manner
that cannot be implemented in the new host. In
these cases, it may be possible to make a cell
factory based on a semi-synthetic or synthetic
pathway with enzymes of known (or predicted)
activities from various species. In this way, it
may be possible to synthesize the desired compound, but in a manner, which is different
from the pathway in the native host. For example, production of the pharmacologically rele10 Filamentous Fungi as Hosts for Heterologous Production of Proteins and Secondary. . .
253