Biomedical Engineering Reference
In-Depth Information
clusters (Severinov et al.
2007
). These studies gave a first glimpse of the chemical
diversity and widespread distribution of lasso peptides in bacteria, and led to the
discovery of capistruin from
Burkholderia thailandensis
E264 in 2008 (Knappe
et al.
2008
), which opened the genome-mining era of lasso peptide discovery.
The rapid expansion of microbial genome sequences allowed the following
in-depth genome-mining studies of lasso peptides. Two distinct methods are com-
monly employed: one is precursor centric and the other is based on McjB homology
search. Link and co-workers used a pattern shared by precursor peptides to search
available genomes by a pattern-matching algorithm (Maksimov et al.
2012
). The
precursor pattern X
5-43
T
X
G
X
6-10
D
/
E
X
5-16
(X denotes any amino acids; the num-
ber of amino acids is indicated in subscript; conserved residues are in bold) takes
into account type II mature peptide features and the requirement of a threonine at
the penultimate position of the leader sequence. The genome context of the puta-
tive precursor genes was subsequently analysed for the presence of open reading
frames (ORFs) that contain conserved motifs of McjB- and McjC-like proteins.
This approach led to the identification of 79 putative lasso clusters out of more
than 3,000 genomes at the time of the study. These clusters are distributed across
nine bacterial phyla and one archaeal phylum. To demonstrate the applicability of
this approach, predicted lasso peptides astexins from
Asticcacaulis excentricus
were successfully produced by heterologous expression in
E. coli
and characterized
(Maksimov et al.
2012
; Maksimov and Link
2013
). By contrast, the discovery of
a number of lasso peptides from proteobacteria (Hegemann et al.
2013a
,
b
2014
;
Zimmermann et al.
2013
) and actinobacteria (Ducasse et al.
2012a
) was based on
McjB homology search. The rationale of this method is that McjB-like proteins that
function as cysteine proteases are unique to lasso gene clusters. McjB-like proteins
can be unambiguously assigned as they conserve the catalytic dyad Cys-His at
the C-terminal domain despite low homology of the N-terminal sequences. Using
McjB as a query and the Position-Specific Iterative Basic Local Alignment Search
Tool (PSI-BLAST) tool followed by manual inspection of the vicinity of the hits,
Marahiel and co-workers identified 102 lasso gene clusters from 82 proteobacterial
genomes (Hegemann et al.
2013b
). Guided by this approach, a total of 23 new lasso
peptides were obtained by heterologous expression in
E. coli
. The above-mentioned
methods are both efficient and systematic. Worth noting, the precursor-centric ge-
nome mining may miss out novel lasso peptides that have unconventional features,
such as first residues different from Gly or Cys. Recently, a powerful mass spec-
trometry (MS)-based peptidogenomic method has been developed (Kersten et al.
2011
), which connects the chemotype to the genotype by matching the tandem MSn
to peptide biosynthetic gene products. Two lasso peptides have been identified in
this way (SSV-2083 and SRO15-2005), although their lasso structures were not
demonstrated by the same study. We independently discovered SSV-2083 (termed
sviceucin in our study) by a classical homology search-based genome-mining ap-
proach (Ducasse et al.
2012a
).
These genome-mining studies allowed notably an appreciation of the diversity
of gene organization of lasso peptide clusters (see Sect. 1.1). One main point is
that an important number of clusters do not encode adenosine triphosphate-binding
Search WWH ::
Custom Search