Construction and screening of vast libraries of natural product-like macrocyclic peptides using in vitro display technologies

Abstract

Macrocyclic structure and backbone N-methylation represent characteristic features of peptidic natural products, which play critical roles in their biological activity. Although natural products have been the traditional source of such peptides, recent developments in synthesizing natural product-like macrocyclic peptides using reconstituted translation systems have enabled us to construct vast trillion-member libraries of non-standard macrocyclic peptides. In addition, a method for displaying such libraries on their corresponding mRNA templates allows us to rapidly screen them for potent ligands against various drug targets. This review describes methodologies for the ribosomal synthesis of novel natural product-like macrocyclic peptides and their recent applications in the discovery of bioactive molecules using in vitro display technologies.
Copyright © 2014 The Authors. Published by Elsevier Ltd.. All rights reserved.

Full text

Construction
and
screening
of
vast
libraries
of
natural
product-like
macrocyclic
peptides
using
in
vitro
display
technologies
Nasir
K
Bashiruddin
and
Hiroaki
Suga
Macrocyclic
structure
and
backbone
N-methylation
represent
characteristic
features
of
peptidic
natural
products,
which
play
critical
roles
in
their
biological
activity.
Although
natural
products
have
been
the
traditional
source
of
such
peptides,
recent
developments
in
synthesizing
natural
product-like
macrocyclic
peptides
using
reconstituted
translation
systems
have
enabled
us
to
construct
vast
trillion-member
libraries
of
non-standard
macrocyclic
peptides.
In
addition,
a
method
for
displaying
such
libraries
on
their
corresponding
mRNA
templates
allows
us
to
rapidly
screen
them
for
potent
ligands
against
various
drug
targets.
This
review
describes
methodologies
for
the
ribosomal
synthesis
of
novel
natural
product-like
macrocyclic
peptides
and
their
recent
applications
in
the
discovery
of
bioactive
molecules
using
in
vitro
display
technologies.
Addresses
Department
of
Chemistry,
Graduate
School
of
Science,
The
University
of
Tokyo,
Tokyo
113-0033,
Japan
Corresponding
author:
Suga,
Hiroaki
(hsuga@chem.s.u-tokyo.ac.jp)
Current
Opinion
in
Chemical
Biology
2015,
24:131–138
This
review
comes
from
a
themed
issue
on
Omics
Edited
by
Benjamin
F
Cravatt
and
Thomas
Kodadek
For
a
complete
overview
see
the
Issue
and
the
Editorial
Available
online
5th
December
2014
http://dx.doi.org/10.1016/j.cbpa.2014.11.011
1367-5931/#
2014
The
Authors.
Published
by
Elsevier
Ltd.
This
is
an
open
access
article
under
the
CC
BY-NC-ND
license
(http://creative-
commons.org/licenses/by-nc-nd/3.0/).
Introduction
Throughout
the
years,
investigations
of
various
clinically
significant
natural
products
have
provided
insight
on
important
structural
features
that
are
necessary
to
afford
high
biological
activity
[1–3].
Amongst
these
features,
macrocyclization
is
arguably
the
most
basic
and
important
structural
property
to
confer,
particularly
in
peptidic
natural
products.
To
date
there
have
been
many
reports
underlying
the
importance
of
having
a
cyclic
structure
over
a
linear
one,
whether
they
be
small
molecules
or
peptides
[4].
The
structural
stability
gained
from
macro-
cyclization
allows
peptides
to
spend
more
time
in
bio-
logically
active
conformations,
therefore
decreasing
the
entropic
cost
upon
binding
their
target
[5].
Amide
bonds
in
such
constrained
cyclic
peptides
also
become
more
difficult
to
access
by
proteases
resulting
in
greater
serum
stability.
In
regards
to
passive
membrane
permeability,
a
macrocyclic
structure
potentially
increases
the
likelihood
of
the
formation
of
intramolecular
hydrogen
bonds
which
can
decrease
interactions
with
water,
thus
increasing
the
potential
for
membrane
permeability
[6].
Many
natural
product
peptides
also
contain
N-methyl
amino
acids
[7].
Similar
to
macrocyclization,
N-methyla-
tion
decreases
conformational
flexibility
around
the
N-
methylated
peptide
bond
through
steric
effects
with
its
neighboring
amino
acid
side
chains
and
stabilizes
the
conformation
via
A
1,3
-like
strain.
N-methylation
of
the
peptide
backbone
also
decreases
the
hydrogen
bond
donating
ability
of
the
peptide
bond
to
water
molecules,
potentially
improving
membrane
permeability.
As
with
most
deviations
from
the
20
proteinogenic
amino
acids,
N-methylation
also
enhances
the
serum
stability
of
such
peptides
through
decreased
accessibility
to
proteases.
Taken
together,
the
combination
of
macrocyclization
and
backbone
N-methylation
of
peptidic
natural
products
significantly
improves
their
drug-like
properties.
Unlike
proteins
that
are
synthesized
by
the
translation
system
consisting
of
the
ribosome
and
translation
factors,
peptidic
natural
products
are
often
synthesized
from
large
multienzyme
complexes,
called
nonribosomal
peptide
synthetases
[8]
(NRPSs).
These
complexes
work
in
con-
cert
through
various
modules
assigned
to
specific
reac-
tions
necessary
for
the
synthesis
of
each
individual
amino
acid
and
their
subsequent
polymerization,
modification
and
cyclization.
The
engineering
of
these
NRPSs
for
the
synthesis
of
diverse
libraries
is
attractive
yet
scientifically
challenging
and
has
met
limited
success
in
producing
libraries
of
diverse
macrocyclic
peptides
due
to
their
mechanistic
and
structural
complexity
[9].
In
contrast,
recent
advances
in
manipulating
cell-free
translation
systems
[10]
have
given
new
opportunities
to
construct
vast
libraries
of
macrocyclic
peptides
and
to
screen
for
bioactive
molecules
by
means
of
in
vitro
display
systems.
This
review
will
discuss
concepts
and
recent
examples
of
novel
macrocyclic
peptides
discovered
by
these
technologies.
Ribosomal
synthesis
of
natural
product-like
macrocyclic
peptides
To
synthesize
macrocyclic
peptides
using
the
translation
machinery,
the
most
basic
approach
is
through
disulfide
Available
online
at
www.sciencedirect.com
ScienceDirect
www.sciencedirect.com
Current
Opinion
in
Chemical
Biology
2015,
24:131–138

bonds
between
cysteine
residues
[11].
However,
their
susceptibility
to
reduction
in
intracellular
environments
makes
them
undesirable
for
some
applications.
There-
fore,
methods
of
forming
a
non-reducible
covalent
bond
for
cyclization
through
simple
chemical
post-translational
modifications
have
been
devised.
Methods
of
bridging
two
primary
amines
between
the
N-terminus
and
a
lysine
sidechain
using
disuccinimidyl
glutarate
(Figure
1a)
or
the
sulfhydryl
group
of
two
cysteine
residues
using
dibromoxylene
(Figure
1b)
have
been
used
to
success-
fully
generate
macrocyclic
peptides
[12,13].
A
similar
method
of
producing
bicyclic
peptides
via
thioether-
crosslinking
of
three
cysteine
residues
has
been
also
reported
(Figure
1c)
[14

].
An
advantage
of
these
methods
is
their
applicability
to
the
standard
proteino-
genic
amino
acids.
However,
when
more
than
two
reac-
tive
residues
appear
in
the
random
regions
of
these
libraries,
the
crosslinking
patterns
become
scrambled
potentially
causing
difficulty
in
deconvoluting
the
out-
come
of
selections
based
on
such
cyclization
methods.
A
technically
more
demanding
method
than
the
above,
but
far
more
reliable
for
the
construction
of
macrocyclic
peptides,
is
based
on
the
concept
of
manipulating
the
genetic
code,
known
as
genetic
code
reprogramming,
where
designated
codons
are
made
vacant
and
then
reassigned
to
nonproteinogenic
amino
acids.
Two
major
methodologies
have
been
reported
to
date,
both
of
which
utilize
custom-made
reconstituted
translation
systems.
One
method
takes
advantage
of
the
mischarging
proper-
ties
of
aminoacyl-tRNA
synthetases
in
the
presence
of
excess
amounts
of
nonproteinogenic
amino
acids,
yield-
ing
the
corresponding
aminoacyl-tRNAs.
Szostak
et
al.
reported
a
method
of
generating
peptides
containing
4-
selenalysine
in
the
peptide
chain
followed
by
the
se-
lective
oxidation
and
concomitant
elimination
of
the
seleno
group
to
yield
a
dehydroalanine
residue
(Figure
2a)
[15].
Dehydroalanine
then
reacts
with
the
sulfhydryl
group
of
cysteine
via
Michael
addition
to
form
a
thioether
bond,
giving
rise
to
lanthionine-like
macro-
cyclic
peptides.
The
other
method
involves
’flexible’
tRNA
acylation
ribozymes,
known
as
flexizymes
[16–18],
developed
by
Suga
et
al.,
which
facilitate
the
preparation
of
a
wide
array
of
nonproteinogenic
aminoacyl
tRNAs
with
nearly
unlim-
ited
choice.
The
combination
of
a
custom-made
in
vitro
translation
system
with
flexizymes,
referred
to
as
the
FIT
(Flexible
In
vitro
Translation)
system,
allows
the
ribosomal
132
Omics
Figure
1
(a) Production method
Cyclization additive
(b)
H2N
H2N
OH OH
OH
O
O
O
O
O
O
O
O
O
O
O
O
O
O
OOH
BrBr
Br
Br
Br
SHSH
SH SH SH
H
N
H
N
S
S
S
S
S
O
N
N
HN NH
Rabbit reticulocyte lysate
PURE system
Phage display
(c)
Current Opinion in Chemical Biology
Post-translational
cyclization
methods
without
the
need
for
genetic
code
reprogramming.
(a)
Disuccinimidyl
glutarate
crosslinking
between
the
amino
terminus
of
a
rabbit
reticulocyte
lysate-produced
peptide
and
a
lysine
residue
[12].
(b)
Dibromoxylene
bridging
of
two
cysteine
residues
of
a
peptide
produced
by
a
reconstituted
bacterial
translation
system
[13].
(c)
Tris
bromomethyl
benzene
bridging
of
three
cysteine
residues
of
a
peptide
displayed
on
phage
resulting
in
a
bicyclic
peptide
[14

].
Current
Opinion
in
Chemical
Biology
2015,
24:131–138
www.sciencedirect.com

synthesis
of
macrocyclic
peptides
using
nonproteinogenic
amino
acids
capable
of
crosslinking
with
other
proteino-
genic
or
nonproteinogenic
residues
(Figure
3a).
The
FIT
system
allows
for
a
wide
variety
of
cyclization
methods,
for
instance,
methyllanthionine-like
macrocyclic
peptides
can
be
synthesized
through
the
incorporation
of
vinylglycine
which
is
thermally
isomerized
to
dehydrobutyrine
which
can
then
form
a
thioether
bond
with
cysteine
residues
(Figure
2b)
[19].
Translation
of
peptides
containing
a
benzylamine
group
designated
by
the
N-terminal
initiating
amino
acid
and
a
downstream
5-hydroxytryptophan
is
a
unique
method
of
mild
oxidative
macrocyclization
forming
a
fluorogenic
indole
linkage
(Figure
2c)
[20].
Moreover,
ribosomal
synthesis
of
head-to-tail
linked
peptides
can
also
be
produced
through
the
C-terminal
Cys-Pro-
HO
G
(glycolic
acid)
sequence
or
programmed
peptidyl-tRNA
drop-off
In
vitro
displays
for
the
screening
of
vast
libraries
Bashiruddin
and
Suga
133
Figure
2
(a) H2N
H2N
H2N
H2NOH
O
MAP
O
O–
H
PDF
S
fMet
O
Se SH
OH
O
OH
O
S
O
50 mM H2O2
pH 7 - 8, RT
SH
OH
OH
HO
S
O
O
O
Cl
O
O
O
OOO
O
O
O
O
O
OH
O
S
SH
O
O
S
HN
HN
HN
OH
O
O
OH
N
N
N
HN
HS
+++
Cys Pro
Diketopiperidine
HOGly
H
N
H
N
OH
O
S
O
OO
HO
NH
N
HN
HS
H
N
OH
N
H
N
N
H
N
H
N
H
N
H
(b)
(c)
(d)
(e)
95°C, 0.5 hr
K3Fe(CN)
6
pH 8, RT
5 min
Current Opinion in Chemical Biology
Peptide
cyclization
methods
made
possible
by
genetic
code
reprogramming.
(a)
Aminoacyl-tRNA
synthetase
mischarging
used
to
incorporate
4-
selenalysine
to
ultimately
form
a
lanthionine
bridge
with
a
cysteine
residue
[15].
(b)–(e)
Cyclization
methods
using
the
FIT
system:
(b)
Incorporation
of
vinylglycine
to
form
a
methyllanthionine
bridge
with
a
cysteine
residue
[19].
(c)
Peptide
translation
initiated
with
N-benzylamine
and
the
incorporation
of
hydroxytryptophan
to
create
a
fluorogenic
benzoxazole
bridge
[20].
(d)
Incorporation
of
glycolic
acid
after
cysteine
and
proline
residues
results
in
spontaneous
conversion
to
diketopiperdine.
The
amino
terminus
produced
by
the
subsequent
treatment
with
peptide
deformylase
and
methionine
aminopeptidase
reacts
with
the
diketopiperidine
to
form
a
backbone
cyclized
peptide
[21–23].
(e)
Peptide
translation
initiated
with
an
N-chloroacetyl
amino
acid
to
form
thioether
bonds
with
cysteine
residues
[24].
www.sciencedirect.com
Current
Opinion
in
Chemical
Biology
2015,
24:131–138

containing
the
C-terminal
Cys-Pro
sequence
(Figure
2d)
[21–23].
In
both
cases,
the
C-terminal
ester
bond
accelerates
the
self-rearrangement
of
N
!
S
migration
to
form
a
C-
terminal
diketopiperadine-thioester,
eventually
yielding
backbone-cyclized
monocyclic
or
disulfide-bridged
bicyclic
peptides.
The
most
convenient
and
reliable
method
of
cyclization
has
been
through
the
translation
of
peptides
with
an
N-
chloroacetyl-amino
acid
initiator
that
can
react
with
a
downstream
cysteine
(Figure
2e)
[24].
The
advantage
of
this
method
is
the
spontaneous
and
selective
thioether
bond
formation
between
the
N-terminal
chloroacetyl
group
and
the
sulfhydryl
group
of
the
closest
cysteine
residue
[25].
A
single
exception
is
that
a
cysteine
residue
adjacent
to
the
N-chloroacetyl-amino
acid
cannot
react
with
the
chloroacetyl
group
due
to
ring
constraint,
thus
leaving
a
free
sulfhydryl
group
at
this
position.
However,
this
selectivity
turns
out
be
a
convenient
way
to
translate
fused-bicyclic
peptides,
having
a
thioether
(sulfide)
bond
between
the
N-terminus
and
the
second
cysteine
and
a
disulfide
bond
between
the
first
and
third
cysteine
resi-
dues.
Importantly,
it
has
been
demonstrated
that
the
FIT
system
facilitates
the
translation
of
peptides
containing
D-
amino
acids
[26],
N-methyl-amino
acids
[27],
N-alkylgly-
cines
[28],
and
those
with
noncanonical
sidechains
[29–33].
Thus,
a
variety
of
peptide
libraries
can
be
readily
prepared
in
an
mRNA-encoded
manner,
leading
to
the
development
of
a
display-based
platform
technology
for
the
discovery
of
bioactive
natural
product-like
macrocyclic
peptides.
134
Omics
Figure
3
(a) Flexizyme Ribosome
tRNA
Aminoacylation of
non-proteinogenic
amino acids
DNA library
PCR
Active
macrocyclic
peptides
Magnetic
Beads
Magnetic
Beads
Target Protein
Target
protein
selection
Counter
selection
30S
50S
Cys
ClAc-aa
(NNK)n
Puromycin
Reverse transcription
Displayed
macrocyclic
peptides
PU
PU
PU
PU
UGCAUGSD
UGC
UGC
AUG
AUG
SD
SD
(b)
(d) (c)
Current Opinion in Chemical Biology
The
RaPID
and
TRAP
systems.
(a)
Nonproteinogenic
amino
acids
are
charged
on
to
tRNAs
by
flexizymes.
(b)
A
DNA
library
is
transcribed
and
the
mRNA
is
attached
to
puromycin
and
then
translated
(attachment
of
puromycin
differs
between
RaPID
and
TRAP
[42,52]).
(c)
Unwanted
mRNA-
peptide
fusions
are
removed
by
counter
selections.
(d)
Target
protein
binding
mRNA-peptide
fusions
are
isolated
and
amplified
for
iterative
rounds
of
selection
or
for
sequencing.
Current
Opinion
in
Chemical
Biology
2015,
24:131–138
www.sciencedirect.com

Display
platforms
for
selecting
bioactive
macrocyclic
peptides
from
vast
libraries
The
methods
discussed
above
can
be
readily
combined
with
appropriate
display
platforms
to
screen
for
active
ligands
against
therapeutic
targets.
A
recent
and
elegant
method
reported
by
Heinis
et
al.
[14

]
has
successfully
produced
potent
thioether-linked
bicyclic
peptide
inhibi-
tors
against
human
proteases
(IC
50
=
20–50
nM)
through
phage
display.
It
should
be
noted,
however,
that
although
the
phage
display
selection
has
yielded
potent
inhibitors
(IC
50
<
100
nM)
for
certain
targets
from
an
initial
selec-
tion,
it
occasionally
yields
less
potent
inhibitors
(IC
50
>
1
mM)
and
thus
reconstruction
of
a
focused
library
based
on
previously
selected
peptide
sequences
becomes
necessary
to
reselect
for
more
potent
inhibitors.
This
is
attributed
to
the
sequence
diversity
of
bacterial
phage
display
libraries,
often
being
<10
10
,
resulting
from
the
transformation
efficiency
of
phages
to
the
host
cells
[34].
mRNA
display,
where
the
C-terminal
residue
of
a
peptide
is
fused
with
puromycin-linked
mRNA
[35],
requires
no
transformation
or
living
organisms
and
thus
has
diversities
in
the
trillions.
In
addition,
genetic
code
reprogramming
is
far
easier
to
perform
in
vitro
rather
than
in
bacteriophage
or
living
organisms,
even
though
a
technology
allowing
for
the
incorporation
of
one
or
two
proteinogenic
amino
acids
into
phage
displayed
peptides
has
been
reported
[36].
To
date
many
successful
selections
of
macrocyclic
pep-
tides
using
mRNA
display
or
its
variants
have
been
reported
[37].
The
Szostak
group
has
reported
two
examples
of
selections
discovering
macrocyclic
peptides
against
thrombin
[12]
(K
D
=
1.5
nM)
and
sortase
A
[38

]
(K
D
=
3
mM),
where
the
former
selection
used
a
post-
translational
thioether-crosslinking
method
while
the
lat-
ter
used
the
lanthionine-like
intramolecular
cyclization
method
described
earlier.
Roberts
et
al.
reported
the
selection
of
macrocyclic
peptides
against
Gai1
which
In
vitro
displays
for
the
screening
of
vast
libraries
Bashiruddin
and
Suga
135
Figure
4
(a)
His
His
Tyr
Tyr
Trp
Phe
Ala
Pro
Asn
Asp His
Leu
Cys Leu
Gly
Gln
Leu
D-Trp
Gly Asp
Val
IIe
Tyr
Tyr
Thr
Thr
Arg
Arg Asn
Cys
Val
KTfa
O
O
O
OO
O
O
O
O
O
O
O
OHO
O
O
OH
O
H
O
O
O
OO
O
O
O
O
O
OO O
S
O
O
O
OO
HO
HO
OH
OH
OH
HO
H
H
O
OO
O
O
O
S
F
N
N
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
HN
HN
HN
HN
HN
HN
HN
N
N
NH
N
H
N
H
N
H
N
H
H
N
H
NH
N
H
N
H
N
NH
NH
NH HN
HN
HN
HN
HN
HN
FF
NH
NH
HN
(b)
(c)
Thioether bond
Thioether bond
(d)
Current Opinion in Chemical Biology
Structures
of
inhibitory
thioether
macrocyclized
peptides
(nonproteinogenic
elements
highlighted
in
yellow).
(a)
Structure
of
the
macrocyclic
peptide
aCAP.
(b)
3D
structures
of
aCAP
binding
and
clamping
CmABCB1
in
an
inward
open
conformation
[46].
(c)
Structure
of
S2iL5
which
contains
an
e-N-trifluoroacetyl
lysine
residue
(K
Tfa
).
(d)
3D
structures
of
the
various
interactions
involved
in
the
inhibition
of
SIRT2
by
S2iL5
[48].
www.sciencedirect.com
Current
Opinion
in
Chemical
Biology
2015,
24:131–138

were
cyclized
via
an
N-hydroxysuccinamide-activated
crosslinking
reagent
between
the
N-terminal
amino
group
in
the
mainchain
and
an
e-amino
sidechain
of
a
lysine
residue
[39]
(K
D
=
2.1
nM).
It
should
be
noted
that
this
anti-Gai1
peptide
selection
involved
translation
of
a
peptide
library
with
N-methylphenylalanine
using
the
amber
codon
suppression
method,
but
no
sequences
containing
N-methylphenylalanine
were
discovered
[40].
The
FIT
system
has
also
been
coupled
with
mRNA
dis-
play,
referred
to
as
the
RaPID
(Random
non-standard
Peptides
Integrated
Discovery)
system
(Figure
3)
[41,42],
which
has
produced
bioactive
macrocyclic
pep-
tides
against
various
targets
[43

–45,47

,49].
By
initiating
peptide
translation
with
N-chloroacetyl-L-amino
acids
or
D-
amino
acids,
thioether
macrocyclized
peptide
libraries
can
be
translated
and
covalently
displayed
on
their
correspond-
ing
mRNAs.
For
instance,
the
RaPID
system
has
been
applied
for
the
selection
of
highly
isoform-selective
inhibi-
tors
against
the
protein
kinase
Akt2
[44]
(IC
50
=
110
nM
and
>250-fold
and
40-fold
higher
IC
50
for
Akt1
and
Akt2,
respectively)
and
a
histone
deacetylase
SIRT2
(Figure
4a,b)
[43,48]
(IC
50
=
3
nM
and
10-fold
and
100-
fold
higher
IC
50
against
SIRT1
and
SIRT3,
respectively;
an
X-ray
co-crystal
structure
of
SIRT2-macrocyclic
peptide
is
available).
Inhibitory
macrocyclic
peptides
were
also
discovered
against
a
bacterial
membrane
drug-transporter
MATE
[45,49,50

]
(multidrug
and
toxin
extrusion)
protein
and
three
macrocyclic
peptides
were
co-crystalized
with
MATE,
revealing
the
first
example
of
a
three-dimensional
structure
of
macrocyclic
peptides
isolated
by
the
RaPID
system.
Two
of
such
peptides
bear
a
lariat
structure
con-
sisting
of
an
N-terminal
‘mini-cycle’
followed
by
a
C-
terminal
‘tail’.
Interestingly,
the
crystal
structures
show
that
the
mini-cycle
was
buried
deep
into
the
cleft
of
the
transporter
with
the
tail
region
occupying
the
rest
of
the
cleft.
The
level
of
target
accommodation
observed
in
this
X-ray
structure
demonstrates
how
the
macrocyclic
portion
fits
to
the
drug-transporting
channel,
indicating
the
benefit
of
using
trillion
member
macrocyclic
peptide
libraries.
More
recently,
macrocyclic
peptide
inhibitors
were
also
selected
against
a
eukaryotic
ABC-drug
transporter
(Figure
4c,d)
[46]
(IC
50
=
65
nM);
the
X-ray
structural
analysis
of
one
of
the
macrocycles
has
revealed
its
distinct
structure
from
anti-MATE
macrocyclic
peptides
and
the
mechanism
for
its
inhibitory
activity.
A
major
strength
of
the
RaPID
system
is
the
ability
of
expressing
N-methyl-macrocyclic
peptides
and
the
selec-
tion
of
active
species.
To
date,
two
examples
of
such
inhibitors
have
been
published:
one
against
E6AP
[47

],
a
HECT
domain
E3
ubiquitin
ligase,
and
another
against
VEGFR2
[51],
a
human
membrane
protein
growth
factor
receptor
(in
this
work
an
updated
method
of
RaPID,
referred
to
as
TRAP
display
[52],
is
used).
Both
cases
have
given
potent
inhibitors
(K
D
=
0.6
nM
and
33
nM
for
E6AP
and
VEGFR2,
respectively).
Future
directions
Manipulations
of
the
translation
system
to
synthesize
vast
libraries
of
macrocyclic
peptides
with
similar
features
seen
in
naturally
occurring
ones
combined
with
screening
via
mRNA
display
methods
have
shown
great
potential
for
accelerating
the
discovery
of
bioactive
natural
pro-
duct-like
macrocycles.
Because
of
the
high
inhibitory
activity
and
specificity
of
the
selected
molecules,
these
technologies
seem
to
be
excellent
platforms
for
the
dis-
covery
of
drug
leads
against
enzymes
and
proteins
involved
in
protein–protein
interactions.
In
terms
of
drug
development,
these
technologies
have
solved
issues
of
proteolytic
instability
and
insufficient
potency
of
traditional
peptidic
drugs,
and
therefore
are
suitable
for
developing
drug
leads
for
extracellular
targets,
which
can
be
delivered
via
intravenous
administration
or
other
related
methods.
However,
some
issues
such
as
insuffi-
cient
cell-membrane
permeability
for
targeting
intra-
cellular
proteins
and
oral
bioavailability
of
selected
macrocyclic
peptides
still
remain
to
be
resolved.
In
the
future,
therefore,
it
will
be
required
to
improve
this
technology
by
adding
more
natural
product-like
struc-
tures,
for
example,
including
heterocycles
in
the
back-
bone,
or
an
appropriate
selective
pressure
that
enables
us
to
discover
macrocyclic
peptides
with
better
drug-like
properties.
Acknowledgements
We
thank
Dr.
Christopher
J
Hipolito,
Prof.
Osamu
Nureki’s
group
at
the
University
of
Tokyo,
and
Prof.
Hiroaki
Kato’s
group
at
Kyoto
University
for
the
collaborations
related
to
this
work.
This
work
was
supported
by
the
Platform
for
Drug
Discovery,
Informatics,
and
Structural
Life
Science
from
MEXT,
Japan
to
HS.
References
and
recommended
reading
Papers
of
particular
interest,
published
within
the
period
of
review,
have
been
highlighted
as:

of
special
interest

of
outstanding
interest
1.
Driggers
EM,
Hale
SP,
Lee
J,
Terrett
NK:
The
exploration
of
macrocycles
for
drug
discovery
—
an
underexploited
structural
class.
Nat
Rev
Drug
Discov
2008,
7:608-624.
2.
Walsh
CT:
The
chemical
versatility
of
natural-product
assembly
lines.
Acc
Chem
Res
2008,
41:4-10.
3.
Craik
DJ,
Fairlie
DP,
Liras
S,
Price
D:
The
future
of
peptide-based
drugs.
Chem
Biol
Drug
Des
2013,
81:136-147.
4.
Dinsmore
CJ,
Bogusky
MJ,
Culberson
JC,
Bergman
JM,
Homnick
CF,
Zartman
CB,
Mosser
SD,
Schaber
MD,
Robinson
RG,
Koblan
KS
et
al.:
Conformational
restriction
of
flexible
ligands
guided
by
the
transferred
noe
experiment:
potent
macrocyclic
inhibitors
of
farnesyltransferase.
J
Am
Chem
Soc
2001,
123:2107-2108.
5.
Khan
AR,
Parrish
JC,
Fraser
ME,
Smith
WW,
Bartlett
PA,
James
MN:
Lowering
the
entropic
barrier
for
binding
conformationally
flexible
inhibitors
to
enzymes.
Biochemistry
1998,
37:16839-16845.
6.
Rezai
T,
Yu
B,
Millhauser
GL,
Jacobson
MP,
Lokey
RS:
Testing
the
conformational
hypothesis
of
passive
membrane
permeability
using
synthetic
cyclic
peptide
diastereomers.
J
Am
Chem
Soc
2006,
128:2510-2511.
136
Omics
Current
Opinion
in
Chemical
Biology
2015,
24:131–138
www.sciencedirect.com

7.
Chatterjee
J,
Gilon
C,
Hoffman
A,
Kessler
H:
N-methylation
of
peptides:
a
new
perspective
in
medicinal
chemistry.
Acc
Chem
Res
2008,
41:1331-1342.
8.
Fischbach
MA,
Walsh
CT:
Assembly-line
enzymology
for
polyketide
and
nonribosomal
peptide
antibiotics:
logic,
machinery,
and
mechanisms.
Chem
Rev
2006,
106:3468-3496.
9.
Weissman
KJ,
Mu
¨ller
R:
Protein–protein
interactions
in
multienzyme
megasynthetases.
ChemBioChem
2008,
9:826-
848.
10.
Shimizu
Y,
Inoue
A,
Tomari
Y,
Suzuki
T,
Yokogawa
T,
Nishikawa
K,
Ueda
T:
Cell-free
translation
reconstituted
with
purified
components.
Nat
Biotechnol
2001,
19:751-755.
11.
Raffler
NA,
Schneider-Mergener
J,
Famulok
M:
A
novel
class
of
small
functional
peptides
that
bind
and
inhibit
human
alpha-
thrombin
isolated
by
mRNA
display.
Chem
Biol
2003,
10:69-79.
12.
Millward
SW,
Takahashi
TT,
Roberts
RW:
A
general
route
for
post-translational
cyclization
of
mRNA
display
libraries.
J
Am
Chem
Soc
2005,
127:14142-14143.
13.
Schlippe
YV,
Hartman
MC,
Josephson
K,
Szostak
JW:
In
vitro
selection
of
highly
modified
cyclic
peptides
that
act
as
tight
binding
inhibitors.
J
Am
Chem
Soc
2012,
134:10469-10477.
14.

Heinis
C,
Rutherford
T,
Freund
S,
Winter
G:
Phage-encoded
combinatorial
chemical
libraries
based
on
bicyclic
peptides.
Nat
Chem
Biol
2009,
5:502-507.
In
this
study,
phage
displayed
peptides
were
bicyclized
with
tris-bromo-
methyl
benzene
and
used
to
select
for
kallikrein
inhibiting
peptides.
Two
iterative
selections
were
required,
one
for
each
cycle,
and
resulted
in
a
bicyclic
peptide
with
1.5
nM
K
i
.
15.
Seebeck
FP,
Ricardo
A,
Szostak
JW:
Artificial
lantipeptides
from
in
vitro
translations.
Chem
Commun
(Camb)
2011,
47:6141-
6143.
16.
Goto
Y,
Katoh
T,
Suga
H:
Flexizymes
for
genetic
code
reprogramming.
Nat
Protoc
2011,
6:779-790.
17.
Murakami
H,
Ohta
A,
Ashigai
H,
Suga
H:
A
highly
flexible
tRNA
acylation
method
for
non-natural
polypeptide
synthesis.
Nat
Methods
2006,
3:357-359.
18.
Passioura
T,
Suga
H:
Flexizymes,
their
evolutionary
history
and
diverse
utilities.
Top
Curr
Chem
2014,
344:331-345.
19.
Goto
Y,
Iwasaki
K,
Torikai
K,
Murakami
H,
Suga
H:
Ribosomal
synthesis
of
dehydrobutyrine-
and
methyllanthionine-
containing
peptides.
Chem
Commun
(Camb)
2009,
23:3419-
3421.
20.
Yamagishi
Y,
Ashigai
H,
Goto
Y,
Murakami
H,
Suga
H:
Ribosomal
synthesis
of
cyclic
peptides
with
a
fluorogenic
oxidative
coupling
reaction.
ChemBioChem
2009,
10:1469-1472.
21.
Kawakami
T,
Ohta
A,
Ohuchi
M,
Ashigai
H,
Murakami
H,
Suga
H:
Diverse
backbone-cyclized
peptides
via
codon
reprogramming.
Nat
Chem
Biol
2009,
5:888-890.
22.
Ohshiro
Y,
Nakajima
E,
Goto
Y,
Fuse
S,
Takahashi
T,
Doi
T,
Suga
H:
Ribosomal
synthesis
of
backbone-macrocyclic
peptides
containing
g-amino
acids.
ChemBioChem
2011,
12:1183-1187.
23.
Kang
TJ,
Hayashi
Y,
Suga
H:
Synthesis
of
the
backbone
cyclic
peptide
sunflower
trypsin
inhibitor-1
promoted
by
the
induced
peptidyl-tRNA
drop-off.
Angew
Chem
Int
Ed
Engl
2011,
50:2159-
2161.
24.
Goto
Y,
Ohta
A,
Sako
Y,
Yamagishi
Y,
Murakami
H,
Suga
H:
Reprogramming
the
translation
initiation
for
the
synthesis
of
physiologically
stable
cyclic
peptides.
ACS
Chem
Biol
2008,
3:120-129.
25.
Iwasaki
K,
Goto
Y,
Katoh
T,
Suga
H:
Selective
thioether
macrocyclization
of
peptides
having
the
N-terminal
2-
chloroacetyl
group
and
competing
two
or
three
cysteine
residues
in
translation
(10
(2012)
5783).
Org
Biomol
Chem
2012,
10:9748.
26.
Fujino
T,
Goto
Y,
Suga
H,
Murakami
H:
Reevaluation
of
the
D-
amino
acid
compatibility
with
the
elongation
event
in
translation.
J
Am
Chem
Soc
2013,
135:1830-1837.
27.
Kawakami
T,
Murakami
H,
Suga
H:
Messenger
RNA-
programmed
incorporation
of
multiple
N-methyl-amino
acids
into
linear
and
cyclic
peptides.
Chem
Biol
2008,
15:32-42.
28.
Kawakami
T,
Murakami
H,
Suga
H:
Ribosomal
synthesis
of
polypeptoids
and
peptoid-peptide
hybrids.
J
Am
Chem
Soc
2008,
130:16861-16863.
29.
Goto
Y,
Murakami
H,
Suga
H::
Initiating
translation
with
D-amino
acids.
RNA
2008,
14:1390-1398.
30.
Sako
Y,
Morimoto
J,
Murakami
H,
Suga
H:
Ribosomal
synthesis
of
bicyclic
peptides
via
two
orthogonal
inter-side-chain
reactions.
J
Am
Chem
Soc
2008,
130:7232-7234.
31.
Nakajima
E,
Goto
Y,
Sako
Y,
Murakami
H,
Suga
H:
Ribosomal
synthesis
of
peptides
with
C-terminal
lactams,
thiolactones,
and
alkylamides.
ChemBioChem
2009,
10:1186-1192.
32.
Sako
Y,
Goto
Y,
Murakami
H,
Suga
H:
Ribosomal
synthesis
of
peptidase-resistant
peptides
closed
by
a
non-reducible
inter-
sidechain
bond.
ACS
Chem
Biol
2008,
3:241-249.
33.
Goto
Y,
Suga
H:
Translation
initiation
with
initiator
tRNA
charged
with
exotic
peptides.
J
Am
Chem
Soc
2009,
131:5040-
5041.
34.
Derda
R,
Tang
SK,
Li
SC,
Ng
S,
Matochko
W,
Jafari
MR:
Diversity
of
phage-displayed
libraries
of
peptides
during
panning
and
amplification.
Molecules
2011,
16:1776-1803.
35.
Roberts
RW,
Szostak
JW:
RNA-peptide
fusions
for
the
in
vitro
selection
of
peptides
and
proteins.
Proc
Natl
Acad
Sci
U
S
A
1997,
94:12297-12302.
36.
Tian
F,
Tsao
ML,
Schultz
PG:
A
phage
display
system
with
unnatural
amino
acids.
J
Am
Chem
Soc
2004,
126:15962-15963.
37.
Ito
K,
Passioura
T,
Suga
H:
Technologies
for
the
synthesis
of
mRNA-encoding
libraries
and
discovery
of
bioactive
natural
product-inspired
non-traditional
macrocyclic
peptides.
Molecules
2013,
18:3502-3528.
38.

Hofmann
FT,
Szostak
JW,
Seebeck
FP::
In
vitro
selection
of
functional
lantipeptides.
J
Am
Chem
Soc
2012,
134:8038-8041.
The
only
example
of
an
in
vitro
selection
for
lanthionine
containing
peptides.
Although
the
selected
lantipeptides
had
modest
binding
affi-
nity,
it
was
shown
that
the
lanthionine
moiety
was
required
for
binding.
39.
Millward
SW,
Fiacco
S,
Austin
RJ,
Roberts
RW:
Design
of
cyclic
peptides
that
bind
protein
surfaces
with
antibody-like
affinity.
ACS
Chem
Biol
2007,
2:625-634.
40.
Frankel
A,
Millward
SW,
Roberts
RW:
Encodamers:
unnatural
peptide
oligomers
encoded
in
RNA.
Chem
Biol
2003,
10:1043-
1050.
41.
Hipolito
CJ,
Suga
H:
Ribosomal
production
and
in
vitro
selection
of
natural
product-like
peptidomimetics:
the
FIT
and
RaPID
systems.
Curr
Opin
Chem
Biol
2012,
16:196-203.
42.
Passioura
T,
Suga
H:
Flexizyme-mediated
genetic
reprogramming
as
a
tool
for
noncanonical
peptide
synthesis
and
drug
discovery.
Chemistry
2013,
19:6530-6536.
43.

Morimoto
J,
Hayashi
Y,
Suga
H:
Discovery
of
macrocyclic
peptides
armed
with
a
mechanism-based
warhead:
isoform-
selective
inhibition
of
human
deacetylase
SIRT2.
Angew
Chem
Int
Ed
Engl
2012,
51:3423-3427.
A
lysine
analog,
e-N-trifluoroacetyl
lysine,
known
to
inhibit
SIRT2
was
incorporated
into
peptides
as
a
‘warhead’
to
direct
the
selection
of
inhibitory
macrocyclic
peptides.
Details
of
the
inhibitory
mechanism
is
discussed
herein.
44.
Hayashi
Y,
Morimoto
J,
Suga
H:
In
vitro
selection
of
anti-Akt2
thioether-macrocyclic
peptides
leading
to
isoform-selective
inhibitors.
ACS
Chem
Biol
2012,
7:607-613.
45.
Hipolito
C,
Tanaka
Y,
Katoh
T,
Nureki
O,
Suga
H:
In
vitro
selection
of
cyclic
peptides
for
cocrystallization
with
a
multidrug
and
toxic
compound
extrusion
protein.
Biopolymers
2013,
100:238-
239.
46.
Kodan
A,
Yamaguchi
T,
Nakatsu
T,
Sakiyama
K,
Hipolito
CJ,
Fujioka
A,
Hirokane
R,
Ikeguchi
K,
Watanabe
B,
Hiratake
J
et
al.:
Structural
basis
for
gating
mechanisms
of
a
eukaryotic
In
vitro
displays
for
the
screening
of
vast
libraries
Bashiruddin
and
Suga
137
www.sciencedirect.com
Current
Opinion
in
Chemical
Biology
2015,
24:131–138

P-glycoprotein
homolog.
Proc
Natl
Acad
Sci
U
S
A
2014,
111:4049-4054.
47.

Yamagishi
Y,
Shoji
I,
Miyagawa
S,
Kawakami
T,
Katoh
T,
Goto
Y,
Suga
H:
Natural
product-like
macrocyclic
N-methyl-peptide
inhibitors
against
a
ubiquitin
ligase
uncovered
from
a
ribosome-expressed
de
novo
library.
Chem
Biol
2011,
18:1562-
1570.
To
date,
this
selection
has
produced
macrocyclic
peptides
with
the
most
N-methyl
amino
acids
per
peptide
(up
to
5)
consisting
of
N-methyl-
glycine,
N-methyl-alanine,
N-methyl-serine
and
N-methyl-phenylalanine.
Several
multiple
N-methylated
macrocyclic
peptides
exhibited
low
to
subnanomolar
binding
affinites
as
well
as
some
in
vitro
inhibitory
acitivity.
48.
Yamagata
K,
Goto
Y,
Nishimasu
H,
Morimoto
J,
Ishitani
R,
Dohmae
N,
Takeda
N,
Nagai
R,
Komuro
I,
Suga
H
et
al.:
Structural
basis
for
potent
inhibition
of
SIRT2
deacetylase
by
a
macrocyclic
peptide
inducing
dynamic
structural
change.
Structure
2014,
22:345-352.
49.
Hipolito
CJ,
Tanaka
Y,
Katoh
T,
Nureki
O,
Suga
H:
A
macrocyclic
peptide
that
serves
as
a
cocrystallization
ligand
and
inhibits
the
function
of
a
MATE
family
transporter.
Molecules
2013,
18:10514-10530.
50.

Tanaka
Y,
Hipolito
CJ,
Maturana
AD,
Ito
K,
Kuroda
T,
Higuchi
T,
Katoh
T,
Kato
HE,
Hattori
M,
Kumazaki
K
et
al.:
Structural
basis
for
the
drug
extrusion
mechanism
by
a
MATE
multidrug
transporter.
Nature
2013,
496:247-251.
The
RaPID
system
was
used
to
select
for
macrocyclic
peptide
cocrys-
tallization
ligands
with
inhibitory
activity.
Peptides
with
macrocyclic
and
‘lariat’
structures
were
discovered
and
their
binding
mechanisms
are
discussed.
51.
Kawakami
T,
Ishizawa
T,
Fujino
T,
Reid
PC,
Suga
H,
Murakami
H:
In
vitro
selection
of
multiple
libraries
created
by
genetic
code
reprogramming
to
discover
macrocyclic
peptides
that
antagonize
VEGFR2
activity
in
living
cells.
ACS
Chem
Biol
2013,
8:1205-1214.
52.
Ishizawa
T,
Kawakami
T,
Reid
PC,
Murakami
H:
TRAP
display:
a
high-speed
selection
method
for
the
generation
of
functional
polypeptides.
J
Am
Chem
Soc
2013,
135:5433-5440.
138
Omics
Current
Opinion
in
Chemical
Biology
2015,
24:131–138
www.sciencedirect.com