of 7
Toolboxes
for
plant
systems
biology
research
Jihyun
Park
1
,
Gozde
S
Demirer
2
,
3
and
Lily
S
Cheung
1
The
terms
‘systems’
and
‘synthetic
biology’
are
often
used
together,
with
most
scientists
striding
between
the
two
fields
rather
than
adhering
to
a
single
side.
Often
too,
scientists
want
to
understand
a
system
to
inform
the
design
of
gene
circuits
that
could
endow
it
with
new
functions.
However,
this
does
not
need
to
be
the
progression
of
research,
as
synthetic
constructs
can
help
improve
our
understanding
of
a
system.
Here,
we
review
synthetic
biology
tool
kits
with
the
potential
to
overcome
pleiotropic
effects,
compensatory
mechanisms,
and
redundancy
in
plants.
Combined
with
-omics
techniques,
these
tools
could
reveal
novel
insights
on
plant
growth
and
development,
an
aim
that
has
gained
renewed
urgency
given
the
impact
of
climate
change
on
crop
productivity.
Addresses
1
School
of
Chemical
and
Biomolecular
Engineering,
Georgia
Institute
of
Technology,
Atlanta,
GA
30332,
USA
2
Department
of
Plant
Biology
and
Genome
Center,
University
of
California
Davis,
Davis,
CA
95616,
USA
3
Department
of
Chemical
Engineering,
California
Institute
of
Technology,
Pasadena,
CA
91125,
USA
Corresponding
author:
Cheung,
Lily
S
(
lily.cheung@gatech.edu
)
Current
Opinion
in
Biotechnology
2022,
75
:xx–yy
This
review
comes
from
a
themed
issue
on
Systems
biology
Edited
by
Mark
P
Styczynski
and
Neda
Bagheri
https://doi.org/10.1016/j.copbio.2022.102692
0958-1669/
ã
2022
The
Author(s).
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/4.0/
).
Introduction
Plant
science
has
experienced
a
surge
of
big
data,
with
techniques
like
laser
capture
microdissection,
Translating
Ribosome
Affinity
Purification
Sequencing
(TRAP-Seq),
Fluorescence-Activated
Cell
Sorting
(FACS),
and
time-
lapse
fluorescence
microscopy
[
1–4
]
providing
extraordi-
nary
tissue/cell-specific
information.
Yet,
our
ability
to
manipulate
genes
with
similar
precision
and
throughput
lacks
behind,
with
most
functional
studies
still
relying
on
knockout
mutants
and
systemic
overexpression
using
the
constitutive
Cauliflower
Mosaic
Virus
(CaMV)
35S
pro-
moter.
These
perturbations
in
single
genes
seldom
pro-
duce
visible
phenotypic
differences,
as
comprehensive
genetic
studies
and
large-scale
genomics
projects
have
shown
[
5
,
6
].
High
similarity
in
coding
sequences
among
plant
gene
families
often
results
in
complete
or
conditional
functional
redundancy,
leading
to
substantial
phenotypic
plasticity
buffering.
Moreover,
in
cases
where
single
gene
perturbations
do
produce
a
visible
phenotype,
pleiotropic
and
compensatory
effects
oftentimes
obscure
the
molecular
mechanisms
responsible
for
changes
in
transcript
or
protein
levels.
Advancing
plant
systems
biology
will
require
precise
perturbations
that
can
overcome
gene
redundancy,
pleio-
tropism,
and
compensatory
mechanisms
in
-omics
stud-
ies.
Such
an
approach
could
be
as
effective
as
experiments
that
used
cell-specific
perturbations
to
validate
predic-
tions
from
single-cell
transcriptomic
analyses,
which
have
revealed
novel
roles
for
known
factors
in
leaves
and
lateral
roots
[
2
,
7
].
However,
such
approach
requires
versatile
genetic
constructs
for
precise
spatiotemporal
control
of
gene
function.
Here,
we
review
recent
tool
kits
for
vector
assembly
and
transcriptional
and
post-translational
regu-
lation
of
gene
products
that
could
be
combined
with
-omics
studies,
focusing
on
those
applicable
to
a
multi-
tude
of
genes.
Vector
assembly
kits
A
crucial
challenge
in
plant
biology
research
is
the
exten-
sive
time
necessary
for
transformation
and
crosses.
Thus,
single-step
delivery
of
whole
gene
circuits
or
multiple
transgenes
is
ideal,
and
numerous
approaches
have
been
developed
to
facilitate
the
task
[
8
,
9
].
One
of
the
most
recently
developed
tools
is
loop
assem-
bly,
a
technique
for
recursive
fabrication
of
large
genetic
circuits
that
can
theoretically
generate
plasmids
with
unlimited
transcription
units
and
length
[
10

].
Loop
assembly
uses
two
Type
IIS
restriction
enzymes
and
corresponding
standardized
vector
sets
(Odd
and
Even
receiver
plasmids).
Assemblies
are
performed
through
iterated
‘loops.’
Two
sets
of
four
plasmid
vectors
are
provided,
which
allow
alternating
assembly
cycles.
The
iterative
process
of
combining
genetic
modules,
four
at
a
time,
can
be
continued
infinitely
by
alternating
between
odd
and
even
Loop
vectors
[
11
].
Standardized
parts
Several
libraries
of
natural
or
synthetic
genetic
parts
are
available
for
plants.
It
is
important
for
these
parts
to
be
orthogonal
and
quantitatively
characterized
[
12
].
This
means
the
parts
should
display
minimal
interaction
between
each
other
and
endogenous
components
and
have
defined
input-output
relations.
Heterologous
and
synthetic
transcriptional
regulators
are
more
likely
to
be
orthogonal
than
plant
derived
ones.
Available
online
at
www.sciencedirect.com
ScienceDirect
www.sciencedirect.com
Current
Opinion
in
Biotechnology
2019,
75
:102692
Transcriptional
activator-like
effectors
from
bacteria
like
Xanthomonas
[
13
]
and
zinc-finger
chimeras
[
14
]
were
some
of
the
earliest
genetic
parts
used
in
plants.
However,
their
engineering
was
laborious.
More
recently,
catalytically
inactive
Cas9
(dCas9)
has
been
shown
to
successfully
activate
and
repress
transcription
of
target
genes
without
appreciable
off-target
effects
[
15
].
The
main
limitation
in
the
generation
of
predictable
gene
circuits
is
still
the
time
necessary
to
test
and
model
the
outputs
of
genetic
parts
in
stably
transformed
plants.
Schaumberg
et
al.
showed
that
quantitative
characteriza-
tion
in
Arabidopsis
protoplasts
can
serve
as
a
good
proxy
for
in
planta
performance
by
testing
128
pairwise
combi-
nations
of
synthetic
promoters
and
repressors
[
16

].
The
system
used
luciferase
as
output
and
Hill
functions
for
modeling.
Variability
due
to
random
transgene
insertion,
common
in
Agrobacterium
-mediated
transformation,
was
a
challenge.
One
option
to
surmount
this
variability
is
to
use
a
ratio-
metric
luciferase
system
[
17
].
Another
option
would
be
to
create
lines
with
predetermined
genomic
landing
sites
like
those
in
Drosophila
melanogaster
[
18
].
The
bacterio-
phage
f
C31
DNA
site-specific
integrase
could
be
used
to
generate
mapped
genomic
locations
for
the
insertion
of
transgenes
with
subtle
differences,
facilitating
compari-
son.
The
f
C31
integration
system
has
already
been
demonstrated
to
work
as
a
memory
switch
in
Nicotiana
benthamiana
[
19
].
Another
tool
plant
biologists
could
borrow
from
fly
geneticists
are
balancer
chromosomes,
which
are
chromosomes
containing
multiple
inverted
regions
capable
of
suppressing
crossovers
during
meiosis
that
would
help
the
visual
tracking
of
chromosomes
carrying
transgenes
during
crosses
[
20
].
Temporal
control
of
gene
expression
and
protein
localization
with
inducible
systems
Chemical
inducers
can
be
used
to
initiate
the
transcrip-
tion
of
transgenes
at
defined
developmental
stages.
Since
the
creation
of
the
tetracycline-inducible
gene
expression
system
[
21
],
numerous
chemically
inducible
systems
for
plants
have
followed.
The
glucocorticoid-inducible
sys-
tem
is
a
primary
example,
which
uses
a
chimeric
tran-
scription
factor
(GVG)
consisting
of
the
GAL4
binding
domain
from
yeast,
the
VP16
activation
domain
from
herpes,
and
the
glucocorticoid
receptor
(GR)
from
rats
[
22
].
This
system
has
been
shown
to
be
tightly
regulated
and
rapidly
induced
in
tobacco,
Arabidopsis
,
rice,
pine,
and
citrus
plants
[
22–26
].
Although
the
glucocorticoid-inducible
system
has
been
demonstrated
effective,
its
orthogonality
has
been
ques-
tioned.
Activation
of
the
system
can
cause
developmental
growth
defects
and
interfere
with
endogenous
gene
expression
[
27
,
28
].
Zuo
et
al.
addressed
these
limitations
with
the
construction
of
the
estrogen-inducible
XVE
system,
which
uses
a
chimeric
transcriptional
factor
con-
sisting
of
the
LexA
binding
domain
from
bacteria,
VP16,
and
the
human
estrogen
receptor
(ER)
[
29
].
Another
alternative
is
the
recently
reported
dexamethasone-
inducible
pOp6/LhGR
system,
which
uses
the
Escherichia
coli
lac
repressor
lacI
His17
,
the
GAL4
transcription-activa-
tion-domain-II,
and
the
rat
GR
ligand
binding
domain
[
30
].
Post-translationally,
the
rapamycin-inducible
KnockSide-
ways
in
Plants
(KSP)
system
can
be
used
to
control
protein
localization.
This
system
relies
on
the
heterodi-
merization
of
the
FKBP
domain
of
HsFKBP12
and
the
FKBP12
rapamycin-binding
domain
of
mTOR.
The
sys-
tem
was
shown
efficient
at
directing
bait
proteins
to
the
plasma
membrane,
mitochondria,
microtubules,
and
nucleus
in
N.
benthamiana
[
31
].
Spatial
control
of
gene
expression
with
tissue/
cell-specific
promoters
In
the
last
decade,
numerous
tissue/cell-specific
promo-
ters
have
been
isolated
from
diverse
plant
species
(e.g.
Arabidopsis
,
rice,
tomato,
soybean)
[
32–35
].
These
pro-
moters
have
typically
been
employed
for
TRAP-Seq
and
developmental
studies
but
are
being
increasingly
used
for
precise
and
controlled
manipulation
of
genes
with
mini-
mum
adverse
effects.
These
promoters
provide
substan-
tial
advantage
over
constitutive
ones
like
CaMV35S,
which
can
cause
pleiotropic
effects
and
reduce
plant
growth.
Two
examples
are
worth
noting.
Decaestecker
et
al.
developed
a
technique
called
CRISPR-TSKO,
which
enables
the
creation
of
somatic
mutations
in
desired
cell
types,
tissues,
and
organs
(
Figure
1
a).
To
achieve
this,
researchers
used
tissue-
specific,
somatic
promoters
to
drive
Cas9
expression
and
demonstrated
root
cap-specific,
stomata-specific,
and
lateral
root-specific
gene
knockouts
in
Arabidopsis
[
36

].
Wang
et
al.
improved
on
this
idea
by
integrating
CRISPR/
Cas9,
the
XVE
system,
and
root-cell-type-specific
pro-
moters,
enabling
temporal
in
addition
to
spatial
control
of
gene
editing
(
Figure
1
b).
This
allowed
the
team
to
trigger
somatic
gene
knockout
in
Arabidopsis
root
meristem
with
estradiol
[
37

].
Engineering
of
synthetic
promoters
for
precise
control
of
transgene
expression
in
a
spatiotemporal
manner
has
been
another
important
advance
in
plant
science.
Readers
interested
in
a
detailed
discussion
of
their
rational
design
are
referred
to
Cai
et
al.
[
38
],
and
for
a
list
of
all
recent
plant
synthetic
promoters
to
Ali
and
Kim
[
39
].
CRISPR/dCas9
for
multigene
regulation
The
activation
and
repression
of
gene
expression
using
dCas9
in
plants
has
become
a
common
and
powerful
2
Systems
biology
Current
Opinion
in
Biotechnology
2019,
75
:102692
www.sciencedirect.com
approach
for
genetic
and
epigenetic
regulation.
In
addi-
tion
to
its
use
for
single-gene
regulation,
CRISPR/dCas9
platforms
offer
unparalleled
multiplex
ability
by
using
multiple
sgRNAs
simultaneously.
There
exist
several
tools
for
multiplexed
activation
in
plants
(dCas9-TV,
dCas9-SunTag,
dCasEV2.1)
[
40–42
].
Recently,
Pan
et
al.
developed
CRISPR-Act3.0
for
highly
efficient
multi-
plexed
gene
activation
and
performed
simultaneous
acti-
vation
of
many
enzyme-encoding
genes
in
rice
as
well
as
multigene
activation
in
Arabidopsis
[
43

].
A
critical
chal-
lenge
of
using
dCas9
for
gene
regulation
is
the
need
for
correct
protospacer
adjacent
motifs
(PAMs)
sequence
proximal
to
the
promoter.
When
using
SpCas9,
it
could
be
challenging
to
find
good
target
sites
with
NGG
PAMs,
given
that
promoters
in
plants
are
often
AT-rich.
To
overcome
this
limitation,
Pan
et
al.
successfully
adapted
the
use
of
dCas12b,
a
protein
that
recognizes
VTTV
PAMs
for
multigene
activation
with
CRISPR-Act3.0.
In
addition,
they
used
the
near-PAM-less
SpCas9
variant,
SpRY,
and
demonstrated
that
dSpRY-Act3.0
is
a
highly
promising
tool
for
multigene
regulation
[
43

].
CRISPR/dCas9
has
also
been
used
for
transcriptional
repression.
However,
to
our
knowledge,
repression
in
plants
has
not
been
done
commonly
in
a
multigene
manner
[
44
],
unlike
the
case
in
bacteria,
yeast,
and
human
cells
[
44–46
].
The
one
study
that
performed
simultaneous
multigene
repression
in
plants
by
Lowder
et
al.
used
a
synthetic
pco-dCas9-3X(SRDX)
transcriptional
repressor
to
reduce
transcript
levels
of
two
microRNAs
in
Arabi-
dopsis
[
47
].
Light
control
of
gene
expression
Optogenetics
uses
light
and
genetically
encoded
photo-
switches
to
alter
gene
expression
reversibly,
thus
offering
an
alternative
in
situations
where
promoters
with
the
desired
tissue/cell-specific
activities
are
unavailable
or
in
cells
with
poor
uptake
of
chemical
inducers.
The
technique
traditionally
uses
photoactivatable
channels
and
light
sensory
parts
from
bacteria,
algae,
and
plants.
However,
because
most
of
these
proteins
respond
to
the
same
wavelengths
of
light
that
control
plant
growth
and
development,
their
use
can
result
in
undesired
side
effects.
The
recently
developed
Plant
Usable
Light-
Switch
Elements
(PULSE)
system
overcomes
this
limi-
tation
and
illustrates
an
elegant
strategy
for
implementing
optogenetics
in
plants
[
48

].
PULSE
is
comprises
two
engineered
proteins
and
a
synthetic
promoter
(P
Opto
)
(
Figure
2
a).
In
this
system,
a
transcriptional
repressor
(B
Off
)
prevents
gene
expression
under
blue
light
(

450
nm),
while
an
activator
(R
On
)
induces
expression
under
red
light
(

660
nm).
Transgene
expression
can
be
tuned
by
controlling
the
intensity
of
monochromatic
red
light
[
48

].
PULSE’s
effectiveness
was
demonstrated
with
transient
expression
in
N.
benthamiana
leaves
and
transgenic
Arabidopsis
.
Gene
expression
was
prevented
in
daylight
(250–800
nm)
by
the
B
Off
repressor
and
inactive
in
dark,
allowing
plants
to
grow
under
normal
photoperiod.
Trans-
fer
to
monochromatic
red
light
induced
expression
of
luciferase
in
Arabidopsis
,
reaching
maximum
levels
after

12
hours;
while
return
to
white
light
decreased
lumi-
nescence,
showing
the
reversibility
of
the
system.
Whether
PULSE
inadvertently
affects
endogenous
sig-
naling
due
to
several
of
its
components
originating
from
plants
remains
to
be
explored.
Nanobodies
and
inducible
systems
for
post-
translational
control
Nanobodies
are
small,
single-domain
antibodies
isolated
from
camelids
that
can
be
transgenically
expressed
to
bind
endogenous
proteins
or
small
molecules.
Research-
ers
demonstrated
the
use
of
nanobody
for
selective
pro-
tein
degradation,
which
can
be
completed
faster
(within
Toolboxes
for
plant
systems
biology
research
Park,
Demirer
and
Cheung
3
Figure
1
(a)
(b)
Temporal control
Spatial control
Tissue/cell-specific
promoters
Promoter
Cas9
Cas9
Cas9
Cas9
Cas9
U6
U6
U6
U6
U6
gRNAs
gRNAs
gRNAs
gRNAs
gRNAs
Inducer
Activator
Promoter
Time
Expression
Nucleus
Gene
Promoter
Promoter
Promoter
Promoter
Current Opinion in Biotechnology
Spatiotemporal
control
of
gene
expression.
(a)
Tissue/cell-specific
promoters
can
enable
spatially
controlled
expression
of
Cas9.
Different
colors
represent
promoters
expressed
in
different
plant
cell/
tissue
types.
(b)
Inducible
Cas9
expression
in
plants
confers
temporal
control
and
can
be
merged
with
cell-specific
promoters
to
provide
both
spatial
and
temporally
manipulated
genes.
Different
colored
peaks
represent
the
ability
of
expressing
genes
at
the
desired
time
using
inducer
molecules,
and
at
the
desired
cells
using
cell-type-
specific
promoters.
www.sciencedirect.com
Current
Opinion
in
Biotechnology
2019,
75
:102692
minutes
to
few
hours)
than
transcriptional
or
RNAi-
mediated
downregulation.
The
technique
fuses
a
nano-
body
with
an
F-box
domain,
resulting
in
a
chimera
that
can
polyubiquitinate
the
proteins
recognized
by
the
nano-
body
and
thus
target
it
for
proteasome
degradation
(
Figure
2
b).
This
prevents
compensation
effects
that
could
otherwise
complicate
the
interpretation
of
pheno-
typic
changes.
Moreover,
when
combined
with
somatic
tissue-specific
or
inducible
promoters
in
multicellular
organisms,
restricted
protein
degradation
can
circumvent
the
lethality
or
sterility
associated
with
knockouts
of
the
target
gene.
An
anti-GFP
nanobody
allows
selective
degradation
of
functional
GFP-tagged
proteins.
Furthermore,
the
pro-
cess
can
be
monitored
in
real-time
by
the
loss
of
fluores-
cence.
The
use
of
this
approach
in
multicellular
organ-
isms
was
first
demonstrated
in
D.
melanogaster
.
The
method,
named
deGradFP,
was
shown
to
titrate
target
proteins
in
less
than
three
hours
and
phenocopy
loss-of-
4
Systems
biology
Figure
2
(
a)
(b
)
O
pto
g
enetic
contro
l
o
f g
ene
e
x
pre
ss
ion
N
ano
b
o
dy
me
d
iate
d-
protein
d
e
g
ra
d
ation
Ant
i
-G
F
P
n
a
no
b
od
y
Ta
rget
prote
i
n
Ubiq
u
i
t
i
n
a
t
i
on
G
F
P
t
a
g
E
3 Li
g
a
se
Proteo
ly
s
i
s
Red
li
g
h
t
L
e
af
ep
i
der
mi
s
R
O
ff
Ub
Ub
Ub
Ub
Ub
Ub
Ub
Ub
R
On
P
Opto
a
nt
i
-G
F
P
E
3
-
Li
g
a
se
Current
Op
i
n
i
on
i
n
B
i
ote
ch
no
l
og
y
Optogenetic
control
of
protein
degradation.
(a)
In
the
PULSE
system,
a
transcriptional
repressor
(B
Off
)
derived
from
the
bacterial
light-regulated
DNA-binding
protein
EL222
and
a
plant
EAR
repression
domain
prevents
gene
expression
under
blue
light
(

450
nm),
while
an
activator
(R
On
)
from
a
plant
phytochrome
B
(PhyB)
and
a
phytochrome-interacting
factor
6
(PIF6)
induces
expression
under
red
light
(

660
nm).
The
synthetic
promoter
(P
Opto
)
is
composed
of
repeated
binding
domains
for
EL222
and
PIF6
upstream
of
a
human
cytomegalovirus
minimal
promoter
and
can
be
used
to
drive
expression
of
anti-GFP
nanobodies.
(b)
Anti-GFP
nanobody
and
E3
Ubiquitin
Ligase
chimeras
can
be
used
to
selectively
target
and
degrade
GFP-tagged
proteins.
Table
1
Summary
of
plant
tools
and
their
availability
Tool
kit
Availability
References
Loop
assembly
Addgene
[
10

,
11
]
uLoop
assembly
Addgene
[
10

,
11
]
Dual
luciferase
ratiometric
reporter
system
Addgene
[
17
]
f
C31
integration
system
Addgene
[
19
]
XVE
inducible
transcription
factor
Addgene
[
29
,
63
]
pOp6/LhGR
gene
expression
system
Nottingham
Arabidopsis
Stock
Centre
(NASC)
[
30
]
Rapamycin-inducible
KnockSideways
in
Plants
(KSP)
VIB-UGent
Center
for
Plant
Systems
Biology
[
31
]
CRISPR-TSKO
Addgene
[
36

]
Inducible
CRISPR/Cas9
system
Addgene
[
37

]
dCas9-TV
Upon
request
to
authors
[
42
]
dCas9-SunTag
Addgene
[
40
]
dSpRY-Act3.0
Addgene
[
43

]
pco-dCas9-3X(SRDX)
Addgene
[
47
]
PULSE
system
Addgene
[
48

]
Anti-GFP
nanobody-based
degradation
system
Upon
request
to
authors
[
50
,
51

]
Anti-GFP
nanobody-based
delocalization
Upon
request
to
authors
[
52
]
Current
Opinion
in
Biotechnology
2019,
75
:102692
www.sciencedirect.com