of 7
Conditional Guide RNAs: Programmable Conditional
Regulation of CRISPR/Cas Function in Bacteria via Dynamic
RNA Nanotechnology
Mikhail H. Hanewich-Hollatz
,
§
, Zhewei Chen
,
§
, Jining Huang
, Lisa M. Hochrein
, and Niles A. Pierce
,
,
,
Division of Biology & Biological Engineering, California Institute of Technology, Pasadena, CA 91125, USA
Division of Engineering & Applied Science, California Institute of Technology, Pasadena, CA 91125, USA
Weatherall Institute of Molecular Medicine, University of Oxford, Oxford OX3 9DS, UK
ABSTRACT:
Dynamic RN
A Nanotechnology
in Living Cells
Programmable
RNA trigger
X
dCas9
Constitutively inactive
(OFF
ON logic)
cgRNA
Y
Programmable
DNA target
Constitutively
active
(ON
OFF
logic)
X
dCas9
Y
cgRNA
A guide RNA (gRNA) directs the function of a CRISPR protein effector to a target gene of
choice, providing a versatile programmable platform for engineering diverse modes of syn-
thetic regulation (edit, silence, induce, bind). However, the fact that gRNAs are constitu-
tively active places limitations on the ability to confine gRNA activity to a desired location
and time. To achieve programmable control over the scope of gRNA activity, here we apply
principles from dynamic RNA nanotechnology to engineer conditional guide RNAs (cgRNAs)
whose activity is dependent on the presence or absence of an RNA trigger. These cgRNAs
are programmable at two levels, with the trigger-binding sequence controlling the scope of
the effector activity and the target-binding sequence determining the subject of the effector
activity. We demonstrate molecular mechanisms for both constitutively active cgRNAs that
are conditionally inactivated by an RNA trigger (ON
OFF logic) and constitutively inac-
tive cgRNAs that are conditionally activated by an RNA trigger (OFF
ON logic). For each
mechanism, automated sequence design is performed using the reaction pathway designer
within NUPACK to design an orthogonal library of three cgRNAs that respond to different
RNA triggers. In
E. coli
expressing cgRNAs, triggers, and silencing dCas9 as the protein
effector, we observe programmable conditional gene silencing with a median dynamic range
of
6-fold for an ON
OFF “terminator switch” mechanism,
15-fold for an ON
OFF
“splinted switch” mechanism, and
3.6-fold for an OFF
ON “toehold switch” mechanism;
the median crosstalk within each cgRNA library is
<
2%,
<
2%, and
20% for the three mechanisms. By providing programmable
control over both the scope and target of protein effector function, cgRNA regulators offer a promising platform for synthetic biology.
KEYWORDS:
small conditional RNA (scRNA), programmable conditional regulators, allosteric regulators, CRISPR/Cas, dy-
namic RNA nanotechnology, molecular programming
INTRODUCTION
Dynamic RNA nanotechnology holds great promise as a
paradigm for introducing synthetic regulatory links into liv-
ing cells and organisms. We envision small conditional RNAs
(scRNAs) that, upon detection of a programmable nucleic
acid input, change conformation to produce a programmable
output that up-regulates or down-regulates the activity of a
biological pathway. In this scenario, the input controls the
scope of regulation and the output controls the target of reg-
ulation, with the scRNA performing signal transduction to
create a logical link between the two.
1,2
Any pathway that
recognizes RNA is a potential candidate for conditional reg-
ulation by scRNAs (e.g., RNA interference, RNase H, PKR,
RIG-1); the CRISPR/Cas pathway is a particularly attractive
candidate because of its functional versatility, high regulatory
dynamic range, and portability between species.
3–5
The repurposing of RNA-guided CRISPR effectors through
development of modified guide RNAs (gRNAs) and CRISPR-
associated (Cas) proteins has yielded a suite of powerful
tools for biological research and synthetic biology. Precision
genome editing has been achieved in a variety of organisms us-
ing gRNAs to direct the nuclease activity of Cas9 and Cas12a
(Cpf1) to a target gene of choice.
3,6–8
Mutation of the nucle-
ase domains to produce a catalytically dead Cas9 (dCas9) has
enabled silencing of genetic expression via inhibition of tran-
scriptional elongation,
4,9
or induction of genetic expression
using dCas9 fusions that incorporate transcriptional regula-
tory domains.
5
Other dCas9 fusions have mediated target-
binding to enable visualization of genomic loci,
10,11
epigenetic
modification,
12
and single-base editing at a specific genomic
locus.
3,13
Hence, gRNA:effector complexes combine the ben-
efits of the rich functional vocabulary of the protein effector
(edit, silence, induce, bind) and the programmability of the
gRNA in targeting effector activity to a gene of choice.
Because gRNAs are constitutively active, additional mea-
sures are needed to restrict effector activity to a desired lo-
cation and time. Temporal control can be achieved by small-
molecule induction of gRNAs
14,15
or Cas9,
16
but this comes
with limitations in terms of multiplexing and spatial con-
trol. Spatiotemporal control has been achieved by regulation
of Cas9 via photoactivation
17
or via tissue-specific promot-
ers
18,19
or microRNAs,
20
which comes with the unwelcome
restriction that all gRNAs are subject to the same regulatory
scope. Systematic mapping of the structure and sequence
properties of functional gRNAs has revealed that Cas9 activ-
ity is tolerant to significant modifications the standard gRNA
structure,
21,22
facilitiating introduction of auxiliary domains
that enable conditional control of gRNA activity via struc-
tural changes induced by small-molecules,
23,24
protein-bound
RNAs,
25
nucleases,
26
or nuclease-recruiting DNAs.
26
Alter-
natively, the activity of standard gRNAs has been modulated
by antisense RNAs
27
or by photolysis of antisense DNAs in-
corporating photocleavable groups.
28
Here, pursuing the scRNA paradigm of programmable con-
ditional regulation based on dynamic RNA nanotechnology,
and leveraging information on tolerated modifications of stan-
dard gRNA structure, we set out to engineer conditional
guide RNAs (cgRNAs) that change conformation in response
to an RNA trigger X to conditionally direct the function
of dCas9 to a target gene Y. Unlike a standard gRNA, a
cgRNA is programmable at two levels, with the trigger-
binding sequence controlling the scope of cgRNA activity and
1
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint
.
http://dx.doi.org/10.1101/525857
doi:
bioRxiv preprint first posted online Jan. 21, 2019;
a
Programmable
trigger
cgRNA
Y
Programmable
target
X
Protein effector
Y
Effector
function
X
Protein effector
Conditional gu
ide RNA (cgRNA) logic and function
...
...
Standard guide RNA (gRNA) structure and interactions
Cas9
handle
S.
pyogenes
terminator
Target-binding
region
gRNA
Target gene Y
Target
PAM
gRNA:dCas9:target
complex
b
Protein ef
fector
dCas9
...
...
Silence
Induce
Edit
Bind
Silence
Induce
Edit
Bind
Constitutively active
(ON
OFF
logic)
Constitutively inactive
(OFF
ON logic)
cgRNA
Figure 1.
Programmable regulators. (a) A conditional guide RNA
(cgRNA) changes conformation in response to a programmable
trigger X to conditionally direct the activity of a protein effec-
tor to a programmable target Y. Top: a constitutively active
cgRNA is conditionally inactivated by X (ON
OFF logic). Bot-
tom: a constitutively inactive cgRNA is conditionally activated by
X (OFF
ON logic). (b) A standard guide RNA (gRNA) is con-
stitutively active, directing the function of protein effector dCas9
to a target gene Y; different dCas9 variants implement different
functions (edit, silence, induce, bind).
the target-binding sequence determining the subject of effec-
tor activity. Functionally, the cgRNA must perform sequence
transduction between X and Y as well as shape transduction
between active/inactive conformations. cgRNA activity can
be engineered to toggle either OFF
ON (as was recently
demonstrated by Siu and Chen
29
) or ON
OFF in response
to a cognate RNA trigger X; this conditional control can be
exerted over dCas9 variants that either edit, silence, induce,
or bind the target Y, emphasizing the broad functional poten-
tial available via interplay between cgRNA logic and protein
effector function (Figure 1a).
RESULTS AND DISCUSSION
Constitutively Active Terminator Switch cgRNA Mech-
anism (ON
OFF logic).
As a starting point, consider the
constitutively active “terminator switch” cgRNA mechanism
of Figure 2a that is conditionally inactivated by RNA trig-
ger X (ON
OFF logic). Compared to a standard gRNA
(Figure 1b), the cgRNA has a modified terminator region
with an extended loop and rationally designed sequence do-
mains “d-e-f”. Hybridization of the RNA trigger X to these
modified domains is intended to form a structure incompati-
ble with cgRNA mediation of dCas9 function. We validated
the cgRNA mechanism
in vivo
in
E. coli
expressing silencing
dCas9
4
as the protein effector and a fluorescent protein re-
porter (mRFP) as the target gene Y. A cell line expressing
the cgRNA exhibits low fluorescence (ON state) while a cell
line expressing both the cgRNA and the cognate RNA trig-
ger exhibit high fluorescence (OFF state), achieving a condi-
tional response of an order of magnitude (Figure 2b). With
the terminator switch mechanism, the sequences of the RNA
0
0.5
1.0
0
10000
20000
A
B
C
X
C
X
A
X
B
X
C
X
A
X
B
c
11.1 ± 0.9
OFF:ON
AF
Trigger
X
A
X
B
-
X
C
-
cgRNA
-
Fluorescence (au)
-0.05
< 0.008
1.00
-0.02
< 0.02
1.0
< 0.02
0.08
1.0
Normalized fluorescence (au)
X
C
X
A
X
B
Trigger
cgRNA
A
B
C
cgRNA
cgRNA:trigger
d
e
d
e
f
d*
e*
f*
f
d*
e*
f*
b
a
Terminator Switch cgRNA
ON
State
OFF
State
Constitutively
Active
RNA trigger X
10
1
10
3
10
5
0
1000
2000
ON
OFF
Fluorescence intensity (au)
Counts
u
Target gene Y
...
...
u*
u
u
6.2 ± 0.8
4.9 ± 0.6
Crosstalk
Autofluorescence
Standard gRNA
cgRNA
cgRNA + trigger
No-target gRNA
Figure 2.
Conditional gene silencing using constitutively active
terminator switch cgRNA in bacteria (ON
OFF logic). (a) Mech-
anism schematic. The constitutively active cgRNA is inactivated
by hybridization of RNA trigger X. Rational sequence design of
cgRNA terminator region (domains “d-e-f” comprising 6 nt linker,
4 nt stem, 30 nt loop) and complementary trigger region (domains
“f*-e*-d*”). (b) Expression of RNA trigger X (40 nt + synthetic
terminator) toggles the cgRNA from ON
OFF, leading to an in-
crease in fluorescence. Single-cell fluorescence intensities via flow
cytometry. Induced expression (aTc) of silencing dCas9 and con-
stitutive expression of mRFP target gene Y and either: standard
gRNA (ideal ON state), cgRNA(ON state), cgRNA + RNA trig-
ger X (OFF state; trigger expression is IPTG-induced), no-target
gRNA that lacks target-binding region (ideal OFF state). Aut-
ofluorescence (AF): cells with no mRFP. (c) Programmable con-
ditional regulation using 3 orthogonal cgRNAs (A, B, C). Left:
ON
OFF conditional response to cognate trigger (OFF:ON ratio
= [OFF
AF]/[ON
AF]). Right: crosstalk between non-cognate
cgRNA/trigger pairs ([non-cognate trigger
no trigger]/[cognate
trigger
no trigger]). Fluorescence via flow cytometry: mean
±
estimated standard error of the mean (with uncertainty propaga-
tion) of median single-cell fluorescence over 20,000 cells,
N
= 3
replicate wells.
trigger X and the silencing target Y are fully independent,
with the cgRNA mediating allosteric regulation – the trigger
down-regulates cgRNA:dCas9 function not by sequestering
the target-binding region (orange in Figure 2a), but by hy-
bridizing to the distal trigger-binding region (blue). Ideally,
a cgRNA would have a strong ON state with activity equiv-
alent to a standard gRNA and a clean OFF state with no
activity. Testing control lines expressing a standard gRNA
targeting Y (ideal ON state) or a no-target gRNA that lacks
the target-binding region (ideal OFF state), we observe a dy-
namic range of two orders of magnitude, demonstrating room
2
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint
.
http://dx.doi.org/10.1101/525857
doi:
bioRxiv preprint first posted online Jan. 21, 2019;
cgRNA:trigger
15 ± 4
0
1000
2000
3000
15 ± 5
8 ± 3
-0.02
< 0.02
1.0
-0.01
0.01
1.0
0.06
0.07
1.0
0
0.5
1.0
cgRNA
RNA trigger X
d
e
d
e
d*
d*
e*
e*
a
Splinted Switch cgRNA
ON
State
OFF
State
Constitutively
Active
b
10
1
10
3
10
5
0
1000
2000
Fluorescence intensity (au)
Counts
ON
OFF
Target gene Y
...
...
u*
u
u
u
A
B
C
X
C
X
A
X
B
X
C
X
A
X
B
c
OFF:ON
AF
Trigger
cgRNA
Fluorescence (au)
Crosstalk
Normalized fluorescence (au)
X
C
X
A
X
B
Trigger
cgRNA
A
B
C
X
A
X
B
-
X
C
-
-
Autofluorescence
Standard gRNA
cgRNA
cgRNA + trigger
No-target gRNA
Figure 3.
Conditional gene silencing using constitutively ac-
tive splinted switch cgRNAs in bacteria (ON
OFF logic). (a)
Mechanism schematic. The constitutively active cgRNA is in-
activated by hybridization of RNA trigger X. Rational sequence
design of the 35 nt Cas9 handle loop (domain “d”) and an ex-
tended 35 nt terminator hairpin loop (domain “e”). (b) Expres-
sion of RNA trigger X (70 nt + synthetic terminator) toggles the
cgRNA from ON
OFF, leading to an increase in fluorescence.
Single-cell fluorescence intensities via flow cytometry. Induced ex-
pression (aTc) of silencing dCas9 and constitutive expression of
sfGFP target gene Y and either: standard gRNA (ideal ON state),
cgRNA(ON state), cgRNA + RNA trigger X (OFF state), no-
target gRNA that lacks target-binding region (ideal OFF state).
Autofluorescence (AF): cells with no sfGFP. (c) Programmable
conditional regulation using 3 orthogonal cgRNAs (A, B, C). Left:
ON
OFF conditional response to cognate trigger (OFF:ON ratio
= [OFF
AF]/[ON
AF]). Right: crosstalk between non-cognate
cgRNA/trigger pairs ([non-cognate trigger
no trigger]/[cognate
trigger
no trigger]). Fluorescence via flow cytometry: mean
±
estimated standard error of the mean (with uncertainty propaga-
tion) of median single-cell fluorescence over 20,000 cells,
N
= 3
replicate wells.
for future improvement of both the ON state and the OFF
state for this cgRNA mechanism. To test programmability,
we used NUPACK
30,31
to design a library of three orthogonal
cgRNA/trigger pairs (Figure 2c), achieving a median
6-fold
conditional ON
OFF response to expression of the cognate
trigger (left) and median crosstalk below 2% between non-
cognate cgRNA/trigger combinations (right).
Constitutively Active Splinted Switch cgRNA Mechanism
(ON
OFF logic).
To test an alternative approach to im-
plementing this ON
OFF conditional logic, we next tested
a constitutively active “splinted switch” cgRNA mechanism
(Figure 3a) that has extended loops in both the Cas9 handle
(domain “d”) and terminator (domain “e”). Hybridization of
RNA trigger X to both loops is intended to form a splint that
is structurally incompatible with cgRNA mediation of dCas9
function. In
E. coli
expressing silencing dCas9 and a fluo-
rescent protein reporter (sfGFP) as the target gene Y, the
splinted switch achieves an order of magnitude ON
OFF
conditional response to expression of RNA trigger X (Fig-
ure 3b). Notably, the ON state is similar to that of a standard
gRNA, implying that dCas9 is more tolerant of the splinted
switch modifications to the handle and terminator loops than
it was to the terminator switch modifications to the termina-
tor linker/stem/loop. However, these gains in ON state are
partially offset by losses in the cleanliness of the OFF state,
which falls short of the no-target gRNA control by an order of
magnitude. Examining a library of three orthogonal splinted
switch cgRNA/trigger pairs designed using NUPACK (Fig-
ure 3c), we observe a median
15-fold ON
OFF conditional
response to expression of the cognate trigger and median
crosstalk below 2% between non-cognate cgRNA/trigger com-
binations. As with the terminator switch mechanism, splinted
switch cgRNAs are allosteric regulators – the trigger down-
regulates cgRNA:dCas9 function by hybridizing to extended
loops (blue in Figure 3a) distal to the target-binding region
(orange). The resulting full sequence independence between
RNA trigger X and target gene Y provides the flexibility for
X to control regulatory scope independent of the choice of Y.
Constitutively Inactive Toehold Switch cgRNA Mecha-
nism (OFF
ON logic).
To reverse the conditional logic, we
then tested a constitutively inactive “toehold switch” cgRNA
mechanism (Figure 4a) that is conditionally activated by
RNA trigger X (OFF
ON logic). The target-binding re-
gion of the cgRNA (domain “u”) is initially sequestered by
a 5
extension to inhibit recognition of target gene Y; hy-
bridization of trigger X to this extension is intended to de-
sequester the target-binding region and enable cgRNA direc-
tion of dCas9 function to target gene Y. In
E. coli
expressing
silencing dCas9 and a fluorescent protein reporter (mRFP)
as the target gene Y, the toehold switch cgRNA achieves less
than an order of magnitude OFF
ON conditional response
to expression of RNA trigger X (Figure 4b). In this case,
the OFF state is degraded by an order of magnitude rela-
tive to the ideal OFF state (no-target gRNA control) and
the ON state is degraded by an order of magnitude rela-
tive to the ideal ON state (standard gRNA control). For
a library of three orthogonal toehold switch cgRNA/trigger
pairs designed using NUPACK (Figure 4c), we observe a me-
dian
3.6-fold OFF
ON conditional response to expression
of the cognate trigger and median crosstalk of
20% between
non-cognate cgRNA/trigger combinations. Recently, Siu and
Chen demonstrated a median
6.6-fold OFF
ON condi-
tional response using toehold switch cgRNAs with subtly dif-
ferent structural details in the sequestration of the target-
binding region.
29
Unlike the terminator switch and splinted
switch mechanisms for ON
OFF logic, toehold switch cgR-
NAs for OFF
ON logic are not allosteric as the cgRNA ini-
tially down-regulates cgRNA:dCas9 function by sequestering
the target-binding region (orange domain “u” in Figure 4a)
with a portion of the trigger-binding region (orange domain
“u*”). As a result, the toehold switch cgRNAs offer only
partial sequence independence between the trigger X and the
target gene Y (“u” is a subsequence of both X and Y). This
partial sequence dependence is not necessarily limiting for
synthetic biology applications where the trigger can be ratio-
nally designed and expressed exogenously, but does pose a
limitation in situations where X and Y are both endogenous
sequences.
Computational Sequence Design of Libraries of Orthog-
onal cgRNAs using NUPACK.
For each cgRNA mechanism
(Figures 2–4), sequence design was performed using the reac-
3
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint
.
http://dx.doi.org/10.1101/525857
doi:
bioRxiv preprint first posted online Jan. 21, 2019;
0
0.5
1.0
X
C
X
A
X
B
X
C
X
A
X
B
1.0
< 0.2
1.0
-0.2
0.2
1.0
Normalized fluorescence (au)
X
C
X
A
X
B
Trigger
cgRNA
A
B
C
Crosstalk
0.2
0.4
0.3
0
10000
20000
A
B
C
4.4 ± 0.4
OFF:ON
AF
Trigger
X
A
X
B
-
X
C
-
cgRNA
-
Fluorescence (au)
1.7 ± 0.2
3.6 ± 0.3
cgRNA:trigger
cgRNA
RNA trigger X
d
u*
d*
u
u
e
u*
d
e
a
Toehold Switch cgRNA
OFF
State
ON
State
Constitutively
Inactive
b
Target gene Y
...
...
u*
u
u
u
c
10
1
10
3
10
5
0
1000
2000
ON
OFF
Fluorescence intensity (au)
Counts
No-target gRNA
cgRNA
cgRNA + trigger
Standard gRNA
Autofluorescence
d*
Figure 4.
Conditional gene silencing using constitutively inac-
tive toehold switch cgRNAs in bacteria (OFF
ON logic). (a)
Mechanism schematic. The constitutively inactive cgRNA is ac-
tivated by hybridization of RNA trigger X. Rational sequence de-
sign of the toehold (domain “d”; 15 nt ) and loop (domain “e”;
8 nt) flanking the sequestration domain “u*” (20 nt). (b) Ex-
pression of RNA trigger X (35 nt + synthetic terminator) toggles
the cgRNA from OFF
ON, leading to a decrease in fluorescence.
Single-cell fluorescence intensities via flow cytometry. Induced ex-
pression (aTc) of silencing dCas9 and constitutive expression of
mRFP target gene Y and either: no-target gRNA that lacks target-
binding region (ideal OFF state), cgRNA (OFF state), cgRNA +
RNA trigger X (ON state), standard gRNA (ideal ON state). Aut-
ofluorescence (AF): cells with no mRFP. (c) Programmable con-
ditional regulation using 3 orthogonal cgRNAs (A, B, C). Left:
OFF
ON conditional response to cognate trigger (OFF:ON ratio
= [OFF
AF]/[ON
AF]). Right: crosstalk between non-cognate
cgRNA/trigger pairs ([non-cognate trigger
no trigger]/[cognate
trigger
no trigger]). Fluorescence via flow cytometry: mean
±
estimated standard error of the mean (with uncertainty propaga-
tion) of median single-cell fluorescence over 20,000 cells,
N
= 3
replicate wells.
tion pathway designer within NUPACK.
30,31
Following Wolfe
et al.,
31
sequence design was formulated as a multistate op-
timization problem using target test tubes to represent re-
actant and product states of cgRNA/trigger hybridization,
as well as to model crosstalk between orthogonal cgRNAs
(Figure 5a). Each reactants tube (Step 0) and products tube
(Step 1) contains a set of desired “on-target” complexes (each
with a target secondary structure and target concentration)
corresponding to the on-pathway hybridization products for
a given step, and a set of undesired “off-target” complexes
(each with a target concentration of 0 nM) corresponding to
on-pathway reactants and off-pathway hybridization crosstalk
for a given step. Hence, these elementary step tubes are de-
signed for full conversion of cognate reactants into cognate
products and against local hybridization crosstalk between
these same reactants. To simultaneously design
N
orthogonal
systems, elementary step tubes are specified for each system
(Figure 5a; left). Furthermore, to design against off-pathway
interactions between systems, a single global crosstalk tube is
specified (Figure 5a; right). In the global crosstalk tube, the
on-target complexes correspond to all reactive species gen-
erated during all elementary steps (
m
= 0
,
1) for all sys-
tems (
n
= 1
, . . . , N
); the off-target complexes correspond
to non-cognate interactions between these reactive species.
Crucially, the global crosstalk tube ensemble omits the cog-
nate products that the reactive species are intended to form
(they appear as neither on-targets nor off-targets). Hence, all
reactive species in the global crosstalk tube are forced to ei-
ther perform no reaction (remaining as desired on-targets) or
undergo a crosstalk reaction (forming undesired off-targets),
providing the basis for minimization of global crosstalk during
sequence optimization.
Sequence design is performed subject to complementarity
constraints inherent to the reaction pathway (Figure 2a; do-
main “d” complementary to “d*”, etc.), as well as to biolog-
ical sequence constraints imposed by the the silencing tar-
get Y (mRFP or sfGFP), the protein effector (dCas9), or
the synthetic terminator; see the constraint shading in Fig-
ure 5a). The sequence is optimized by reducing the ensemble
defect quantifying the average fraction of incorrectly paired
nucleotides over the multi-tube ensemble.
31,34,35
Within the
ensemble defect, defect weights were applied to prioritize de-
sign effort.
31
Optimization of the ensemble defect implements
both a positive design paradigm, explicitly designing for on-
pathway elementary steps, and a negative-design paradigm,
explicitly designing against off-pathway crosstalk.
31
Figure 5b displays the Reactants and Products tubes for a
completed sequence design (cgRNAs of Figure 2). For cgRNA
A (left panel), on-target complexes are predicted to form
with quantitative yield at the target concentrations, but with
some unintended base-pairing (nucleotides not shaded dark
red). These structural defects within the ensemble of on-
target complexes reflect the real-world challenges of design-
ing a cgRNA that satisfies biological sequence constraints,
changes conformation in response to a cognate RNA trigger,
and operates orthogonally to a library of other cgRNAs. For
the corresponding library of orthogonal cgRNAs (A, B, C),
each cgRNA is predicted to interact appreciably only with its
cognate RNA trigger (right panel).
Conceptual Opportunities for cgRNA-Enabled Biological
Research Tools, Therapeutics, and Synthetic Biology.
The
ability to rationally design cgRNAs suggests a conceptual
framework for enabling biologists to exert spatiotemporal
control over regulatory perturbations in living organisms us-
ing CRISPR/Cas technology. In principle, Cas activity could
be restricted to a desired cell type, tissue, or organ by select-
ing an endogenous RNA trigger X with the desired spatial and
temporal expression profile. To shift conditional regulation to
a different tissue or developmental stage, the cgRNA would
be reprogrammed to recognize a different trigger sequence.
Signal transduction with cgRNAs would also have attractive
therapeutic potential, with trigger X as a programmable dis-
ease marker and target Y as an independent programmable
therapeutic target, enabling selective treatment of diseased
cells. Synthetic biology provides another attractive arena
for use of cgRNAs. Traditional synthetic biology regulators
have relied on protein:protein and protein:DNA interactions
mined from existing genomes, placing limits on scalability due
to crosstalk and the limited number of available regulators.
4
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint
.
http://dx.doi.org/10.1101/525857
doi:
bioRxiv preprint first posted online Jan. 21, 2019;
Sequence design
Equilibrium probability
1.0
0.8
0.6
0.4
0.2
0.0
Reactants
Global Crosstalk
(All systems, 10nM)
(Step 1)
(Step 0)
Products
a
cgRNA-A
10nM
Trigger-A
10nM
cgRNA-A:trigger-X
A
10nM
Sequence analysis
b
Constrained by target gene
cgRNA-A
10.0nM
Trigger-X
A
10.0nM
cgRNA-A:trigger-X
A
10.0nM
Reactants
Products
Computational orthogonality
Constrained by dCas9
Constrained by synthetic terminator
Designed sequence for cgRNA A & trigger X
A
Designed sequence for cgRNA B & trigger X
B
Designed sequence for cgRNA C & trigger X
C
X
C
X
A
X
B
X
C
X
A
X
B
X
C
X
A
X
B
Trigger
cgRNA
A
B
C
0
5
10
[cgRNA:trigger] (nM)
Figure 5.
Computational cgRNA sequence design using NU-
PACK.
30,31
(a) Target test tubes for design of 3 orthogonal cgR-
NAs A, B, C (terminator switch mechanism for Figure 2). Left:
Elementary step tubes. Reactants tube (Step 0): cgRNA and trig-
ger. Products tube (Step 1): cgRNA:trigger. Each target test
tube contains a set of desired “on-target” complexes (each with
the depicted target secondary structure and a target concentration
of 10 nM) corresponding to the on-pathway hybridization prod-
ucts for a given step and a set of undesired “off-target” complexes
(all complexes of up to 2 strands, each with a target concentra-
tion of 0 nM; not depicted) corresponding to on-pathway reactants
and off-pathway hybridization crosstalk for a given step. To de-
sign 3 orthogonal systems, there are two elementary step tubes
for each system A, B, C. Right: Global crosstalk tube. Contains
the depicted on-target complexes corresponding to reactive species
generated during Steps 0 and 1 (each with the depicted target sec-
ondary structure and a target concentration of 10 nM) as well as
off-target complexes (all complexes of up to 2 strands, each with a
target concentration of 0 nM; not depicted) corresponding to off-
pathway interactions between these reactive species. To design 3
orthogonal systems, the global crosstalk tube contains a set of on-
targets and off-targets for each system A, B, C. (b) Analysis of de-
sign quality.
30,32
Left: Tubes depict the target structure and pre-
dicted concentration for each on-target complex with nucleotides
shaded to indicate the probability of adopting the depicted base-
pairing state at equilibrium. For this design, all on-targets are
predicted to form with quantitative yield at the 10 nM target con-
centration but some nucleotides have unwanted base-pairing inter-
actions (nucleotides not shaded dark red). Right: Computational
orthogonality study. Predicted equilibrium concentration of each
cgRNA:trigger complex for the 3 orthogonal systems of Figure 2
(one cgRNA species and one RNA trigger species per tube). RNA
at 37
C in 1M Na
+
.
33
cgRNA regulators offer a promising platform for scalable syn-
thetic biology.
Comparison of cgRNAs to other scRNAs.
It is interesting
to compare the present work engineering cgRNAs (a partic-
ular class of scRNAs with notable properties) to the scR-
NAs we previously engineered toward the goal of conditional
RNA interference (RNAi).
1,2
In both cases, the scRNAs are
intended to perform signal transduction between detection
of a programmable RNA input and production of a biologi-
cally active programmable output. In the case of conditional
RNAi, the scRNAs detect an mRNA input X and interact
to produce a substrate that is processed by Dicer to pro-
duce an siRNA output targeting independent silencing target
mRNA Y for destruction. Because Dicer substrates are struc-
turally simple, comprising predominantly a duplex containing
the target-binding sequence,
36
signal transduction between
X and Y and inactive/active states is performed by scRNAs
upstream of formation of the biologically active Dicer sub-
strate. For example, the simplest mechanism we have devised
to date involves a dimer scRNA that conditionally generates
a monomer Dicer substrate anti-Y upon detection of mRNA
X.
1,2
By contrast, the standard gRNAs that serve as sub-
strates for Cas9 protein effectors are not only structurally
more complex than Dicer substrates (involving multiple du-
plexes, loops, and tails), but Cas9 also appears to be more
permissive of modifications to the standard structure, pro-
viding hooks for engineering programmable conditional reg-
ulation. As a result, it is possible to do signal transduc-
tion between X and Y and inactive/active states (for either
ON
OFF or OFF
ON logic) all within a single cgRNA (i.e.,
a single monomer scRNA). A benefit of this mechanistic sim-
plicity is that monomer cgRNAs can be readily expressed,
while expression of well-formed multimer scRNAs such as
those developed for conditional Dicer substrate formation ap-
pears more challenging, possibly necessitating delivery with
chemical reagents.
CONCLUSIONS
The present work represents only a first step toward our long-
term goal of engineering programmable conditional regula-
tors that function robustly in living organisms. Here, we
made progress on several fronts: 1) demonstrating cgRNA
mechanisms
in vivo
in bacteria for both constitutively ac-
tive cgRNAs that are inactivated by a cognate RNA trig-
ger (ON
OFF logic) and constitutively inactive cgRNAs
that are conditionally activated by a cognate RNA trigger
(OFF
ON logic), 2) demonstrating the applicability and ver-
satility of dynamic RNA nanotechnology in living cells, and
3) demonstrating automated sequence design using NUPACK
to engineer libraries of orthogonal cgRNA/trigger pairs.
In order to develop cgRNAs into a versatile platform for
biological research, a number of major improvements are
needed. First, standard gRNAs routinely achieve two orders
of magnitude in regulatory dynamic range,
4
and it is desir-
able to engineer improved cgRNA mechanisms that exploit
this full dynamic range. Toward this end, further understand-
ing of the structure/function relationships between cgRNAs,
triggers, and Cas effectors are needed to ascertain how to ro-
bustly achieve both a strong ON state and a clean OFF state
depending on the presence/absence of the cognate trigger.
Second, to enable tissue-selective regulation in living organ-
isms, it is critical that cgRNAs are able to efficiently detect
a trigger that is a subsequence of a longer endogenous RNA
(e.g., a subsequence of an mRNA). Detection of a subsequence
of a full-length mRNA poses significant additional challenges
relative to detection of a short RNA trigger,
2,29
increasing
the degree of difficulty in achieving a conditional response
5
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint
.
http://dx.doi.org/10.1101/525857
doi:
bioRxiv preprint first posted online Jan. 21, 2019;
that exploits the full dynamic range. Third, in common with
the terminator switch and splinted switch mechanisms stud-
ied here (but unlike the toehold switch mechanisms studied
here and elsewhere
29
), it is important that cgRNA regula-
tors be allosteric, so that the sequence of the target gene Y
places no restriction on the sequence of the RNA trigger X,
enabling independent control over the regulatory scope (us-
ing X) and the regulatory target (using Y). Significant effort
and innovation are needed to achieve these goals and develop
cgRNAs that operate as plug-and-play programmable con-
ditional regulators within endogenous biological circuits in
living organisms.
METHODS SUMMARY
For each mechanism, orthogonal cgRNA/trigger pairs were
designed using the reaction pathway engineering tools
within NUPACK (
nupack.org
).
30,31
Control gRNA and
cgRNA/trigger constructs were generated by inserting se-
quence modifications into a previously described pgRNA-
bacteria vector.
4
A modified
E. coli
MG1655 strain contain-
ing genomically incorporated mRFP and sfGFP
4
was used
for all fluorescence assays. Sequence verified strains were
grown overnight in EZ-RDM (Teknova) containing carbeni-
cillin and chloramphenicol and seeded at 100
×
dilution in
fresh medium and grown to midlog phase (
4 h), then further
diluted
100
×
to normalize cell density with fresh medium
containing aTc for induction of Cas9 expression (and 5mM
IPTG for terminator switch experiments only). Induced cells
were grown at 37
C with continuous shaking for 12 h, with
end-point fluorescence measured via flow cytometry (20,000
live cell counts per well).
ASSOCIATED CONTENT
Supporting Information.
Methods, sequences, plasmids, schemat-
ics, flow cytometry replicates. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
niles@caltech.edu
Author Contributions
§
M.H.H.-H. and Z.C. contributed equally.
Notes
The authors declare competing financial interests in the form of
pending patents.
ACKNOWLEDGEMENTS
We thank S. Qi for the gift of plasmids and the gift of
E. coli
ex-
pressing mRFP and sfGFP, N.J. Porubsky for assistance with re-
action pathway engineering using NUPACK, A. Hou and J. Kishi
for performing preliminary studies, and R. Phillips for discussions
on allosteric regulation. This work was funded by the Defense Ad-
vanced Research Projects Agency (HR0011-17-2-0008; the findings
are those of the authors and should not be interpreted as represent-
ing the official views or policies of the US Government), by the Cal-
tech Center for Environmental Microbial Interactions (CEMI), by
the National Institutes of Health (5T32GM112592), by the Rosen
Bioengineering Center at Caltech, by the National Science Foun-
dation Molecular Programming Project (NSF-CCF-1317694), by a
Professorial Fellowship at Balliol College (University of Oxford),
and by the Eastman Visiting Professorship at the University of
Oxford.
REFERENCES
(1) Hochrein, L. M., Schwarzkopf, M., Shahgholi, M., Yin, P.,
and Pierce, N. A. (2013) Conditional Dicer substrate forma-
tion via shape and sequence transduction with small condi-
tional RNAs.
J. Am. Chem. Soc.
135, 17322–17330.
(2) Hochrein, L. M., Ge, T. J., Schwarzkopf, M., and Pierce,
N. A. (2018) Signal transduction in human cell lysate via
dynamic RNA nanotechnology.
ACS Synth. Biol.
7, 2796–
2802.
(3) Knott, G. J., and Doudna, J. A. (2018) CRISPR-Cas guides
the future of genetic engineering.
Science
361, 866.
(4) Qi, L. S., Larson, M. H., Gilbert, L. A., Doudna, J. A., Weiss-
man, J. S., Arkin, A. P., and Lim, W. A. (2013) Repurpos-
ing CRISPR as an RNA-guided platform for sequence-specific
control of gene expression.
Cell
152, 1173–1183.
(5) Gilbert, L. A., Larson, M. H., Morsut, L., Liu, Z., Brar, G. A.,
Torres, S. E., Stern-Ginossar, N., Brandman, O., Whitehead,
E. H., Doudna, J. A., Lim, W. A., Weissman, J. S., and Qi,
L. S. (2013) CRISPR-mediated modular RNA-guided regula-
tion of transcription in eukaryotes.
Cell
154, 442–451.
(6) Cong, L., Ran, F. A., Cox, D., Lin, S., Barretto, R., Habib,
N., Hsu, P. D., Wu, X., Jiang, W., Marraffini, L. A.,
and Zhang, F. (2013) Multiplex genome engineering using
CRISPR/Cas systems.
Science
, 819–823.
(7) Mali, P., Yang, L., Esvelt, K. M., Aach, J., Guell, M., Di-
Carlo, J. E., Norville, J. E., and Church, G. M. (2013) RNA-
guided human genome engineering via Cas9.
Science
339,
823.
(8) Zetsche, B., Heidenreich, M., Mohanraju, P., Fedorova, I.,
Kneppers, J., DeGennaro, E. M., Winblad, N., Choudhury,
S. R., Abudayyeh, O. O., Gootenberg, J. S., Wu, W. Y.,
Scott, D. A., Severinov, K., van der Oost, J., and Zhang,
F. (2016) Multiplex gene editing by CRISPR–Cpf1 using a
single crRNA array.
Nat. Biotechnol.
35, 31.
(9) Larson, M. H., Gilbert, L. A., Wang, X., Lim, W. A.,
Weissman, J. S., and Qi, L. S. (2013) CRISPR interference
(CRISPRi) for sequence-specific control of gene expression.
Nat. Protoc.
8, 2180–2196.
(10) Chen, B., Gilbert, L. A., Cimini, B. A., Schnitzbauer, J.,
Zhang, W., Li, G.-W., Park, J., Blackburn, E. H., Weiss-
man, J. S., Qi, L. S., and Huang, B. (2013) Dynamic imag-
ing of genomic loci in living human cells by an optimized
CRISPR/Cas system.
Cell
155, 1479–1491.
(11) Knight, S. C., Tjian, R., and Doudna, J. A. (2017) Genomes
in focus: Development and applications of CRISPR-Cas9
imaging technologies.
Angew. Chem. Int. Ed.
57, 4329–4337.
(12) Hilton, I. B., D’Ippolito, A. M., Vockley, C. M., Thakore,
P. I., Crawford, G. E., Reddy, T. E., and Gersbach, C. A.
(2015) Epigenome editing by a CRISPR-Cas9-based acetyl-
transferase activates genes from promoters and enhancers.
Nat. Biotechnol.
33, 510.
(13) Gaudelli, N. M., Komor, A. C., Rees, H. A., Packer, M. S.,
Badran, A. H., Bryson, D. I., and Liu, D. R. (2017) Pro-
grammable base editing of A*T to G*C in genomic DNA
without DNA cleavage.
Nature
551, 464.
(14) Aubrey, B. J., Kelly, G. L., Kueh, A. J., Brennan, M. S.,
O’Connor, L., Milla, L., Wilcox, S., Tai, L., Strasser, A.,
and Herold, M. J. (2015) An inducible lentiviral guide RNA
platform enables the identification of tumor-essential genes
and tumor-promoting mutations in vivo.
Cell Rep.
10, 1422–
1432.
(15) Bertero, A., Pawlowski, M., Ortmann, D., Snijders, K., Yian-
gou, L., Cardoso de Brito, M., Brown, S., Bernard, W. G.,
Cooper, J. D., Giacomelli, E., Gambardella, L., Hannan, N.
R. F., Iyer, D., Sampaziotis, F., Serrano, F., Zonneveld, M.
C. F., Sinha, S., Kotter, M., and Vallier, L. (2016) Optimized
inducible shRNA and CRISPR/Cas9 platforms for in vitro
studies of human development using hPSCs.
Development
143, 4405.
(16) Chen, T., Gao, D., Zhang, R., Zeng, G., Yan, H., Lim, E., and
Liang, F.-S. (2017) Chemically controlled epigenome editing
through an inducible dCas9 system.
J. Am. Chem. Soc.
139,
11337–11340.
(17) Nihongaki, Y., Otabe, T., and Sato, M. (2018) Emerging ap-
proaches for spatiotemporal control of targeted genome with
inducible CRISPR-Cas9.
Anal. Chem.
90, 429–439.
(18) Shen, Z., Zhang, X., Chai, Y., Zhu, Z., Yi, P., Feng, G.,
Li, W., and Ou, G. (2014) Conditional knockouts generated
by engineered CRISPR-Cas9 endonuclease reveal the roles of
6
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint
.
http://dx.doi.org/10.1101/525857
doi:
bioRxiv preprint first posted online Jan. 21, 2019;
coronin in C. elegans neural development.
Dev. Cell
30, 625–
636.
(19) Ablain, J., Durand, E. M., Yang, S., Zhou, Y., and Zon, L. I.
(2015) A CRISPR/Cas9 vector system for tissue-specific gene
disruption in zebrafish.
Dev. Cell
32, 756–764.
(20) Hirosawa, M., Fujita, Y., Parr, C. J. C., Hayashi, K.,
Kashida, S., Hotta, A., Woltjen, K., and Saito, H. (2017) Cell-
type-specific genome editing with a microRNA-responsive
CRISPR–Cas9 switch.
Nucleic Acids Res.
45, e118.
(21) Briner, A. E., Donohoue, P.D., Gomaa, A.A., Selle, K., Slo-
rach, E.M., Nye, C.H., Haurwitz, R.E., Beisel, C.L., May,
A.P., and Barrangou, R. (2014) Guide RNA functional mod-
ules direct Cas9 activity and orthogonality.
Mol. Cell
56,
333–339.
(22) Nowak, C. M., Lawson, S., Zerez, M., and Bleris, L. (2016)
Guide RNA engineering for versatile Cas9 functionality.
Nu-
cleic Acids Res.
44, 9555–9564.
(23) Liu, Y., Zhan, Y., Chen, Z., He, A., Li, J., Wu, H., Liu,
L., Zhuang, C., Lin, J., Guo, X., Zhang, Q., Huang, W., and
Cai, Z. (2016) Directing cellular information flow via CRISPR
signal conductors.
Nat. Methods
13, 938.
(24) Tang, W., Hu, J. H., and Liu, D. R. (2017) Aptazyme-
embedded guide RNAs enable ligand-responsive genome edit-
ing and transcriptional activation.
Nat. Commun.
8, 15939.
(25) Lee, Y. J., Hoynes-O’Connor, A., Leong, M. C., and Moon,
T. S. (2016) Programmable control of bacterial gene expres-
sion with the combined CRISPR and antisense RNA system.
Nucleic Acids Res.
44, 2462–2473.
(26) Ferry, Q. R. V., Lyutova, R., and Fulga, T. A. (2017) Rational
design of inducible CRISPR guide RNAs for de novo assembly
of transcriptional programs.
Nat. Commun.
8, 2109.
(27) M ̈uckl, A., Schwarz-Schilling, M., Fischer, K., and Sim-
mel, F. C. (2018) Filamentation and restoration of nor-
mal growth in Escherichia coli using a combined CRISPRi
sgRNA/antisense RNA approach.
PLOS ONE
13, e0198058.
(28) Jain, P. K., Ramanan, V., Schepers, A. G., Dalvie, N. S.,
Panda, A., Fleming, H. E., and Bhatia, S. N. (2016) De-
velopment of light-activated CRISPR using guide RNAs with
photocleavable protectors.
Angew. Chem. Int. Ed.
55, 12440–
12444.
(29) Siu, K.-H., and Chen, W. (2018) Riboregulated toehold-
gated gRNA for programmable CRISPR–Cas9 function.
Nat.
Chem. Biol.
10.1038/s41589–018–0186–1.
(30) Zadeh, J. N., Steenberg, C. D., Bois, J. S., Wolfe, B. R.,
Pierce, M. B., Khan, A. R., Dirks, R. M., and Pierce, N. A.
(2011) NUPACK: Analysis and design of nucleic acid systems.
J. Comput. Chem.
32, 170–173.
(31) Wolfe, B. R., Porubsky, N. J., Zadeh, J. N., Dirks, R. M., and
Pierce, N. A. (2017) Constrained multistate sequence design
for nucleic acid reaction pathway engineering.
J. Am. Chem.
Soc.
139, 3134–3144.
(32) Dirks, R. M., Bois, J. S., Schaeffer, J. M., Winfree, E., and
Pierce, N. A. (2007) Thermodynamic analysis of interacting
nucleic acid strands.
SIAM Rev.
49, 65–88.
(33) Serra, M. J., and Turner, D. H. (1995) Predicting thermody-
namic properties of RNA.
Methods Enzymol.
259, 242–261.
(34) Zadeh, J. N., Wolfe, B. R., and Pierce, N. A. (2011) Nucleic
acid sequence design via efficient ensemble defect optimiza-
tion.
J. Comput. Chem.
32, 439–452.
(35) Wolfe, B. R., and Pierce, N. A. (2015) Sequence design for
a test tube of interacting nucleic acid strands.
ACS Synth.
Biol.
4, 1086–1100.
(36) Kim, D. H., and Rossi, J. J. (2007) Strategies for silencing
human disease using RNA interference.
Nat. Rev. Genet.
8,
173–184.
7
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint
.
http://dx.doi.org/10.1101/525857
doi:
bioRxiv preprint first posted online Jan. 21, 2019;