Article
Synthetic protein circuits for programmable control
of mammalian cell death
Graphical abstract
Highlights
d
Synthetic synpoptosis protein circuits proteolytically control
apoptosis and pyroptosis
d
Circuits override cell death biases and induce mixed death
modes in populations
d
Synpoptosis circuits can be coupled to inputs to enable cell-
state-specific activation
d
Engineered sender cells can kill other cells through
intercellular circuit transmission
Authors
Shiyu Xia, AndrewC. Lu,VictoriaTobin,...,
Margaret Sui, Felix Horns,
Michael B. Elowitz
Correspondence
melowitz@caltech.edu
In brief
Naturally inspired protein engineering
produces synpoptosis circuits that
programmably control user-selectable
death programs in target
mammalian cells.
Xia et al., 2024, Cell
187
, 2785–2800
May 23, 2024
ª
2024 The Author(s). Published by Elsevier Inc.
https://doi.org/10.1016/j.cell.2024.03.031
ll
Article
Synthetic protein circuits for programmable
control of mammalian cell death
Shiyu Xia,
1,2
Andrew C. Lu,
1,2,3,6
Victoria Tobin,
1,2,4,6
Kaiwen Luo,
1,2,6
Lukas Moeller,
1,2
D. Judy Shon,
1,2
Rongrong Du,
1,2
James M. Linton,
1,2
Margaret Sui,
1,2,5
Felix Horns,
1,2
and Michael B. Elowitz
1,2,7,
*
1
Division of Biology and Biological Engineering, California Institute of Technology, Pasadena, CA 91125, USA
2
Howard Hughes Medical Institute, California Institute of Technology, Pasadena, CA 91125, USA
3
UCLA-Caltech Medical Scientist Training Program, University of California, Los Angeles, CA 90095, USA
4
UC Davis-Caltech Veterinary Scientist Training Program, University of California, Davis, CA 95616, USA
5
Present address: Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm 17165, Sweden
6
These authors contributed equally
7
Lead contact
*Correspondence:
melowitz@caltech.edu
https://doi.org/10.1016/j.cell.2024.03.031
SUMMARY
Natural cell death pathways such as apoptosis and pyroptosis play dual roles: they eliminate harmful cells
and modulate the immune system by dampening or stimulating inflammation. Synthetic protein circuits
capable of triggering specific death programs in target cells could similarly remove harmful cells while appro-
priately modulating immune responses. However, cells actively influence their death modes in response to
natural signals, making it challenging to control death modes. Here, we introduce naturally inspired ‘‘synpop-
tosis’’ circuits that proteolytically regulate engineered executioner proteins and mammalian cell death. These
circuits direct cell death modes, respond to combinations of protease inputs, and selectively eliminate target
cells. Furthermore, synpoptosis circuits can be transmitted intercellularly, offering a foundation for engineer-
ing synthetic killer cells that induce desired death programs in target cells without self-destruction. Together,
these results lay the groundwork for programmable control of mammalian cell death.
INTRODUCTION
Mammalian systems use distinct cell death programs to eliminate
harmful cells and shape immunity.
1–4
Apoptosis is immunologi-
cally silent or ‘‘cold.’’
5
,
6
By contrast, pyroptosis is immunologi-
cally ‘‘hot’’ and involves the substantial release of damage-asso-
ciated molecular patterns (DAMPs).
7–18
Apoptosis and pyroptosis can each be advantageous, de-
pending on immunological context. The immunostimulatory na-
ture of pyroptosis can promote cell killing. For example, inducing
pyroptosis in a 15% minority of cells was sufficient to clear an
entire tumor by boosting anti-tumor immunity.
19
Consistently,
expression of the gasdermin (GSDM) family of pore-forming pro-
teins, the executioners of pyroptosis, positively correlates with
cancer patient survival, and cytotoxic lymphocytes upregulate
GSDM expression in cancer cells.
20
To escape pyroptosis, can-
cer cells generate loss-of-function GSDM mutations, silence
GSDM expression, and express non-pyroptotic GSDM vari-
ants.
21–24
Although beneficial in some contexts, pyroptosis can
lead to pathological inflammation if triggered excessively.
25
Therefore, it would be desirable to controllably induce apoptosis
or pyroptosis and tune their relative frequencies.
Existing cell-killing approaches cannot fully direct the mode of
cell death. Cytotoxic drugs are often limited to triggering
apoptosis in cold tumors.
26
,
27
CAR-T cells can effectively target
cells expressing either a single antigen or multiple antigens.
28–31
However, to kill target cells, CAR-T cells typically use gran-
zymes, which may induce either apoptosis or pyroptosis.
7
Gran-
zyme-independent approaches have also been attempted,
including engineered TRAIL-presenting cells
32
and synthetic cir-
cuits that regulate caspases, BID, or BAX.
33–35
However, these
approaches are similarly restricted to induction of apoptosis.
To enable tailored control of cell death, we need a set of syn-
thetic circuits with the following features: first, the circuits should
allow the activation and repression of both apoptosis and pyrop-
tosis. Second, they should steer the mode of cell death in various
cell contexts. Third, they should allow the integration and compu-
tation of multiple input signals. Fourth, they should be able to
selectively kill target cells. Fifth, they should support cell-cell
transmission, offering the potential to engineer synthetic killer
cells that use designed death programs to eliminate other cells.
Here, we present synthetic protein-level cell death circuits,
collectively termed ‘‘synpoptosis’’ circuits, that demonstrate
the above features. To engineer these circuits, we took inspira-
tion from natural cell death pathways that use regulated proteol-
ysis along with protein-level caging and degradation mecha-
nisms. Synpoptosis circuits provide a foundation for rationally
designed, programmable control of mammalian cell death.
ll
OPEN ACCESS
Cell
187
, 2785–2800, May 23, 2024
ª
2024 The Author(s). Published by Elsevier Inc.
2785
This is an open access article under the CC BY license (
http://creativecommons.org/licenses/by/4.0/
).
AB
CD
E
(legend on next page)
ll
OPEN ACCESS
2786
Cell
187
, 2785–2800, May 23, 2024
Article
RESULTS
Synpoptosis circuits control user-selectable cell death
programs
Due to the inherent tendency of cells toward apoptosis or pyrop-
tosis, natural inducers do not completely direct the mode of cell
death. For example, when cytotoxic lymphocytes deliver gran-
zymes into target cells, target cells lacking GSDMs undergo
apoptosis, whereas those expressing functional GSDMs un-
dergo pyroptosis.
20
,
21
This lack of control allows target cells to
activate undesirable death programs that favor their own survival
or trigger systemic toxicity.
7
To design synpoptosis circuits capable of steering the mode of
cell death (
Figure 1
A), we drew inspiration from naturally occur-
ring cell death programs. Caspases, as proteases, contribute to
apoptosis by activating the caspase-activated DNase and the
XKR8 phospholipid scramblase, among other pro-apoptotic fac-
tors.
36–39
Caspases can also cleave GSDMs for pore formation in
pyroptosis.
7
,
9
Additionally, in synthetic biology, proteases repre-
sent a ‘‘common currency,’’ as they can be engineered to
perform various signal-processing tasks.
40–44
Therefore, prote-
olysis is a promising mode of synthetic regulation of cell death
(
Figures 1
B–1D).
In most experiments, we used human embryonic kidney 293
(HEK293) cells because they express low levels of endogenous
GSDMs, providing a clear background for synthetically induced
cell death (
Figure S1
A). We transiently co-transfected HEK cells
with DNA encoding the circuits, often along with Cherry, a fluo-
rescent protein marker, and analyzed cells by flow cytometry
16–24 h later. To quantify cell death, we stained the cells with An-
nexin, Sytox, or both. Annexin stains apoptotic cells by binding
to phosphatidylserine exposed on the outer leaflet and pyrop-
totic cells by binding to phosphatidylserine in the inner leaflet af-
ter membrane permeabilization.
45
By contrast, Sytox, a mem-
brane-impermeable dye, primarily stains pyroptotic cells.
46
Therefore, apoptotic cells are typically Annexin-high and Sy-
tox-low, while pyroptotic cells are Annexin-high and Sytox-
high (
Figure S1
B). For simplicity, in experiments involving only
a single death mode, we often used a single relevant dye. To
focus on cells influenced by the circuits, we calculated the frac-
tion of Annexin- or Sytox-positive cells after gating on the fluo-
rescent co-transfection marker (
Figure S1
C; see
STAR
Methods
).
Previous work showed that the tobacco etch virus protease
(TEVP) could activate a modified caspase-3 whose natural
cleavage site between its large and small subunits was replaced
by a TEVP cleavage site.
33
Using Annexin staining, we verified
the ability of the modified caspase-3 to kill HEK cells when co-
transfected with TEVP (
Figure 1
C, first module). Further, the cells
only modestly took up Sytox, suggesting apoptosis (
Figure 1
E).
To enable more complex functions for control of apoptosis
downstream, we engineered additional caspase-3 variants that
can be modulated by TEVP cleavage. We first fused the large
and small subunits of caspase-3 to heterodimerizing leucine zip-
pers to produce a constitutively active non-covalent complex.
47
Then, we appended a TEVP-removable dihydrofolate reductase
degron to the small subunit. This degron is constitutively
active,
48
substantially reducing caspase-3 activity in the
absence of TEVP. When TEVP was expressed, proteolytic
removal of the degron efficiently activated apoptosis (
Figure 1
C,
second module). To enable deactivation of apoptosis through
proteolysis, we fused a TEVP site-caged N-degron
49
,
50
to the
large subunit of caspase-3. The caged N-degron is inactive
because its N-terminal destabilizing residue, a tyrosine, is pre-
ceded by the TEVP site sequence. Upon TEVP cleavage, the de-
stabilizing residue is exposed, activating the N-degron. As such,
TEVP cleavage repressed apoptosis (
Figure 1
C, third module).
Next, we sought to engineer proteolytic control of pyroptosis.
We inserted the cleavage sites of TEVP, tobacco vein mottling vi-
rus protease (TVMVP), and hepatitis C virus protease (HCVP) into
a linker region between the N-terminal (pore-forming) and C-ter-
minal (auto-inhibitory) domains of three mammalian GSDMs.
Each engineered GSDM effectively triggered pyroptosis in HEK
cells in the presence of the cognate protease, as shown by Sytox
staining (
Figures 1
D first module, 1E, and
S1
D). We focused on
GSDMA in subsequent experiments because it is thought to be
orthogonal to endogenous host pathways.
51–53
As with caspase-3, we designed additional GSDM variants
that allow positive and negative control over pyroptosis. GSDM
structures showed that the N termini of GSDMs need to be
accessible to lipids for pore formation.
54–58
Therefore, we caged
the activity of the GSDMA N-terminal domain by blocking its N
terminus with a bulky maltose-binding protein. Insertion of a
TEVP cleavage site allowed proteolytic removal of the bulky
tag to restore pore-forming activity (
Figure 1
D, second module).
In a complementary approach, we fused a constitutively active
degron to the GSDMA N-terminal domain to lower its activity.
In this configuration, removal of the degron by TEVP also
induced pyroptosis (
Figure 1
D, third module). To proteolytically
switch off GSDM activity, we initially mimicked a natural mecha-
nism in which the GSDMD N-terminal domain is inactivated
through cleavage by caspase-3.
59
,
60
However, inserting the
Figure 1. Synpoptosis circuits control user-selectable cell death programs
(A) Synthetic cell death synpoptosis circuits steer the mode of cell death by operating orthogonally to cell-intrinsic death programs.
(B) Molecular building blocks of synpoptosis circuits include caspase-3 subunits, gasdermin domains, viral proteases such as TEVP, degrons, malto
se-binding
protein, and leucine zippers.
(C) Synthetic apoptosis modules use viral proteases, such as TEVP, to activate or repress engineered variants of caspase-3. We transiently transfec
ted HEK cells
with plasmid DNA encoding the synthetic modules and then quantified cell death by staining and flow cytometry. Throughout the figure, the gray window ind
icates
the fractions established by negative and positive transient DNA transfection controls; dots represent biological replicates (distinct culture w
ells).
(D) Synthetic pyroptosis modules similarly use TEVP to regulate engineered GSDMA.
(E) Apoptosis and pyroptosis exhibit different staining patterns with Annexin and Sytox. Annexin stains both apoptotic and pyroptotic cells, while
Sytox primarily
stains pyroptotic cells. We calculated the fraction of cells that stained positive for each dye after gating on the transfected cells based on fluoresc
ence of a co-
transfected marker. Data represent three independent experiments.
See also
Figure S1
.
ll
OPEN ACCESS
Cell
187
, 2785–2800, May 23, 2024
2787
Article
AB
CDE
FGH
(legend on next page)
ll
OPEN ACCESS
2788
Cell
187
, 2785–2800, May 23, 2024
Article
TEVP cleavage site at an equivalent location within GSDMA
abolished its activity (
Figure 1
D, fourth module). As an alternative
strategy, we fused the GSDMA N-terminal domain to a TEVP
site-caged N-degron. This design allowed protease-mediated
suppression of pyroptosis (
Figure 1
D, fifth module). We note
that cells also naturally use loop architectures to downregulate
GSDM activity.
61–63
These loops are at present challenging to
construct synthetically but motivate future synpoptosis circuit
designs that fine-tune the penetrance of cell death.
Together, these results demonstrate that synthetic regulated
proteolysis circuit modules can bidirectionally control apoptosis
and pyroptosis.
Synpoptosis circuits lead to canonical features of
cell death
Cell death induced by synpoptosis circuits exhibited Annexin
and Sytox staining patterns indicative of apoptosis and pyropto-
sis. To further characterize circuit-induced cell death, we looked
for additional canonical features associated with naturally occur-
ring cell death.
Drugs and death ligands typically induce cell death asynchro-
nously within a cell population.
64–66
To assess whether synpop-
tosis is similarly asynchronous, we transfected HEK cells with
TEVP-activated caspase-3 and GSDMA circuits and then dou-
ble-stained the cells with Annexin and Sytox for flow cytometry
analysis over a time span. Both synthetic apoptosis and pyrop-
tosis exhibited asynchrony (
Figures 2
A and 2B). In the case of
apoptosis, Annexin-positivity preceded Sytox-positivity, indi-
cating a sequence of apoptosis followed by eventual cell lysis
(
Figure 2
A). By contrast, the pyroptotic cells displayed a simulta-
neous increase in Annexin and Sytox signals (
Figure 2
B). Similar
dynamic patterns were observed without gating on the co-trans-
fected Cherry, demonstrating robust circuit effects on a popula-
tion level (
Figures 2
A and 2B). Because the apoptotic cells even-
tually progress to cell lysis, we confined our flow cytometry
analysis to a window between 16 and 24 h post-transfection to
distinguish between apoptosis and pyroptosis (
Figure S1
B).
Next, considering that naturally occurring cell death is influ-
enced by the concentration of executioner molecules,
67
we
investigated whether synpoptosis is concentration-sensitive.
Indeed, engineered auto-inhibited caspase-3 and GSDMA,
when expressed at high levels indicated by co-transfected Cher-
ry, induced noticeable cell death even without the activating
TEVP (
Figure S2
A). Further, titrating the amount of plasmid
DNA encoding the synpoptosis circuits enabled DNA dose-
dependent control over killing fractions (
Figure S2
B). At a
constant amount of plasmid DNA, tuning could achieved by
adjusting mRNA dosage (
Figure S2
C) through a synthetic
miRNA-based incoherent feedforward loop (IFFL) motif,
68
,
69
which allows DNA dosage-independent control of protein
expression.
70
,
71
Then, because naturally occurring cell death can be modu-
lated by small-molecule compounds, we investigated the effect
of Q-VD-OPh, a caspase inhibitor, on synpoptosis circuits (
Fig-
ure 2
C). In line with expectations, Q-VD-OPh markedly attenu-
ated apoptosis induced by the caspase-3 circuit but did not
affect pyroptosis mediated by the GSDMA circuit. We also asked
whether cells treated with synpoptosis circuits stain positive for
TO-PRO-3, a dye that enters apoptotic cells through pannexin
channels and pyroptotic cells through permeabilized mem-
branes. As anticipated, cells treated with either circuit showed
positive staining for TO-PRO-3 (
Figure 2
D).
A physiologically important feature of pyroptosis is the release
of pro-inflammatory cytokines, notably the interleukin (IL)-1 fam-
ily including IL-1
b
and IL-18.
72–74
For demonstration purposes,
we generated a HEK cell line that stably expresses IL-1
b
and
IL-18. We observed an obvious increase in supernatant levels
of IL-1
b
and IL-18 in cells treated with the GSDMA circuit,
compared with cells treated with the caspase-3 circuit or the
mock transfection control (
Figure 2
E).
Furthermore, we examined the morphological characteristics
of cells treated with synpoptosis circuits (
Figure S2
D). We
analyzed cell shapes by phase-contrast microscopy. Mock-
transfected cells displayed a mostly flat and extended appear-
ance. By contrast, cells transfected with the caspase-3 circuit
showed extensive grape-like membrane blebbing and cell
shrinkage, characteristic of apoptosis. On the other hand, cells
transfected with the GSDMA circuit showed cell rounding and
parachute-like membrane swelling, indicative of pyroptosis.
Additionally, the nuclear-localized H2B-Cherry marker allowed
us to visualize nuclear fragmentation in cells treated with the
caspase-3 circuit.
Lastly, mRNA represents a rapidly growing modality for tran-
sient therapeutic protein expression.
75
,
76
To determine whether
the mRNA versions of the synpoptosis circuits could generate
the expected killing effects, we generated mRNA by
in vitro
tran-
scription, used it to transfect cells, and then read out three indi-
cators of cell death. A luciferase-based assay to quantify
Figure 2. Synpoptosis circuits lead to canonical features of cell death
(A) Transient transfection of HEK cells with plasmid DNA encoding the TEVP-activated caspase-3 circuit triggered asynchronous apoptosis, shown by
flow
cytometry. Between 16 and 24 h after transfection (gray window), Sytox remains low in apoptotic cells and therefore can reliably distinguish between
apoptotic
and pyroptotic cells. Throughout the paper, dots represent biological replicates (distinct culture wells).
(B) Similarly, the TEVP-activated GSDMA circuits triggered asynchronous pyroptosis.
(C) Q-VD-OPh suppressed circuit-induced apoptosis but not pyroptosis, shown by flow cytometry. The gray window indicates the fractions established
by
negative and positive transient DNA transfection controls.
(D) TO-PRO-3 stains cells killed by the apoptosis and pyroptosis circuits, shown by flow cytometry.
(E) The apoptosis and pyroptosis circuits differently triggered the release of IL-1
b
and IL-18 from engineered HEK cells that stably express these cytokines.
(F) Transient transfection using
in vitro
-transcribed mRNA transcripts of the synpoptosis circuits led to loss of ATP-based cell viability.
(G) The mRNA version of the pyroptosis circuit triggered more LDH release than the apoptosis circuit.
(H) The mRNA version of the pyroptosis circuit triggered more ATP release than the apoptosis circuit. The same wells were repeatedly captured across a
time
course.
See also
Figure S2
.
ll
OPEN ACCESS
Cell
187
, 2785–2800, May 23, 2024
2789
Article
adenosine triphosphate (ATP) levels in the cell culture, an indica-
tor of viable cells, revealed that mRNA-encoded synpoptosis cir-
cuits metabolically inactivated cells after killing (
Figure 2
F). The
GSDMA circuit resulted in substantial release of lactate dehydro-
genase (LDH), a pyroptotic signature (
Figure 2
G), as well as ATP,
a small-molecule DAMP (
Figure 2
H).
Together, these results demonstrate that synpoptosis circuits
induce canonical features of cell death programs and function
expectedly whether delivered as DNA or mRNA.
Synpoptosis circuits direct the mode of cell death
Instead of leaving the choice of death mode up to cells, we
sought to use synpoptosis circuits to actively direct cell death.
Specifically, we aimed to drive pyroptosis in apoptosis-prone
cells, promote apoptosis in pyroptosis-prone cells, and trigger
either death program in cells capable of both apoptosis and py-
roptosis. Because these experiments involve mixed programs,
we double-stained the cells with Sytox and Annexin to distin-
guish between apoptosis and pyroptosis. Given that Annexin la-
bels both apoptotic and pyroptotic cells, we used it as a proxy for
total cell death.
To drive pyroptosis in apoptosis-prone cells, we established a
synthetic system that mimics a GSDM-negative cell context,
with TEVP activating caspase-3 to induce apoptosis. In this
background, ectopically expressing TEVP-activatable GSDMA
converted apoptosis to pyroptosis (
Figure 3
A), suggesting that
GSDMA dominated over caspase-3. Although this dominance
provided a straightforward approach to drive pyroptosis in
GSDM-negative cells, it could be exploited by cells seeking to
evade apoptosis. Indeed, cells that express GSDME, a natural
substrate for caspase-3, undergo pyroptosis downstream of
caspase-3 activation.
26
,
27
This effect can be recapitulated by
ectopically expressing wild-type (WT) GSDME, which alone did
not induce cell death but caused pyroptosis in response to
TEVP-mediated caspase-3 activation (
Figure 3
B).
To promote apoptosis in cells expressing GSDME, we
searched for a protein-level GSDME inhibitor. Recent studies re-
vealed that GSDMB has alternatively spliced variants, with the
non-pyroptotic variants inhibiting the pyroptotic ones.
23
,
24
Moti-
vated by this
trans
inhibition mechanism, we tested a panel of
GSDME N-terminal domain mutants, including some associated
with cancer,
21
and identified I217N as defective in inducing py-
roptosis and capable of inhibiting the WT counterpart (
Fig-
ure S3
A). Then, we co-transfected this mutant along with WT
GSDME, TEVP, and TEVP-activatable caspase-3. The mutant
expectedly suppressed pyroptosis, as evidenced by reduced
Sytox uptake, while permitting caspase-3-induced apoptosis,
as read out by high Annexin levels (
Figure 3
C).
The results above demonstrate that synpoptosis circuits can
guide cells toward apoptosis or pyroptosis, irrespective of their
intrinsic preferences. However, to leverage the pro-inflamma-
tory benefits of pyroptosis without triggering excessive inflam-
mation, or conversely, the immunosuppressive benefits of
apoptosis without curbing beneficial inflammation, it would be
ideal to be able to adjust the relative levels between the two
modes of cell death. By titrating down the GSDME inhibitor,
we could attenuate pyroptosis to various degrees while still al-
lowing activated caspase-3 to induce apoptosis (
Figure S3
B).
Further quantification using the Annexin signal as a measure
of total cell death and the Sytox signal as an indicator of pyrop-
tosis revealed the tunable ratio between the two death pro-
grams (
Figure 3
D).
HEK cells are pyroptosis-incompetent without ectopically
introduced GSDMs. To assess the efficacy of synpoptosis cir-
cuits in cells with more sophisticated endogenous death cir-
cuitry, we selected two widely used immune cell lines, Jurkat
and THP-1. We transfected Jurkat and THP-1 cells with mRNA
encoding TEVP-activated caspase-3 and GSDMA circuits and
found that the cells died through the anticipated programs (
Fig-
ure 3
E). Further, to validate the orthogonality of synpoptosis cir-
cuits to endogenous death circuitry, we tested the circuits in
GSDMD-knockout (KO) THP-1 cells. As expected, the circuits
killed the GSDMD-KO THP-1 cells similarly to the WT cells
(
Figure S3
C).
Together, these results demonstrate that synpoptosis circuits
can direct the mode of cell death in various cell contexts and
tune the ratios between apoptosis and pyroptosis.
Synpoptosis circuits perform combinatorial
computation
Although engineered executioners enable control of cell death
modes, taking advantage of upstream inputs would allow syn-
poptosis circuits to target specific cells. Perhaps for similar rea-
sons, natural cell death pathways respond to logical combina-
tions of inputs. For instance, either caspase-1 or caspase-11
activates GSDMD, functioning as an OR-like gate.
77
Both
GSDMD cleavage and its lipidation are necessary for pyroptosis,
forming an AND-like gate.
78
,
79
The activation and repression synpoptosis modules devel-
oped above can be combined to achieve such combinatorial
logic gating functions. We focused on three biologically relevant
gates: triggering cell death in the presence of inputs 1 AND 2; in
the presence of input 1 OR 2; and in the presence of input 1 AND
the absence of input 2. Although the first and third gates increase
specificity by killing cells with the two inputs in the right combi-
nations, the second gate broadens specificity and could mitigate
antigen escape.
80
To trigger apoptosis in the presence of inputs 1 (TEVP) AND 2
(TVMVP), we fused TEVP-activatable caspase-3 to a TVMVP-
removable degron (
Figure 4
A, inputs 1 AND 2). In this design,
input 1 releases the linker constraint between the large and small
subunits of the engineered caspase-3, while input 2 protects the
caspase-3 from degradation. By contrast, to trigger apoptosis in
the presence of input 1 OR 2, we placed tandem TEVP and
TVMVP cleavage sites at the inter-subunit linker of caspase-3
(
Figure 4
A, input 1 OR 2). In this configuration, either protease
is sufficient to activate the engineered caspase-3. Similar princi-
ples can be used to trigger apoptosis in the presence of input 1
AND the absence of input 2 by adding a TEVP-removable degron
to the small subunit and a TVMVP-activatable N-degron to the
large subunit (
Figure 4
A, input 1 AND NOT 2). By extension, using
different configurations of degrons and cleavage sites, we con-
structed synthetic apoptosis executioners that perform more bi-
nary logic operations (
Figure S4
A). With some modifications, the
gate designs were transferable to pyroptosis programs
(
Figures 4
B and
S4
B).
ll
OPEN ACCESS
2790
Cell
187
, 2785–2800, May 23, 2024
Article