Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic Gene Expression Heterogeneity

Abstract

Intratumoral heterogeneity is associated with aggressive tumor behavior, therapy resistance, and poor patient outcomes. Such heterogeneity is thought to be dynamic, shifting over periods of minutes to hours in response to signaling inputs from the tumor microenvironment. However, models of this process have been inferred from indirect or post-hoc measurements of cell state, leaving the temporal details of signaling-driven heterogeneity undefined. Here, we developed a live-cell model system in which microenvironment-driven signaling dynamics can be directly observed and linked to variation in gene expression. Our analysis reveals that paracrine signaling between two cell types is sufficient to drive continual diversification of gene expression programs. This diversification emerges from systems-level properties of the EGFR-RAS-ERK signaling cascade, including intracellular amplification of amphiregulin-mediated paracrine signals and differential kinetic filtering by target genes including Fra-1, c-Myc, and Egr1. Our data enable more precise modeling of paracrine-driven transcriptional variation as a generator of gene expression heterogeneity. A record of this paper’s transparent peer review process is included in the Supplemental Information.

Full text

Article
Systems-Level Properties of EGFR-RAS-ERK
Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity
Graphical Abstract
Highlights
dMalignant breast cancer cells drive variable ERK activity
through AREG secretion
dNon-malignant cells receiving paracrine AREG increase in
gene expression variation
dDifferential filtering of dynamic ERK activity diversifies gene
expression states
dParacrine signals promote cellular exploration of gene
expression space
Authors
Alexander E. Davies, Michael Pargett,
Stefan Siebert, ..., Gerald Quon,
Mina J. Bissell, John G. Albeck
Correspondence
davies.474@osu.edu (A.E.D.),
jgalbeck@ucdavis.edu (J.G.A.)
In Brief
This work establishes a fluorescent live-
cell reporter system to monitor local
communication between malignant and
non-malignant breast cancer cells. This
system reveals that the EGFR-RAS-ERK
signaling network amplifies small ligand
fluctuations to generate dynamically
heterogeneous gene expression states,
which may provide an adaptive
advantage for tumor cells.
Amphiregulin paracrine gradients Single cell ERK dynamics
ERK target gene heterogeneity
ERK Activit
y
Ti
m
e
ERK Activity
Time
ERK Activity
Time
ERK Activit
y
Tim
e
ERK Activit
y
Tim
e
ERK Translocation Reporters
Erk docking site
Phospho sites
mTurquoise
Dynamic
gene expression
heterogeneity
Fra-1
Egr1
Dynamic Filtering
Fra-1 mCherry
Endogenous gene tagging
Single cell Fra-1 Expression
Time
Stochastic “Reciever” cell ERK activity
Malignant
“Sender”
Non-malignant
“Receiver”
c-Fos
c-Myc
Malignant
“Sender”
Non-malignant
“Receiver”
mVenus
Davies et al., 2020, Cell Systems 11, 1–15
August 26, 2020 ª2020 The Authors. Published by Elsevier Inc.
https://doi.org/10.1016/j.cels.2020.07.004 ll

Article
Systems-Level Properties of EGFR-RAS-ERK
Signaling Amplify Local Signals to Generate
Dynamic Gene Expression Heterogeneity
Alexander E. Davies,
1,2,4,
*Michael Pargett,
1
Stefan Siebert,
1
Taryn E. Gillies,
1
Yongin Choi,
1
Savannah J. Tobin,
1,3
Abhineet R. Ram,
1
Vaibhav Murthy,
3
Celina Juliano,
1
Gerald Quon,
1
Mina J. Bissell,
2
and John G. Albeck
1,5,
*
1
Department of Molecular and Cellular Biology, University of California, Davis, Davis, CA 95616, USA
2
Division of Biological Systems and Engineering, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
3
Department of Veterinary Biosciences, College of Veterinary Medicine, the Ohio State University, Columbus, OH 43210, USA
4
Present address: Department of Veterinary Biosciences, College of Veterinary Medicine, the Ohio State University, Columbus, OH
43210, USA
5
Lead Contact
*Correspondence: davies.474@osu.edu (A.E.D.), jgalbeck@ucdavis.edu (J.G.A.)
https://doi.org/10.1016/j.cels.2020.07.004
SUMMARY
Intratumoral heterogeneity is associated with aggressive tumor behavior, therapy resistance, and poor pa-
tient outcomes. Such heterogeneity is thought to be dynamic, shifting over periods of minutes to hours in
response to signaling inputs from the tumor microenvironment. However, models of this process have
been inferred from indirect or post-hoc measurements of cell state, leaving the temporal details of
signaling-driven heterogeneity undefined. Here, we developed a live-cell model system in which microenvi-
ronment-driven signaling dynamics can be directly observed and linked to variation in gene expression. Our
analysis reveals that paracrine signaling between two cell types is sufficient to drive continual diversification
of gene expression programs. This diversification emerges from systems-level properties of the EGFR-RAS-
ERK signaling cascade, including intracellular amplification of amphiregulin-mediated paracrine signals and
differential kinetic filtering by target genes including Fra-1, c-Myc, and Egr1. Our data enable more precise
modeling of paracrine-driven transcriptional variation as a generator of gene expression heterogeneity. A re-
cord of this paper’s transparent peer review process is included in the Supplemental Information.
INTRODUCTION
Cellular heterogeneity is a prominent feature of many tumors,
including breast, colorectal, and brain (Gao et al., 2016;Stingl
and Caldas, 2007). While some heterogeneity can be attributed
to the genetic mosaicism of tumors, much variation arises non-
genetically and involves the ability of cells to reversibly shift their
gene expression profiles over time (Gupta et al., 2011). This ‘‘dy-
namic heterogeneity’’ provides an adaptive advantage for can-
cer cells, contributing to metastasis and drug resistance
(Sharma et al., 2010). Potential drivers of dynamic heterogeneity
are multifactorial (Meacham and Morrison, 2013) and include
both intrinsic stochastic processes and temporal shifts in
extrinsic signaling or adhesive inputs received from the tumor
microenvironment (TME) (Friedl and Alexander, 2011;Tam and
Weinberg, 2013).
In current models of heterogeneity, each cancer cell receives
unique signaling inputs due to spatial variation in the TME; these
inputs can then drive differential gene expression programs,
which lead to phenotypic heterogeneity (Davies and Albeck,
2018;Cazet et al., 2018;Lu et al., 2014). However, these models
are qualitative, and key quantitative and kinetic aspects of this
process remain uncharacterized. First, the scope of TME hetero-
geneity needed to drive variable gene expression is not known.
TME signals acting on tumor cells emanate from multiple cell
types, extracellular matrices (ECMs), and mechanical forces,
and while a paracrine signal is an obvious candidate to drive dif-
ferences, it is not clear whether it alone can generate widespread
transcriptional variation. Second, it is not clear whether signaling
pathways act as linear transmitters of heterogeneous TME sig-
nals (Nunns and Goentoro, 2018) or instead reshape inputs
through amplification or feedback (Tyson et al., 2003). Finally,
it is not known how frequently gene expression states fluctuate
in response to TME variation, which is essential for understand-
ing the population dynamics of drug-resistant cells and for
designing the most effective time course of drug treatments.
While in vivo studies can verify the physiological importance of
TME-driven heterogeneity, a defined in vitro model is still needed
to address these questions quantitatively.
Heterogeneity plays a prominent role in basal-like breast can-
cer (BLBC), an aggressive malignancy in which cells interchange
between multiple states that vary in tumor initiating capacity and
Cell Systems 11, 1–15, August 26, 2020 ª2020 The Authors. Published by Elsevier Inc. 1
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
ll
OPEN ACCESS
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

A
ERK activity
reporters
S1
(non-malignant)
T4-2
(malignant)
ERKTR-
mTurquoise
OR
ERKTR-
mVenus
Ligand
E
G
F
R
E
G
F
R
Grb2
SOS
Ras
Raf
MEK
ERK
ERK-target
genes
P
P
PP
Ligand
H
DCB
GFE
Mean Derivative/Mean
S1 Control
S1 EGF
T4-2 Control
T4-2 EGF
0
0.04
0.08
0.12
0.16
0.20 *NS
NS
IK
J
L
Mono-Culture Co-Culture
02468
10
20
30
40
0 0.5 11.5 22.5
10
20
30
0 0.5 1 1.5
10
30
50
02468
10
20
30
40
Frequency (pulse/hour)
0 0.5 11.5 22.5
10
20
30
0 0.5 1 1.5
10
30
50 S1
T4-2
Peak Amplitude (a.u.)
% cells
Duration (hours)
% cells% cells
0.6
0.9
0.3 0.6
0.6
0.7
0.5
0.5
0.02
0.0045
0.02
0.02
S1
ERKTR C/N
Control
Pre Post
0.5
2.0
-1 0 1 2 3 4
Time (hr)
ERKTR C/N
Pre Post
100nM MEKi
0.5
2.0
-1 0 1 2 3 4
Time (hr)
S1
0.5
2.0
-1 0 1 2 3 4
Time (hr)
S1
ERKTR C/N
Pre Post
20ng/ml EGF
Pre Post
T4-2
-1 0 1 2 3 4
Time (hr)
0.5
2.0
ERKTR C/N
Control 100nM MEKi
Pre Post
-1 0 1 2 3 4
Time (hr)
0.5
2.0
ERKTR C/N
T4-2
20ng/ml EGF
Pre Post
-1 0 1 2 3 4
Time (hr)
0.5
2.0
ERKTR C/N
T4-2
0.5
2.0
-1 0 1 2 3 4
Time (hr)
0.5
2.0
20ng/ml EGF
Pre Post
S1co ERKTR C/N
T4-2co ERKTR C/N
0.5
2.0
-1 0 1 2 3 4
Time (hr)
0.5
2.0
2μM EGFRi
Pre Post
S1co ERKTR C/N
T4-2co ERKTR C/N
0.5
2.0
-1 0 1 2 3 4
Time (hr)
0.5
2.0
Control
Pre Post
S1co ERKTR C/N
T4-2co ERKTR C/N
mean mean mean
mean mean mean
mean mean mean
M
N
O
EGFRi
01234567
Time (hrs)
0.7
0.8
0.9
1
1.1
1.2
ERKTR C/N
01234567
Time (hrs)
0.5
1
1.5
ERKTR C/N
T4-2m/S1m Control T4-2co/S1co Control T4-2co/S1co 0.5uM EGFRi
T4-2co/S1co 2uM EGFRi T4-2co/S1co 4uM EGFRi
T4-2 S1
+treatment +treatment
TACEi
T4-2m/S1m Control T4-2co/S1co Control T4-2co/S1co 12.5uM TACEi
T4-2co/S1co 25uM TACEi T4-2co/S1co 50uM TACEi
01234567
Time (hrs)
0.7
0.8
0.9
1
1.1
1.2
ERKTR C/N
01234567
Time (hrs)
0.5
1
1.5
ERKTR C/N
T4-2 S1
+treatment +treatment
T4-2m/S1m Control T4-2co/S1co Control T4-2co/S1co 5ug/mL FBAb
T4-2co/S1co 10ug/mL FBAb T4-2co/S1co 20ug/mL FBAb
AREG FBAb
01234567
Time (hrs)
0.7
0.8
0.9
1
1.1
1.2
ERKTR C/N
01234567
Time (hrs)
0.5
1
1.5
ERKTR C/N
T4-2 S1
+treatment +treatment
mean mean mean
Figure 1. Progression to Malignancy Is Associated with Stochastic RAS-ERK Signaling Dynamics and Heterogeneous Target Gene
Expression
(A) Schematic of the EGFR-RAS-ERK signaling pathway in mammalian cells and ERKTR.
(legend continued on next page)
ll
OPEN ACCESS Article
2Cell Systems 11, 1–15, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

drug resistance (Brooks et al., 2015). Several findings suggest a
role for epidermal growth factor receptor (EGFR) signaling via the
proto-oncogene RAS and extracellular signal-regulated kinase
(ERK) in BLBC heterogeneity. First, genes stimulated by this
pathway (ERK target genes, hereafter ETGs) include transcrip-
tion factors such as c-Myc and Fra-1 that have been implicated
as drivers of breast cancer malignancy (Belguise et al., 2005;
Berns et al., 1992;Tam et al., 2013). Second, while BLBC tumors
do not frequently carry mutations in the RAS cascade, they often
overexpress EGFR (Reis-Filho et al., 2006) or show protein phos-
phorylation profiles consistent with receptor tyrosine kinase ac-
tivity (Hochgr€
afe et al., 2010). Third, live-cell reporters have re-
vealed that paracrine signaling generates local, highly dynamic
RAS-ERK activation (Albeck et al., 2013;Aoki et al., 2013;Hirat-
suka et al., 2015) and that the dynamics of ERK influence the
expression of ETGs (Bugaj et al., 2018;Wilson et al., 2017).
Together, these findings suggest that paracrine signaling
through EGFR-RAS-ERK may drive dynamic ETG expression,
resulting in heterogeneous populations of BLBC cells, but this
association has not been tested at a mechanistic level.
In this study, we investigated the link between microenviron-
mental heterogeneity, signaling dynamics, and gene expression
heterogeneity in a cell line model of BLBC, HMT-3522 (Rizki
et al., 2008). These cells were originally derived from reduction
mammoplasty tissue and subjected to multiple rounds of
in vitro and in vivo selection for tumorigenic behavior (Madsen
et al., 1992). The non-malignant cells, termed S1, were sponta-
neously immortalized in culture using defined media, give rise
to polarized acinar structures, exhibit growth arrest, and do
not generate tumors in mouse models. Malignant T4-2 cells
were derived from S1 cells through serial passage in the
absence of exogenous growth factor, followed by passage
through a mouse model to isolate tumor-forming invasive cells.
Here, we developed a co-culture model of S1 and T4-2 cells,
expressing color-coded live-cell reporters for ERK. This
enabled us to track dynamic signaling profiles driven by para-
crine signaling and link them to gene expression profiles of in-
dividual cells. We show that paracrine signaling between these
cell types is sufficient to drive heterogeneity in gene expression
similar to that found in BLBC tumors. Furthermore, we find that
the pulsatile nature of ERK signaling, coupled to differentially
responding target genes, expands the repertoire of transcrip-
tional states and drives changes in ETG expression over time.
Our findings validate a model in which EGFR-RAS-ERK
signaling amplifies paracrine variation to generate dynamic
gene expression heterogeneity.
RESULTS
A Model of Paracrine Signaling-Induced Heterogeneity
in EGFR-RAS-ERK Signaling
To investigate cell-cell signaling in a simplified microenviron-
ment, we utilized the HMT-3522 cell line series. A defining feature
of progression from S1 (non-malignant) to T4-2 (malignant) is
increased production and secretion of the EGFR ligand amphir-
egulin (AREG), which promotes proliferation in the absence of
exogenous growth factor (Kenny and Bissell, 2007). To monitor
ERK activity stimulated via AREG, we generated S1 or T4-2 cells
with genetically encoded fluorescent ERK translocation re-
porters (ERKTRs) (Figure 1A) (Regot et al., 2014;Sparta et al.,
2015). These reporters contain a tandem nuclear import and
export sequence that is also a substrate for ERK; phosphoryla-
tion of the reporter by ERK suppresses shuttling from nucleus
to cytoplasm. Cytoplasmic shuttling is opposed by phospha-
tase-mediated dephosphorylation of the reporter (Regot et al.,
2014), and ERK kinase activity is, thus, measured as the ratio
of cytosolic (C) to nuclear (N) fluorescence (hereafter ERKTR
C/
N
). The different versions of the reporter are functionally equiva-
lent, varying only in color of the fluorescent protein, facilitating
identification of individual cell types in our experiments.
We measured ERK activity under: (1) baseline imaging me-
dium lacking growth factor, (2) treatment with MEK inhibitor,
PD0325901 (MEKi), or (3) stimulation with EGF. Under baseline
conditions, S1 cells exhibited few fluctuations in ERKTR
C/N
(Figure 1B; Video S1A), with infrequent low amplitude pulses.
Addition of MEKi had no effect on ERKTR
C/N
, indicating that
ERK activity in S1 cells was below the detectable limit for the re-
porter (Figure 1C). At low EGF concentrations (0.2–2 ng/mL) we
observed ERKTR
C/N
responses ranging from undetectable to
low amplitude pulses (Figure S1), while addition of 20 ng/mL
EGF resulted in sustained activation of ERK signaling (Figures
1D and S1;Video S1A). These results indicate that S1 cells
depend on growth factor stimulation for ERK activation, similar
to other non-malignant mammary epithelial cells (Gillies
et al., 2017).
In contrast to S1, T4-2-ERKTR cells exhibited elevated and
variable ERKTR signals under baseline conditions (Figure 1E;
Video S1B), which were reduced by MEKi (Figure 1F; Video
(B–G) Measurements of ERK activity by ERKTR
C/N
in S1 or T4-2 cells following treatment with exogenous growth factors or MEK inhibitor. In each panel, mean
ERKTR
C/N
is shown at bottom with the 25
th
–75
th
interquartile range (IQR) depicted by shading. Unshaded traces represent individual cells and red vertical lines
indicate the addition of growth factor, inhibitor, or vehicle. n > 2,400 cells per condition.
(H) Variability of ERKTR
C/N
in control versus EGF integrated over time (n = 8,372). Boxes indicate IQR, and whiskers the range. NS, not significant, *p <0.01 by
unpaired t test.
(I–K) Measurements of ERK activity by ERKTR
C/N
in co-cultured S1 (turquoise lines) and T4-2 cells (yellow lines). Annotations as in (B–G). n > 1,900 cells per
condition.
(L) Distributions of single-cell pulse frequency (per hour), duration (hours), and amplitude (arbitrary units) of ERKTR
C/N
for S1 (turquoise) and T4-2 (yellow) (3,497
total cells analyzed). Numbers upper right correspond to median value color-coded to each cell type.
(M) Mean ERKTR
C/N
traces of T4-2m, S1m, or T4-2co/S1co treated with EGFR inhibitor (erlotinib) at the indicated concent rations. Numbers of cells analyzed per
condition (T4-2/S1): monoculture, control 2187/692; co-culture, control 244/122; 0.5 mM erlotinib 199/89; 2 mM erlotinib 239/78, 4 mM erlotinib 281/105.
(N) Mean ERKTR
C/N
traces of T4-2m, S1m, or T4-2co/S1co treated with TACE inhibitor (TACEi). Numbers of cells analyzed per condition (T4-2/S1): monoculture,
control 2374/2036; co-culture, control 249/122; 12.5 mM TACEi 233/116; T4-2/S1co 25 mM TACEi; 256/110, 50 mM TACEi 238/90.
(O) Mean ERKTR
C/N
traces of T4-2m, S1m, or T4-2co/S1co treated with AREG function-blocking antibody. Numbers of cells analyzed per condition (T4-2/S1):
monoculture, control 2374/2036; co-culture, control 243/55; 5 mg FBAb 237/56; 10 mg FBAb 222/76; 20 mg FBAb 261/74.
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15, August 26, 2020 3
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

S1C). EGF provoked a maximal ERK response similar in magni-
tude to stimulated S1 cells; however, the amplitude of this
response relative to baseline was much smaller than in S1 due
to the pre-existing high level of ERK activity (Figures 1G and
1H; Video S1B). These results confirm that progression to malig-
nancy in HMT-3522 is associated with elevated baseline ERK
signaling, which is stochastic, self-sustaining, and influenced
weakly by additional stimuli (Kenny and Bissell, 2007).
To evaluate the effects of paracrine signaling, we co-cultured
S1 cells with T4-2 cells at different ratios and measured
ERKTR
C/N
signals (Figures 1I–1K; Videos S2A–S2G). For clarity,
we refer to co-cultured cells as S1co and T4-2co, and to mono-
cultured cells as S1m and T4-2m. In S1co, compared with S1m,
we observed increased ERKTR
C/N
that varied over time within
each cell (Figure 1I; Videos S2A and S2D). A 30:70 ratio of S1
to T4-2 produced the most reliable activation in S1co, inducing
ERKTR
C/N
signals that were quantitatively similar in pulse fre-
quency, amplitude, and duration to the pulse characteristics of
T4-2m cells (Figure 1L). Consistent with paracrine signaling as
the source of ERK activity, EGFR inhibitor (EGFRi) blocked
ERK signaling in both S1co and T4-2co cells with similar potency
to MEKi (Figure 1J; Video S3A). The response of S1co cells to
exogenous EGF was dampened relative to S1m cells and similar
to that of T4-2m (Figure 1K; Video S3B), in agreement with re-
ceptor occupancy by paracrine ligands.
To confirm that ERK activity in S1co and T4-2m cells is medi-
ated by paracrine ligands, we treated cells with TAPI-0, an in-
hibitor of the AREG and TGF-ashedding enzyme (TACE),
compared to EGFRi (Figure 1M) or MEKi (Figures S1B–S1D).
TACEi produced an initial suppression of ERK activity in both
S1co and T4-2m, followed by a gradual return to dynamic
signaling (Figures 1N and S1C). We found that TACEi exerts a
prolonged effect if culture medium is changed immediately
prior to TACEi addition (Figure S1E), suggesting that residual
AREG in the microenvironment accounts for the gradual return
to dynamic signaling in the absence of media washout. An
AREG function-blocking antibody (FBAb) also strongly sup-
pressed ERKTR
C/N
for 1 h, followed by a gradual return of
ERK activity (Figures 1O and S1C), which is consistent with
saturation of the FBAb over time by continuous AREG secretion
from T4-2 cells. High concentrations of TGF-aFBAb had no
detectable effect on ERKTR
C/N
(Figure S1F). These results indi-
cate that paracrine release of AREG by T4-2 cells is the primary
driver of stochastic ERK signaling and are also consistent with
the role of AREG as a low-affinity EGFR ligand that can diffuse
freely to adjacent cells (DeWitt et al., 2001,2002). Finally, we
compared these observations with a co-culture lacking para-
crine signaling. MCF10A-CA1D cells, which carry an activating
RAS mutation but are not known to secrete AREG, showed
similar ERKTR
C/N
levels to T4-2m cells but did not efficiently
drive ERK signaling in S1co cells (Figure S1G). These results
establish a defined system in which S1co cells serve as a
receiver for paracrine EGFR-mediated signals produced by
T4-2co cells.
Paracrine Signaling Drives Heterogeneous Expression
of EGFR-RAS-ERK-Regulated Genes
We investigated gene expression heterogeneity using single-cell
mRNA sequencing (scRNA-seq), comparing S1m cells with or
without growth factor, S1co/T4-2co mixtures, and T4-2m cells
(Figures 2A and S2A–S2D). We assessed global gene expression
profiles as a function of cell type and/or conditions (Figure 2B).
Clustering analysis of S1m, S1m treated with 100 ng/mL AREG
(S1m
+AREG
), and T4-2m cells found that cells broadly sort ac-
cording to type and conditions (Figure 2B). Similarly, S1co and
T4-2co cells segregated into distinct clusters, even without
filtering by mRNA expression of mCherry or mVenus. We found
eleven clusters of cells with unique but partially overlapping sig-
natures (Figures 2C and S2E; Data S1). T4-2m and T4-2co cells
were each comprised 3 clusters enriched for genes involved in
replication and cell division, as well as growth factors, reactive
oxygen enzyme SOD2, and TNF-related apoptosis-inducing
ligand (TNFSF10) (Figures 2C and S2E). S1m cells treated with
growth factor segregated into two clusters: the bulk population
of cells (cluster 2) showed broad expression changes, with a
notable increase in MAPK-regulated genes compared with unsti-
mulated controls (Figures 2C and S2E), while the second cluster
(cluster 8) showed increased expression of cell-cycle-associ-
ated genes. S1co cells formed a separate cluster (cluster 9)
with an expression profile that partially overlapped with S1m
and T4-2m cells but also included a diverse set of genes distinct
from other conditions, including inflammation-associated genes
(Figure S2E; Data S1).
Since distances in tSNE plots can distort the similarity of clus-
ters, we performed additional analysis to determine the relative
gene expression similarity between S1m, S1m
+AREG
, S1co,
and T4-2 cells. Using pairwise correlation (Figure 2D) and prin-
cipal component analysis (Figure 2E) we found that S1m and
T4-2m cells exhibit divergent gene expression profiles, whereas
S1m
+AREG
and S1co cells exhibit strong correlation in gene
expression profiles. These findings imply that AREG is the pre-
dominant factor driving gene expression changes under co-cul-
ture conditions and that co-culture alone is not sufficient to fully
convert global S1 gene expression to a malignant-like expres-
sion profile.
We next compared transcriptional states by quantifying cell-
to-cell variance in gene expression. Gene expression variance
typically increases with the mean; we removed this bias by
calculating, per gene, the ‘‘excess dispersion,’’ i.e., the variance
exceeding the fitted relationship between mean and variance. In
both S1 and T4-2 cells, co-culture conditions promoted an in-
crease in excess dispersion for the majority of genes (69% and
90%, respectively, Figures 2F and 2G, p = 1.7 310
61
).However,
excess dispersion was more widespread in S1m than in T4-2m
cells (Figure 2H), while treatment of S1m cells with AREG
reduced excess dispersion in nearly all genes, relative to un-
treated S1m (Figure 2I). The ordering of variability in S1 cells
(S1co > S1m > S1
+AREG
) is consistent with a model in which
exogenous growth factor addition suppresses expression vari-
ability, because cells receive uniformly high levels of stimulation,
whereas co-culture enhances variability, as cells receive variable
levels of stimulation. The observation that excess dispersion is
increased between S1m and S1co cells but decreased between
S1m and T4-2m cells suggests that paracrine signaling can have
a stronger influence on transcriptional variability than progres-
sion to a malignant phenotype, which may select for a
narrowed gene expression space. The increase in excess
dispersion in T4-2co relative to T4-2m cells could reflect the
ll
OPEN ACCESS Article
4Cell Systems 11, 1–15, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

effects of local regions of high S1 density, which have a suppres-
sive effect on T4-2 signaling (Video S2), potentially by acting as a
sink for secreted AREG.
To ask whether the variable ERK dynamics observed in T4-2
cells are associated with greater gene expression variance, we
refined our analysis to known ETGs (c-Fos, c-Myc, Fra-1, and
Egr1). We found that 60% of ETGs showed increased excess
dispersion in T4-2m relative to S1m cells, with a number of
genes, including Egr1, falling well outside the bulk distribution
(Figure 3A; Data S1). The known classes of ETGs that are sensi-
tive to different patterns of ERK activity (Uhlitz et al., 2017)
showed similar increases in excess dispersion (Figure 3B).
Thus, the variable ERK activity observed in both T4-2m and
S1co correlates with increased variability in expression of
ETGs, many of which are transcription factors linked to malig-
nancy. To confirm these results, we measured Fra-1, Egr1, c-
Fos, and c-Myc, protein expression by immunofluorescence
(Figure 3C). Expression was minimal in S1m, but elevated in
T4-2m, with coefficients of variance (CVs) ranging from 50%–
100% (Figures 3C and 3D). Similarly, under co-culture condi-
tions, ETG protein levels and heterogeneity increased in S1co
cells (Figures 3E and 3F).
We compared ETG heterogeneity driven by paracrine signals
in T4-2m cells to models of BLBC driven by RAS mutations,
MCF10A-CA1D and MB-MDA-231. ERKTR
C/N
displayed similar
temporal variability in all cell types that was reduced by MEKi
(Figure S3A and S3B). Accordingly, Fra-1 expression varied
from cell to cell with CVs >45% for all cell types (Figure S3C).
As expected, TACEi treatment greatly reduced Fra-1 expression
in T4-2 cells but not in the RAS-driven cell types. A combination
of TACEi and AREG restored Fra-1 expression in T4-2 cells, but
with lower variance (CV = 29%), while the same treatment did not
significantly alter Fra-1 variance in the other cell lines (Fig-
ure S3A). Thus, RAS- and paracrine-driven ERK signaling can
both promote variance in ETG expression but are distinct in their
dependence on extracellular ligands.
Surprisingly, despite their co-regulation by ERK, ETG mRNAs
showed discordant expression in single cells, particularly be-
tween FOSL1 (Fra-1) and EGR1 (R = 0.043) and between
FOSL1 and FOS (R = 0.026) (Figure 3G). Co-staining of Fra-1
B
C
FI
A
−20
0
20
40
−50 −25 0 25
tSNE_1
tSNE_2
S1m
T4-2m
T4-2co
S1co
S1m+AREG
−20
0
20
40
−50 −25 0 25
tSNE_1
tSNE_2
12
0
7
6
5
10
9
3
4
8
S1
Non-malignant
cells
T4-2
Malignant cells
Co-culture
Drop-seq single-cell
RNA sequencing
tSNE analysis Single-cell
global expression
variance analysis
Single-cell pathway-specific
expression variance analysis
GH
−20
0
20
40
−50 −25 0 25
tSNE_1
tSNE_2
S1m
S1co
−20
0
20
40
−50 −25 0 25
tSNE_1
tSNE_2
T4-2m
T4-2co
tSNE_1
−20
0
20
40
−50 −25 0 25
tSNE_2
S1m
T4-2m −20
0
20
40
−50 −25 0 25
tSNE_1
tSNE_2
S1m
S1m+AREG
−2.5 0.0 2.5 5.0 7.5 10.0
S1m log mean
S1m − S1co
delta excess dsipersion
−1
0
1
31%dispersion S1m
69%dispersion S1co
−2.5 0.0 2.5 5.0 7.5 10.0
T4-2m log mean
T4-2m − T4-2co
delta excess dispersion
−2
−1
0
1
2
10%dispersion T4-2m
90%dispersion T4-2co
−2.5 0.0 2.5 5.0 7.5 10.0
S1m log mean
S1m − T4-2m
delta excess dispersion
−2
−1
0
1
2
dispersion T4-2m 14%
dispersion S1m 86%
−2.5 0.0 2.5 5.0 7.5 10.0
S1m log mean
S1m − S1m+AREG
delta excess dispersion
−1
0
1
2
dispersion S1m+AREG 5%
dispersion S1m 95%
S1m
S1m+AREG
S1co
T4−2m
T4−2m
S1co
S1m+AREG
S1m
0
0.2
0.4
0.6
Correlation Coefficient
D
E
−0.1 0.0 0.1 0.2
Principal Component 1
Cell Type
S1m
S1m+AREG
S1co
T4−2m
Figure 2. Induction of Global Gene Expression Variance by Cellular Crosstalk
(A) Schematic depicting cell culture conditions and analyses for S1, T4-2, and S1/T4-2 co-cultures for single-cell RNA sequencing. Red: S1 cells expressing
ERKTR-mCherry; yellow: T4-2 cells expressing ERKTR-mVenus.
(B and C) tSNE plots of single-cell transcript profiles, colored by identified cell type and condition (B) or by clustering of transcriptional profiles (C). n > 1,900 cells
per condition.
(D) Pairwise correlation of the indicated cells and conditions using raw count data.
(E) Scatterplot of first principal component weights per the indicated cells and conditions.
(F–I) Top row: tSNE plots highlighting the groups of cells compared by excess dispersion. Bottom row: differences in excess dispersion between treatment
conditions (>1,800 genes sampled); each dot represents the relative excess dispersion for one gene. Gene s are color-coded according to which condition shows
greater excess dispersion. Percentages indicate the proportion of genes falling on either side of the horizontal line.
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15, August 26, 2020 5
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

E
Flurorescence Intensity (A.U.)
% Cells (T4-2m)
40 290 550
0
20
40 Egr1
CV = 99%
70 390 710
0
5
10
15 Fra-1
CV = 54%
100 290 480
0
5
10
15
20 c-Fos
CV = 71%
110 530 940
0
5
10
15 c-Myc
CV = 62%
AB D
Breast Tumor
DAPI Fra-1
Region 1Region 2
DAPI Egr1
Region 3Region 4
J
−0.6
−0.3
0.0
0.3
0.6
0246
S1m log mean
S1m − T4-2m
delta excess dispersion
●
●
●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
DEGIEG ILG
ETG dispersion by class
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
−0.6
−0.3
0.0
0.3
0.6
0246
43%
57%
DEG only
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
−0.6
−0.3
0.0
0.3
0.6
0246
29%
71%
ILG only
●
0.6
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
−0.6
−0.3
0.0
0.3
0246
36%
64%
IEG only
G
ERKTR
Egr1
Control ControlMEKi MEKi
% Cells
40 290 550
0
20
40
60
40 290 550
0
20
40
60
Flurorescence Intensity (A.U.)
CV = 25%
CV = 21%
CV = 69%
CV = 17%
S1m S1co / T4-2co
% Cells ERKTR
Fra-1
70 390 710
0
20
40
60
80
70 390 710
0
20
40
60 CV = 49%
CV = 32%
CV = 36%
CV = 34%
Control ControlMEKi MEKi
Flurorescence Intensity (A.U.)
S1m S1co / T4-2co
100nM MEKi OverlapControl
% Cells
110 530 940
0
10
20
30
40 CV = 51%
CV = 41%
110 530 940
0
20
40
60 CV = 36%
CV = 30%
Flurorescence Intensity (A.U.)
Control ControlMEKi MEKi
S1m S1co / T4-2co
ERKTRc-Myc
% Cells ERKTR
c-Fos
100 290 480
0
10
20
30
40 CV = 40%
CV = 31%
100 290 480
0
20
40
CV = 30%
CV = 28%
Flurorescence Intensity (A.U.)
Control ControlMEKi MEKi
S1m S1co / T4-2co
F
H
Egr1 Fra-1
c-Fos
c-Myc
ETG DAPI
S1m
ETG DAPI
T4-2m
Co-culture
S1m
S1co Control
T4-2co Control
S1co MEKi
T4-2co MEKi
0
100
200
300
400
500
600
700
800
900
1000
1100
Fra-1 Intensity (A.U.)
*
*
*
*
Fra-1
0
100
200
300
400
500
600
700
800
900
1000
1100
Egr1 Intensity (A.U.)
*
*
*
*
Egr1
Co-culture
S1m
S1co Control
T4-2co Control
S1co MEKi
T4-2co MEKi
Fra-1Egr1MERGE DAPI
Fra-1
c-FosMERGE DAPI
T4-2m T4-2m
I
−0.6
−0.3
0.0
0.3
0.6
0246
S1m log mean
S1m − T4-2
delta excess dispersion
ETG Dispersion
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
EGR1
FOSL1
dispersion S1m
dispersion T4-2m
C
Pairwise ETG correlation
0.043
0.31
0.22
0.026
0.071 0.11
FOSFOSL1
MYC
EGR1
FOS FOSL1
4
3
2
1
0
2
1
0
-1
2
1
0
-1
-1 0 1 2 -1 0 1 201234
Control
MEKi
Fra-1
DAPI Egr1
Control
T4-2m
Control
MEKi
Figure 3. Induction of ETG Expression Variance by Cellular Crosstalk
(A) Comparison of variance in ETG transcript levels between S1 and T4-2 cells. Red dots indicate genes that display excess dispersion in T4-2 cells compared
to S1.
(B) The same graph in (A) is color coded by known ETG expression classes: immediate-early genes (IEG, light blue), immediate-late genes (ILG, purple), and
delayed-early genes (DEG, orange). Insets depict each of the classes alone. Annotated as in Figures 2F–2I.
(C) Immunofluorescence of ETGs Fra-1, Egr1, c-Fos, and c-Myc (red) and nuclei (blue) in S1 and T4-2 cells in the absence of exogenous growth factors. Scale
bar, 20 mm.
(D) Histograms depicting the fluorescence intensity distribution of ETG proteins under baseline conditions in T4-2 cells. n> 4,000 cells per ETG measured.
(E) S1m (turquoise) and S1co and T4-2co (turquoise and yellow, respectively) cells stained for ETG expression (red) under control or MEK inhibitor conditions
with Fra-1 and Egr1 protein expression distributions plotted as histograms. Control treated (black), MEKi-treated (pale red), and overlapping data (dark red)
(n = 25,169). Scale bar, 20 mm.
(F) Box and whisker plots comparing mean Fra-1 and Egr1 protein levels between S1m cells co-cultured cells (n = 30,370). *p < 0.01.
(G) Pairwise single-cell gene expression correlation for the indicated ETGs. Red numbers indicate the corre lation coefficient (Pearson correction).
(H) Co-staining of Fra-1 and Egr1 or Fra-1 and c-Fos in mono-cultured T4-2 cells under baseline conditions. Scale bar, 20 mm.
(I) Mono-cultured T4-2 cells stained for Fra-1 or Egr1 under baseline conditions or treatment with MEKi. Scale bar, 20 mm.
(J) Formalin-fixed resected human invasive ductal carcinoma samples stained for Fra-1 or Egr1.
ll
OPEN ACCESS Article
6Cell Systems 11, 1–15, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

A
B
C
D
EF
G
H
I
J
Figure 4. Paracrine AREG and Gene-Specific Expression Kinetics Coordinately Induce Heterogeneous ETG Expression
(A and B) Time-dependent expression of Fra-1 (red) and Egr1 (green) in S1m or T4-2m cells following exposure to 100 ng/mL AREG or 4 mM erlotinib for the
indicated times. Scale bar, 20 mm.
(C) S1m and T4-2m scatter plots of single-cell Fra-1 or Egr1 signal intensities and correlations (red lines). n > 5,900 cells per condition.
(D) T4-2 single-cell gene expression analysis of preprocessed (normalized and scaled) single-cell RNA sequencing data partitioned as a function of single-cel l
FOSL1 and EGR1 expression.
(legend continued on next page)
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15, August 26, 2020 7
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

and Egr1, or Fra-1 and c-Fos proteins in T4-2m cells further
corroborated their lack of correlation (Figure 3H). While this
divergence could be explained by decoupling of these genes
from ERK signaling in malignant cells, this possibility was ruled
out by treatment of T4-2m cells with MEKi, which suppressed
both Fra-1 and Egr1 expression (2.3-fold and 2.5-fold) (Figure 3I).
We confirmed the relevance of these findings by staining human
breast cancer tissues, finding that both Fra-1 and Egr1 show
similar degrees of heterogeneity in patient samples (Figure 3J).
Thus, co-culture of S1 and T4-2 cells recapitulates the heteroge-
neity in gene expression previously observed in BLBC (Nguyen
et al., 2016) and identifies new features of this heterogeneity.
Dynamic Properties of ERK-ETG Signaling Enhance
Heterogeneity in Target Gene Expression
Paracrine signaling reduces cell-to-cell variability in many sys-
tems (Rand et al., 2012;Shalek et al., 2014;Handly et al.,
2015) making it unclear why paracrine AREG signaling enhances
ETG variability in S1co and T4m cells. One model consistent with
discordant ETG expression is that, as ERK activity changes over
time (Figures 1G and 1K), distinct induction and turnover param-
eters (Amit et al., 2007;Uhlitz et al., 2017) favor expression of
each ETG at different times within individual cells. We, therefore,
examined the temporal response of these genes in the absence
of paracrine signaling using S1m cells, which displayed low or
undetectable Fra-1 and Egr1 in baseline conditions (Figure 4A)
compared with T4-2m (Figure 4B). Fra-1 and Egr1 increased in
a correlated manner at 1.5 h after AREG stimulation (R
2
= 0.69,
Figures 4A–4C). However, with extended stimulation, Fra-1
continued to increase while Egr1 levels peaked and then
decreased. This loss of correlation at the cellular level (R
2
=
0.44, Figures 4A–4C) was similar to expression patterns
observed in T4-2m cells under control and stimulated conditions
(R
2
= 0.46 and 0.44, respectively; Figures 4B and 4C). Similar
patterns of variance were also observed for c-Fos and c-Myc
(Figure S4A), indicating that this behavior is shared among other
ETGs. Thus, decoupled expression due to distinct induction and
turnover kinetics can account for the low correlation of ETGs in
T4-2m cells (Figures 3G and 3H), which experience asynchro-
nous paracrine-mediated ERK activation.
To explore the functional significance of different single-cell
ETG expression states, we correlated single-cell ETG expres-
sion with cell states using the MSigDB (Liberzon et al., 2015)
Hallmark sets and the NCI pathway interaction database (PID)
(Figures 4D–4F and S5;Data S2). T4-2 cells expressing high
FOSL1 and EGR1 contained EMT and MYC target signatures
and increased expression of ERBB, MAPK, and AP1 pathways,
consistent with high ERK activity (Figure 4E) compared with
FOSL1-low/EGR1-low cells. Additionally, FOSL1-high/EGR1-
high cells display increased TGF-bsignaling pathway expression
relative to FOSL1-high/EGR1-low cells and higher expression of
cell division-associated pathways relative to FOSL1-low/EGR1-
high cells (Figure 4F). These findings suggest that heterogeneity
in Fra-1 and Egr1 gene and protein expression within single cells
is correlated with functional differences in cell behaviors.
To evaluate the significance of ETG expression heterogeneity
in the context of chemotherapeutic response, we examined
the DNA damage response to carboplatin, a cytotoxic agent
commonly used in BLBC. Under baseline conditions we found
carboplatin induced minimal levels of the DNA damage marker
p-ɣH2AX in non-malignant MCF10A cells, which have low
levels of Fra-1 and Egr1 (Figure 4G), whereas T4-2 cells re-
sponded with a larger increase (Figure 4H). Both MCF10A
and T4-2 cells showed a greatly increased p-ɣH2AX response
in the presence of EGF, which was attenuated after treatment
with EGFRi. At the single-cell level, p-ɣH2AX was heteroge-
neous, leading us to ask whether the ETG expression status
of each cell correlated with its DNA damage response. We
used partial least squares regression analysis (PLSR) to model
each cell’s p-ɣH2AX response as a function of its ETG protein
abundance (Figure 4I–4J). At the single-cell level, ETG abun-
dance was highly predictive of p-ɣH2AX, more so than treat-
ment with growth factor (GF) and carboplatin, as indicated by
the relative model weights. In particular, Fra-1 staining was
highly correlated with p-ɣH2AX, whereas Egr1 staining was
only weakly correlated (Figure S4). FOSL1 expression is corre-
lated with upregulation of the Hallmark DNA repair pathway set
(Data S2), further strengthening its correlation with DNA dam-
age response characteristics. Although a direct causal relation-
ship between Fra-1 and DNA damage response cannot be
strictly assigned based on the aggregate data, there is a strong
correlation, indicating a potential link between cell-to-cell vari-
ation in ETG expression and heterogeneous responses to DNA-
damaging chemotherapy.
Ligand-Specific ERK Dynamics Modulate ETG
Expression Heterogeneity
Because different EGFR ligands induce distinct ERK activation
dynamics (Nakakuki et al., 2010;Sparta et al., 2015), we exam-
ined how these ligands impact the heterogeneous expression
of ETGs, comparing ERK activity kinetics induced by AREG or
EGF in S1 cells. At saturating EGF (20 ng/mL), >80% of cells re-
sponded with sustained ERK activation (Figures 5A and 5C). At
lower levels of EGF (2 ng/mL), only a fraction of S1 cells
(E) Differential gene expression analysis based on FOSL1 and EGR1 expression partitioning. A complete list of upregulated molecular signature database
(MSigDB) hallmark and NCI pathway database (PID) pathways in FOSL1 or EGR1 high- versus low-expressing cells.
(F) As in (E), differential gene expression analysis based on ETG expression was analyzed with MSigDB and PID. The list represents uniquely upregulated
pathways for the indicated comparisons.
(G) Analysis of p-gH2AX, Fra-1, and Egr1 staining levels in MCF10A cells under the indicated conditions: control, +GF (growth factor, 100 ng/mL AREG), or 5 mM
erlotinib, and 50 mM carboplatin. n > 1,000 cells per condition. NS, not significant; *p < 0.01 unpaired t test. Scale bar, 20 mm.
(H) Analysis of pH2AX, Fra-1, and Egr1 staining levels in T4-2 cells under the indicated conditions: control, +GF (growth factor, 20 ng/mL EGF), or 5 mM erlotinib
and 50 mM carboplatin. n > 5,000 cells per condition. NS, not significant; *p < 0.01 by unpaired t test. Scale bar, 20 mm.
(I) PLSR analysis on MCF10A cells indicating model fit to p-gH2AX signals (predicted versus measured) and parameter weighting by Egr1, Fra-1, erlotinib (EGFRi),
carboplatin, time of fixation (Fix), replicates (Rep), or intraexperiment replicates (Well). n = 56,300.
(J) PLSR analysis on T4-2 cells indicating model fit to p-gH2AX (predicted versus measured) and parameter weighting by Egr1, Fra-1, erlotinib (EGFRi), car-
boplatin, time of fixation (Fix), replicates (Rep), or intra experiment replicates (Well). n = 209,940.
ll
OPEN ACCESS Article
8Cell Systems 11, 1–15, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

responded (20%, Figures 5A and 5C), showing intermittent
pulses of ERK activity. By comparison, 10 ng/mL AREG (func-
tionally equivalent to 2 ng/mL EGF; Harris et al., 2003;Macdon-
ald-Obermann and Pike, 2014) resulted in activation of 80% of
S1 cells (Figures 5B and 5C). However, ERK activity decreased
within 2 h of AREG addition and occurred subsequently as sto-
chastic pulses of activity (Figures 5B and 5D). Treatment with
100 ng/mL AREG resulted in a similar fraction of responding cells
(Figure 5C), but produced a maximal initial response with
continual pulsatile activity (Figure 5B), unlike the sustained activ-
ity produced by 20 ng/mL EGF (Figure 5A). At an even higher
level of AREG (1,000 ng/mL), essentially all cells responded
with a high level of ERKTR
C/N
that was sustained for hours
without pulsing (Figures 5B and 5C). We measured the AREG
secreted into the media by T4-2 cells to be 18.7 ng/mL at equi-
librium, falling within the range of tested conditions, and we
found that AREG protein levels varied between individual T4-2
cells in a population (Figure 5E). Together, these observations
explain the high degree of cell-to-cell ERK signaling variability
in T4-2m and S1co cells, as secreted AREG is both spatially var-
iable and prone to drive pulsatile ERK responses.
We next tested whether the different patterns of ERK activity
induced by EGF and AREG impact the expression of ETGs. At
an early time point (1.5 h) following addition of growth factor,
A
C
DAPI
AREG
T4-2m
FG
E
B
D
0 5 10 15 20 25
Time (hrs)
0.85
0.9
0.95
1
1.05
1.1
Mean ERKTR C/N
S1m +AREG
0 10 100 1000ng/ml
11%
17%
19%
15%
Control
AREG 10ng/mL
AREG 100 ng/mL
EGF 2 ng/mL
EGF 20 ng/mL
0
20
40
60
80
100
% Responders
% Responding cells
AREG 1000 ng/mL
20ng/ml EGF
0
2.5
Control 2ng/ml EGF
S1m ERKTR C/N
2 6 10 14 18
Time (hr)
2 6 10 14 18
Time (hr)
2 6 10 14 18
Time (hr)
mean mean mean
1000ng/ml AREG10ng/ml AREG 100ng/ml AREGControl
0.5
2.0
S1m ERKTR C/N
2 6 10 14 18
Time (hr)
22 2 6 10 14 18
Time (hr)
22 2 6 10 14 18
Time (hr)
22 2 6 10 14 18
Time (hr)
22
mean mean mean mean
0.85
0.9
0.95
1
1.05
1.1
Mean ERKTR C/N
S1m +EGF
0 0.2 2 20ng/ml
10%
18%
19%
16%
0 5 10 15 20 25
Time (hrs)
DAPI Merge
24 Hours (S1m cells)
20 2 0.2
EGF (ng/mL)
AREG (ng/mL)
1000 100 10 Control
0.66
0.47
0.67
0.55
0.43
0.55
Fra-1
81%
84%
76%
60%
75%
76%
Egr1
102%
97%
122%
122%
112%
78%
Fra-1
97%
94%
72%
84%
81%
76%
20 2 0.2
EGF (ng/mL)
Control
DAPI Merge
1.5 Hours (S1m cells)
AREG (ng/mL)
1000 100 10
0.68
0.71
0.77
0.69
0.71
0.88
Egr1
104%
122%
120%
78%
79%
112%
Figure 5. AREG Drives Stochastic ERK Signaling to Induce Heterogeneous ETG Expression
(A and B) Single cell and mean traces of ERKTR
C/N
in S1m cells exposed to the indicated EGFR ligand and concentrations for a duration of >18 h. Vertical red lines
indicate the time of ligand or vehicle addition, the bottom plot shows the mean ERKTR
C/N
in bold, with IQR shaded. Above the mean trace, 5 repres entative single-
cell measurements of ERKTR
C/N
are shown. >500 cells were for each condition.
(C) Bar graphs depict the percent of cells responding to the indicated dose of EGFR ligand within 30 min of treatment. n > 500 cells per condition
(D) Mean ERKTR
C/N
traces for S1 cells receiving the indicated EGF or AREG treatments. Percentages represent the temporal variability score for each condition.
n > 200 cells per condition.
(E) Immunofluorescence imaging of AREG expression in T4-2 cells, AREG (red) and nuclei (blue).
(F and G) Co-staining of Fra-1 and Egr1 in S1m cells at the indicated timepoints and conditions. Percentages indicate the coefficient of variation for Fra-1 or Egr1
respectively. Numbers inset on ‘‘merge’’ images indicate the R
2
value for Fra-1 and Egr1. n > 1,000 cells per condition. Scale bar, 20 mm.
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15, August 26, 2020 9
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

Egr1 and Fra-1 levels increased uniformly as a function of ligand
concentration (Figure 5F), regardless of the stimulating ligand.
By contrast, at a later time point (24 h), ETG expression hetero-
geneity increased (Figure 5G), and the relationships of Fra-1 and
Egr1 to ligand concentration diverged. Fra-1 expression
increased monotonically with the concentration of either ligand,
while Egr1 expression peaked at intermediate ligand concentra-
tions and decreased for ligand concentrations capable of stimu-
lating a sustained ERK response. These results indicate that
pulsed versus sustained ERK activity induce differential re-
sponses in Fra-1, which integrates ERK intensity over time,
and Egr1, which responds transiently and is insensitive to sus-
tained activity.
Paracrine EGFR Signaling Drives Changes in
Transcriptional State on the Scale of Hours
Our results imply temporal variation in ERK-driven expression in
each cell. To explore how gene expression may vary over time,
0
100
200
300
400
Protein conc. (au)
Egr1
Fra-1
0 1 2 3 4 5 6 8 10 12 14 16 18 20 22
Time (hrs)
0
0.5
1
ERK (au)
Egr1
Fra-1
0
100
200
300
400
Protein conc. (au)
Egr1
Fra-1
0123456 8 10 12 14 16 18 20 22
Time (hrs)
0
0.5
1
ERK (au)
Egr1
Fra-1
0
100
200
300
400
Protein conc. (au)
Egr1
Fra-1
0 1 2 3 4 5 6 8 10 12 14 16 18 20 22
Time (hrs)
0
0.5
1
ERK (au)
Egr1
Fra-1
A
G
H
D
E
F
mRNA
ERK
TF TFP
P PP
ØØØ
kpTF
kdTF
kb
kmkØmkØPkØPP
kdP
kpP
kP
KD
τmτP
0
100
200
300
400
500
600
Protein conc. (au)
Egr1
Fra-1
0123456 8 10 121416182022
Time (hrs)
0
0.5
1
ERK (au)
Egr1
Fra-1
BC
Time (hrs)
0
100
200
300
400
Protein conc. (au)
Egr1
Fra-1
0 1 2 3 4 5 6 8 10 12 14 16 18 20 22
0
0.5
1
ERK (au)
Egr1
Fra-1
0
100
200
300
400
Protein conc. (au)
Egr1
Fra-1
0 1 2 3 4 5 6 8 10 12 14 16 18 20 22
Time (hrs)
0
0.5
1
ERK (au)
Egr1
Fra-1
0
100
200
300
Protein conc. (au)
Egr1
Fra-1
0 1 2 3 4 5 6 8 10 12 14 16 18 20 22
Time (hrs)
0
0.5
1
ERK (au)
Egr1
Fra-1
400
I
0
100
200
300
400
500
600
Protein conc. (au)
Egr1
Fra-1
0 1 2 3 4 56 8 10 12 14 16 18 20 22
Time (hrs)
0
0.5
1
ERK (au)
Egr1
Fra-1
J
S1m T4-2m
S1m
S1m
T4-2m
T4-2m
ERK Actvity
1) TAPI +
AREG FBAb
2) EGF 3) MEKi
Measure Egr1/
Fra-1 expression
-2 -1.5 -0.5 02
Time (hrs)
00.5 11.5 2
ERKTRFra-1MERGEDAPI Egr1
Control
ETG Protein Expression
1.3x
Fra-1 Egr1
1.7x
0.8x
1.2x
Control
0 0.5 1 1.5 2
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
Intensity (A.U.)
Control
0 0.5 1 1.5 2
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
Time (hrs) Time (hrs)
Figure 6. Mathematical Simulation of ETG Expression Heterogeneity
(A) Model diagram describing ERK-dependent modification of a transcription factor (TF), expression of mRNA, and a protein (P) product. Superscript P indicates
phosphorylation, clock icons indicate a time delay t, all lowercase k’s indicate rate parameters (see Table S1 for definitions and values used), and uppercase K
indicates a dissociation constant for feedback effects.
(B and C) Simulated responses to hypothetical square wave ERK signals and corresponding ETG expression levels (presented in arbitrary units ).
(B–I) Insets show the trajectory over time in expression space, starting at the circle and ending at the X.
(D–I) Simulated responses to measured single-cell ERK activity traces from S1m (D and F) or T4-2m (G–I) cells in control conditions (D, E, G, and H) or treated with
1,000 ng/mL AREG (F and I). Insets show the modeled trajectory over time in expression space, starting at the circle and ending at the X.
(J) Temporal ETG responses to a single pulse of ERK activity in T4-2 cells. Fixed images were taken at the indicated time points, in hours, and compared relative to
control. Images presented were equally scaled to time point 0 to prevent over-scaling of peak intensity. Graphs correspond with mean ETG intensity and numbers
represent the fold change in ETG intensity relative to control. Scale bar, 20 mM.
ll
OPEN ACCESS Article
10 Cell Systems 11, 1–15, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

1. Spatial heterogeneity
2. Local temporal fluctuation
3. Intracellular amplification
4. Differential filtering
5. Expression program changes
EGFR
ERK
Fra-1
Egr1
c-Fos
P
PP
1.
2.
3.
4.
5.
Expression
time
gene 2
gene 3
gene 1
AREG
time
ERK
time
AREG
A
D
B
E
F
Fra-1 Temporal Variability
10 20 30 40 50 60 70 80
% Cells
0
2
4
6
Mean Derviative/Mean
P=1.065e-21
MCF10A-m
MCF10A-co
Overlap
MCF10A-m
0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
EKAR
FRET Ratio
70:30
50:50
30:70
MCF10A-co : T4-2-co Ratio
*
*
***
40
60
80
100
120
140
Fra-1
Fluorescence (A.U.)
MCF10A-m
70:30
50:50
30:70
MCF10A-co : T4-2-co Ratio
*
*
NS
MCF10A-m
+AREG
70:30 50:50 30:70
Fra-1
mCherry
EKAR
EKAR
ERKTR
Fra-1
mCherry
MCF10A-m
MCF10A-co / T4-2-co C
G
H
Fra-1
Fluorescence (A.U.)
Time (hrs)
MCF10A-m+AREG
30
50
70
90
110
0612
CV= 28%
I
40
50
60
70
80
90
012
MCF10A-m
70:30
50:50
30:70
Mean Fra-1
012
30
50
70
90
110 MCF10A-m Fra-1
CV = 27%
012
30
50
70
90
110 70:30 MCF10A-co Fra-1
CV = 29%
012
30
50
70
90
110 50:50 MCF10A-co Fra-1
CV = 32%
012
Time (hrs)
30
50
70
90
110 30:70 MCF10A-co Fra-1
CV = 35%
Fra-1 Fluorescence (A.U.)
Time (hrs)
40
50
60
70
0612
MCF10A
+AREG
+MEKi
MCF10A-m
Fra-1
Fluorescence (A.U.)
Time (hrs)
60
70
80
90
0612
+AREG
30:70
+MEKi
MCF10A-co
Fra-1
Fluorescence (A.U.)
J
012012
30
180
MCF10A-m MCF10A-co
Time (hrs) Time (hrs)
Fra-1 Fluorescence (A.U.)
mean mean
Figure 7. Live-Cell Imaging of Paracrine-Induced Fra-1 Temporal Expression Variability in Co-culture
(A) Box plots depicting the mean FRET ratio signal in MCF10A cells carrying the ERK reporter, EKAR, in mono-culture (n = 2,428) and in co-culture with T4-2 at the
indicated ratios (70:30 n = 2,526, 50:50 n = 2,014, 30:70 n = 998). Boxes indicate IQR and whiskers the range. *p < 0.01 by unpaired t test.
(B) Still frames of MCF10A EKAR Fra1::mCherry cells in mono- and co-culture. Upper panels depict EKAR expression (turquoise) and tagged endogenous Fra-1
(red); T4-2 cells are not shown in co-culture images. Middle panels show Fra-1::mCherry expression only (red). Lower panels show EKAR expressing 10A cells
(turquoise) and ERKTR expressing T4-2 cells (yellow). Scale bar, 20 mm.
(C) Top plot, mean Fra-1 intensity over time for each ratio of T4-2 co-culture (MCF10A [mono-cultured], 70:30, 50:50, and 30:70 MCF10A to T4-2 ratios) as labeled
on the graph. Below, plots depict the same mean traces with variance (shaded, IQR) and CV (numbers upper right).
(D) Box plots depicting the mean Fra-1 expression levels under the indicated conditions. Boxes indicate the IQR and whiskers the range. MCF-10A (n = 157),
70:30 (n = 182), 50:50 (n = 160), 30:70 (n = 66). *p < 0.01 by unpaired t test.
(E) Mean Fra-1::mCherry expression in mono-cultured, AREG stimulated MCF10A cells. Plot depicts the mean traces with variance (shaded, IQR) and CV.
(F) Fra-1::mCherry expression in mono-cultured, unstimulated MCF10A cells (purple, re-plotted from C), with exogenous AREG at 20 ng/mL (dark gray), and
100 nM MEKi (light gray).
(G) Fra-1::mCherry expression in MCF10A cells co-cultured at a 30:70 ratio of MCF10A to T4-2 cells (orange, re-plotted from C), in comparison with MCF10A cells
at the same co-culture ratio treated with 100 ng/mL AREG (dark gray), or 100 nM MEKi (light gray). See Video S4 for live-cell examples. For panels (A–G), mono-
cultured MCF10A n > 500 cells per condition. Co-cultured MCF10A 70:30 N > 622 cells per condition.
(legend continued on next page)
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15, August 26, 2020 11
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

we constructed a mathematical model extending our previous
work (Gillies et al., 2017). For any given ERK activity time series,
the model estimates target protein expression over time, based
on mRNA and protein decay rates, as well as parameters for
ERK-induced production, stabilization, and negative feedback
(Figure 6A). Based on their observed kinetics, Fra-1 and Egr1
represent biological examples near the extreme behaviors ex-
pected from this model. The slow kinetics and relatively low het-
erogeneity of Fra-1 place it conceptually as a low-pass filter with
a large time constant, which reflects average recent ERK activity.
In contrast, the rapid response of Egr1 and its apparent negative
feedback inhibition under long-term stimulation place it as a
high-pass filter. We constructed a model to represent these
ETG archetypes, selecting example parameters based on previ-
ously published kinetic data for mRNA and protein expression of
Fra-1 and Egr1 (Gillies et al., 2017;Schwanh€
ausser et al., 2011;
Uhlitz et al., 2017). We simulated responses to artificial square
wave inputs (Figures 6B and 6C) and to single-cell ERK traces
measured from S1 and T4-2 cells (Figures 6D–6I). Consistent
with our findings in fixed S1m and T4-2m cells, modeled expres-
sion levels of Fra-1 change slowly, reflecting average recent ERK
activity regardless of rapid fluctuations. Conversely, simulated
Egr1 preferentially rises when ERK activity pulses (Figures 6E,
6F, 6H, and 6I) but remains suppressed in both inactive (Fig-
ure 6D) and highly active (Figure 6G) conditions.
Using these simulations, we plotted the ETG expression space
that a single cell occupies, driven by the stochastic ERK condi-
tions observed in our experiments (Figures 6B–6I, insets). In
both T4-2 and S1 cells, our model demonstrates that stochastic
fluctuations in ERK induce larger excursions in ETG expression
state over time (Figures 6E and 6H), compared with sustained
ERK activity (Figures 6F and 6I). This model demonstration is
consistent with our experimental findings that pulses in ERK ac-
tivity lead to rapid changes in ETG expression (<30 min) followed
by decay in protein levels (Figure 6J). Thus indicating that para-
crine-induced expression is not persistent and instead highly
sensitive to temporal changes in signaling. These findings illus-
trate how pulses in a single microenvironmental signal can
enable cells to take on a range of expression states, as repre-
sented by two differently responding genes.
To directly observe the ETG fluctuations predicted by the
model, we used MCF10A cells carrying an mCherry fusion at
the endogenous FOSL1 locus (Fra-1::mCherry) and the ERK
sensor EKAR3 (Gillies et al., 2017)(Harvey et al., 2008). Thus,
ERK activity and the expression of Fra-1 protein can be visual-
ized in a cell line that acts as a ‘‘receiver,’’ similar to S1 cells,
for paracrine signals from T4-2 cells (Figure 7A; Videos S4A
and S4B). The mean and variance of Fra-1::mCherry expression
under each condition were consistent with the level of EKAR
activity (Figures 7A–7D). While 20 ng/mL AREG treatment
increased the mean Fra-1 intensity in MCF10Am cells by 24%
(Figure 7E), relative variation remained essentially constant
(CV = 27% untreated, 28% AREG-treated, Figures 7E and 7C),
which is consistent with the sustained ERK response to AREG
in MCF10A (Gillies et al., 2017). However, in co-cultured cells,
mean Fra-1::mCherry intensity was similar to the maximum
induced by AREG (Figures 7F and 7G) but displayed a higher de-
gree of variation (Figures 7C and 7D, CV = 35% at 30:70 ratio). At
the single-cell level, temporal fluctuations in Fra-1::mCherry
expression were evident under co-culture conditions, in compar-
ison to mono-cultured cells (Figure 7H). To exclude high fre-
quency noise, we filtered our data to focus on changes consis-
tent with expression regulation (>30 min). We then calculated
the mean derivative, normalized by the mean Fra-1::mCherry
intensity on a single-cell basis, to measure variance of
Fra-1::mCherry expression over time (Fra-1 temporal variability
index) in mono- versus co-culture. By this measure, time-depen-
dent variability in Fra-1::mCherry increased by 20% in co-culture
relative to mono-culture (Figure 7I, p = 1.065e21). Thus, consis-
tent with model, Fra-1 expression varies dynamically in individual
cells over the scale of hours, resulting in the observed heteroge-
neity in Fra-1 expression.
DISCUSSION
By exploring diverse cellular phenotypes, tumor cells gain a se-
lective survival advantage in adverse physiological environments
(Michor and Polyak, 2010). Genetic mutations are important
drivers of tumor cell diversification but are constrained to oper-
ate on timescales of weeks or longer. Immediate cell survival in
response to the time-varying stress conditions within a tumor
may require a more rapid and flexible means of diversification.
Our analysis demonstrates how such variation can be generated
by the EGFR-RAS-ERK signaling pathway at levels equivalent to
those observed in human breast cancer tissues. This expression
variance results from a specific mode of operation in which het-
erogeneity in the TME is amplified by intracellular signal process-
ing in the EGFR-RAS-ERK pathway to drive fluctuation of cancer
related genes on the timescale of hours (Figure 7J). While this
system is simplified relative to actual tumors, these data estab-
lish a potentially important source of variation in the expression
of key tumor genes, driven by spatial variation within the TME.
The mixture of malignant (T4-2) and non-malignant (S1 or
MCF10A) cells within our co-culture model represents an
approximation of the cellular and genetic diversity within a tu-
mor, where clones containing different mutational burdens
would interact. Although the arrangement of the cells in vitro
does not specifically replicate the cellular organization of a tu-
mor, we expect AREG gradients to occur on similar length scales
in vivo because tumor cells are likely to express AREG heteroge-
neously, as we observe for T4-2 cells. The proficiency of T4-2
cells to self-stimulate through AREG secretion is consistent
with previous work (Fischer et al., 2003;Madsen et al., 1992),
and we find that this mode of RAS-ERK signaling induces
both a high magnitude of ERK activity as well as time-dependent
variation that propagates to the level of gene expression.
(H) Fra-1::mCherry fluorescence over time (12 h) in individual MCF10A-m (purple) or MCF10A-co cells (orange). The bottom trace represents mean Fra-1 signal
with IQR shaded; the traces above depict representative single-cell traces of Fra-1 expression over time. N > 700 cells per condition.
(I) Distribution of temporal variability in individual cells, defined as the absolute value of the derivative Fra-1 signal divided by the mean Fra-1 signal for each cell.
(J) A model of heterogeneity generation by paracrine-driven EGFR-RAS-ERK signaling. Molecular processes are shown at left, and the information processing
function performed by each biochemical step is shown at right (numbered 1–5).
ll
OPEN ACCESS Article
12 Cell Systems 11, 1–15, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

Co-cultured non-malignant cells respond with similar character-
istics of RAS-ERK activity and ETG expression, even in the
absence of activating mutations in the RAS-ERK pathway. Within
our experimental system, the low level of constitutive ERK
signaling and high capacity for stimulation of these cells makes
them ideal receivers to quantify the paracrine stimuli from the
nearby malignant cells. An intriguing but unresolved question is
whether, within the context of a real tumor, such receiver cells
represent a population that is important for tumor progression.
In principle, such cells, supported by paracrine signals from
more highly mutated cells nearby, could represent a reservoir
of uncommitted possibilities with higher adaptive potential than
heavily mutated cells that cannot tolerate further genetic alter-
ations. Such flexibility could play a role in tumor cell survival
when new stresses, such as cytotoxic chemotherapeutics, are
encountered.
The general process we demonstrate, by which spatial fluctu-
ations in paracrine ligands are amplified and converted to tem-
poral variability in gene expression, may also apply in other
signaling pathways, such as IL-6 (Hartman et al., 2013;He
et al., 2013) or TGF-b(Oft et al., 1996;Wang et al., 2014). How-
ever, the kinetic parameters of the AREG-EGFR-RAS-ERK
pathway appear especially well suited for generating global di-
versity in gene expression. The amplification characteristics of
the EGFR pathway are capable of driving high-intensity ERK ac-
tivity bursts in response to even sub-saturating concentrations of
ligand, and the attenuation of these pulses by negative feedback
is rapid, resulting in ERK output that magnifies external changes
in AREG concentration. Subsequently, the genes responding to
ERK activity do so with a wide range of kinetic properties (Uhlitz
et al., 2017), continuously translating the temporal dynamics of
the pathway into different gene expression profiles. It is intriguing
to speculate that the prevalence and potency of the RAS-ERK
pathway in tumorigenesis may be a consequence not only of
its target functions (as other, less prevalent pathways, also stim-
ulate cell proliferation and migration) but also of its ability to act
as a generator of cellular diversity. It is also possible that the ca-
pacity to diversify gene expression profiles may be an important
aspect of the function of RAS-ERK signaling within the develop-
mental processes in which it acts. In vivo data indicate that tran-
sient gradients of EGFR ligand signaling on the scale of 50 mm
drive pulsatile ERK activity in C. elegans vulval patterning (de la
Cova et al., 2017), or in the mouse epidermis (Hiratsuka et al.,
2015). These dynamic patterns could create variation in ERK-
controlled gene expression similar to that observed here, which
could play a role in determining cell fate (Hamilton et al., 2019). A
key question yet to be resolved is how the ligand gradients within
a tumor differ quantitatively from those in normal developing tis-
sues. Our data suggest that ETGs, such as EGR1, FOSL1, MYC,
and FOS, could help to address this question by creating an
expression signature that reflects the dynamic status of ERK
signaling, which may be useful in identifying EGFR signaling gra-
dients in vivo using transcriptional profiling.
Our system makes it possible to model and quantify EGFR-
mediated generation of gene expression variability in vitro.
However, it by no means recapitulates the full complexity of tu-
mor physiology. Many additional sources of variability are pre-
sent in the TME, including other cytokines, metabolites, ions,
and ECMs. Futhermore, many genes other than Egr1 and Fra-1
contribute to tumor cell survival and drug response. Nonethe-
less, this experimental system can be used to further investigate
particular mechanisms behind spatial heterogeneity and to eval-
uate applicable candidate inhibitors for their ability to reduce tu-
mor cell heterogeneity, a property that is likely closely related to
their efficacy (Shea et al., 2018), but for which no direct measure-
ment has yet been available. It remains important to determine
which forms of gene expression heterogeneity play a functional
role in cancer cell spread or survival and which forms are simply
an epiphenomenon of deregulated tumor cell states. Our simula-
tions characterize the behavior of cells within a two-gene expres-
sion landscape, but overall transcriptional variability occurs in a
highly multidimensional space. An essential, though challenging,
next step relies on the effort to identify, at the single-cell level, the
relationship between variation in these many dimensions and
relevant cellular phenotypes. We expect that the data and model
presented here will help to guide experiments in which heteroge-
neity is manipulated in vivo on the appropriate time and length
scales.
Key Changes Prompted by Reviewer Comments
In response to the reviewer comments, we added additional
analysis of single-cell gene expression data and performed anal-
ysis of DNA damage responses in relation to ETG expression.
We reorganized the presentation of the computational model
and reworded our interpretation regarding the correlation be-
tween Fra-1 and DNA damage response to indicate that they co-
vary and a direct relationship cannot be established in the cur-
rent context. We also added a section to the STAR Methods
denoting the cell density and conditions used to determine the
concentration of free AREG in our experiments by ELISA. For
context, the complete transparent peer review record is included
within the Supplemental Information.
STAR+METHODS
Detailed methods are provided in the online version of this paper
and include the following:
dKEY RESOURCES TABLE
dRESOURCE AVAILABILITY
BLead Contact
BMaterials Availability
BData and Code Availability
dEXPERIMENTAL MODEL AND SUBJECT DETAILS
BCell Culture and Media
BHuman Tissue Samples
dMETHOD DETAILS
BReporter Line Construction
BLive-Cell Microscopy and Co-culture Conditions
BImmunofluorescence Microscopy
BDNA Damage Response Assay
BAmphiregulin ELISA
dQUANTIFICATION AND STATSTICAL ANALYSIS
BImaging, Data Processing, and Statistics and Normal-
ization
BDynamic Modeling
BSingle Cell RNA Sequencing and Analysis
BSequencing Strategy
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15, August 26, 2020 13
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

BReference for Read Alignment
BDrop-seq Pipeline and Generation of the Gene Expres-
sion Matrix
BSpecies Mixing Experiment
BCell QC and Clustering
BExcess Variance Calculation and Delta-Excess
Variance
SUPPLEMENTAL INFORMATION
Supplemental Information can be found online at https://doi.org/10.1016/j.
cels.2020.07.004.
ACKNOWLEDGMENTS
Funding for this work was provided by an American Association for Cancer
Research Stand Up To Cancer Innovative Research Grant (SU2C-AACR-
IRG-01-16) and by the National Institute for General Medical Sciences (R01-
GM115650) to J.G.A. Stand up to Cancer (SU2C) is a program of the Entertain-
ment Industry Foundation. Research grants are administered by the American
Association for Cancer Research, the scientific partner of SU2C. A.E.D. was
funded by the California Institute of Regenerative Medicine/ Children’s Hospi-
tal of Oakland Clinical Research Fellowship and the Ohio State University
Comprehensive Cancer Center and the National Institutes of Health under
grant number (P30 CA016058). Funding was also provided by NSF CAREER
award 1846559 to G.Q. Y.C. was supported by an NIH T32 training grant in Mo-
lecular and Cellular Biology grant number T32 GM007377 from NIH NIGMS.
A.E.D. and M.J.B. were also funded by the Breast Cancer Research Founda-
tion and the Woodland Fund. We would like to thank the members of the Bis-
sell, Juliano, Quon, and Albeck Labs for their helpful suggestions on the
manuscript.
AUTHOR CONTRIBUTIONS
A.E.D., J.G.A., and M.J.B. conceptualized the study, analyzed data, and wrote
the manuscript. A.E.D., S.J.T., and A.R.R. performed imaging experime nts.
S.S. performed single-cell RNA sequencing and analysis with assistance
from A.E.D. A.E.D. and T.E.G. performed data processing and statistical anal-
ysis. M.P. performed statistical analysis and model generation. G.Q. and Y.C.
performed excess variance calculations and statistical analysis on single-cell
sequencing data. M.J.B., V.M., and C.J. analyzed data and contributed to
writing the manuscript.
DECLARATION OF INTERESTS
The authors declare no competing interests.
Received: October 11, 2019
Revised: May 6, 2020
Accepted: July 2, 2020
Published: July 28, 2020
REFERENCES
Albeck, J.G., Mills, G.B., and Brugge, J.S. (2013). Frequency-modulated
pulses of ERK activity transmit quantitative proliferation signals. Mol. Cell 49,
249–261.
Amit, I., Citri, A., Shay, T., Lu, Y., Katz, M., Zhang, F., Tarcic, G., Siwak, D.,
Lahad, J., Jacob-Hirsch, J., et al. (2007). A module of negative feedback reg-
ulators defines growth factor signaling. Nat. Genet. 39, 503–512.
Aoki, K., Kumagai, Y., Sakurai, A., Komatsu, N., Fujita, Y., Shionyu, C., and
Matsuda, M. (2013). Stochastic ERK activation induced by noise and cell-to-
cell propagation regulates cell density-dependent proliferation. Mol. Cell 52,
529–540.
Belguise, K., Kersual, N., Galtier, F., and Chalbos, D. (2005). FRA-1 expression
level regulates proliferation and invasiveness of breast cancer cells. Oncogene
24, 1434–1444.
Berns, E.M., Klijn, J.G., van Putten, W.L., van Staveren, I.L., Portengen, H., and
Foekens, J.A. (1992). c-myc amplification is a better prognostic factor than
HER2/neu amplification in primary breast cancer. Cancer Res. 52, 1107–1113.
Briand, P., Petersen, O.W., and Van Deurs, B. (1987). A new diploid nontumori-
genic human breast epithelial cell line isolated and propagated in chemically
defined medium. In Vitro Cell Dev. Biol. 23, 181–188.
Brooks, M.D., Burness, M.L., and Wicha, M.S. (2015). Thera peutic implications
of cellular heterogeneity and plasticity in breast cancer. Cell Stem Cell 17,
260–271.
Bugaj, L.J., Sabnis, A.J., Mitchell, A., Garbarino, J.E., Toettcher, J.E., Bivona,
T.G., and Lim, W.A. (2018). Cancer mutations and targeted drugs can disrupt
dynamic signal encoding by the RAS-Erk pathway. Science 361, eaao3048.
Cazet, A.S., Hui, M.N., Elsworth, B.L., Wu, S.Z., Roden, D., Chan, C.L.,
Skhinas, J.N., Collot, R., Yang, J., Harvey, K., et al. (2018). Targeting stromal
remodeling and cancer stem cell plasticity overcomes chemoresistance in tri-
ple negative breast cancer. Nat. Commun. 9, 2897.
Cong, Le, Ran, F. Ann, Lin, Shuailiang, Barretto, Robert, Naomi, Habib, Hsu,
Patrick D., Wu, Xuebing, Jiang, Wenyan, Marraffini, Luciano A., and Zhang,
Feng (2013). Multiplex genome engineering using CRISPR/Cas systems.
Science 339, 819–823.
Davies, A.E., and Albeck, J.G. (2018). Microenvironmental signals and
biochemical information processing: cooperative determinants of intratumoral
plasticity and heterogeneity. Front. Cell Dev. Biol. 6, 44.
DeWitt, A., Iida, T., Lam, H.Y., Hill, V., Wiley, H.S., and Lauffenburger, D.A.
(2002). Affinity regulates spatial range of EGF receptor autocrine ligand bind-
ing. Dev. Biol. 250, 305–316.
de la Cova, Claire, Townley, Robert, Regot, Sergi, and Greenwald, Iva (2017). A
Real-Time Biosensor for ERK Activity Reveals Signaling Dynamics during C.
elegans Cell Fate Specification. Dev. Cell 42, 542–553.
DeWitt, A.E., Dong, J.Y., Wiley, H.S., and Lauffenburger, D.A. (2001).
Quantitative analysis of the EGF receptor autocrine system reveals cryptic
regulation of cell response by ligand capture. J. Cell Sci. 114, 2301–2313.
Fischer, O.M., Hart, S., Gschwind, A., and Ullrich, A. (2003). EGFR signal trans-
activation in cancer cells. Biochem. Soc. Trans. 31, 1203–1208.
Friedl, P., and Alexander, S. (2011). Cancer invasion and the microenviron-
ment: plasticity and reciprocity. Cell 147, 992–1009.
Gao, R., Davis, A., McDonald, T.O., Sei, E., Shi, X., Wang, Y., Tsai, P.C.,
Casasent, A., Waters, J., Zhang, H., et al. (2016). Punctuated copy number
evolution and clonal stasis in triple-negative breast cancer. Nat. Genet. 48,
1119–1130.
Gillies, T.E., Pargett, M., Minguet, M., Davies, A.E., and Albeck, J.G. (2017).
Linear integration of ERK activity predominates over persistence detection in
Fra-1 regulation. Cell Syst. 5, 549–563.e5.
Gupta, P.B., Fillmore, C.M., Jiang, G., Shapira, S.D., Tao, K., Kuperwasser, C.,
and Lander, E.S. (2011). Stochastic state transitions give rise to phenotypic
equilibrium in populations of cancer cells. Cell 146, 633–644.
Hamilton, William B., Mosesson, Yaron, Monteiro, Rita S., Emdal, Kristina B.,
Knudsen, Teresa E., Francavilla, Chiara, Barkai, Naama, Olsen, Jesper, and
Brickman, Joshua M. (2019). Dynamic lineage priming is driven via direct
enhancer regulation by ERK. Nature 575, 355–360.
Handly, L.N., Pilko, A., and Wollman, R. (2015). Paracrine communication
maximizes cellular response fidelity in wound signaling. eLife 4, e09652.
Harris, R.C., Chung, E., and Coffey, R.J. (2003). EGF receptor ligands. Exp.
Cell Res. 284, 2–13.
Hartman, Z.C., Poage, G.M., den Hollander, P., Tsimelzon, A., Hill, J.,
Panupinthu, N., Zhang, Y., Mazumdar, A., Hilsenbeck, S.G., Mills, G.B., and
Brown, P.H. (2013). Growth of triple-negative breast cancer cells relies upon
coordinate autocrine expression of the proinflammatory cytokines IL-6 and
IL-8. Cancer Res. 73, 3470–3480.
Harvey, Christopher D., Ehrhardt, Anka G., Cellurale, Cristina, Zho ng, Haining,
Yasuda, Ryohei, Davis, Roger J., and Svoboda, Karel (2008). A genetically en-
coded fluorescent sensor of ERK activity. PNAS 105, 19264–19269.
He, G., Dhar, D., Nakagawa, H., Font-Burgada, J., Ogata, H., Jiang, Y.,
Shalapour, S., Seki, E., Yost, S.E., Jepsen, K., et al. (2013). Identification of
ll
OPEN ACCESS Article
14 Cell Systems 11, 1–15, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

liver cancer progenitors whose malignant progression depends on autocrine
IL-6 signaling. Cell 155, 384–396.
Hiratsuka, T., Fujita, Y., Naoki, H., Aoki, K., Kamioka, Y., and Matsuda, M.
(2015). Intercellular propagation of extracellular signal-regulated kinase acti-
vation revealed by in vivo imaging of mouse skin. eLife 4, e05178.
Hochgr€
afe, F., Zhang, L., O’Toole, S.A., Browne, B.C., Pinese, M., Porta
Cubas, A., Lehrbach, G.M., Croucher, D.R., Rickwood, D., Boulghourjian, A.,
et al. (2010). Tyrosine phosphorylation profiling reveals the signaling network
characteristics of Basal breast cancer cells. Cancer Res. 70, 9391–9401.
Hochman-Mendez, C., Cantini, M., Moratal, D., Salmeron-Sanchez, M., and
Coelho-Sampaio, T. (2014). A fractal nature for polymerized laminin. PLoS
One 9, e109388.
Janes, Kevin A., Wang, Chun-Chao, Holmberg, Karin J., Cabral, Kristin, and
Brugge, Joan S. (2010). Identifying single-cell molecular programs by stochas-
tic profiling. Nature Methods 7, 311–317.
Jaqaman, Khuloud, Loerke, Dinah, Mettlen, Marcel, Kuwata, Hirotaka,
Grinstein, Sergio, Schmid, Sandra L., and Danuser, Gaudenz (2008). Robust
single-particle tracking in live-cell time-lapse sequences. Nat. Methods 5,
695–702.
Kenny, P.A., and Bissell, M.J. (2007). Targeting TACE-dependent EGFR ligand
shedding in breast cancer. J. Clin. Invest. 117, 337–345.
Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. (2009). Ultrafast and
memory-efficient alignment of short DNA sequences to the human genome.
Genome Biol. 10, R25.
Liberzon, Arthur, Birger, Chet, Thorvaldsdo
´ttir, Helga, Ghandi, Mahmoud,
Mesirov, Jill P., and Tamayo, Pablo (2015). The Molecular Signatures
Database Hallmark Gene Set Collection. Cell Systems 1, 417–425.
Lu, H., Clauser, K.R., Tam, W.L., Fro
¨se, J., Ye, X., Eaton, E.N., Reinhardt, F.,
Donnenberg, V.S., Bhargava, R., Carr, S.A., and Weinberg, R.A. (2014). A
breast cancer stem cell niche supported by juxtacrine signalling from mono-
cytes and macrophages. Nat. Cell Biol. 16, 1105–1117.
Macdonald-Obermann, J.L., and Pike, L.J. (2014). Different epidermal growth
factor (EGF) receptor ligands show distinct kinetics and biased or partial ago-
nism for homodimer and heterodimer formation. J. Biol. Chem. 289,
26178–26188.
Macosko, E.Z., Basu, A., Satija, R., Nemesh, J., Shekhar, K., Goldman, M.,
Tirosh, I., Bialas, A.R., Kamitaki, N., Martersteck, E.M., et al. (2015). Highly par-
allel genome-wide expression profiling of individual cells using nanoliter drop-
lets. Cell 161, 1202–1214.
Madsen, M.W., Lykkesfeldt, A.E., Laursen, I., Nielsen, K.V., and Briand, P.
(1992). Altered gene expression of c-myc, epidermal growth factor receptor,
transforming growth factor-alpha, and c-erb-B2 in an immortalized human
breast epithelial cell line, HMT-3522, is associated with decreased growth fac-
tor requirements. Cancer Res. 52, 1210–1217.
Meacham, C.E., and Morrison, S.J. (2013). Tumour heterogeneity and cancer
cell plasticity. Nature 501, 328–337.
Michor, F., and Polyak, K. (2010). The origins and implications of intratumor
heterogeneity. Cancer Prev. Res. (Phila) 3, 1361–1364.
Nakakuki, T., Birtwistle, M.R., Saeki, Y., Yumoto, N., Ide, K., Nagashima, T.,
Brusch, L., Ogunnaike, B.A., Okada-Hatakeyama, M., and Kholodenko, B.N.
(2010). Ligand-specific c-Fos expression emerges from the spatiotemporal
control of ErbB network dynamics. Cell 141, 884–896.
Nguyen, A., Yoshida, M., Goodarzi, H., and Tavazoie, S.F. (2016). Highly var-
iable cancer subpopulations that exhibit enhanced transcriptome variability
and metastatic fitness. Nat. Commun. 7, 11246.
Nunns, H., and Goentoro, L. (2018). Signaling pathways as linear transmitters.
eLife 7, e33617.
Oft, M., Peli, J., Rudaz, C., Schwarz, H., Beug, H., and Reichmann, E. (1996).
TGF-beta1 and Ha-RAS collaborate in modulating the phenotypic plasticity
and invasiveness of epithelial tumor cells. Genes Dev. 10, 2462–2477.
Pargett, M., and Albeck, J.G. (2018). Live-cell imaging and analysis with mul-
tiple genetically encoded reporters. Curr. Protoc. Cell Biol. 78, 4.36.1–4.36.19.
Pargett, M., Gillies, T.E., Teragawa, C.K., Sparta, B., and Albeck, J.G. (2017).
Single-cell imaging of ERK signaling using fluorescent biosensors. Methods
Mol. Biol. 1636, 35–59.
Rand, U., Rinas, M., Schwerk, J., No
¨hren, G., Linnes, M., Kro
¨ger, A., Flossdorf,
M., Ka
´ly-Kullai, K., Hauser, H., Ho
¨fer, T., and Ko
¨ster, M. (2012). Multi-layered
stochasticity and paracrine signal propagation shape the type-I interferon
response. Mol. Syst. Biol. 8, 584.
Regot, S., Hughey, J.J., Bajar, B.T., Carrasco, S., and Covert, M.W. (2014).
High-sensitivity measurements of multiple kinase activities in live single cells.
Cell 157, 1724–1734.
Reis-Filho, J.S., Pinheiro, C., Lambros, M.B., Milanezi, F., Carvalho, S.,
Savage, K., Simpson, P.T., Jones, C., Swift, S., Mackay, A., et al. (2006).
EGFR amplification and lack of activating mutations in metaplastic breast car-
cinomas. J. Pathol. 209, 445–453.
Rizki, A., Weaver, V.M., Lee, S.Y., Rozenberg, G.I., Chin, K., Myers, C.A.,
Bascom, J.L., Mott, J.D., Semeiks, J.R., Grate, L.R., et al. (2008). A human
breast cell model of preinvasive to invasive transition. Cancer Res. 68,
1378–1387.
Satija, R., Farrell, J.A., Gennert, D., Schier, A.F., and Regev, A. (2015). Spatial
reconstruction of single-cell gene expression data. Nat. Biotechnol. 33,
495–502.
Schwanh€
ausser, B., Busse, D., Li, N., Dittmar, G., Schuchhardt, J., Wolf, J.,
Chen, W., and Selbach, M. (2011). Global quantification of mammalian gene
expression control. Nature 473, 337–342.
Shalek, A.K., Satija, R., Shuga, J., Trombetta, J.J., Gennert, D., Lu, D., Chen,
P., Gertner, R.S., Gaublomme, J.T., Yosef, N., et al. (2014). Single-cell RNA-
seq reveals dynamic paracrine control of cellular variation. Nature 510,
363–369.
Sharma, S.V., Lee, D.Y., Li, B., Quinlan, M.P., Takahashi, F., Maheswaran, S.,
McDermott, U., Azizian, N., Zou, L., Fischbach, M.A., et al. (2010). A chro-
matin-mediated reversible drug-tolerant state in cancer cell subpopulations.
Cell 141, 69–80.
Shea, M.P., O’Leary, K.A., Fakhraldeen, S.A., Goffin, V., Friedl, A., Wisinski,
K.B., Alexander, C.M., and Schuler, L.A. (2018). Antiestrogen therapy in-
creases plasticity and cancer stemness of prolactin-induced ERa+ mammary
carcinomas. Cancer Res. 78, 1672–1684.
Sparta, B., Pargett, M., Minguet, M., Distor, K., Bell, G., and Albeck, J.G.
(2015). Receptor level mechanisms are required for epidermal growth factor
(EGF)-stimulated extracellular signal-regulated kinase (ERK) activity pulses.
J. Biol. Chem. 290, 24784–24792.
Stingl, J., and Caldas, C. (2007). Molecular heterogeneity of breas t carcinomas
and the cancer stem cell hypothesis. Nat. Rev. Cancer 7, 791–799.
Tam, W.L., Lu, H., Buikhuisen, J., Soh, B.S., Lim, E., Reinhardt, F., Wu, Z.J.,
Krall, J.A., Bierie, B., Guo, W., et al. (2013). Protein kinase C alpha is a central
signaling node and therapeutic target for breast cancer stem cells. Can cer Cell
24, 347–364.
Tam, W.L., and Weinberg, R.A. (2013). The epigenetics of epithelial-mesen-
chymal plasticity in cancer. Nat. Med. 19, 1438–1449.
Tyson, J.J., Chen, K.C., and Novak, B. (2003). Sniffers, buzzers, toggles and
blinkers: dynamics of regulatory and signaling pathways in the cell. Curr.
Opin. Cell Biol. 15, 221–231.
Uhlitz, F., Sieber, A., Wyler, E., Fritsche-Guenther, R., Meisig, J., Landthaler,
M., Klinger, B., and Bl€
uthgen, N. (2017). An immediate-late gene expression
module decodes ERK signal duration. Mol. Syst. Biol. 13, 928.
Wang, C.C., Bajikar, S.S., Jamal, L., Atkins, K.A., and Janes, K.A. (2014). A
time- and matrix-dependent TGFBR3-JUND-KRT5 regulatory circuit in single
breast epithelial cells and basal-like premalignancies. Nat. Cell Biol. 16,
345–356.
Wilson, M.Z., Ravindran, P.T., Lim, W.A., and Toettcher, J.E. (2017). Tracing
information flow from erk to target gene induction reveals mechanisms of dy-
namic and combinatorial control. Mol. Cell 67, 757–769.e5.
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15, August 26, 2020 15
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

STAR+METHODS
KEY RESOURCES TABLE
REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
Anti-Fra-1, clone C-12 Santa Cruz Biotechnology Cat#sc28310; RRID: AB_627632
Anti-Egr-1, clone 15F7 Cell signaling Cat#4153; RRID: AB_2097038
Anti-c-Fos, clone 9F6 Cell signaling Cat#2250; RRID: AB_2247211
Anti-c-Myc Cell Signaling Cat#9402; RRID: AB_2151827
Anti-pH2AX Millipore Sigma Cat#05-636; RRID: AB_2755003
TGFa function blocking antibody R&D Systems Cat#AF-239; RRID: AB_2201779
Amphiregulin function blocking antibody R&D Systems Cat#MAB262; RRID: AB_2060676
Biological Samples
Human Invasive Ductal Carcinoma Comparative Human Tissue Network, NCI De-identified
Chemicals, Peptides, and Recombinant Proteins
Epidermal growth factor Peprotech Cat#AF-100-15
Amphiregulin Peprotech Cat# 100-55B
TAPI-0 Tocris Cat#5523
Erlotinib Selleck Biochemicals Cat#S1023
Gefitinib (ZD1839) Selleck Biochemicals Cat#S1025
PD0325901 Selleck Biochemicals Cat#S1036
Sodium Selenite Sigma-Aldrich Cat#S5261
Beta-Estradiol Sigma-Aldrich Cat#E2758
Transferrin Sigma-Aldrich Cat#T8158
Prolactin Sigma-Aldrich Cat#L6520
Cholera Toxin Sigma-Aldrich Cat#C8052
Hydrocortisone Sigma-Aldrich Cat#H0888
Insulin Sigma-Aldrich Cat#I9278
Bovine Serum Albumin Sigma-Aldric h Cat#A7906
Heat Inactivated Horse Serum Life Technologies Cat#26050
10mM Dragon Green Bangs Laboratory Cat# FC06F-10163
Puromycin Life Technologies Cat#A113803
Laminin-111 Life Technologies Cat#23017015
Fugene HD Promega Cat#E2311
Cell Culture Media
DMEM/F-12 1:1 Life Technologies Cat#11320
Critical Commercial Assays and Sequencing Regents
Amphiregulin ELISA R&D Systems Cat#DY262
DSQ 3x9 array microfluidic device Nanoshift N/A
Barcoded Beads SeqB ChemGenes Corp. MACOSKO-2011-10B
Agencourt AMPure XP beads Beckman Coulter Cat#A63881
Nextera XT (Illumina) sample preparation kit Illumina FC-131-1024
Deposited Data
Raw sequencing data GEO GSE118312
Experimental Models: Cell Lines
Human: MCF-10A, clone 5E Joan Brugge, Harvard Medical School
(Janes et al., 2010)
RRID:CVCL_0598
Human: 5E/Fra1::mCherry/EKAR3 (Gillies et al., 2017) Available on request
(Continued on next page)
ll
OPEN ACCESS Article
e1 Cell Systems 11, 1–15.e1–e5, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

RESOURCE AVAILABILITY
Lead Contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, John
Albeck (jgalbeck@ucdavis.edu).
Materials Availability
Plasmids generated in this study are forthcoming to Addgene. Plasmids will be are available upon request from the Lead Contacts.
Data and Code Availability
MATLAB code and R markdown files are available at https://www.mcb.ucdavis.edu/faculty-labs/albeck/data.htm. Imaging data re-
quests will be fulfilled by the Lead Contacts. Sequencing data have been uploaded to the GEO repository: GEO accession
GSE118312.
EXPERIMENTAL MODEL AND SUBJECT DETAILS
Cell Culture and Media
HMT-3522 cell lines were maintained in DMEM/ F12 supplemented with prolactin, insulin, sodium selenite, hydrocortisone, b-estro-
gen, transferrin, and EGF (S1 only), as previously described (Kenny and Bissell, 2007). MCF10A-5E cell lines were maintained in
DMEM/F12 media supplemented with 5% horse serum, insulin, cholera toxin, hydrocortisone, and EGF. All cell lines were maintained
in 5% C02 at 37C.
Human Tissue Samples
De-identified human breast cancer tissue samples were provided by the Cooperative Human Tissue Network, as described in STAR
Methods, in accordance with IRB protocols. Other investigators may have received samples from these same tissue specimens.
METHOD DETAILS
Reporter Line Construction
Reporter cell lines were created using lentiviral transduction from plasmid DNA containing the ERK Translocation Reporter (ERKTR)
(Regot et al., 2014) fused to fluorescent proteins as indicated below. T4-2 reporter lines were constructed with ERKTR-mVenus. S1
Continued
REAGENT or RESOURCE SOURCE IDENTIFIER
HMT-3522 Mina Bissell, Lawrence Berkeley National
Laboratory (Briand et al., 1987;
Rizki et al., 2008)
Available on request
Recombinant DNA
Plasmid: pPBJ-EKAR3nes-puro (Sparta et al., 2015) Addgene # forthcoming
Plasmid: pLJM1-ERKTR-mCherry-puro This paper Addgene # forthcoming
Plasmid: pLJM1-ERKTR-mTurquoise2-puro This paper Addgene # forthcoming
Plasmid: pLJM1-ERKTR-mVenus-puro This paper Addgene # forthcoming
Plasmid: pAAV-Fra1-mCherry (Gillies et al., 2017) Addgene # forthcoming
pX330-U6-Chimeric_BB-CBh-hSpCas9 (Cong et al., 2013) Addgene #
42230
Plasmid: pX330-FOSL1 (Gillies et al., 2017) Addgene # forthcoming
Software and Algorithms
NIS-Elements AR ver. 4.20 Nikon RRID:SCR_014329
Bio-Formats ver. 5.1.1 (May 2015) OME RRID:SCR_000450
uTrack 2.0 (Jaqaman et al., 2008)http://www.utsouthwestern.edu/labs/
danuser/software/
MATLAB Mathworks SCR_001622
Seurat Satija Laboratory https://satijalab.org/seurat/
R Studio R Studio https://rstudio.com/
Other
Glass Bottom Plates, #1.5 cover glass In Vitro Scientific Cat#P24-1.5H-N, P96-1.5H-N
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15.e1–e5, August 26, 2020 e2
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

reporter lines were constructed using ERKTR-mTurquoise2 or ERKTR-mCherry. Reporter expressing cells were selected with puro-
mycin followed by sorting using flow cytometry. Flow sorting was conducted using a wide gating strategy to maximally preserve the
inherent heterogeneity of S1 and T4-2 cell populations. MCF10A-EKAR-Fra-1::mCherry cell lines were generated previously using
the EKAR3 reporter inserted into MCF10A cells carrying mCherry fused to the C terminus of endogenous FOSL1 (Fra-1) locus (Gillies
et al., 2017).
Live-Cell Microscopy and Co-culture Conditions
Live-cell microscopy was conducted on multi-well plates with #1.5 glass bottoms that were coated with laminin-111 (Invitrogen,
Carlsbad, CA). 50ug/ml Laminin-111 was deposited onto glass wells overnight in 20mM sodium acetate buffer containing 1mM
CaCl
2
to generate a fractal laminin-coated surface (Hochman-Mendez et al., 2014). Immediately prior to plating, the wells were
washed once with PBS to remove excess buffer and laminin. For mono-culture experiments, cells were plated at a density of
9000 cells/ well for both S1 and T4-2 cells. For co-culture experiments, total cell density was kept constant and the ratio of
S1:T4-2 cells adjusted for a particular experiment (e.g. 30:70 S1 to T4-2 ratio corresponds to a combination of 2700 and 6300 cells
per well, respectively). Plates were then incubated for 24 h in complete media. After 24 h, plates were washed twice in media con-
taining no additives then placed in custom imaging media (DMEM/ F12 without phenol red, riboflavin, and folic acid) containing hy-
drocortisone, b-estrogen, transferrin, and Hoechst stain (1:100,000), then incubated overnight prior to live-cell microscopy. Following
preparation, plates were imaged on a Nikon Ti-E inverted microscope fitted with an environmental chamber. A single stage position
was chosen within each well of the plate and time lapse images were captured every 6 min under the indicated conditions with
20X 0.75 NA objective and Andor Zyla 5.5 scMOS camera. Automated imaging was performed using NIS-Elements AR software.
Immunofluorescence Microscopy
For fixed staining experiments cells were plated exactly as described for live cell imaging experiments. Following treatment and in-
cubation under the indicated conditions cells were fixed in 4% paraformaldehyde in phosphate buffered saline for 20 min. Wells were
then blocked with buffer containing 0.1% Triton and 4% bovine serum albumin. Primary antibodies were used at 1:400 dilution. Sec-
ondary antibodies were used at a 1:200 dilution. Nuclei were stained with Hoechst-33342 and imaged using a DAPI filter set; nuclear
stain images are labeled ‘DAPI’ for brevity.
DNA Damage Response Assay
MCF10A or T4-2 cells were treated with vehicle control, amphiregulin or EGF, respectively, and with erlotinib or carboplatin. Media
were then washed out at 12 h and cells fixed at time scales of 0,1,2, and 3 h post-washout using paraformaldehyde. Cells were then
co-stained with pH2AX-Alexa 488, Fra-1, and Egr1 antibodies for imaging and analysis.
Amphiregulin ELISA
To ascertain the concentration of free AREG present in our imaging experiments, media was removed from multiple wells of an im-
aging plate at a time point corresponding with the start live cell imaging experiments and subjected to ELISA. Seeding density was
9000 cells per well. Measurement of media amphiregulin levels were made using the Quantikine human amphiregulin ELISA kit per
the manufacturer instructions.
QUANTIFICATION AND STATSTICAL ANALYSIS
Imaging, Data Processing, and Statistics and Normalization
For all experiments >500 cells were analyzed per condition unless otherwise indicated. Each imaging dataset presented is represen-
tative of at least two independent replicate experiments. Image, data processing, and statistics were performed using custom
MATLAB software as previously described (Pargett and Albeck, 2018;Pargett et al., 2017). Linear regression was performed using
a Theil-Sen estimator as previously described (Gillies et al., 2017). Temporal Variability Index (TVI) was calculated by taking the ab-
solute value of the mean differential of filtered Fra-1 intensity ðFra1FÞover time divided by the mean.
TVI =
d
dt ðFra1FÞFra1F
Fra-1 filtering was performed in MATLAB employing an infinite input response and Butterworth filter design set to lowpass filtering
to remove high frequency noise with a periodicity < 30 min. All t-tests were calculated using an unpaired approach, significance was
considered P<0.05.
Partial least squares regression (PLSR) analysis was performed as previously described (Gillies et al., 2017) by incorporating all
replicates and fixed time point data into the analysis.
Dynamic Modeling
The model of ERK dependent gene expression (Equations 1,2,3, and 4) was constructed from a mass action approximation of four
steps: (1) phosphorylation of a transcription factor by ERK (TFP), (2) transcription of target mRNA (mRNA), (3) translation of target
ll
OPEN ACCESS Article
e3 Cell Systems 11, 1–15.e1–e5, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

protein (P), and (4) potential stabilization of target protein by ERK-dependent phosphorylation (PP). Additionally, a regulatory term is
included in the transcription process allowing negative feedback from target protein onto its own production. Stabilization and nega-
tive feedback are included as at least several ERK target genes are known to be phosphorylated by ERK, inhibiting their degradation,
and feedback is a common feature in gene expression, evidenced for ERK target genes by decreasing mRNA expression after long
term stimulus (Gillies et al., 2017;Uhlitz et al., 2017). The model is formulated as a delay differential equation to account for the effec-
tive lag times of transcription and translation without explicitly addressing the complex processes involved. Parameters, and their
values for Fra-1 and Egr1, are listed in Table S1.
d
dt TFPðtÞ=kpTF ERK ðtÞTFTTF PðtÞkdTF TFPðtÞ(Equation 1)
d
dt mRNAðtÞ=kb+kmTFPðttmÞ
PðttmÞ+PPðttmÞ
KDy
+1
kBmmRNAðtÞ(Equation 2)
d
dt PðtÞ=kPmRNAðttPÞ+kdP PPðtÞðkBP+kpP ÞPðtÞ(Equation 3)
d
dtPPðtÞ=kpP ERKðtÞPðtÞðkBPP+kdPÞPPðtÞ(Equation 4)
Single Cell RNA Sequencing and Analysis
Cell Lines Used for Sequencing and Treatments
To conduct single cell RNA sequencing experiments the following cells were grown under standard culture conditions for 4 days then
dissociated with 0.25% trypsin/ EDTA and rinsed into phosphate buffered saline immediately before Drop-seq. S1 and T4-2 mono-
culture cells were grown as described previously with the media changed daily. S1/T4-2 cells were grown under the same conditions
except that no exogenous growth factors were added. S1a cells received new media daily containing 100ng/ml amphiregulin. Spe-
cies mixing experiments were performed using mouse embryonic fibroblasts (MEFs) (Figure S2A).
Drop-seq
Droplets were generated using a DSQ 3x9 array microfluidic part (Nanoshift, Emeryville, CA, USA) using a Drop-seq setup according
to Macosko et al. (2015) (Online-Dropseq-Protocol-v.3.1-Dec-2015, http://mccarrolllab.com/dropseq/). Droplet size was determined
using fluorescent beads (P=S/2%) as described (Measuring-Droplet-Volume-in-Home-Made-Devices, http://mccarrolllab.com/
dropseq/). Barcoded beads and cells were loaded at concentrations specified in (Figure S2C). Prior to the collection, cell syringes
and tubing were blocked using PBS + 0.1% BSA. A magnetic mixing disc was inserted into the cell syringe to allow for manual
cell mixing during the run and the cell pump was used in a vertical position. Droplets were collected in 50mL Falcon tubes and
the target volume of aqueous flow varied in between 1-1.2mL of the cell suspension. Droplets were broken immediately after collec-
tion and reverse transcription, exonuclease-treatment and further processing was conducted as described previously (Macosko
et al., 2015). For each library, three test PCRs (50ml) each containing a bead equivalent of 100 STAMPs were conducted to determine
the optimal cycle number for library amplification. 35ml of each test PCR were purified using Agencourt AMPure XP beads (21ml beads
(0.6X) and 7mlofH
2
O for elution) and the DNA concentrations were determined using a Qubit 4 Fluorometer. A concentration in be-
tween 400-1000pg/ml was taken as optimal. A variable number of PCR reactions was conducted to amplify the available 1st strand
cDNA also with 100 STAMP bead equivalents per reaction with the optimal cycle number (50 ml volume; 4 + 8-11 cycles, Figure S2C).
12-ml fractions of each PCR reaction were pooled, then double-purified with 0.6X volumes of Agencourt AMPure XP beads and eluted
in H
2
O using 1/3rd of the bead volume. 1ml of the amplified cDNA libraries were quantified Qubit 4 Fluorometer and library size dis-
tribution verified on a BioAnalyzer High Sensitivity Chip (Agilent). 600 pg cDNA of the library was fragmented and amplified (12 cycles)
using the Nextera XT (Illumina) sample preparation kit. The libraries were double-purified with 0.63volumes of AMPure XP Beads and
quantified.
Sequencing Strategy
Nextera libraries were sequenced on Illumina Nextseq 500 sequencers using the NextSeq High Output v2 kit (75 cycles), using a
custom primer and a custom paired end sequencing strategy, R1 20bp, index 8bp, R2 remaining bp (Macosko et al., 2015).
The five Nextera libraries were pooled and a total of 9.9k anticipated STAMPs were loaded on the flow cell with the following
library portions; S1/MEF: 5%, T4: 17.1%, T4/S1: 30.3%, S1: 17.2%, S1 amphiregulin: 13.1%. The pool was sequenced on two
NextSeq 500 runs. Raw sequencing data have been submitted to the GEO repository and are available under GEO accession
GSE118312.
ll
OPEN ACCESS
Article
Cell Systems 11, 1–15.e1–e5, August 26, 2020 e4
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004

Reference for Read Alignment
For the species mixing experiment we used a mixed reference (human+mouse) available at GEO accession GSE63269 (Macosko
et al., 2015). For mapping of T4 and S1 reads we used the human genome primary assembly (GRCh38) available at https://www.
ensembl.org/ and release 92 gene models. Sequences for transgenic markers mCherry and mVenus were added to the reference.
Drop-seq Pipeline and Generation of the Gene Expression Matrix
All Drop-seq data were processed using Drop-seq Tools v1.12 as described previously (http://mccarrolllab.com/wp-content/
uploads/2016/03/DropseqAlignmentCookbookv1.2Jan2016.pdf)(Macosko et al., 2015). Bowtie2 (v2.2.6)(Langmead et al., 2009)
was used for read alignment with the settings –phred33 –very-sensitive -N 1.
Species Mixing Experiment
To determine whether single cells were generated, we mixed S1 cells and MEF (mouse) cells, then mapped the resultant data against
a hybrid reference containing both human and mouse genes. Cells were loaded at approximately 100 cells/uL (50cells/ml final), which
would be expected to give 2.5% doublet rate, similar to the 1.6% doublet rate observed in the experiment (Figure S2A). This library
was sequenced very shallow since its sole purpose was doublet assessment.
Cell QC and Clustering
Cells with more than 500/2k and less than 6k/30k genes/UMIs were considered in the analyses (Figure S2B DropseqStats). Cells with
more than 10% of mitochondrial expression were filtered out and excluded from downstream analyses. 10,212 cells remained after
filtering with medians of 2,653 genes and 6,283 UMIs. Analyses were conducted using R package Seurat (v2.2.1)(Satija et al., 2015).
Excess Variance Calculation and Delta-Excess Variance
UMI counts for each cell were converted to transcripts per million (TPM) by scaling all cell-wise counts to 1,000,000. For each cell
population iand each gene g, under the assumption of a negative binomial distribution, we computed its mean expression mi;g
and variance vi;g, then calculated gene-wise dispersion parameters ai;gfrom the expression vi;g=mi;g+ai;gm2
i;g. We observed a linear
trend between log mi;gand log ai;gfor all cell populations, indicating as the mean expression of a gene increased, the overdispersion
tends to decrease. Therefore, to measure change in dispersion for each gene between two populations (when the mean may also
change), we instead compare ‘‘excess variance’’. For every pair of populations iand jbeing compared, we fit an ‘‘excess variance’’
term Ei;g=ðmi;gbmi;gÞ, where the fitted overdispersion estimate bmi;gis generated by performing Local Polynomial Regression Fitting
(loess) using R version 3.4.4 with span=1, simultaneously regressing log ai;gon log mi;gand log aj;gon log mj;gover all genes g. Sta-
tistical significance was calculated using a paired, two-sided Wilcoxon signed rank test between all Ei;gand Ej;g. See Data S1 for a
complete list of genes and relative excess variance.
Parameters were selected to match previously data collected on average mRNA and protein expression kinetics (Gillies et al.,
2017;Schwanh€
ausser et al., 2011;Uhlitz et al., 2017).
ll
OPEN ACCESS Article
e5 Cell Systems 11, 1–15.e1–e5, August 26, 2020
Please cite this article in press as: Davies et al., Systems-Level Properties of EGFR-RAS-ERK Signaling Amplify Local Signals to Generate Dynamic
Gene Expression Heterogeneity, Cell Systems (2020), https://doi.org/10.1016/j.cels.2020.07.004