Secretome mediated interactions between sensory neurons and breast cancer cells

Abstract

The role of the nervous system in aiding cancer progression and metastasis is an important aspect of cancer pathogenesis. Interaction between cancer cells and neurons in an in vitro platform is a simple and robust method to further understand this phenomenon. In our study, we aimed to examine in vitro reciprocal effect between breast cancer cells and cancer‐sensitized peripheral primary sensory neurons. Secretome obtained from either cultured DRG neurons from tumor‐burdened rats, or MRMT1 breast cancer cells were used to study neuronal and cancer cell reciprocity. We utilized neurite analysis, modified cell migration assay and cell signaling pathway inhibitors to determine neurite growth patterns and cell migration in PC12/DRG neurons and MRMT1 cells, respectively. MRMT1 secretome was found to induce significant neurite outgrowth in PC12 and primary sensory neurons. Secretome‐induced neurite growth in PC12 cells was partly mediated by PI3K and ERK pathways, but not by adenylyl cyclase. Conversely, secretome from tumor‐sensitized sensory neuron cultures induced increased rate of migration in cultured MRMT1 cells. Results from our study provide additional support to the hypothesis that both breast cancer cells and nerve terminals secrete signaling messengers that have a reciprocal effect on each other.

Full text

MOLECULAR CANCER BIOLOGY
Molecular Cancer Biology
Secretome mediated interactions between sensory neurons
and breast cancer cells
Chinnu Jerard | Pallavi Madhusudanan | Aditi Swamy | Karishma Ravikumar |
Sahadev A. Shankarappa
Center for Nanosciences & Molecular
Medicine, Amrita Vishwa Vidyapeetham,
Kochi, Kerala, India
Correspondence
Sahadev A. Shankarappa, Center for
Nanosciences and Molecular Medicine, Amrita
Institute of Medical Sciences and Research
Center, Amrita Vishwa Vidyapeetham, Kochi,
682041, Kerala, India.
Email: sahadevs@acnsmm.aims.amrita.edu,
sahadevs@icloud.com
Funding information
Council of Scientific and Industrial Research,
India, Grant/Award Number: 09/963(0046)-
2K19EMR-1; Department of Biotechnology,
Ministry of Science and Technology, India,
Grant/Award Number: BT/PR24515/
MED/30/1926/2017; Department of Science
and Technology, Ministry of Science and
Technology, India, Grant/Award Number:
DST/NM/NS-282/2019(G)
Abstract
The role of the nervous system in aiding cancer progression and metastasis is an
important aspect of cancer pathogenesis. Interaction between cancer cells and neu-
rons in an in vitro platform is a simple and robust method to further understand this phe-
nomenon. In our study, we aimed to examine in vitro reciprocal effect between breast
cancer cells and cancer-sensitized peripheral primary sensory neurons. Secretome
obtained from either cultured DRG neurons from tumor-burdened rats, or MRMT1
breast cancer cells were used to study neuronal and cancer cell reciprocity. We utilized
neurite analysis, modified cell migration assay and cell signaling pathway inhibitors to
determine neurite growth patterns and cell migration in PC12/DRG neurons and
MRMT1 cells, respectively. MRMT1 secretome was found to induce significant neurite
outgrowth in PC12 and primary sensory neurons. Secretome-induced neurite growth in
PC12 cells was partly mediated by PI3K and ERK pathways, but not by adenylyl cyclase.
Conversely, secretome from tumor-sensitized sensory neuron cultures induced
increased rate of migration in cultured MRMT1 cells. Results from our study provide
additional support to the hypothesis that both breast cancer cells and nerve terminals
secrete signaling messengers that have a reciprocal effect on each other.
KEYWORDS
axon, cancer nerve crosstalk, conditioned media, neurite, tumor microenvironment
What's new?
Recent reports have pointed toward a key role of peripheral nerves in cancer initiation, progres-
sion and metastasis. However, how neurons and tumor cells interact within the tumor microen-
vironment to facilitate cancer progression remains unclear. Here, the authors observed
increased neurite growth induced by breast cancer cells and demonstrated the role of the PI3K
pathway in cancer cell-induced neuritogenesis in PC12 cells. Reciprocally, tumor-sensitized neu-
rons from tumor-burdened animals promoted breast cancer cell migration in vitro. The results
provide additional support to the hypothesis that both breast cancer cells and nerve terminals
secrete signaling messengers with a reciprocal effect on each other.
Abbreviations: AC, adenylyl cyclase; BSA, bovine serum albumin; CM, conditioned media; DMEM, Dulbecco's modified Eagle's medium; DRG, dorsal root ganglion; ERK, extracellular signal-
regulated kinase; FBS, fetal bovine serum; HBSS, Hank's balanced salt solution; H-F12, Ham's F-12 nutrient mix; HS, horse serum; MRMT1, syngeneic breast cancer cells; NGF, nerve growth
factor; PDL, poly-D-lysine hydrobromide; PFA, paraformaldehyde; PI3K, phosphatidylinositol-3 kinase; PS, penicillin-streptomycin mix; RPMI, Rosewell Park Memorial Institute medium; SD,
Sprague-Dawley; TCM, tumor conditioned media; TME, tumor microenvironment.
Received: 20 June 2022 Revised: 7 March 2023 Accepted: 20 March 2023
DOI: 10.1002/ijc.34529
Int. J. Cancer. 2023;1–10. wileyonlinelibrary.com/journal/ijc ©2023 UICC. 1

1|INTRODUCTION
Nerve terminals within the tumor microenvironment (TME) have
been long suspected to play a role in tumor growth and metasta-
sis.
1
Cancerous lesions in the breast, prostate and pancreas are
notable examples where tumor mass has rich neural innervation.
2
Clinically, the presence of neural innervation in tumor tissue is
generally associated with poor outcomes,
3,4
although the exact
mechanism of how nerve-cancer cell interactions cause cancer
aggravation is not completely clear. Recent reports, however, point
toward a key role played by peripheral nerves in cancer initiation,
progression and metastasis.
5-8
Aprevailingviewhasbeentherole
of nervous system in conveying tumor-associated signals to the
brain, which in turn modulates the neuroendocrine-immune sys-
tem to promote tumorigenesis.
9
Although a central mechanism
nicely supports the overall involvement of the nervous system in
tumor progression, a more local effect involving the peripheral
nerve has also been examined. It has been hypothesized that
tumor cells form physical contacts with nerve endings and receive
neurotransmitter-mediated intracellular signals, resulting in tumor
proliferation, apoptosis and metastasis.
6,10
These cancer cell-nerve
contacts have also been referred to as “neuro-neoplastic
synapses,”
5
although further evidence on the nature of such con-
tacts needs to be better examined. In addition, invasion of cancer
cells along the perineural sheath of peripheral nerves has been
established as a route for metastasis.
11
Interestingly, in what
seems like cellular reciprocity, the tumor tissue is also known to
induce neural growth and increase nerve innervation within the
TME, via trophic factor-mediated pathways (eg, NGF, BDNF,
VEGF).
12
Recent studies have found that tumor tissue in the breast is
richly innervated by sympathetic, parasympathetic and sensory
nerves.
13-15
Activation of adrenergic sympathetic neurons is asso-
ciated with tumorigenesis, while activation of the cholinergic para-
sympathetic neurons reduced cancer cell proliferation.
14
Interestingly, the exact role of sensory neurons in breast cancer
tissue is not completely clear, although some studies have indi-
cated that sensory neurons may promote tumorigenesis and
induce hyperalgesia in patients.
16
Tumor samples derived from
breast cancer patients show enhanced neurite sprouting, while
in vitro studies have demonstrated breast cancer cell-mediated
increase in neurite sprouting in immortalized PC12 cells.
17
How-
ever, it remains not clear how cancer cells in the breast affect sen-
sory neurons, and what effect cancer-primed sensory neurons may
have on breast cancer cells. Connectedly, breast cancer cell-
induced signaling pathways in neuronal cells are also an understu-
died aspect of breast tumorigenesis.
In our study, we attempt to understand the nature of recipro-
cal relationship between sensoryneuronsandbreastcancercells
in an in vitro setting. We have utilized MRMT1 breast cancer cell
line, PC12 cell line and dissociated dorsal root ganglion sensory
neurons to determine the effect of breast cancer cells and neurons
on each other.
2|MATERIALS AND METHODS
2.1 |Cells and reagents
PC12 cell line (RRID: CVCL_0481) was procured from American Type
Culture Collection, USA, and MRMT1 cell line (RRID:CVCL_5156) was
purchased from Riken Cell Bank, Japan. All experiments were per-
formed in mycoplasma-free cells. Hank's balanced salt solution
(HBSS), Dulbecco's modified Eagle's medium (DMEM), Ham's F-12
nutrient mix (H-F12), Rosewell Park Memorial Institute medium
(RPMI-1640) were purchased from Lonza chemicals, USA. Trypan blue
was purchased from Loba Chemie, India. Collagenase type I, dispase II,
fetal bovine serum (FBS), penicillin-streptomycin mix (PS) and horse
serum (HS) were acquired from Gibco, NY. Paraformaldehyde (PFA),
bovine serum albumin (BSA), Papain, poly-D-lysine hydrobromide
(PDL) and collagen type IV were purchased from Sigma-Aldrich, USA.
Hydrochloric acid, TRIS-buffer, Triton X-100, sodium chloride, potas-
sium chloride, disodium hydrogen phosphate and potassium dihydro-
gen phosphate, sodium hydroxide were from Merck Chemicals, USA.
Antibody β-III tubulin was purchased from Cell Signaling Technology,
MA. Hoechst stain was purchased from Santa Cruz Biotechnology,
TX. Secondary anti-mouse IgG antibody, conjugated to Dylight
488 was purchased from Vector Labs, USA. Calcein-AM dye was pur-
chased from Thermo Fisher Scientific, MA. Phosphatidylinositol-3
kinase (PI3K) inhibitor (LY294002), MEK inhibitor (U0126) and adeny-
lyl cyclase (AC) inhibitor (SQ-22536) were obtained from Tocris Bio-
science, Bristol, UK.
2.2 |Animal care
Adult female Sprague-Dawley (SD) rats weighing 200 to 250 g were
housed in pairs, allowed standard rat diet and water ad libitum, and
maintained in 10 hours/14 hours light/dark cycle. For extraction of
dorsal root ganglion, rats were euthanized by CO
2
inhalation and tis-
sue dissected under sterile conditions as explained below.
2.3 |Animal model of metastasized breast cancer
in the bone
Tumors representing metastasized breast cancer in the rat were
induced by injecting syngeneic breast cancer cells (MRMT1) into the
proximal end of the tibial bone, as described previously.
18
Briefly,
MRMT1 cells were cultured in medium containing RPMI-1640, 10%
(v/v) FBS and 1% (v/v) PS. Confluent cultures were trypsinized, centri-
fuged at 200gfor 5 minutes, re-suspended in HBSS at a concentration
of 3 !10
7
cells/mL and placed on ice until injection. Rats were anes-
thetized by intraperitoneal administration of ketamine (100 mg/kg)
and xylazine (5 mg/kg), followed by a 5-mm incision on the anterior
surface of the left knee. MRMT1 cells (3 !10
5
cells in 10 μL) were
injected into the upper part of the tibia in the anterolateral region, just
underneath the articular cartilage. The surgical wound was closed
2JERARD ET AL.

using tissue adhesive and the animals were allowed to recover. Tumor
growth was clinically monitored and was observed to become palpa-
ble by week 1. Animals were euthanized for dorsal root ganglion
(DRG) extraction between 2 and 3 weeks post-tumor induction.
2.4 |PC12 cell culture and neurite assay
PC12 cells were seeded on a collagen-coated (100 μg/mL) 96-well
plate with seeding density of 7800 cells/cm
2
, and maintained at 37"C
in 5% CO
2
under humidified condition for 24 hours in medium con-
taining RPMI-1640 with 10% (v/v) HS, 5% (v/v) FBS and 1% (v/v)
PS. Cultured cells were then exposed to conditioned media obtained
from MRMT1 cells for 72 hours, followed by intracellular loading with
the membrane-permeable calcein dye (1 μg/mL) and imaged under a
fluorescence microscope (Leica DMI 3000B, !100 magnification).
Five random images were acquired per well. Neurite length, number
of neurite-bearing cells and neurite branching pattern were analyzed
using the neurite outgrowth software application module (Molecular
Devices, USA).
2.5 |Signaling pathway inhibitors
PC12 cells were cultured for 24 hours, followed by treatment with
50% (v/v) MRMT1-derived conditioned media containing either PI3K
inhibitor (LY294002, 2.5 μM), extracellular signal-regulated kinase
(ERK) inhibitor (U0126, 10 μM) or adenylyl cyclase inhibitor
(SQ-22536, 25 μM). All drugs were dissolved in 0.1% DMSO. Treated
cultures were then maintained at 37"Cin5%CO
2
under humidified con-
ditions for 48 hours followed by calcein staining and neurite analysis.
2.6 |Dorsal root ganglion neurons
Lumbar DRG's were harvested from euthanized rats and immediately
immersed in ice-cold Ca
2+
Mg
2+
free HBSS. DRG cells were disso-
ciated using a modified procedure that has been previously
reported.
19
For tumor-burdened rats, upper lumbar DRG's were
harvested because axons innervating the tumor-ridden epiphyseal
region of the tibia arise from neuronal cell bodies located in the
L1-L3 spinal segment.
20
Briefly, excised 2 to 4 DRG's were serially
dissociated in papain (1.3 units/mL) and collagenase (type I)-
dispase (type II) enzyme cocktail (3 and 4.6 mg/mL, respectively)
for 30 minutes each, at 37"Cundershakingcondition.Theenzy-
matically dissociated cells were then triturated using fire-polished
pipettes and seeded on poly-D-lysine coated plates with medium
containing DMEM with 10%(v/v) Ham's F-12 supplement, 10%
(v/v) FBS and 1% (v/v) PS, and maintained at 37"Cin5%CO
2
under humidified conditions for 5 days before fixation and stain-
ing. DRG's in the experimental groups received MRMT1-derived
conditioned media (10% and 50%), along with comparable media-
type controls.
2.7 |Immunocytochemistry
DRG cultures were fixed with 4% (w/v) PFA for 30 minutes at room
temperature, followed by PBS wash and permeabilization with 0.2%
(v/v) triton-X for 30 minutes in a 24-well plate. Cells associated nonspe-
cific interactions were reduced by blocking with 3% (w/v) BSA for
30 minutes, followed by overnight incubation with mouse anti-βIII
tubulin (0.43 μg/mL) at 4"C. Cultures were then washed and incubated
with Dylight conjugated anti-mouse IgG antibody (10 μg/mL) for
2 hours at room temperature and counterstained with Hoechst (5 μg/mL)
nuclear stain. Five random images per well were acquired using Leica
DMI 3000B microscope fitted with a CCD camera (Leica DFC 3000G)
at !200 magnification. Number of neurite-bearing cells, neurite
length and neurite branching pattern were quantified using neurite
outgrowth software application module (Molecular Devices, USA).
2.8 |Total RNA isolation and cDNA synthesis
Total RNA was extracted from CM treated PC12 cells that were main-
tained in culture conditions as described above. RNA extraction was
performed as per manufacturer's protocol using the Nucleospin RNA
Plus kit (740 984, Takara Bio, Japan). RNA purity was assessed by the
ratio of absorbance at 260 and 280 nm using a NanoDrop spectro-
photometer. Only sample with absorbance ratio above 2.0 were con-
sidered for cDNA synthesis. PrimeScript RT Reagent Kit (RR037A,
Takara Bio, Japan) was used to synthesize cDNA from 1 μg RNA sam-
ple as per the manufacturer's protocol in a CFX96 Real-Time system
thermal cycler (Bio-Rad, USA). cDNA samples were stored at #20"C
till further use.
2.9 |Quantitative real-time PCR
Quantitative RT-PCR (qRT-PCR) was performed using the TBGreen Pre-
mix Ex TaqII (RR820A, Takara Bio, Japan) as per the manufacturer's proto-
col. Primers were preformulated for Synapsin-1 (sense 50-CCAA
TCATAAAGAGATGCTC-30and antisense 50-CAATATCCTGGAAGTCA
TGT-30), GAP-43 (sense 50-AAGCTGAGGAGGAGAA-AGAA-30and anti-
sense 50-GCATGTTCTTGGTCAGCC-30)andthehousekeepinggene
(β-actin).
21
Thermal cycling conditions of denaturation at 95"Cfor
1minute,followedby40cyclesat95
"Cfor15seconds,withannealingat
54"C, 30 seconds and extension phase at 72"Cfor30seconds,wasused.
All RT-qPCR experiments were performed on CFX96 Real-Time system
thermal cycler (Bio-Rad, USA). Relative mRNA expression level was calcu-
lated by 2
#ΔΔCq
(Livak) method where CT of target gene was normalized
to that of reference gene (β-actin) for both test and calibrator samples.
2.10 |Cell migration assay
MRMT1 cultures (10 000 cells/cm
2
)treatedwithconditionedmedia
(50%) derived from 3-day DRG cultures harvested from naive, or tumor-
JERARD ET AL.3

burdened rats were subjected to a modified scratch assay. Briefly, 80% to
90% confluent MRMT1 cells cultured in medium containing RPMI-1640,
10% (v/v) FBS and 1% (v/v) PS, were scrapped off cells using a cell
scraper with 1.8-cm wide sterile blade, and the cell plates were imaged
every 4th hour for 12 hour. The distance traveled by the cells relative to
apremarkedreferenceline(whitedottedline,Figure5)onthewellplate
was measured using a Nikon EclipseTE2000-Umicroscopefittedwitha
Nikon digital sight camera at !40 magnification. Only cells migrating
beyond 200 μmfromtheedgeoftheoriginalscratchwereconsideredas
migratory and used for analysis. The analysis was performed using ImageJ
(NIH) software. Serum depletion, a strategy utilized by some studies in
migration assay, was not adopted in our study since we observed reduced
MRMT1 viability in serum-depleted medium.
2.11 |Statistical analysis
Data are represented as mean ± SD. Statistical analysis between experi-
mental groups with parametric data sets was made using ordinary one-
way ANOVA. P<.05wasconsideredstatisticallysignificantineachcase.
3|RESULTS
3.1 |Effect of MRMT1 conditioned medium on
PC12 differentiation
First, we wanted to determine whether breast cancer cells have an
effect on neuronal differentiation in vitro. To test this, we measured
neurite outgrowth in PC12 cells that were exposed to conditioned
media (CM) derived from MRMT1 breast cancer cells. The PC12 cell line
was chosen as a “neuron-like cell model”since it acquires neuron-like
features upon nerve growth factor treatment. PC12 cells cultured for
6 days in native media demonstrated a spherical morphology
(Figure 1A), while those cultured in either 10% or 50% MRMT1 CM
were more likely to be nonspherical with sprouting thin neurites
(Figure 1B,C). Even though MRMT1 CM-treated PC12 cells demon-
strated an altered morphology compared to controls, the number of
PC12 cells, however, was similar in all groups (Figure 1D;P> .05). Fur-
ther analysis of differentiating PC12 cells revealed that treatment with
50% MRMT1 CM resulted in an increase in the number of cells sprout-
ing neurites (Figure 1E), and the length of sprouted neurites was signifi-
cantly longer than controls (Figure 1F)(P< .05). However, the
branching of neurites was similar across all groups (Figure 1G). It is also
interesting to note that CM-derived from unrelated noncancerous cells
(L929, mouse fibroblast cells) did not induce neurite outgrowth in PC12
cells, suggesting cancer secretome specificity (Figure S1). These obser-
vations suggest that the secretome from MRMT1 cells induces neurito-
genesis in cultured PC12 cells but does not affect cell proliferation.
Further, expression of Synapsin-1 and GAP-43 mRNA which indicate
cellular differentiation, were significantly upregulated in PC12 cells trea-
ted with MRMT1 CM (Figure 1H;P< .05).
10% TCM
Control
50% TCM
0
10
20
30
40
No of PC12 cells pf
10 500
MRMT1 cell CM (%)
ns
ns
50 μm
50 μm
50 μm
0
20
40
60
80
100
Neurite bearing cells (%)
10 500
MRMT1 cell CM (%)
ns
*
0
100
200
300
400
500
600
Neurite length pf (μm)
*
*
ns
MRMT1 cell CM (%)
10 500
0
5
10
15
20
25
30
Total n o o f b r a nche s p f
10 50 CM CM
CC
SYN-1 GAP-43
(H)
0.0
0.5
1.0
1.5
2.0
2.5
Relative fold change
0
3
6
9
12
*
*
0
MRMT1 cell CM (%)
ns
ns
ns
(A)
(C)
(E)
(G)
(F)
(D)
(B)
FIGURE 1 Conditioned medium from cultured breast cancer
cells induce enhanced neurite growth in PC12 cells.
Representative images of PC12 cells treated with naive medium
(A), or 10% (B) and 50% (C) MRMT1-derived tumor conditioned
medium (TCM) for 72 hours exhibiting neurite growth (arrows)
from individual cells. Neurite growth from PC12 cells was
quantified as shown in (D-G). Data shown are mean ± SD obtained
from three separate experiments performed in triplicate.
Comparative analysis between indicated groups were performed
using one-way ANOVA with Bonferroni's multiple comparison
tests, * indicates P< .05.-fold change in mRNA level of
differentiation marker genes after normalization to β-actin gene,
performed in triplicate is shown in (H). Comparative analysis
between control (C) and 50% MRMT1 CM (CM) groups were
performed using unpaired Students t-test, * indicates P<.05.
[Color figure can be viewed at wileyonlinelibrary.com]
4JERARD ET AL.

3.2 |Activation of signaling pathways in PC12
cells by MRMT1 conditioned medium
Further, to determine the mechanism behind MRMT1 CM-induced
neuritogenesis in PC12 cells, we explored the role of phospha-
tidylinositol-3 kinase (PI3K), ERK and AC signaling pathways. Cells cul-
tured in MRMT1 CM with (vehicle) or without (naive) DMSO were
used as comparative controls. We observed that the presence of sig-
naling inhibitors did not alter MRMT1-induced neurite outgrowth in
PC12 cells, but compared to naive PC12 cultures, there was a notice-
able reduction (P< .05) in the overall length of neurites in cells treated
with PI3K inhibitor (PI3Ki; Figure 2A-H). Length of neurites was also
reduced in cultures treated with ERK inhibitor (ERKi), but
Neurite bearing cells
(normalized to naive)
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
MRMT1 cell CM (50%)
Naive Veh PI3Ki ERKi ACi
(G) (H)
(F)
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
MRMT1 cell CM (50%)
*
Neurite length
(normalized to naive)
Naive Veh PI3Ki ERKi ACi
1.2
0.8
0.4
0.0
MRMT1 cell CM (50%)
Naive Veh PI3Ki ERKi ACi
1.6
2.0
2.4
Branches pf
(normalized to naive)
(A) Naive Veh
PI3Ki ERKi
ACi
(B)
(D)
(E)
(C)
50 μm
50 μm
50 μm
50 μm
50 μm
FIGURE 2 Effect of cell signaling inhibitors on neurite growth in
PC12 cells treated with conditioned medium derived from MRMT1
cells. Representative images of PC12 cells treated with naive medium
(A), vehicle (veh) (B), PI3Ki (C), ERKi (D) and ACi (E) for 48 hours. Bar
graph depicting the number of neurite-bearing cells (F), along with
neurite length (G) and neurite branching (H) from analyzed PC12 cells.
Naive group indicates cells that did not receive any inhibitors, while
the vehicle group was exposed to 0.1% DMSO. Data shown are
mean ± SD obtained from three separate experiments done in
triplicate. Comparative analysis between indicated groups were
performed using ordinary one-way ANOVA, * indicates P< .05. [Color
figure can be viewed at wileyonlinelibrary.com]
WF
<3 kDa
>3 kDa
0
0.5
1.5
0.5
1.5
1
2
No of PC12 cells
<3 kDa
>3 kDa
WF
Neurite bearing cells
Neurite length
Branches pf
(A)
(C)
(E)
(G)
(F)
(D)
(B)
(normalized toWF)
0
1
2
<3 kDa
>3 kDaWF
(normalized toWF)
0
1
2
<3kDa
>3kDaWF
(normalized toWF)
0
1
2
<3 kDa>3 kDa
WF
(normalized toWF)
***
***
0.5
1.5
0.5
1.5
50 μm 50 μm
50 μm
ns
ns ns
ns
FIGURE 3 Conditioned medium fraction of >3 kDa molecular
weight cut-off, induced enhanced neurite growth in PC12 cells.
Representative images of PC12 cells treated with unfractionated
MRMT1 CM (whole fraction, WF) (A), >3 kDa fraction (B) and <3 kDa
fraction (C) for 72 hours. Neurite growth from PC12 cells was
quantified as shown in (D-G). Data obtained from CM fractions were
normalized to parameters from the whole fraction group, performed
in triplicate. Comparative analysis between indicated groups was
performed using one-way ANOVA with Bonferroni's multiple
comparison tests, * indicates P< .05. [Color figure can be viewed at
wileyonlinelibrary.com]
JERARD ET AL.5

the difference in mean neurite length between ERK inhibitor
and vehicle-treated or naive cells was not statistically significant
(P> .05; Figure 2G). Interestingly, ERK inhibitors reduced neurite
branching in MRMT1 CM treated cells (Figure 2H), but adenylyl
cyclase inhibitor (ACi) did not have any effect. These observations
strongly suggest that MRMT1 CM-induced neurite outgrowth in
PC12 cells is partly mediated by PI3K and ERK pathways.
3.3 |Effect of conditioned medium fractions on
PC12 neurite growth
To further assess the molecular components in the MRMT1 CM
responsible for PC12 neuritogenesis, we treated cultured PC12 cells
with CM fractions that were obtained by centrifugal filtration with a
molecular weight cut-off of 3 kDa (Figure 3A-G). Both the obtained
CM fractions were not detrimental to cell growth, and there was no
significant difference in cell number in either of the groups. Interest-
ingly, we observed that cells treated with CM fraction of >3 kDa dem-
onstrated neurite activity, in terms of neurite length and branching,
that was quite similar to that of cells treated with unfractionated
MRMT1 CM (Figure 3A,B;P> .05). The CM fraction from the <3 kDa
cut-off induced only small or no neurite outgrowths (Figure 3C), that
was significantly lower than PC12 cells treated with unfractionated
CM (Figure 3F,G;P< .05). These results suggest that the
MRMT1 secreted molecular component/s responsible for neurito-
genic activity in PC12 cells are likely to be larger proteins and not
smaller peptides and cell-secreted metabolites.
3.4 |Effect of MRMT1 conditioned medium on
peripheral sensory neuron differentiation
To further determine the effect of MRMT1-derived CM on dissoci-
ated sensory neurons, lumbar DRG neurons harvested from naive
adult rats (Figure 4A) were exposed to MRMT1 CM for 5 days. Based
on our previous experiments with PC12 cells, we chose to use 10%
and 50% v/v of MRMT1 CM to test DRG neurite responsiveness.
Compared to controls, visible difference in neurite sprouting could be
clearly observed in DRGs treated with MRMT1 CM (Figure 4B-D).
There was $20% to 25% increase in neurite-bearing neurons in DRG
cultures that were treated with MRMT1 CM as compared to control
media. We observed increased number of neurite-bearing DRG neu-
rons in cultures exposed to both 10% and 50% MRMT1 CM (P< .05,
Figure 4E). In addition, the length of neurites and the number of neur-
ite branching in DRG neurons exposed to 50% MRMT1 CM were also
significantly increased (P< .05, Figure 4F,G), suggesting cancer
secretome-mediated neuritogenic effect on peripheral sensory
neurons.
3.5 |Effect of tumorigenic DRG cell conditioned
medium on cancer cell migration
To investigate the nature of cellular reciprocity between sensory neu-
rons and cancer cells, we examined the effect of CM obtained from
(E)
Neurite bearing cells (%)
125
100
75
50
25
0
10% 50%
***
(G)
Neurite length pf (μm)
500
600
400
300
200
100
0
*
10% 50%
700
10% 50%
250
200
150
100
50
0
*
Tota l n o o f bra n c hes p f
(F)
Control media
MRMT1 cell CM (%)
(B)
10% TCM 50% TCM
(C) (D)
Control(A)
20 μm20 μm
20 μm
FIGURE 4 Conditioned medium from cultured breast cancer cells
induce enhanced neurite growth in dorsal root ganglion sensory
neurons. Representative phase contrast image of dissociated DRG
cells with neurons (solid arrows) and glial cells (broken arrows) in (A).
Representative fluorescent images of dissociated sensory neurons
treated with control (B) or 10%, 50% MRMT1CM (C,D), showing
neurite outgrowth (arrows). Bar graphs depicting number of neurite-
bearing cells (D), along with neurite length (F) and neurite branching
(G) in sensory neurons, treated with 10% or 50% MRMT1-derived
CM. Data shown are mean ± SD obtained from three separate
experiments done in triplicate. Comparative analysis between
indicated groups were performed using ordinary one-way ANOVA, *
indicates P< .05. [Color figure can be viewed at
wileyonlinelibrary.com]
6JERARD ET AL.

culturing DRG cells that were harvested from tumor-burdened rats.
Since the tumor-inoculated area in the proximal end of the tibia is
innervated by neurites from the lumbar DRGs, care was taken to har-
vest and culture only such DRGs for these experiments. Because DRG
cells were cultured in medium that was different from medium
required for MRMT1 cells, appropriate controls were also included in
the experiment (Figure 5). The scratch assay performed on monolayer
of MRMT1 cells exposed to CM from tumor-burdened DRG cells
demonstrated enhanced MRMT1 cell motility compared to matched
controls (P< .05, Figure 5) at all time points tested. These results sug-
gest a positive cell migration effect of tumor-sensitized neurons on
cultured cancer cells. Interestingly, migration of MRMT1 cells exposed
to CM obtained from PC12 cells primed with MRMT1 CM was not
different from controls (Figure S2).
4|DISCUSSION
The interactions between cancer cells and axonal nerve endings have
emerged as an important component that contributes to the tumor
microenvironment and promotes cancer pathogenesis.
22
Evidence for
nerve-tumor interactions arises from studies that report tumor regres-
sion and tumor growth inhibition in models of nerve denervation in
prostate cancer
23
and tumor of tongue.
24
Just as denervation inhibits
0
10
20
30
40
Migrating MRMT1 cells pf
4h 8h 12h
Native media
Control media
DRG CM from non-tumor (NT) rats
DRG CM from tumor (T) rats
✱✱✱ ✱✱✱
Native
media
Control
media
DRG-CM
from NT rats
DRG CM
from T rats
0 h
4 h
8 h
12 h
200μm
200μm
200μm
200μm
200μm
200μm
200μm
200μm
200μm
200μm
200μm
200μm200μm
200μm
200μm
200μm
(A)
(B)
FIGURE 5 Conditioned medium from
cultured DRG neurons harvested from
tumor-burdened rats enhance migration of
breast cancer cells. Representative images
(A) and bar graph (B) depicting the number
of migrating MRMT1 cells in the presence
of 50% CM from cultured DRG neurons
(from tumor-burdened rats) after a scratch
injury. Data shown are mean ± SD
obtained from 3 to 6 separate experiments
done in duplicate. Comparative analysis
between indicated groups were performed
using one-way ANOVA, with Bonferroni's
multiple comparisons test, ***
indicates P< .001.
JERARD ET AL.7

cancer pathogenesis to a certain extent, increased neural signaling is
thought to promote tumorigenesis.
25
Understanding this crosstalk
between cancer cells and nerves can provide new insights into carci-
nogenesis and could potentially aid in the development of neurogenic
strategies to combat cancer. However, what remains unclear is how
neurons and tumor cells interact within the tumor microenvironment
to facilitate cancer progression. In vitro platforms provide an excellent
and cost-effective approach to understand such mechanisms, but a
reproducible and measurable effect between cancer cells and neurons
needs to be established. In our study, using an in vitro system, we pri-
marily sought to understand possible reciprocal effects of cultured
breast cancer cells and sensory neurons on each other.
Here, we employed CM from MRMT1, or rat DRG sensory neu-
rons to understand possible crosstalk between breast cancer cells and
nerve cells, and also to assess possible signaling pathways. Utilization
of CM for understanding the global effects of secreted proteins has
been well-established and used in several reports.
26,27
In our study,
CM derived from MRMT1 cells significantly increased the likelihood
of cells sprouting neurites and enhanced neurite length in PC12 cells.
This finding was in agreement with previous studies where CM from
MDA MB 231 breast cancer cells induced expansion of neurite
growth cone
16
and neurite length. Since neurite growth is mainly due
to the production of new actin and microtubule structures within the
neuronal filopodia,
28
secretome from MRMT1 cells in our study could
essentially trigger expression of these proteins. We further found
enhanced mRNA expression of markers associated with axonogenesis
and neuronal growth (Synapsin-1 and GAP-43), that further supports
the neuritogenic role of MRMT1 CM.
29
Interestingly, cell-secreted
VEGF-A, GM-CSF and GDNF have been reported to be involved in
breast cancer cell-mediated neurite expansion.
16,30,31
Even in the cur-
rent study, it is likely that the observed increase in neurite growth is
mediated by MRMT1 cell secreted tropic factors because MRMT1
CM fraction rich in trophic factors (>3 kDa fraction) increased PC12
neurite growth. Additional support for a possible trophic factor-
mediated neuritogenesis lies in our observation where PI3K and ERK
signaling pathway inhibitors partially blocked MRMT1 CM-induced
neuritogenesis in PC12 cells. Since stimulation of the nerve growth
factor (NGF) receptor, TrkA is known to promote neurite growth in
PC12 cells via activation of Raf/ERK and PI3K/AKT signaling path-
ways, it is likely that NGF may play an important role in MRMT1 cell-
mediated neurite outgrowth in PC12 cells as well. Relatedly, NGF
mediated increase in neurite density has also been reported in gastric
cancer cells, where neurite expansion was attributed to NGF released
by acetylcholine-treated gastric cancer cells.
32
Similarly, breast cancer
epithelial cells including MCF-7, T47D, MDA-MB-231 and BT20 have
also been shown to secrete NGF
33
in culture conditions.
Similar to our findings with PC12 cells, we found that MRMT1
CM induced enhanced neurite growth in DRG sensory neurons, in a
concentration-dependent manner. Our finding with DRG sensory neu-
rons was consistent with other studies where cancer-derived condi-
tioned media induced sensory neuronal outgrowth.
16,34,35
However,
in a separate in vivo study in our laboratory, we have noticed reduced
number of peripheral neurites from DRG neurons innervating MRMT1
tumor in the rat tibia. This seemingly contradictory observation in
neuritogenesis in the animal model could perhaps be attributed to the
displacement and insufficient adaptation of innervating fibers in the
tumor vicinity due to the fast-growing MRMT1 cells within a 2-week
time period. More studies are warranted to understand this
incongruency.
Conversely, CM from DRG cultures harvested from tumor-
burdened rats, induced enhanced migration of MRMT1 cells. Sensory
neuron-driven cell migration has been previously attributed to the
expression of axon guidance receptors like the Plexin receptor in
tumor cells.
36,37
Additionally, the plexin family of guidance receptors
is known to be key regulators of several cancer phenotypes.
38,39
Intriguingly in our study, we observed enhanced migration of cancer
cells treated with CM from DRG cultures harvested from tumor-
burdened animals, but not from naive animals. However, it should be
noted that satellite glial cells and Schwann cells are an inherent part
of a DRG culture and are required for normal functioning of sensory
neurons.
40
Hence, from our study, it is not quite apparent whether
the DRG secretome-mediated effect on MRMT1 cell migration is
purely due to sensory neurons or, due to the combined effect of glial
and neuronal cells. Additionally, it is important to note that MRMT1
primed PC12 cells did not alter MRMT1 cell migration in our study,
which perhaps could be because the PC12 cell line originates from
pheochromocytoma cells and may not completely mimic the effects of
neurons.
41
However, more studies are warranted to further under-
stand the mechanism by which cancer-sensitized peripheral sensory
neurons, and/or glial cells drive changes in cancer cell migration. This
will open new ways to disrupt nerve-cancer interactions to possibly
target breast cancer metastasis, though preliminary evidence indicates
that beta-blockers and tricyclic drugs can indeed reduce metastasis
and increase survival rate in cancer patients.
42,43
Results from our
study further support these observations and are in line with the
evolving view that the tumor microenvironment is a milieu of neuro-
transmitters, neuropeptides and neurotrophins that are constantly
acting on nerve terminals and cancer cells engaging in a feed-forward
mechanism.
In conclusion, evidence from our study supports the hypothesis
that both tumor cells and nerve terminals secrete tropic messengers
that have a reciprocal effect on each other. Further studies are
needed to evaluate the nature of these secreted factors, which may
eventually lead to the identification of targets for addressing nerve-
dependent metastasis.
AUTHOR CONTRIBUTIONS
Chinnu Jerard: Investigation, methodology, data curation, visualiza-
tion, writing-original draft. Pallavi Madhusudanan: investigation, data
curation. Aditi Swamy: Investigation. Karishma Ravikumar: Investiga-
tion. Sahadev A. Shankarappa: Conceptualization, supervision, project
administration, funding acquisition, writing-review & editing and for-
mal analysis. The work reported in the article has been performed by
the authors, unless clearly specified in the text.
8JERARD ET AL.

ACKNOWLEDGEMENTS
The authors would like to acknowledge Mr. Arun, Mr. Vishnu,
Mr. Sajith, Mr. Sunil, Mrs. Shanthini, Mrs. Sunitha, Dr. Sreekumar
Kanoth and Dr. A.K.K Unni from the central lab animal facility, AIMS,
for their enormous help with animal tissue harvest. We also
thank Dr. Krishnakumar Menon, Dr. Neeraj Katiyar, Mr. Vignesh,
Mrs. Gayathri Raju and Ms. Anushka Banerjee for scientific discus-
sions and valuable suggestions. We thank Dr. Sanjay Pal for gener-
ously providing the centrifugal filter units when most needed.
FUNDING INFORMATION
This work was supported in part, by grants from the Department of
Biotechnology—BT/PR24515/MED/30/1926/2017 and from The
Nanomission, Department of Science and Technology—DST/NM/NS-
282/2019 (G), Government of India, to SS. PM received support from
CSIR-SRF fellowship (09/963[0046]-2K19EMR-1).
CONFLICT OF INTEREST STATEMENT
Authors declare no competing interest in this work.
DATA AVAILABILITY STATEMENT
All data depicted in this article, including raw data from replicate
experiments, analysis and statistical testing have been uploaded into a
freely accessible repository. This information can be accessed at the fol-
lowing link: https://doi.org/10.6084/m9.figshare.c.6049100. Further
information is available from the corresponding author upon request.
ETHICS STATEMENT
The study was conducted using protocols approved by the Institu-
tional Animal Ethical Committee, Amrita Institute of Medical Sciences,
Kochi, India in accordance with guidelines set forth by the Committee
for the Purpose of Control and Supervision of Experiments on Ani-
mals, Government of India.
ORCID
Chinnu Jerard https://orcid.org/0000-0003-4111-4867
Sahadev A. Shankarappa https://orcid.org/0000-0002-7302-6362
TWITTER
Sahadev A. Shankarappa @Sahadev_Shankar
REFERENCES
1. Wang W, Li L, Chen N, et al. Nerves in the tumor microenvironment:
origin and effects. Front Cell Dev Biol. 2020;8:1-16.
2. Li X, Peng X, Yang S, et al. Targeting tumor innervation: premises,
promises, and challenges. Cell Death Discov. 2022;8(1):1-12.
3. Huang D, Su S, Cui X, et al. Nerve fibers in breast cancer tissues indi-
cate aggressive tumor progression. Medicine. 2014;93(27):1-10.
4. Ferdoushi A, Griffin N, Marsland M, et al. Tumor innervation and clini-
cal outcome in pancreatic cancer. Sci Rep. 2021;11(1):7390.
5. Li S, Sun Y, Gao D. Role of the nervous system in cancer metastasis
(review). Oncol Lett. 2013;5(4):1101-1111.
6. Kuol N, Stojanovska L, Apostolopoulos V, Nurgali K. Role of the ner-
vous system in cancer metastasis. J Exp Clin Cancer Res. 2018;37(1):
1-12.
7. Zahalka AH, Frenette PS. Nerves in cancer. Nat Rev Cancer. 2020;
20(3):143-157.
8. Wang H, Zheng Q, Lu Z, et al. Role of the nervous system in cancers:
a review. Cell Death Discov. 2021;7(1):1-12.
9. Ondicova K, Mravec B. Role of nervous system in cancer aetiopatho-
genesis. Lancet Oncol. 2010;11(6):596-601.
10. Hunt PJ, Kabotyanski KE, Calin GA, Xie T, Myers JN, Amit M. Inter-
rupting neuron-tumor interactions to overcome treatment resistance.
Cancers (Basel). 2020;12(12):1-19.
11. Chen SH, Zhang BY, Zhou B, Zhu CZ, Sun LQ, Feng YJ. Perineural
invasion of cancer: a complex crosstalk between cells and
molecules in the perineural niche. Am J Cancer Res. 2019;9(1):
1-21.
12. Jobling P, Pundavela J, Oliveira SMR, Walker MM, Hondermarck H.
Nerve-cancer cell cross-talk: a novel promoter of tumor progression.
Cancer Res. 2015;75(9):1777-1781.
13. Szpunar MJ, Belcher EK, Dawes RP, Madden KS. Sympathetic inner-
vation, norepinephrine content, and norepinephrine turnover in
orthotopic and spontaneous models of breast cancer. Brain Behav
Immun. 2016;53:223-233.
14. Reavis HD, Chen HI, Drapkin R. Tumor innervation: cancer has some
nerve. Trends Cancer. 2020;6(12):1059-1067.
15. Kamiya A, Hayama Y, Kato S, et al. Innervation and activity and its
effect on breast. Nat Neurosci. 2019;22:1289-1305.
16. Austin M, Elliott L, Nicolaou N, Grabowska A, Hulse RP. Breast cancer
induced nociceptor aberrant growth and collateral sensory axonal
branching. Oncotarget. 2017;8:76606-76621.
17. Pundavela J, Roselli S, Faulkner S, et al. Nerve fibers infiltrate the
tumor microenvironment and are associated with nerve growth factor
production and lymph node invasion in breast cancer. Mol Oncol.
2015;9:1626-1635.
18. Mathew SE, Madhusudanan P, Shankarappa SA. Effect of peritumoral
bupivacaine on primary and distal hyperalgesia in cancer-induced
bone pain. J Pain Res. 2020;13:1305-1313.
19. Shankarappa SA, Piedras-Rentería ES, Stubbs EB. Forced-exercise
delays neuropathic pain in experimental diabetes: effects on
voltage-activated calcium channels. JNeurochem. 2011;118(2):
224-236.
20. Matsuo K, Ji S, Miya A, et al. Innervation of the tibial epiphysis
through the intercondylar foramen. Bone. 2019;120:297-304.
21. Banno Y, Nemoto S, Murakami M, et al. Depolarization-induced dif-
ferentiation of PC12 cells is mediated by phospholipase D2 through
the transcription factor CREB pathway. J Neurochem. 2008;104(5):
1372-1386.
22. Shurin MR, Shurin G, Zlotnikov SB, Bunimovich YL. The neuroim-
mune axis in the tumor microenvironment. J Immunol. 2020;204(2):
280-285.
23. Magnon C, Hall SJ, Lin J, et al. Autonomic nerve development con-
tributes to prostate cancer progression. Science. 2013;341(6142):
713-714.
24. Restaino AC, Vermeer PD. Neural regulations of the tumor microenvi-
ronment. FASEB Bioadv. 2022;4(1):29-42.
25. Humsa V, Michelle M. Neuronal activity in ontogeny and oncology.
Trends Cancer. 2017;3(2):89-112.
26. Hathout Y. Approaches to the study of the cell secretome. Expert Rev
Proteomics. 2007;4(2):239-248.
27. da Cunha BR, Domingos C, Buzzo Stefanini AC, et al. Cellular interac-
tions in the tumor microenvironment: the role of secretome. J Cancer.
2019;10(19):4574-4587.
28. Sainath R, Gallo G. Cytoskeletal and signaling mechanisms of neurite
formation. Cell Tissue Res. 2014;359(1):267-278.
29. Amaral LD, Santos NAGD, Sisti FM, Del Bel E, Santos ACD. The anti-
biotic doxycycline mimics the NGF signaling in PC12 cells: a relevant
mechanism for neuroprotection. Chem Biol Interact. 2021;341:
109454.
JERARD ET AL.9

30. Schweizerhof M, Stösser S, Kurejova M, et al. Hematopoietic colony-
stimulating factors mediate tumor-nerve interactions and bone cancer
pain. Nat Med. 2009;15(7):802-807.
31. He S, Chen CH, Chernichenko N, et al. GFRα1 released by nerves
enhances cancer cell perineural invasion through GDNF-RET signal-
ing. Proc Natl Acad Sci U S A. 2014;111(19):E2008-E2017.
32. Hayakawa Y, Sakitani K, Konishi M, et al. Nerve growth factor pro-
motes gastric tumorigenesis through aberrant cholinergic signaling.
Cancer Cell. 2017;31(1):21-34. doi:10.1016/j.ccell.2016.11.005
33. Demont Y, Corbet C, Page A, et al. Pro-nerve growth factor induces
autocrine stimulation of breast cancer cell invasion through
tropomyosin-related kinase A (TrkA) and sortilin protein. J Biol Chem.
2012;287(3):1923-1931.
34. Park SH, Eber MR, Fonseca MM, et al. Usefulness of the measure-
ment of neurite outgrowth of primary sensory neurons to study
cancer-related painful complications. Biochem Pharmacol. 2021;188:
114520. doi:10.1016/j.bcp.2021.114520
35. Okui T, Hiasa M, Ryumon S, et al. The HMGB1/RAGE axis induces
bone pain associated with colonization of 4T1 mouse breast cancer in
bone. J Bone Oncol. 2021;26:100330. doi:10.1016/j.jbo.2020.100330
36. Gurrapu S, Tamagnone L. Semaphorins as regulators of phenotypic
plasticity and functional reprogramming of cancer cells. Trends Mol
Med. 2019;25(4):303-314.
37. Saxena S, Hayashi Y, Wu L, et al. Pathological and functional signifi-
cance of Semaphorin-5A in pancreatic cancer progression and metas-
tasis. Oncotarget. 2018;9(5):5931-5943.
38. Tamagnone L. Emerging role of Semaphorins as major regulatory sig-
nals and potential therapeutic targets in cancer. Cancer Cell. 2012;
22(2):145-152.
39. Zhang X, Shao S, Li L. Characterization of class-3 semaphorin recep-
tors, neuropilins and plexins, as therapeutic targets in a pan-cancer
study. Cancers (Basel). 2020;12(7):1-24.
40. Madhusudanan P, Reade S, Shankarappa SA. Neuroglia as targets for
drug delivery systems: a review. Nanomedicine. 2017;13(2):667-679.
doi:10.1016/j.nano.2016.08.013
41. Greene LA, Tischlert AS. Establishment of a noradrenergic clonal line
of rat adrenal pheochromocytoma cells which respond to nerve
growth factor (sympathetic neurons/cell culture/catecholamines/dif-
ferentiation/neurites). Cell Biol. 1976;73:2424-2428.
42. Fitzgerald PJ. Beta blockers, norepinephrine, and cancer: an epidemi-
ological viewpoint. Clin Epidemiol. 2012;4(1):151-156.
43. Bielecka AM, Obuchowicz E. Antidepressant drugs as a complemen-
tary therapeutic strategy in cancer. Exp Biol Med. 2013;238(7):
849-858.
SUPPORTING INFORMATION
Additional supporting information can be found online in the Support-
ing Information section at the end of this article.
How to cite this article: Jerard C, Madhusudanan P, Swamy A,
Ravikumar K, Shankarappa SA. Secretome mediated
interactions between sensory neurons and breast cancer cells.
Int J Cancer. 2023;1‐10. doi:10.1002/ijc.34529
10 JERARD ET AL.

Secretome mediated interactions between sensory neurons and
breast cancer cells
Chinnu Jerard1, Pallavi Madhusudanan1, Aditi Swamy1, Karishma Ravikumar1
Sahadev A Shankarappa1 *
Table of Contents
Supplementary Materials and Methods 2
Supplementary Figure 1 3
Supplementary Figure 2 4

Supplementary Materials and Methods
PC12 neurite assay in fibroblast conditioned medium
PC12 cells were seeded on collagen-coated (100 µg/mL) 96-well plate with seeding
density of 7800 cells/cm2, and maintained at 37ºC in 5% CO2 under humidified
condition for 24h. All cells were maintained in medium containing RPMI-1640 with 10%
(v/v) HS, 5% (v/v) FBS and 1% (v/v) PS. Cultured cells were then exposed to
conditioned media obtained from L929 cells for 72h, followed by intracellular loading
with the membrane permeable calcein dye (1 μg/mL), and imaged under a
fluorescence microscope (Leica DMI 3000B, 100x magnification). Five random images
were acquired per well. Neurite length, number of neurite bearing cells, and neurite
branching pattern were analysed using the neurite outgrowth software application
module (Molecular Devices, USA).
Cell migration assay in response to MRMT1 primed PC12 cell secretome
MRMT1 cultures (10,000 cells/cm2) treated with conditioned media (50%) derived from
3-d PC12 cultures primed with MRMT1 CM were subjected to a modified scratch
assay. Briefly, 80-90% confluent MRMT1 cells cultured in medium containing RPMI–
1640, 10% (v/v) FBS, and 1% (v/v) PS, were scrapped off cells using a cell scraper
with 1.8 cm wide sterile blade, and the cell plates were imaged every 4th hour for 12h.
The distance travelled by the cells relative to a pre-marked reference line (white dotted
line, figure 5) on the well plate was measured using a Nikon Eclipse TE2000-U
microscope fitted with a Nikon digital sight camera at 40 times magnification. Only cells
migrating beyond 200 µm from the edge of the original scratch were considered as
migratory and used for analysis. The analysis was performed using ImageJ (NIH)
software. Serum depletion, a common strategy utilized by some studies for migration
assay, was not adopted in this study since we observed reduced MRMT1 viability in
serum depleted medium.

Supplementary Figure 1
Supplementary Figure 1. Conditioned medium from cultured mouse fibroblast cells
did not induce PC12 cell neurite outgrowth. PC12 cells were treated with naïve
medium, 10% and 50% L929-derived CM for 72h. Bar graph depicting the number of
neurite bearing cells (left) along with neurite length (right) from analysed PC12 cells.
Data shown are mean ± SD obtained from 3 separate experiments performed in
triplicate. Comparative analysis between indicated groups were done using one-way
ANOVA with Bonferroni’s multiple comparisons tests.

Supplementary Figure 2
Supplementary Figure 2. Conditioned medium from MRMT1 sensitized PC12 cells
did not alter the migration of scratch-injured MRMT1 cells. MRMT1-sensitized PC12
CM was obtained by culturing PC12 cells in 50% MRMT1 CM for 72 h. Fresh 80-90%
confluent MRMT1 cells cultured in 12-well plates with 50% cancer-sensitized PC12
CM were subjected to scratch injury. Bar graph depicting the number of migrating
MRMT1 cells after a scratch injury is shown above. Data shown are mean ± SD
obtained from one individual experiment done in triplicate. Comparative analysis
between indicated groups were performed using unpaired Students’ t test.