Content uploaded by Stefano Fais
Author content
All content in this area was uploaded by Stefano Fais on Mar 18, 2019
Content may be subject to copyright.
Tumour cell cannibalism is a long- standing
story that was virtually neglected for up to
a century but is now receiving increasing
attention. The first reports describing the
presence of cells within other cells in tumour
tissues emerged approximately 120 years
ago1,2. For decades after these few reports,
cannibalism was considered only a curiosity
or simply a matter for histopathologists,
with the literature consisting mostly of
cytopathological reports. From a mere
cytological point of view, cannibalism
was recognized when cell preparations
or tumour tissues contained cells with a
crescent- shaped nucleus in turn containing
another smaller cell, often contained within
a large vacuole. The crescent shape of the
nucleus was due to the cell- containing
vacuole displacing the nucleus towards
the cell periphery. This morphological
appearance of cannibalism led to the
emergence of imaginative terminology such
as ‘bird’s- eye cells’3 and the more common
‘cell- in- cell’4.
While the cytological appearance of
cannibalistic cells that are common in cancer
specimens remains a key defining feature
(BOX1), within the past decade, numerous
mechanistic studies have provided evidence
that cancer cell cannibalism actually results
from a set of complex and distinct cell
processes that are utilized by cancer cells
to gain advantages that promote malignant
progression. Cell- in-cell structures can form
as a result of the activation of cannibalistic
mechanisms that are, in some ways,
cancers13. Scavenging activity for proteins
present in the microenvironment through
macropinocytic engulfment has been shown
to support cancer cell proliferation when
amino acids from vasculature are lacking14.
Similarly, like unicellular organisms, cancer
cells can also engulf whole cells from their
surroundings to scavenge extracellular
nutrients in bulk.
Metastatic human melanoma cells were
first shown to exert potent engulfment
activity directed against both inert material
and apoptotic cells15 (FIG.1a). The fact that
this activity was a property of metastatic
cells, while virtually absent in melanoma cells
deriving from primary tumours, piqued
interest in the field that engulfment activity
could be selected during cancer progression.
Metastatic melanoma cells were able to
engulf and digest even live cells, an activity
seemingly unrelated to macrophage
behaviour. This observation suggested that
these malignant cells might have acquired
a broad activity to engulf both live and
dead cells, potentially in order to feed, an
activity that was called cannibalism. When
metastatic melanoma cells were co- cultured
with live lymphocytes, they were found
to cannibalize them at high rates, which
allowed the metastatic cells to survive even
under serum- starved conditions, while
the ingestion of latex beads had no effect
to promote survival, and cancer cells died
quickly7. These findings established cell
cannibalistic activity, directed towards dead
cells, live cells and even cells of a different
cell type (Tcells), as an important metabolic
adaptation of cancers.
Recently, a similar cannibalism activity
was demonstrated in the breast cancer
cell line MDA- MB-231, which was shown
to cannibalize mesenchymal stem cells
(MSCs) at high rates when mixed in 3D
co- cultures designed to mimic metastasis
to bone16. Cannibalism of MSCs, like Tcell
cannibalism by metastatic melanoma cells,
could support the survival of breast cancer
cells cultured under starvation conditions
in low serum and further promoted the
acquisition of phenotypic traits resembling
dormancy16. Thus, cannibalistic behaviour
by cancer cells directed towards cells of
other types consistently confers metabolic
advantages, which may be the reason for the
occurrence of cannibalism in cancer.
reminiscent of macrophage phagocytosis5
or by distinct activities that involve the
invasion of cells into each other rather than
engulfment6. These processes each result
in the formation of similar cytological
cell- in-cell structures but are referred
to by names that denote their distinct
cellular mechanisms, including those that
are thought to occur primarily through
engulfment, which have been called
cannibalism7, homotypic cell cannibalism8
and phagoptosis9, and distinct processes
that instead occur through an invasion
mechanism, called entosis6, suicidal
emperipolesis10 and emperitosis11. Here,
wereview recent advances in elucidating
theroles in cancer progression of two
of these related, yet distinct, cell- in-cell
processes: cannibalism and entosis.
Cannibalism: cancer cells feed
The mentor of modern immunology,
Metchnikoff, in his initial studies on
phagocytosis, started from investigations
of unicellular organisms, which actually
ingest other microorganisms in order to
feed upon the ingested material. It was only
later that he discovered the existence of
circulating cells that used phagocytosis to
protect the higher organisms from external
agents, a milestone discovery of modern
immunology12. Cancer cells, like unicellular
organisms, have a need to scavenge nutrients
from their environment, particularly under
conditions where tumour vasculature
is deficient, which is common in many
OPINION
Cell- in-cell phenomena in cancer
StefanoFais and MichaelOverholtzer
Abstract | Cell- in-cell structures are reported in numerous cancers, and their
presence is an indicator for poor prognosis. Mechanistic studies have identified
how cancer cells manage to ingest whole neighbouring cells to form such
structures, and the consequences of cell- in-cell formation on cancer progression
have been elucidated. In this Opinion article, we discuss how two related cell-
in-cell processes, cell cannibalism and entosis, are regulated and how these
mechanisms promote cancer progression. We propose that cannibalistic activity
is a hallmark of cancer that results in part from selection by metabolic stress and
serves to feed aggressive cancer cells.
PERSPECTIvES
Nature reviews
|
CanCer
www.nature.com/nrc
PersPectives
Mechanism of cell cannibalism. The
mechanism underlying cannibalistic activity
in cancer cells resembles phagocytosis to
some extent: both processes can be directed
against apoptotic cells, cellular debris or
inert material such as latex beads (FIG.1a,b),
both are associated with expression within
engulfing cells of a common cell surface
marker, CD68, and both involve actin-
dependent engulfment activity. Whereas
phagocytosis is typically directed towards
dying or dead cells, cannibalistic cells could
engulf live cells as well, suggesting a distinct
mechanism. Cancer cell cannibalism shares
mechanistic similarities with phagocytosis
Box 1 | Published reports of cell- in-cell activity in cancer
Cancer in general
•Steinhaus(1891)1
•Stroebe(1892)2
•Guptaetal.(2003)56
•Sharmaetal.(2011)77
•Singhaletal.(2011)78
•Bansaletal.(2011)55
•Chenetal.(2013)70
•Singhetal.(2013)79
•Ferreiraetal.(2015)80
•Kale(2015)81
•Schwegleretal.(2015)57
•Huangetal.(2015)82
•Melendezetal.(2015)83
Acute megakaryoblastic leukaemia
•Morales-Camachoetal.(2016)84
Angiosarcoma
•Jeonetal.(2004)85
Breast carcinoma
•Fujiietal.(1986)86
•Abodiefetal.(2006)87
•Overholtzeretal.(2007)6
•Krajcovicetal.(2011)69
•Almeida(2015)88
•Durganetal.(2017)33
•Kinoshitaetal.(2018)89
Carcinoma in bile fluid
•Nakajimaetal.(1994)90
Central giant cell granuloma
•Sarodeetal.(2014)91
•Azzietal.(2017)92
Cervical carcinoma
•Ngetal.(2003)93
•Gamezetal.(2009)94
•Abdullgaffar(2017)95
Chronic periapical lesions
•Kaleleetal.(2016).96
Gallbladder carcinoma
•Carusoetal.(1991)97
Gastric carcinoma
•Carusoetal.(2002)98
•Carusoetal.(2013)99
•Barresi(2015)100
Giant cell tumours of bone
•Sarodeetal.(2017).101
Head and neck cancer
•Schenkeretal.(2017)58
Lung carcinoma
•DeSimoneetal.(1980)102
•Brouweretal.(1984)103
•Conwayetal.(2013)104
•Mackayetal.(2018)46
Medulloblastoma
•Escamillaetal.(2015)105
Melanoma
•Beatyetal.(1997)106
•Luginietal.(2003)15
•Luginietal.(2006)7
Mesothelioma
•Ehya(1986)107
•Stevensetal.(1992)108
•Ylaganetal.(2005)109
•Kimuraetal.(2009)110
•Cakiretal.(2009)111
•Matsumotoetal.(2013).112
Metastatic epithelioid haemangioendothelioma
•Abatietal.(1994)113
Oral squamous cell carcinoma
•Sarodeetal.(2012)114
•Sarodeetal.(2014)115
•Joseetal.(2014)116
•Jain(2015)117
•Jainetal.(2017)118
•Sarodeetal.(2017)119
Ovarian steroid cell tumour, malignant
•Kosakaetal.(2017)120
Pancreatic carcinoma
•Silvermanetal.(1988)121
•Silvermanetal.(1989)122
•Guptaetal.(1992)123
•Khayyata(2005)124
•Canoetal.(2012)8
Prostate carcinoma
•Wenetal.(2013)125
•Gilloteauxetal.(2016)126
Renal carcinoma
•Kongetal.(2015)127
Salivary duct carcinoma
•Aryaetal.(2011)128
Soft tissue tumour
•Fernandez-Flores(2012)129
Squamous cell carcinoma
•HuangandMichael(2014)130
Urinary tract cancer
•Logothetou-Rella(1994)131
•Kojimaetal.(1998)132
•Deyetal.(2005)133
•Hattorietal.(2007)134
•Ohsakietal.(2010)135
•Ahmed(2015)136
a Breast cancer
b Breast cancer
c
d Head and neck
cancer
e Oral squamous
cell carcinoma
e Oral squamous
cell carcinoma
Salivary duct
carcinoma
Inthefigure,cell-in-cellstructuresfromhuman(parta)andfeline(partb)breastcanceraredepicted.Imaging
inpartashowsdifferentialinterferencecontrastandimmunofluorescencestaining(greenisβ-catenin,and
redisphospho-histoneH3serine10(pHH3);scalebar,10μm),andpartbshowshaematoxylinandeosin(H&E)
staining.Neutrophilcannibalisminhumansalivaryductcarcinoma,withH&Estaining,isshowninpartc.
Incell-in-cellstructuresinheadandneckcancer(partd),immunohistochemicalstainingforE-cadherinis
shown.Cell-in-cellstructuresinhumanoralsquamouscellcarcinoma,withH&Estaining,areshowninparte.
PartaadaptedfromREF.33,CC-BY-4.0.PartbadaptedfromREF.80,CC-BY-4.0.PartcadaptedfromArya,P.,
Khalbuss,W.E.,Monaco,S.E.&Pantanowitz,L.Salivaryductcarcinomawithstrikingneutrophil-tumorcell
cannibalism.Cytojournal8,15(2011),WoltersKluwerMedknowPublications.PartdadaptedfromREF.57,
CC-BY-3.0.ParteadaptedwithpermissionfromREF.114,Elsevier.
in the unicellular organism Dictyostelium
discoideum, which engulfs microorganisms
as a primary means of supporting
metabolism. The gene phg1a (which
encodes putative phagocytic receptor1a)
was discovered to control the phagocytic
activity of D. discoideum by regulating the
binding between the amoeba cells and
bacteria, a recognized early step of the
phagocytic process17. Phg1A is also called
transmembrane 9 protein (TM9) in other
organisms and belongs to a highly conserved
family of transmembrane proteins that
includes three members in D. discoideum
(Phg1A, Phg1B and Phg1C) and yeasts
(TM9 superfamily member1 (Tmn1)Tmn3)
and four members in Drosophila
melanogaster and vertebrates, including
humans (TM9 superfamily member 1
(TM9SF1)–TM9SF4). In D.melanogaster
TM9 proteins have been shown to control
phagocytosis by haemocytes18,19. In
humans, the TM9SF3 gene is upregulated
inpaclitaxel- resistant cell lines and in gastric
cancer, where it is associated with increased
local invasion and tumour progression20,21.
TM9SF4 is overexpressed in acute myeloid
leukaemia and myelodysplastic syndromes22
and in human melanoma cells derived
from metastatic lesions23. Importantly, in
metastatic melanoma cells, TM9SF4 is
required for engulfment activity against
either yeast or live lymphocytes, with its
knockdown abrogating the cannibalistic
activity of melanoma23.
While the molecular mechanism
underlying how TM9 proteins control
phagocytosis and cannibalism awaits
further characterization, intriguingly, a
recent study has shown that phg1-knockout
D.discoideum are also unable to efficiently
kill bacteria, probably owing to a loss
of lysosomal lytic enzymes24. TM9SF2
and TM9SF4 (the closest orthologue of
Phg1A) are also expressed on endosomes
in mammalian cells and exert a role in
phagocytosis of bacteria25. TM9SF4 has
been shown to colocalize with the vacuolar-
type H+-ATPase (V- ATPase), cooperating
with the V- ATPase for pH regulation23,26.
Downregulation of TM9SF4 in cancer cells,
or of Phg1A in D. discoideum, also leads
to an altered pH gradient in cells, with a
decrease in cytosolic pH and an increase in
extracellular pH25,26, suggesting that TM9
proteins exert homeostatic control of pH at
the cellular level. pH gradients play a key
role in many cellular functions, including
proteolytic enzyme activation, cell- to-cell
interaction, cell- to-extracellular vesicle
interaction, receptor- to-ligand interactions,
drug uptake and additional endocytic
and phagocytic pathways27. It is therefore
possible that TM9SF4 contributes to an
acidic cancer microenvironment27,28, which
also could induce cannibalistic activity7.
Overall, how cannibalism differs from
phagocytic activity is an important, ongoing
question. The mechanism of cannibalism
has also been shown to involve the cell
cortical protein ezrin15, which is required
for metastatic cells to feed, as well as actin15
and vesicles of the caveolar network7 (FIG.1a).
How these components act in concert with
TM9 proteins will be important to uncover.
Entosis: cancer cells compete
Entosis (derived from the Greek ἐντός
(entos), meaning within, and -ωσις (-osis),
meaning disease)6 is a different way for
cancer cells to form cell- in-cell structures
(FIG.1c). Cell uptake during entosis in cancer
cells occurs via E- cadherin and P- cadherin.
Because cadherins primarily bind each
other in a homotypic manner, entosis is
thought to occur mostly between cells of the
same type, independent of their malignant
state, which is unlike cannibalistic activity
in metastatic melanoma that was directed
against Tcells. After establishing cell–cell
adhesions, the engulfed cells actively
drive their own uptake into neighbouring
cells (FIG.1d). This aspect of entosis has
remained enigmatic but is a feature
Nature reviews
|
CanCer
PersPectives
a
Cannibalism bc
d
Entosis
Live
cell
Latex
bead
Ezrin
Actin
Caveolin
Dead
cell
Cancer
cell
• Low energy
• Aberrant mitosis
• Matrix detachment
Loser
cell
Winner
cell
MRTF
MRTF
RHO
ROCK
KRAS
↓ATP
↑Tension ↑ATP
↓Tension
E-cadherin or
P-cadherin
PDZ-
RHOGEF
Ezrin
DIA1
p190-
RHOGAP
Fig. 1 | Regulation of cannibalism and entosis. a | Cancer cells can cannibalize live cells (for example,
Tcells, neutrophils and natural killer cells), dead cells and non- living material (for example, beads and
other particles) through a mechanism involving actin, ezrin and caveolin 1. b | Metastatic melanoma
cells with cannibalistic activity , with ingested latex beads are shown. Red represents immunostain-
ing for CD68, and blue represents 4ʹ,6-diamidino-2-phenylindole (DAPI)-stained nuclei.
c | Entosis between breast cancer cells is shown. Blue arrows indicate ingested and killed cells, white
arrows indicate ingested live cells, and red arrows show partially internalized cells. Green represents
E- cadherin immunostaining, red represents lysosomal membrane staining (lysosomal- associated
membrane protein 1 (L AMP1)), and blue represents DAPI- stained nuclei. d | Entosis involves the inva-
sion of one cell (the loser cell, green) into another (the winner cell, blue) induced in response to any of
the following stimuli: low energy states and activation of AMP- activated protein kinase (AMPK) in
losers, aberrant shape regulation in mitotic cells and detachment of cells from the extracellular matrix.
The right schematic shows a partial entotic structure. Increased cell tension regulated by RHO–RHO-
associated coiled- coil-containing protein kinase (ROCK) and/or diaphanous- related formin 1
(DIA1) –actomyosin in losers drives uptake following establishment of cell–cell adhesion (E- cadherin
or P-cadherin). Tension is distributed to the back of loser cells in part by p190-RHO GTPase- activating
protein (p190-RHOGAP) at cell–cell junctions and PDZ domain- containing RHO guanine nucleotide
exchange factor (PDZ- RHOGEF) at the cortex30,31. Entosis involves transcriptional upregulation of ezrin
through myocardin- related transcription factor (MRTF) that is recruited to the nucleus in response to
increased cortical tension32. KRAS activation and relative high- energy states are linked to winner cell
identity owing to a relative reduction in cell tension compared with losers.
supported by multiple lines of evidence
across numerous studies6,29–37. Dying cells
engulfed by phagocytosis also participate
actively in their clearance by presenting
the ‘eat- me’ signal phosphatidylserine in
an ATP- dependent manner. Rather than
exposing eat- me signals, entotic cells
instead provide the key driving force,
through actin polymerization and myosin
contraction, to promote the ingestion
process itself 6. Actomyosin contraction in
invading cells is regulated by key controllers
of cell tension, including RHOA, RHO-
associated coiled- coil-containing protein
kinase (ROCK)6 and diaphanous- related
formin 1 (DIA1)31 (FIG.1d). The invading
cells, and not engulfing cells, accumulate
actin and myosin at the cell cortex,
opposite the cell adhesions that form at
the engulfment interface, and the resulting
mechanical tension drives the cell- in-cell
invasion process6. This mechanism may
have substantial energetic implications for
entosis occurring in cancer cell populations
as compared with other mechanisms of
cell engulfment. The active involvement
of invading cells in this process also
distinguishes entosis from cannibalistic
or phagocytic forms of engulfment, as
thismechanism excludes dead cells or
non- living material.
Once ingested, internalized entotic
cells are killed by their hosts following
maturation of the endocytic membrane,
called the entotic vacuole, that surrounds
them6. Maturation of this vacuole involves
modification by autophagy pathway
proteins that direct lipidation of the
autophagosomal protein microtubule-
associated protein 1 light chain 3 (LC3)
onto the entotic vacuole, which is followed
by lysosome fusion and internalized cell
death and degradation38. Inthis manner,
autophagy pathway proteins, which
normally function to recycle intracellular
components through lysosomal digestion,
play a key role in the lysosomal scavenging
of bulk extracellular nutrients derived from
ingested entotic cells.
While this mechanism of cell death,
called entotic cell death, is the most
common fate for cells that are ingested
by entosis, other fates are also possible.
Internalized cells can undergo apoptosis,
particularly when lysosome function
or autophagy proteins are disrupted38.
Internalized cells can also escape from
their hosts and emerge unharmed, a fate
that is upregulated substantially when both
lysosome function or autophagy genes and
apoptosis are disabled6,38. Internalized cells
in clinical cancer specimens show evidence
of DNA fragmentation (visualized by TdT
dUTP nick end labelling (TUNEL)) in the
absence ofcleavage of caspase 3, which is a
hallmarkof apoptosis, suggesting that non-
apoptotic cell death may be a common fate
for entotic cells in human cancers6.
Entotic induction and competition — a
driver for cancer evolution? Recent
evidence has shown that entosis can
promote competition between cancer cells,
which can affect the evolution of cancer cell
populations29. Serial detachment of mixed
cancer cell populations from extracellular
matrix, the first known inducer of this
process6,39, promoted high rates of entosis,
which led to the progressive elimination of
ingested cells, or losers, as the engulfing, or
winner, cells accumulated29. Loser cells in
this context were identified by mechanical
tension and were stiffer than winners. Ahigh
rate of entosis therefore selects for cells
with increased mechanical deformability,
a known property of malignant cells, and
indeed cancer cells preferentially eliminate
non- cancer cells by entosis when cultured in
mixed populations29.
The concept of cell competition was first
revealed from studies of mosaic tissues in
D.melanogaster more than 40years ago.
During development, it was observed that
the individual cells within a tissue could
sense relative fitness. Winner cells, identified,
for example, by a normal complement of
ribosome genes, overpopulated tissues as a
result of eliminating loser cells with ribosome
gene knockout. Wild- type cells could also
be recognized as losers and eliminated
when neighbouring cells overexpressed
oncogenes, such as Myc, which rendered
them ‘supercompetitors’40–42.
While apoptosis and cell extrusion are
thought to be the major mechanisms that
eliminate loser cells in D. melanogaster, and
these also eliminate less- fit mammalian
cells, our findings revealed entosis as an
additional mechanism that can promote
competition in mammalian cells and
cancers43,44. Oncogene expression in
the form of activated KRAS29 or MYC45
as well as mutation of the p53 tumour
suppressor46 results in winner cell behaviour
in cancer cell lines where entosis is a major
competitive mechanism. Conceivably,
entosis could contribute to not only
evolution within cancers but also a pre-
cancerous effect called ‘field cancerization’,
where pre- cancerous cells expand within
normal tissues by competition, amplifying
the pool of abnormal cells with oncogene
or tumour suppressor mutations43. The
fact that cancer cells have been shown to
preferentially engulf non- cancer cells by
entosis suggests that this process could
contribute to such clonal spreading29.
In addition to matrix detachment,
entosis is regulated by nutrient signalling.
Nutrient starvation (serum, amino acids and
glucose) induces high rates of this process,
even when cells are adherent to matrix36.
Glucose was determined to be the major
nutrient that suppresses entosis induction
by inhibiting activity of the energy-
sensing AMP- activated protein kinase
(AMPK). After long periods of starvation
(for example, 48–72hours), entosis rates
increase significantly in a manner requiring
AMPK activity, which is induced to the
highest levels, intriguingly, in loser cells
(FIG.1). AMPK activity was also found to
be required in losers for entosis to occur36.
These findings identified AMPK as a major
inducer of this process linked directly to the
energy status of cells.
The regulation of entosis by AMPK
in starved cancer cell populations leads
to competitive interactions between cells
in which those with the lowest energy
levels, and concomitantly the highest
levels of AMPK activity, are sacrificed to
feed those with lowered AMPK activity.
This may be counterintuitive when
considered from a cell- autonomous point
of view, but a potentially rational choice
for population- scale survival. Long- term
survival of a population calls for the fittest
cells, potentially even under conditions of
chronic nutrient deprivation, to benefit
by feeding off of those that are less fit, a
concept central to the original findings
of cell competition in developing tissues.
These findings also suggest that cancer
cellsmay switch their behaviour after
prolonged periods of stress to select for
population survival rather thanto rescue
individual cells47.
AMPK is activated by direct binding of
AMP or ADP and responds to an increase
in the AMP to ATP or ADP to ATP ratio in
cells, which is elevated during starvation.
While entosis was known previously to
allow winner cells to scavenge nutrients
from ingested losers48, these new findings
reveal that entosis is induced in cancer cell
populations as a starvation response. During
chronic glucose starvation, cell ingestion by
entosis supports the outgrowth of selected
winner cells, which are able to continue
to divide at least in part by feeding off of
ingested losers36. This selected winner cell
population exhibits reduced levels of AMPK
activity and an ability to ingest naive cells
at high rates, demonstrating selection for
competitive activity.
www.nature.com/nrc
PersPectives
Actomyosin contraction and energy
expenditure may be distinctive features
of entosis. It may be important to consider
that the entosis mechanism differs from
phagocytosis or cannibalism in one key
aspect related to bioenergetics: the actin-
based regulation of the cell shape changes
that drive entosis occur within the ingested
cells rather than the engulfing cells. This
may have important implications for
entosis that could in part underlie its utility
in starved populations. Phagocytosis has
been shown to require energy expenditure
from engulfing cells, which is estimated
to be on the order of at least 1.5-fold to
2-fold that of the total cell complement of
ATP for macrophages ingesting numerous
red blood cells49. In large part, this is
thought to be due to ATP utilization for
the actin polymerization and myosin
contraction that drive the phagocytic
process, although endosome maturation
and degradative lysosomal functions that
require, for example, the ATP- dependent
V- ATPase proton pump also contribute.
Glycolysis has been shown to be required
for efficient phagocytosis by macrophages50,
presumably to support the energetic
demands of this process, and energy from
creatine phosphate stores has further been
shown to mobilize to phagocytic cups to
support the local energetic requirements
of engulfment51. For entosis, the energetic
demands of theactin- dependent processes
that drive the engulfment appear to
be largely handled by the invading
cells6, and the energy demands may be
substantial, involving not only actomyosin-
based tension but also transcriptional
upregulation of the cortical protein ezrin,
which is needed to maintain tension to
drive completion of the process32 (FIG.1d).
Whereas phagocytosis requires glycolysis,
as it is blocked by treatment of macrophages
with the non- metabolizable glucose
analogue 2-deoxyglucose49, by contrast,
entosis is induced under conditions
where glycolysis is inhibited by glucose
starvation36. In this context the expenditure
of energy from invading cells may spare
winners from exertion when the starved
population overall is limited for nutrients
and ATP and when nutrient recovery
from ingested cells is needed to support
cell growth and proliferation. Through
this mechanism, it is conceivable that the
metabolic gains for winners derived from
ingested cells may offset the expenditures,
when the engulfment process itself is
controlled largely by losers.
Other cancer cell- in-cell phenomena
Cannibalism and entosis are not the only
mechanisms that can lead to the ingestion
of whole live cells. Other processes include
suicidal emperipolesis, emperitosis and
phagoptosis. Emperipolesis was originally
described as the presence of an intact
cell within the cytoplasm of another cell,
conceivably owing to the active penetration
of one cell by another, and can refer to
the active penetration of Tcells within
others52. Suicidal emperipolesis accordingly
results in the ingestion and clearance of
CD8+ Tlymphocytes after degradation by
lysosomal proteolytic enzymes10, which
is thought to represent a mechanism to
eliminate autoreactive Tcells by ingestion
into hepatocytes. Emperitosis (derived
from emperipolesis and apoptosis) is
a caspase 3-dependent cell- in-cell apoptotic
cell death differing from entotic cell death
and typically appearing as a dying cytotoxic
T lymphocyte or natural killer cell within
a tumour cell vacuole11,53. Emperitosis could
contribute to immune cell evasion by cancers,
but its pathophysiological importance is not
yet well explored. Phagoptosis on the other
hand may be important for cancer therapy.
By this mechanism, phagocytic cells
(for example, macrophages) can ingest and
kill whole live target cells in a manner
resembling cancer cell cannibalism9. But
unlike cannibalism, phagoptosis is thought
to occur in cancers mostly as a result of
therapeutic interventions that involve the
targeting of cancer cells with antibodies that
engage phagocytosis. Notably, antibodies
designed to block the ‘don’t eat me’ signal
CD47 on cancer cells have been shown
to induce phagoptosis and potent cancer
regression in mouse models, suggesting
a potential therapeutic approach54. While
entosis and cannibalism by contrast occur
within cancers as they evolve, phagoptosis
may represent a largely therapeutic strategy
that harnesses the ability of phagocytic cells
to ingest live target cancer cells and kill them.
Conclusion
Among the open questions on the
relationships between cell- in-cell
phenomena and cancer, we believe that one
— are cell- in-cell structures a hallmark of
malignant cancers? — has been answered.
Cell- in-cell structures have been reported
for decades in clinical specimens4, and
their presence has been used as a marker
of malignancy55,56 with prognostic
significance46,57,58. Moreover, researchers
have now identified this cell behaviour
occurring in numerous cancer types in
humans or animal models (BOX1). Cell
behaviour leading to engulfment of cells
is therefore a common event occurring
during malignant transformation, and
we consider here that this is a hallmark of
cancer cell behaviour.
If cell- in-cell processes are a cancer
hallmark, why do cells do this? We posit
here that cell ingestion in cancer arises
at least in part owing to a common need
among all cancers to feed in order to
support the metabolism of individual cells.
Cannibalistic behaviour was first shown to
feed metastatic melanoma cells7 and recently
shown to feed breast cancer cells as well16.
Now, entosis has also been demonstrated
to occur at high levels between cells in
breast cancer populations in response to
nutrient starvation, where it is required
for population outgrowth36. A stringent
requirement to support cell metabolism
in cancers that often lack sufficient blood
supply may underlie selection for multiple
mechanisms, for example, cannibalism and
entosis, to enable this type of cell feeding
behaviour that increases intracellular
nutrient pools to support cell survival,
growth and proliferation (FIG.2).
Nature reviews
|
CanCer
PersPectives
Glossary
Cannibalism
The engulfment of live or dead cells or debris by cancer
cells as described in metastatic melanoma; also used
frequently as a general term for engulfment.
Cell- in-cell
A general term used to describe the appearance of
whole, typically live, cells ingested into other cells.
Emperipolesis
A general term that describes the uptake of live cells
and their movement inside host cells. Suicidal
emperipolesis is the mechanism of invasion of live
Tcells into hepatocytes.
Emperitosis
A mechanism resembling entosis and involving natural
killer cells that invade into cancer cells and die by
granzyme B- mediated cell death.
Entosis
The uptake of live cells into other cells through an
invasive mechanism; entosis typically involves cell–cell
adhesion proteins and actomyosin- mediated
contraction within invading cells regulated by RHO
GTPases and RHO- associated coiled- coil-containing
protein kinase (ROCK).
Homotypic cell cannibalism
A term describing a homotypic engulfment mechanism
suggested to involve pancreatic cancer cells.
Phagocytosis
The engulfment of dying or dead cells or
microorganisms through receptor- mediated
mechanisms.
Phagoptosis
The uptake and killing of live cells through phagocytosis.
By what mechanisms do cell- in-cell
structures form? We have reviewed here
two of the best- characterized mechanisms
underlying cell- in-cell activity in cancers:
cannibalism and entosis. While these
mechanisms clearly differ, more remains
to be learned about how these are
executed. Docannibalism and entosis
share mechanistic similarities? Recently,
a homotypic mechanism of cannibalism
was reported in pancreatic cancers called
homotypic cell cannibalism. While this
process, like entosis, targeted cells for
homotypic ingestion, it appeared to differ
from the entosis mechanism and did not
involve ROCK activity8. On the other hand,
cannibalistic activity by MDA- MB-231
breast cancer cells directed towards MSCs,
a heterotypic form of cell ingestion, was
argued not to occur by entosis and yet
intriguingly required ROCK16, suggesting
mechanistic parallels. Another heterotypic
mechanism that drives natural killer cell
uptake into tumour cells has been shown,
intriguingly, to require E- cadherin and
ezrin, which is known to regulate entosis
as well as cell cannibalism7,32,59,60. Finally,
MDA- MB-468 breast cancer cells, which
are deficient for entosis owing to loss of
expression of the cell adhesion protein
α- catenin61, have been shown to ingest
their neighbours at high rates when
overexpressing the LIP isoform of the
C/EBPβ transcription factor, and this
process also required ROCK activity62.
Howthese different cell ingestion
mechanisms may resemble or differ from
each other remains to be explored in
further studies, for example, by probing
the requirement of cell adhesion proteins,
ROCK, ezrin or caveolin for cell uptake or
by examining potentially active roles of the
internalizing cells in driving cell- in-cell
formation. How cannibalistic mechanisms
relate to phagocytosis also remains an
important, open question. For all cell-
in-cell processes, continued elucidation of
molecular machinery will be essential to
further define the roles of these processes in
cancer, particularly for further examination
of the roles of these processes invivo.
Thepresence of entosis, for example6,30, and
its consequences on cell populations6,47 are
typically defined by dependence on RHOA,
ROCK and cadherins, but loss of function
for these regulators can have pleiotropic
effects. The identification of additional
machinery controlling cell- in-cell processes
will be important in further studies.
How many different inducers of
cannibalism or entosis exist? High
levels of reactive oxygen species63, as well
as treatment with ionizing radiation or
chemotherapeutic agents such as cisplatin
and paclitaxel64, were recently shown to
induce cell- in-cell formation in a variety
of cancer cells, suggesting that other
stress conditions, in addition to matrix
detachment and nutrient starvation, could
engage cells in these processes. Entosis is
now known to be induced by numerous
signals that impinge upon actomyosin,
including altered cell polarity signalling35,
changes to the lipid content of the cell
membrane34 and aberrant shape regulation
in mitotic cells, which leads to the ingestion
and death of cells progressing through cell
division33, suggesting that entosis could
function to clear dysfunctional cells and in
this manner act as a tumour- suppressive
mechanism. Perhaps consistent with
this, expression of tumour suppressor
proteins p16 and p19ARF, which are both
encoded by CDKN2A, was recently shown
to inhibit entosis induction, suggesting
that this cell- in-cell mechanism could
function to clear cells that escape from
proper cell cycle control65. Growth factor
signalling through epidermal growth
factor (EGF)46, cytokine signalling through
interleukin-8 (IL-8)66 and lysophosphatidic
acid (LPA)31 are also known to induce
entosis, suggesting potential interplay
between cell competition and signalling
factors in the tumour microenvironment,
an idea to be explored in further studies.
Beyond promoting loser cell death, it is
also possible that entosis could benefit the
internalized cells in some circumstances,
as they can sometimes avoid cell death
and escape from their hosts. Whether
entosis, or other cell- in-cell mechanisms,
could potentially provide internalized cells
with a beneficial or protected niche, for
example, from immune detection, remains
an open question6. Further discoveries of
triggers of cell- in-cell behaviour may allow
deeper insight into the selective pressures
that engage these processes during cancer
progression. However entosis or cannibalism
is induced, we propose that this behaviour
contributes to a microevolutionary process
that progressively selects cells armed to
survive the hostile conditions associated
with cancer microenvironments, including
deficient blood and nutrient supply as well
as potentially hypoxia and changes in pH.
Cell- in-cell behaviour could contribute
substantially as an important piece of the
emerging cancer evolution puzzle67,68. Future
studies may uncover whether cell- in-cell
mechanisms in addition to entosis can
promote competition between individual
cells in cancer cell populations.
Can cell- in-cell activity be utilized
therapeutically? For entosis, as cancer cells
ingest each other through this mechanism,
the induction of high levels of this process
could conceivably contribute positively
to cancer responses in combination
www.nature.com/nrc
PersPectives
a
b
• Metastatic potential
• Intracellular nutrients
• Ploidy and/or aneuploidy
• Intracellular nutrients
1 32
Cannibalism
Entosis
Lysosome
1 32
Cell
tension
Loser status Winner status
Fig. 2 | Consequences of cannibalism and entosis. a | Cannibalism occurs at high frequency in meta-
static cells and is a marker of malignancy. (1) Cannibalism of live or dead cells in the microenviron-
ment (for example, Tcells) followed by lysosomal digestion of cannibalized cells (2) increases the
intracellular nutrient levels in cancers (3), which can rescue cannibalizing cells from the effects of
starvation. b | Entosis occurs between neighbouring cancer cells, or between normal cells and cancer
cells, where non- cancer cells are losers (green). Loser cells invade into winners (1) and are killed and
degraded by lysosomal enzymes (2) to provide intracellular nutrients (3). Entosis selects for lowered
levels of cell tension, as stiffer cells internalize into softer cells. Entosis is also known to disrupt proper
cytokinesis, leading to cell division failures and increases in cell ploidy and aneuploidy.
therapies. However, high levels of entosis
also promote ploidy changes46,69,70, feed
winner cells in the population36 and select
for mechanical deformability29, suggesting
multiple potential negative consequences
that could result from promoting this
process. The activity of cannibalism on the
other hand could potentially be leveraged
against cancers as a therapeutic strategy in a
potentially more straight- forward manner.
The ingestion of high amounts of inert
particles, for example, could conceivably
mask cell uptake and deny certain cancer
cells the ability to feed through this
mechanism7 (FIG.1), an idea related to the
ingestion of nanomaterials that we and
others have shown can lead to cancer cell
death71. How cannibalistic activity relates
to the ingestion of therapeutic particles,
and also to the secretion and ingestion
of endogenous nanoparticles, such as
exosomes72, remains an outstanding
question that could relate to future
therapeutic strategies. Exosomes can deliver
drugs with a defined antitumour activity73,74,
and the ingestion of these particles, like
cell ingestion, may be facilitated in cancers,
representing a future anticancer approach.
Intriguingly, cannibalistic structures
themselves may also have cancer- inhibiting
potential in some contexts, as shown for
MDA- MB-231 breast cancer cells that failed
to survive or form cancers in mice when
admixed with human umbilical cord MSCs,
whose cannibalization led to cell death
of the engulfing cancer cells75. Similarly, a
Tcell- derived cell line called HOZOT was
shown to efficiently become ingested by
cancer cells and to be capable of delivering
oncolytic virus, thereby inducing cancer
cell death that was shown to inhibit tumour
growth in mice76. It is our hope that further
studies of cell- in-cell mechanisms may not
only reveal new insights into the biology
of cancer but also uncover mechanisms
related to cell- in-cell formation that can be
leveraged for cancer therapy.
StefanoFais1* and MichaelOverholtzer2*
1Department of Oncology and Molecular Medicine,
National Institute of Health, Rome, Italy.
2Cell Biology Program, Memorial Sloan Kettering
Cancer Center, New York, NY, USA.
*e- mail: stefano.fais@iss.it; overhom1@mskcc.org
https://doi.org/10.1038/s41568-018-0073-9
Published online xx xx xxxx
1. Steinhaus, J. Ueber carcinoma- einschlusse. Virchows
Arch. 126, 533–535 (1891).
2. Stroebe, H. Zur Kenntniss verschiedener cellularer
Vorgange und Erscheinungen in Geschwulsten.
Beitrage Pathol. 11, 1 (1892).
3. Bauchwitz, M. The bird’s eye cell: cannibalism or
abnormal division of tumor cells. Acta Cytol. 25, 92
(1981).
4. Overholtzer, M. & Brugge, J. S. The cell biology
of cell- in-cell structures. Nat. Rev. Mol. Cell Biol. 9,
796–809 (2008).
5. Fais, S. Cannibalism: a way to feed on metastatic
tumors. Cancer Lett. 258, 155–164 (2007).
6. Overholtzer, M. etal. A nonapoptotic cell death
process, entosis, that occurs by cell- in-cell invasion.
Cell 131, 966–979 (2007).
7. Lugini, L. etal. Cannibalism of live lymphocytes by
human metastatic but not primary melanoma cells.
Cancer Res. 66, 3629–3638 (2006).
8. Cano, C. E. etal. Homotypic cell cannibalism,
a cell- death process regulated by the nuclear
protein1, opposes to metastasis in pancreatic cancer.
EMBO Mol. Med. 4, 964–979 (2012).
9. Brown, G. C. & Neher, J. J. Eaten alive! Cell death by
primary phagocytosis: ‘phagoptosis’. Trends Biochem.
Sci. 37, 325–332 (2012).
10. Benseler, V. etal. Hepatocyte entry leads to
degradation of autoreactive CD8 Tcells. Proc. Natl
Acad. Sci. USA 108, 16735–16740 (2011).
11. Wang, S. etal. Rapid reuptake of granzyme B leads to
emperitosis: an apoptotic cell- in-cell death of immune
killer cells inside tumor cells. Cell Death Dis. 4, e856
(2013).
12. Underhill, D. M., Gordon, S., Imhof, B. A., Nunez, G.
& Bousso, P. Elie Metchnikoff (1845–1916):
celebrating 100 years of cellular immunology and
beyond. Nat. Rev. Immunol. 16, 651–656 (2016).
13. Finicle, B. T., Jayashankar, V. & Edinger, A. L. Nutrient
scavenging in cancer. Nat. Rev. Cancer https://doi.org/
10.1038/s41568-018-0048-x (2018).
14. Palm, W. & Thompson, C. B. Nutrient acquisition
strategies of mammalian cells. Nature 546, 234–242
(2017).
15. Lugini, L. etal. Potent phagocytic activity
discriminates metastatic and primary human
malignant melanomas: a key role of ezrin. Lab. Invest.
83, 1555–1567 (2003).
16. Bartosh, T. J., Ullah, M., Zeitouni, S., Beaver, J.
& Prockop, D. J. Cancer cells enter dormancy after
cannibalizing mesenchymal stem/stromal cells (MSCs).
Proc. Natl Acad. Sci. USA 113 , E6447–E6456 (2016).
17. Cornillon, S. etal. Phg1p is a nine- transmembrane
protein superfamily member involved in dictyostelium
adhesion and phagocytosis. J. Biol. Chem. 275,
34287–34292 (2000).
18. Fais, S. & Fauvarque, M. O. TM9 and cannibalism:
how to learn more about cancer by studying amoebae
and invertebrates. Trends Mol. Med. 18, 4–5 (2012).
19. Bergeret, E. etal. TM9SF4 is required for Drosophila
cellular immunity via cell adhesion and phagocytosis.
J. Cell Sci. 121, 3325–3334 (2008).
20. Chang, H. etal. Identification of genes associated with
chemosensitivity to SAHA/taxane combination
treatment in taxane- resistant breast cancer cells.
Breast Cancer Res. Treat. 125, 55–63 (2011).
21. Oo, H. Z. etal. Identification of novel transmembrane
proteins in scirrhous- type gastric cancer by the
Escherichia coli ampicillin secretion trap (CAST)
method: TM9SF3 participates in tumor invasion
and serves as a prognostic factor. Pathobiology 81,
138–148 (2014).
22. Mackinnon, R. N. etal. The paradox of 20q11.21
amplification in a subset of cases of myeloid
malignancy with chromosome 20 deletion. Genes
Chromosomes Cancer 49, 998–1013 (2010).
23. Lozupone, F. etal. The human homologue of
Dictyostelium discoideum phg1A is expressed by
human metastatic melanoma cells. EMBO Rep. 10,
1348–1354 (2009).
24. Le Coadic, M. etal. Phg1/TM9 proteins control
intracellular killing of bacteria by determining cellular
levels of the Kil1 sulfotransferase in Dictyostelium.
PLOS ONE 8, e53259 (2013).
25. Perrin, J. etal. The nonaspanins TM9SF2 and
TM9SF4 regulate the plasma membrane localization
and signalling activity of the peptidoglycan recognition
protein PGRP- LC in Drosophila. J. Innate Immun. 7,
37–46 (2015).
26. Lozupone, F. etal. TM9SF4 is a novel V- ATPase-
interacting protein that modulates tumor pH
alterations associated with drug resistance and
invasiveness of colon cancer cells. Oncogene 34,
5163–5174 (2015).
27. Fais, S., Venturi, G. & Gatenby, B. Microenvironmental
acidosis in carcinogenesis and metastases: new
strategies in prevention and therapy. Cancer
Metastasis Rev. 33, 1095–1108 (2014).
28. Schwartz, L., Seyfried, T., Alfarouk, K. O.,
Da Veiga Moreira, J. & Fais, S. Out of Warburg effect:
an effective cancer treatment targeting the tumor
specific metabolism and dysregulated pH. Semin.
Cancer Biol. 43, 134–138 (2017).
29. Sun, Q. etal. Competition between human cells by
entosis. Cell Res. 24, 1299–1310 (2014).
30. Sun, Q., Cibas, E. S., Huang, H., Hodgson, L.
& Overholtzer, M. Induction of entosis by epithelial
cadherin expression. Cell Res. 24, 1288–1298
(2014).
31. Purvanov, V., Holst, M., Khan, J., Baarlink, C.
& Grosse, R. G- Protein-coupled receptor signaling and
polarized actin dynamics drive cell- in-cell invasion.
eLife https://doi.org/10.7554/eLife.02786 (2014).
32. Hinojosa, L. S., Holst, M., Baarlink, C. & Grosse, R.
MRTF transcription and Ezrin- dependent plasma
membrane blebbing are required for entotic invasion.
J. Cell Biol. 216, 3087–3095 (2017).
33. Durgan, J. etal. Mitosis can drive cell cannibalism
through entosis. eLife 6, e27134 (2017).
34. Ruan, B. etal. Cholesterol inhibits entotic cell- in-cell
formation and actomyosin contraction. Biochem.
Biophys. Res. Commun. 495, 1440–1446 (2018).
35. Wan, Q. etal. Regulation of myosin activation during
cell- cell contact formation by Par3-Lgl antagonism:
entosis without matrix detachment. Mol. Biol. Cell 23,
2076–2091 (2012).
36. Hamann, J. C. etal. Entosis is induced by glucose
starvation. Cell Rep. 20, 201–210 (2017).
37. Xia, P. etal. Aurora A orchestrates entosis by
regulating a dynamic MCAK- TIP150 interaction.
J.Mol. Cell. Biol. 6, 240–254 (2014).
38. Florey, O., Kim, S. E., Sandoval, C. P., Haynes, C. M.
& Overholtzer, M. Autophagy machinery mediates
macroendocytic processing and entotic cell death by
targeting single membranes. Nat. Cell Biol. 13,
1335–1343 (2011).
39. Ishikawa, F., Ushida, K., Mori, K. & Shibanuma, M.
Loss of anchorage primarily induces non- apoptotic cell
death in a human mammary epithelial cell line under
atypical focal adhesion kinase signaling. Cell Death
Dis. 6, e1619 (2015).
40. Diaz, B. & Moreno, E. The competitive nature of cells.
Exp. Cell Res. 306, 317–322 (2005).
41. Claveria, C. & Torres, M. Cell competition: mechanisms
and physiological roles. Annu. Rev. Cell Dev. Biol. 32,
411–439 (2016).
42. Merino, M. M., Levayer, R. & Moreno, E. Survival
of the fittest: essential roles of cell competition in
development, aging, and cancer. Trends Cell Biol. 26,
776–788 (2016).
43. Sun, Q., Huang, H. & Overholtzer, M. Cell- in-cell
structures are involved in the competition between
cells in human tumors. Mol. Cell. Oncol. 2, e1002707
(2015).
44. Huang, H., Chen, Z. & Sun, Q. Mammalian cell
competitions, cell- in-cell phenomena and their
biomedical implications. Curr. Mol. Med. 15,
852–860 (2015).
45. Patel, M. S., Shah, H. S. & Shrivastava, N.
c- Myc-dependent cell competition in human cancer
cells. J. Cell. Biochem. 118, 1782–1791 (2017).
46. Mackay, H. L. etal. Genomic instability in mutant p53
cancer cells upon entotic engulfment. Nat. Commun.
9, 3070 (2018).
47. Hamann, J. C. & Overholtzer, M. Entosis enables a
population response to starvation. Oncotarget 8,
57934–57935 (2017).
48. Krajcovic, M., Krishna, S., Akkari, L., Joyce, J. A.
& Overholtzer, M. mTOR regulates phagosome and
entotic vacuole fission. Mol. Biol. Cell 24, 3736–3745
(2013).
49. Loike, J. D., Kozler, V. F. & Silverstein, S. C. Increased
ATP and creatine phosphate turnover in phagocytosing
mouse peritoneal macrophages. J. Biol. Chem. 254,
9558–9564 (1979).
50. Mazur, M. T. & Williamson, J. R. Macrophage
deformability and phagocytosis. J. Cell Biol. 75,
185–199 (1977).
51. Kuiper, J. W. etal. Creatine kinase- mediated ATP
supply fuels actin- based events in phagocytosis.
PLOSBiol. 6, e51 (2008).
52. Humble, J. G., Jayne, W. H. & Pulvertaft, R. J.
Biological interaction between lymphocytes and other
cells. Br. J. Haematol. 2, 283–294 (1956).
53. Salvesen, G. S. Dying from within: granzyme B converts
entosis to emperitosis. Cell Death Differ. 21, 3–4 (2014).
54. Chao, M. P., Weissman, I. L. & Majeti, R. The CD47-
SIRPalpha pathway in cancer immune evasion and
potential therapeutic implications. Curr. Opin. Immunol.
24, 225–232 (2012).
55. Bansal, C., Tiwari, V., Singh, U., Srivastava, A.
& Misra, J. Cell cannibalism: a cytological study in
effusion samples. J. Cytol. 28, 57–60 (2011).
Nature reviews
|
CanCer
PersPectives
56. Gupta, K. & Dey, P. Cell cannibalism: diagnostic marker
of malignancy. Diagn. Cytopathol. 28, 86–87 (2003).
57. Schwegler, M. etal. Prognostic value of homotypic cell
internalization by nonprofessional phagocytic cancer
cells. Biomed Res. Int. 2015, 359392 (2015).
58. Schenker, H., Buttner- Herold, M., Fietkau, R.
& Distel, L. V. Cell- in-cell structures are more potent
predictors of outcome than senescence or apoptosis
inhead and neck squamous cell carcinomas. Radiat.
Oncol. 12, 21 (2017).
59. Wang, S. etal. Internalization of NK cells into tumor
cells requires ezrin and leads to programmed cell- in-cell
death. Cell Res. 19, 1350–1362 (2009).
60. Wang, X. Cell- in-cell phenomenon: a new paradigm in
life sciences. Curr. Mol. Med. 15, 810–818 (2015).
61. Wang, M. etal. Impaired formation of homotypic
cell- in-cell structures in human tumor cells lacking
alpha- catenin expression. Sci. Rep. 5, 12223 (2015).
62. Abreu, M. & Sealy, L. Cells expressing the C/EBPbeta
isoform, LIP, engulf their neighbors. PLOS ONE 7,
e41807 (2012).
63. Balvan, J. etal. Oxidative stress resistance in
metastatic prostate cancer: renewal by self- eating.
PLOS ONE 10, e0145016 (2015).
64. Martins, I. etal. Anticancer chemotherapy and
radiotherapy trigger both non- cell-autonomous and
cell- autonomous death. Cell Death Dis. 9, 716 (2018).
65. Liang, J. etal. CDKN2A inhibits formation of homotypic
cell- in-cell structures. Oncogenesis 7, 50 (2018).
66. Ruan, B. etal. Expression profiling identified IL-8 as a
regulator of homotypic cell- in-cell formation. BMB Rep.
51, 412–417 (2018).
67. Ibrahim- Hashim, A. etal. Defining cancer
subpopulations by adaptive strategies rather than
molecular properties provides novel insights into
intratumoral evolution. Cancer Res. 77, 2242–2254
(2017).
68. Zhang, J., Cunningham, J. J., Brown, J. S.
& Gatenby, R. A. Integrating evolutionary dynamics
into treatment of metastatic castrate- resistant
prostate cancer. Nat. Commun. 8, 1816 (2017).
69. Krajcovic, M. etal. A non- genetic route to aneuploidy
in human cancers. Nat. Cell Biol. 13, 324–330 (2011).
70. Chen, Y. H. etal. Prevalence of heterotypic tumor/
immune cell- in-cell structure invitro and invivo leading
to formation of aneuploidy. PLOS ONE 8, e59418
(2013).
71. Kim, S. E. etal. Ultrasmall nanoparticles induce
ferroptosis in nutrient- deprived cancer cells and
suppress tumour growth. Nat. Nanotechnol. 11,
977–985 (2016).
72. Zhao, H. etal. The key role of extracellular vesicles in
the metastatic process. Biochim. Biophys. Acta 1869,
64–77 (2018).
73. Federici, C. etal. Exosome release and low pH belong
to a framework of resistance of human melanoma
cellsto cisplatin. PLOS ONE 9, e88193 (2014).
74. Iessi, E. etal. Acridine Orange/exosomes increase
thedelivery and the effectiveness of Acridine Orange
in human melanoma cells: a new prototype for
theranostics of tumors. J. Enzyme Inhib. Med. Chem.
32, 648–657 (2017).
75. Chao, K. C., Yang, H. T. & Chen, M. W. Human
umbilical cord mesenchymal stem cells suppress
breast cancer tumourigenesis through direct cell- cell
contact and internalization. J. Cell. Mol. Med. 16,
1803–1815 (2012).
76. Onishi, T. etal. Tumor- specific delivery of biologics by
a novel Tcell line HOZOT. Sci. Rep. 6, 38060 (2016).
77. Sharma, N. & Dey, P. Cell cannibalism and cancer.
Diagn. Cytopathol. 39, 229–233 (2011).
78. Singhal, N., Handa, U., Bansal, C. & Mohan, H.
Neutrophil phagocytosis by tumor cells — a cytological
study. Diagn. Cytopathol. 39, 553–555 (2011).
79. Singh, G., Mathur, S. R., Iyer, V. K. & Jain, D.
Cytopathology of neoplastic meningitis: a series of 66
cases from a tertiary care center. Cytojournal 10, 13
(2013).
80. Ferreira, F. C. etal. Four cases of cell cannibalism in
highly malignant feline and canine tumors. Diagn.
Pathol. 10, 199 (2015).
81. Kale, A. Cellular cannibalism. J. Oral Maxillofac.
Pathol. 19, 7–9 (2015).
82. Huang, H. etal. Detecting cell- in-cell structures in
human tumor samples by E- cadherin/CD68/CD45
triple staining. Oncotarget 6, 20278–20287 (2015).
83. Melendez- Lazo, A. etal. Cell cannibalism by malignant
neoplastic cells: three cases in dogs and a literature
review. Vet. Clin. Pathol. 44, 287–294 (2015).
84. Morales- Camacho, R. M. etal. Leukaemic blast
cannibalism in acute megakaryoblastic leukaemia with
myeloid sarcoma. Br. J. Haematol. 173, 336 (2016).
85. Jeon, Y. K., Kim, H. W., Choi, H. J. & Park, I. A. Fine
needle aspiration cytology of epithelioid angiosarcoma.
Report of a case with nuclear grooves and indentations.
Acta Cytol. 48, 223–228 (2004).
86. Fujii, M. etal. Cytologic diagnosis of male breast cancer
with nipple discharge. A case report. Acta Cytol. 30,
21–24 (1986).
87. Abodief, W. T., Dey, P. & Al- Hattab, O. Cell cannibalism
in ductal carcinoma of breast. Cytopathology 17,
304–305 (2006).
88. Almeida, S. M. & Rotta, I. Cerebrospinal fluid cell
cannibalism in metastatic breast adenocarcinoma.
Arq. Neuropsiquiatr. 73, 469 (2015).
89. Kinoshita, M. etal. Cytological diagnostic clues in
poorly differentiated squamous cell carcinomas of the
breast: streaming arrangement, necrotic background,
nucleolar enlargement and cannibalism of cancer cells.
Cytopathology 29, 22–27 (2018).
90. Nakajima, T. etal. Multivariate statistical analysis of
bile cytology. Acta Cytol. 38, 51–55 (1994).
91. Sarode, S. C. & Sarode, G. S. Cellular cannibalism in
central and peripheral giant cell granuloma of
theoralcavity can predict biological behavior
of the lesion. J. Oral Pathol. Med. 43, 459–463
(2014).
92. Azzi, L. etal. A giant- cell lesion with cellular
cannibalism in the mandible: case report and review
of brown tumors in hyperparathyroidism. Case Rep.
Dent. 2017, 9604570 (2017).
93. Ng, W. K. etal. Thin- layer cytology findings of small
cell carcinoma of the lower female genital tract. Review
of three cases with molecular analysis. Acta Cytol. 47,
56–64 (2003).
94. Gamez, R. G., Jessurun, J., Berger, M. J.
& Pambuccian, S. E. Cytology of metastatic cervical
squamous cell carcinoma in pleural fluid: report of a
case confirmed by human papillomavirus typing.
Diagn. Cytopathol. 37, 381–387 (2009).
95. AbdullGaffar, B. Clear cell carcinoma first suspected
inPap smear. The value of neutrophil cannibalism
bytumor cells. Diagn. Cytopathol. 45, 176–178
(2017).
96. Kalele, K. P., Patil, K. P., Nayyar, A. S. & Sasane, R. S.
Atypical lymphocytes and cellular cannibalism: a
phenomenon, first of its kind to be discovered in
chronic periapical lesions. J. Clin. Diagn. Res. 10,
ZC01–4 (2016).
97. Caruso, R. A., Famulari, C., Giuffre, G. & Mazzeo, G.
Pleomorphic carcinoma of the gallbladder: report of a
case. Tumori 77, 523–526 (1991).
98. Caruso, R. A., Muda, A. O., Bersiga, A., Rigoli, L. &
Inferrera, C. Morphological evidence of neutrophil-
tumor cell phagocytosis (cannibalism) in human
gastric adenocarcinomas. Ultrastruct. Pathol. 26,
315–321 (2002).
99. Caruso, R. A. etal. Neutrophil- rich gastric carcinomas:
light and electron microscopic study of 9 cases with
particular reference to neutrophil apoptosis.
Ultrastruct. Pathol. 37, 164–170 (2013).
100. Barresi, V. etal. Phagocytosis (cannibalism) of
apoptotic neutrophils by tumor cells in gastric
micropapillary carcinomas. World J. Gastroenterol.
21, 5548–5554 (2015).
101. Sarode, G. S. etal. Cellular cannibalism in giant cells
of central giant cell granuloma of jaw bones and giant
cell tumors of long bones. J. Investig. Clin. Dent. 8,
e12214 (2017).
102. DeSimone, P. A., East, R. & Powell, R. D. Jr. Phagocytic
tumor cell activity in oat cell carcinoma of the lung.
Hum. Pathol. 11, 535–539 (1980).
103. Brouwer, M., de Ley, L., Feltkamp, C. A., Elema, J.
& Jongsma, A. P. Serum- dependent “cannibalism”
and autodestruction in cultures of human small cell
carcinoma of the lung. Cancer Res. 44, 2947–2951
(1984).
104. Conway, A. B., Hart, M. K., Jessurun, J.
& Pambuccian, S. E. “Cannonballs” and psammoma
bodies: unusual cytologic features of metastatic
pulmonary small- cell carcinoma in a pleural effusion.
Diagn. Cytopathol. 41, 247–252 (2013).
105. Escamilla, V. etal. Bone marrow cellular cannibalism
by medulloblastoma. Am. J. Hematol. 90, 466–467
(2015).
106. Beaty, M. W., Fetsch, P., Wilder, A. M., Marincola, F.
& Abati, A. Effusion cytology of malignant melanoma.
A morphologic and immunocytochemical analysis
including application of the MART-1 antibody. Cancer
81, 57–63 (1997).
107. Ehya, H. The cytologic diagnosis of mesothelioma.
Semin. Diagn. Pathol. 3, 196–203 (1986).
108. Stevens, M. W., Leong, A. S., Fazzalari, N. L.,
Dowling, K. D. & Henderson, D. W. Cytopathology of
malignant mesothelioma: a stepwise logistic regression
analysis. Diagn. Cytopathol. 8, 333–341 (1992).
109. Ylagan, L. R. & Zhai, J. The value of ThinPrep and
cytospin preparation in pleural effusion cytological
diagnosis of mesothelioma and adenocarcinoma.
Diagn. Cytopathol. 32, 137–144 (2005).
110 . Kimura, N., Dota, K., Araya, Y., Ishidate, T.
& Ishizaka, M. Scoring system for differential diagnosis
of malignant mesothelioma and reactive mesothelial
cells on cytology specimens. Diagn. Cytopathol. 37,
885–890 (2009).
111. Cakir, E., Demirag, F., Aydin, M. & Unsal, E.
Cytopathologic differential diagnosis of malignant
mesothelioma, adenocarcinoma and reactive
mesothelial cells: a logistic regression analysis. Diagn.
Cytopathol. 37, 4–10 (2009).
112 . Matsumoto, S. etal. Morphology of 9p21 homozygous
deletion- positive pleural mesothelioma cells analyzed
using fluorescence insitu hybridization and virtual
microscope system in effusion cytology. Cancer
Cytopathol. 121, 415–422 (2013).
113 . Abati, A., Cajigas, A. & Hijazi, Y. M. Metastatic
epithelioid hemangioendothelioma in a pleural
effusion: diagnosis by cytology. Diagn. Cytopathol. 11,
64–67 (1994).
114 . Sarode, G. S., Sarode, S. C. & Karmarkar, S. Complex
cannibalism: an unusual finding in oral squamous cell
carcinoma. Oral Oncol. 48, e4–e6 (2012).
115 . Sarode, S. C. & Sarode, G. S. Neutrophil- tumor cell
cannibalism in oral squamous cell carcinoma. J. Oral
Pathol. Med. 43, 454–458 (2014).
116 . Jose, D. etal. Evaluation of cannibalistic cells:
a novel entity in prediction of aggressive nature of oral
squamous cell carcinoma. Acta Odontol. Scand. 72,
418–423 (2014).
117 . Jain, M. An overview on “cellular cannibalism” with
special reference to oral squamous cell carcinoma.
Exp. Oncol. 37, 242–245 (2015).
118 . Jain, M. etal. Assessment of tumor cell cannibalism
as a predictor of oral squamous cell carcinoma — a
histopathologic correlation. Gulf J. Oncolog. 1, 52–56
(2017).
119 . Sarode, S. C., Sarode, G. S., Chuodhari, S. & Patil, S.
Non- cannibalistic tumor cells of oral squamous cell
carcinoma can express phagocytic markers. J. Oral
Pathol. Med. 46, 327–331 (2017).
120. Kosaka, N. etal. Cytological findings of ascitic fluid
with a malignant ovarian steroid cell tumor: a case
report and literature review. Acta Cytol. 61, 165–171
(2017).
121. Silverman, J. F., Dabbs, D. J., Finley, J. L.
& Geisinger, K. R. Fine- needle aspiration biopsy of
pleomorphic (giant cell) carcinoma of the pancreas.
Cytologic, immunocytochemical, and ultrastructural
findings. Am. J. Clin. Pathol. 89, 714–720 (1988).
122. Silverman, J. F., Finley, J. L., Berns, L. & Unverferth, M.
Significance of giant cells in fine- needle aspiration
biopsies of benign and malignant lesions of the
pancreas. Diagn. Cytopathol. 5, 388–391 (1989).
123. Gupta, R. K. & Wakefield, S. J. Needle aspiration
cytology, immunocytochemistry, and electron
microscopic study of unusual pancreatic carcinoma
with pleomorphic giant cells. Diagn. Cytopathol. 8,
522–527 (1992).
124. Khayyata, S., Basturk, O. & Adsay, N. V. Invasive
micropapillary carcinomas of the ampullo-
pancreatobiliary region and their association with
tumor- infiltrating neutrophils. Mod. Pathol. 18,
1504–1511 (2005).
125. Wen, S., Shang, Z., Zhu, S., Chang, C. & Niu, Y.
Androgen receptor enhances entosis, a non- apoptotic
cell death, through modulation of Rho/ROCK pathway
in prostate cancer cells. Prostate 73, 1306–1315
(2013).
126. Gilloteaux, J., Ruffo, C., Jamison, J. M. & Summers, J. L.
Modes of internalizations of human prostate carcinoma
(DU145) cells invitro and in murine xenotransplants.
Ultrastruct. Pathol. 40, 231–239 (2016).
127. Kong, Y., Liang, Y. & Wang, J. Foci of entotic nuclei in
different grades of noninherited renal cell cancers.
IUBMB Life 67, 139–144 (2015).
128. Arya, P., Khalbuss, W. E., Monaco, S. E. &
Pantanowitz, L. Salivary duct carcinoma with striking
neutrophil- tumor cell cannibalism. Cytojournal 8, 15
(2011).
129. Fernandez- Flores, A. Cannibalism in a benign soft
tissue tumor (giant- cell tumor of the tendon sheath,
localized type): a study of 66 cases. Rom. J. Morphol.
Embryol. 53, 15–22 (2012).
130. Huang, C. C. & Michael, C. W. Cytomorphological
features of metastatic squamous cell carcinoma in
serous effusions. Cytopathology 25, 112–119 (2014).
www.nature.com/nrc
PersPectives
131. Logothetou- Rella, H. Glycosaminoglycan- sac formation
invitro. Interactions between normal and malignant
cells. Histol. Histopathol. 9, 243–249 (1994).
132. Kojima, S., Sekine, H., Fukui, I. & Ohshima, H.
Clinical significance of “cannibalism” in urinary
cytology of bladder cancer. Acta Cytol. 42,
1365–1369 (1998).
133. Dey, P., Amir, T., Jogai, S. & Al Jussar, A. Fine- needle
aspiration cytology of metastatic transitional cell
carcinoma. Diagn. Cytopathol. 32, 226–228
(2005).
134. Hattori, M. etal. Cell cannibalism and nucleus-
fragmented cells in voided urine: useful parameters for
cytologic diagnosis of low- grade urothelial carcinoma.
Acta Cytol. 51, 547–551 (2007).
135. Ohsaki, H. etal. Can cytological features differentiate
reactive renal tubular cells from low- grade urothelial
carcinoma cells? Cytopathology 21, 326–333 (2010).
136. Ahmed Wani, F. & Bhardwaj, S. Cytological evaluation
and significance of cell cannibalism in effusions and
urine cytology. Malays. J. Pathol. 37, 265–270 (2015).
Acknowledgements
This work was supported by grants from the Ministry of
Health, Italy (S.F.) and the US National Institutes of Health
(CA154649; M.O.). The authors included many different
examples of published reports of cell- in-cell activity in cancer
in Box1 of this review to show the breadth of this activity; the
authors apologize to those whose work was not included
owing to space limitations.
Author contributions
Both authors contributed equally to this work.
Competing interests
The authors declare no competing interests.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
Reviewer information
Nature Reviews Cancer thanks E. Moreno, M. Olson and
A. Thorburn for their contribution to the peer review of
thiswork.
Nature reviews
|
CanCer
PersPectives
