\makeatletter
\@ifundefined{HCode}
{
\documentclass[CRBIOL,Unicode,screen,biblatex]{cedram}
\addbibresource{crbiol20260189.bib}
\newenvironment{noXML}{}{}
\let\citep\parencite
\let\citet\textcite
\def\xcitealp#1#2{\citeauthor{#1}, \citelink{#1}{#2}}
\newcommand*{\citelink}[2]{\hyperlink{cite.\therefsection @#1}{#2}}
\def\defcitealias#1#2{}
\let\citepalias\parencite
\let\citetalias\textcite
\newenvironment{Table}{\begin{table}}{\end{table}}
\def\thead{\noalign{\relax}\hline}
\def\endthead{\noalign{\relax}\hline}
\def\tabnote#1{\vskip4pt\parbox{.9\linewidth}{#1}}
\def\tsup#1{$^{{#1}}$}
\def\tsub#1{$_{{#1}}$}
\RequirePackage{etoolbox}
\usepackage[most]{tcolorbox}
\def\jobid{crbiol20260189}
%\graphicspath{{/tmp/\jobid_figs/web/}}
\graphicspath{{./figures/}}
\newcounter{runlevel}
\let\MakeYrStrItalic\relax
\def\refinput#1{}
\def\back#1{}
\usepackage{multirow}
\def\nrow#1{\@tempcnta #1\relax%
\advance\@tempcnta by 1\relax%
\xdef\lenrow{\the\@tempcnta}}
\def\morerows#1#2{\nrow{#1}\multirow{\lenrow}{*}{#2}}
\def\botline{\\\hline}
\def\hyphen{\text{-}}
\def\0{\phantom{0}}
\def\xsection#1{}
\def\og{\guillemotleft}
\def\fg{\guillemotright}
\def\xsection#1{}
\def\back#1{}
\skip\footins=11pt
\usepackage{upgreek}
\DOI{10.5802/crbiol.194}
\datereceived{2026-03-24}
\dateaccepted{2026-04-02}
\ItHasTeXPublished
}
{
\PassOptionsToPackage{authoryear}{natbib}
\documentclass[crbiol]{article}
\def\CDRdoi{10.5802/crbiol.194}
\usepackage[T1]{fontenc}
\def\xmorerows#1#2{\morerows{#1}{#2}}
\def\citelink#1#2{\citeyear{#1}}
\def\xcitealp#1#2{\citealp{#1}}
\def\href#1#2{\url[#1]{#2}}
\def\xsection#1{}
\makeatletter
\def\CDRsupplementaryTwotypes#1#2{}
\newcommand\@coi{}
\newcommand\COI[1]{\gdef\@coi{#1}}
\newcommand\printCOI{\ifx\@coi\@empty\else%
\section*{Declaration of interests}
\@coi\fi
}
}
\makeatother

\usepackage{upgreek}

\COI{The authors do not work for, advise, own shares in, or receive
funds from any organization that could benefit from this article, and
have declared no affiliations other than their research organizations.}

\dateposted{2026-05-13}
\begin{document}

%\dateposted{2026-02-16}

\begin{noXML}

\CDRsetmeta{articletype}{review}


\makeatletter
\def\@setafterauthor{%
  \vglue3mm%
\begingroup\hsize=10.5cm\advance\hsize\abstractmarginL\raggedright
\noindent
   \leftskip\abstractmarginL
  \normalfont\Small
  \@afterauthor\par
\endgroup
\vskip-2pt plus 3pt minus 1pt
}
\makeatother

\title{Adhesion-mediated force transmission regulates cell competition
in epithelia}

\alttitle{La transmission de force par adh\'{e}rence r\'{e}gule la
comp\'{e}tition cellulaire dans les \'{e}pith\'{e}liums}

\author{\firstname{Andreas} \lastname{Schoenit}\CDRorcid{0000-0003-3714-5636}\IsCorresp}
\address{Universit\'{e} Paris Cit\'{e}, CNRS, Institut Jacques Monod,
75013 Paris, France}
\email[A. Schoenit]{andreas.schonit@ijm.fr}

\author{\firstname{Ren\'{e} Marc} \lastname{M\`{e}ge}\CDRorcid{0000-0001-8128-5543}\IsCorresp}
\addressSameAs{1}{Universit\'{e} Paris Cit\'{e}, CNRS, Institut Jacques
Monod, 75013 Paris, France}
\email[R. M. M\`{e}ge]{rene-marc.mege@ijm.fr}

\author{\firstname{Beno\^{i}t} \lastname{Ladoux}\CDRorcid{0000-0003-2086-1556}\IsCorresp}
\addressSameAs{1}{Universit\'{e} Paris Cit\'{e}, CNRS, Institut Jacques
Monod, 75013 Paris, France}
\address{Max-Planck Zentrum f\"{u}r Physik und Medizin Erlangen, Germany}
\address{Department of Physics, Friedrich-Alexander-Universit\"{a}t
Erlangen-N\"{u}rnberg, Germany}
\email[B. Ladoux]{benoit.ladoux@ijm.fr}

\shortrunauthors

\keywords{\kwd{Force transmission}\kwd{Cell adhesion}\kwd{Cell
competition}\kwd{Cell extrusion}\kwd{Tissue
mechanics}\kwd{Mechanobiology}}

\altkeywords{\kwd{Transmission de force}\kwd{Adh\'{e}rence
cellulaire}\kwd{Comp\'{e}tition cellulaire}\kwd{Extrusion
cellulaire}\kwd{M\'{e}canique tissulaire}\kwd{M\'{e}canobiologie}}
\editornote{Andreas Shoenit received the Grandes Avancées en Biologie 2025 award.}
\alteditornote{Andreas Shoenit est lauréat 2025 du prix des Grandes avancées françaises en biologie}
\begin{abstract} 
Organs and tissues consist of a precise arrangement of different cell
types, all playing a specific role to fulfil the biological function of
the tissue. Small changes in cellular phenotypes or behaviour can lead
to developmental defects, tissue malfunctions or the emergence of
diseases. Therefore, tissue integrity, health and function are
maintained through different quality control mechanisms. One highly
conserved mechanism is cell competition, through which cells of reduced
fitness are eliminated. Cells can employ various strategies to
eliminate each other. Those include the exertion of mechanical forces,
but its role in determining the competition outcome remains unclear.
Here, we report that heterogeneities in force transmission capabilities
mediated by cell--cell adhesion differences lead to cell competition.
We show that increased force transmission endows collectives of cells
with a fitness advantage, as it provides increased resistance to
elimination forces. Elimination forces are generated from large stress
fluctuations, emerging at the interfaces of competing cell populations.
Besides promoting the removal of unfit cells in a wide range of
biological conditions where local cell--cell adhesion heterogeneities
are observed, this mechanism might be of general importance for the
generation and maintenance of tissue boundaries. 
\vspace*{-2pt}
\end{abstract} 

\begin{altabstract} 
Les organes et les tissus sont constitu\'{e}s d'un agencement
pr\'{e}cis de diff\'{e}rents types de cellules, qui jouent toutes un
r\^{o}le sp\'{e}cifique pour assurer la fonction biologique du tissu.
De l\'{e}g\`{e}res modifications des ph\'{e}notypes ou du comportement
cellulaire peuvent entra\^{i}ner des anomalies de d\'{e}veloppement,
des dysfonctionnements tissulaires ou l'apparition de maladies. C'est
pourquoi l'int\'{e}grit\'{e}, la sant\'{e} et le fonctionnement des
tissus sont maintenus gr\^{a}ce \`{a} diff\'{e}rents m\'{e}canismes de
contr\^{o}le qualit\'{e}. L'un de ces m\'{e}canismes, tr\`{e}s
conserv\'{e}, est la comp\'{e}tition cellulaire, qui permet
d'\'{e}liminer les cellules dont la capacit\'{e} d'adaptation est
r\'{e}duite. Les cellules peuvent recourir \`{a} diverses
strat\'{e}gies pour s'\'{e}liminer mutuellement. Parmi celles-ci figure
l'exercice de forces m\'{e}caniques, dont le r\^{o}le dans la
d\'{e}termination de l'issue de la comp\'{e}tition reste cependant
incertain. Nous pr\'{e}sentons ici des r\'{e}sultats indiquant que les
h\'{e}t\'{e}rog\'{e}n\'{e}it\'{e}s dans les capacit\'{e}s de
transmission des forces, induites par des diff\'{e}rences
d'adh\'{e}sion cellulaire, conduisent \`{a} une comp\'{e}tition
cellulaire. Nous montrons qu'une transmission de force accrue
conf\`{e}re aux collectifs de cellules un avantage en termes de
fitness, car elle leur procure une r\'{e}sistance accrue aux forces
d'\'{e}limination. Ces forces d'\'{e}limination sont
g\'{e}n\'{e}r\'{e}es par d'importantes fluctuations de contrainte,
apparaissant aux interfaces entre les populations cellulaires en
comp\'{e}tition. Ce m\'{e}canisme favorise l'\'{e}limination des
cellules inaptes dans de nombreuses conditions biologiques
pr\'{e}sentant des disparit\'{e}s locales d'adh\'{e}sion cellulaire, et
pourrait \'{e}galement rev\^{e}tir une importance g\'{e}n\'{e}rale pour
la formation et le maintien des fronti\`{e}res tissulaires.
\end{altabstract} 

%\input{CR-pagedemetas} 

\maketitle

\vspace*{-2pt}

\twocolumngrid

\end{noXML}

\section{Introduction}

\begin{figure*}
\vspace*{2pt}
\includegraphics{fig01}
\vspace*{2pt}
\caption{\label{fig1}Cell competition shapes tissues by comparing
relative cellular fitness levels. (A)~General concept of cell
competition: a population of cells shows higher fitness (winners) and
is expanding at the cost of the population of lower fitness (losers).
Loser cells are removed through cell elimination. (B)~Cells compare
their relative fitness to their surroundings. Multiple factors can
influence which population wins and which one loses. (C)~Loser cells
are removed from the tissue via cell extrusion. This requires the
coordinated exertion of forces within the removed cell and its
neighbours. (D)~Aim of this study: potential impact of force
transmission differences on cell competition. Force transmission is
mediated by cell--cell adhesion, which depends on the level of cadherin
proteins.}
\end{figure*}

\looseness=-1
Organs and tissues consist of a precise arrangement of various cell
types, working in concert to ensure optimal tissue function. Even small
deviations from the tissue blueprint, for example in the composition of
cell types or cellular phenotypes and behaviour, can lead to
developmental defects, tissue malfunctions or the emergence of
diseases. To prevent this, tissue quality control mechanisms facilitate
the selective removal of defective cells.\ One~such cell elimination
mechanism is cell competition, through which cells of relatively lower
fitness than their surroundings are removed \citep{1,2,3}
(Figure~\ref{fig1}A). Competitive dynamics shape tissues during the
whole animal's lifespan. They are first observed at the blastocyst
stage \citep{4}, ensure robust pattern formation during morphogenesis
\citep{5} and may be an important factor in ageing-related processes
\citep{6}. Furthermore, cell competition can prevent the
emergence of cancer through the removal of transformed cells
\citep{7,8}. At the same time, mutated cells gaining\pagebreak{} a fitness
advantage can eliminate the surrounding normal cells through
competitive interactions, promoting cancer progression \citep{7,8}.

Those various examples highlight the need to understand what determines
winning and losing, i.e.\ understanding how fitness changes are sensed
and how they can lead to cell elimination through different mechanisms
\citep{9}. Cellular fitness generally depends on the tissue context and
can be influenced by multiple factors. Those include proliferation,
genetics, protein expression, metabolic state, intrinsic resistance to
death, but also the ways cells interact with their extracellular
environment \citep{1,2,3,4,5,6,7,8}. They all could provide a
competitive advantage or disadvantage conveying winner or loser status,
respectively (Figure~\ref{fig1}B). Loser cells are often eliminated via
cell extrusion, which is an inherited mechanical process and requires
the coordinated exertion of forces within the eliminated cell and its
neighbors \citep{10,11} (Figure~\ref{fig1}C). However, frameworks
integrating determinants of the loser cell status with the mechanism of
cell elimination are often lacking.

Mechanical cell competition describes scenarios where cellular fitness
and the elimination of loser cells are determined by tissue mechanics
\citep{12}. There, compressive stresses lead to cell death and
subsequent removal of loser cells. Compression can generally promote
cell extrusion \citep{13,14} and can be mediated by collective
migration \citep{15,16}, tissue crowding \citep{17}, differential
proliferation \citep{18}, growth \citep{19} and homeostatic cell
densities \citep{20,21} within mixed cell populations. Changing tissue
mechanics has led to contradicting outcomes, e.g.\ when the mechanical
properties of the extracellular environment were modulated
\citep{22,23} or when contractility was changed through the
overexpression of Ras oncogenes \citep{17,24,25,26}.\ Thus, there is so
far no consensus on how collective mechanical interactions and the
cellular mechanical phenotype determine the competition outcome. Both
critically depend on the fundamental process of force transmission, but
its role in cell competition has not been explored yet. 

In epithelia, intercellular force transmission, but also force sensing
and force exertion, are mainly mediated by the adherens junction
complex \citep{27}. A key component of it is E-cadherin (E-cad), a
transmembrane protein that links neighbour cells to one another
\citep{28}. It has been shown to be crucial for efficient mechanical
coupling between cells, transmission of mechanical information and
collective tissue behaviour \citep{29,30,31,32,33}. Consequently, we
hypothesized that modulating force transmission through changing
cadherin-mediated cell--cell adhesion strength may affect cellular
fitness and lead to the emergence of cell competition
(Figure~\ref{fig1}D). 

\section{Summary}
\subsection{Force transmission provides a competitive advantage}
\label{sec1.1}

In our recent work \citep{34}, we studied the impact of force
transmission differences on competitive interactions within tissues.
Importantly, such differences can be found in pathological conditions,
for example within metaplastic breast cancers, which are a subtype of
highly aggressive triple-negative breast cancers. They present a major
therapeutic challenge because they consist of at least two cellular
subpopulations, one epithelial-like and the other mesenchymal-like
\citep{35}.\ While E-cad is absent in the mesenchymal cells, it is
strongly expressed in the epithelial-like subpopulation
(Figure~\ref{fig2}A). Due to this intratumoral force transmission
heterogeneity, we assessed the emergence of potential cell competition
through cultivating and monitoring patient-derived xenografts. We found
that the E-cad positive epithelial-like cells expanded over time,
thereby removing the E-cad negative mesenchymal-like cells from the
tissue (\mbox{Figure}~\ref{fig2}B). This indeed suggested that force
transmission differences lead to intratumoral cell competition, and
that increased cell--cell adhesion might increase cellular fitness.

\begin{figure*}
\includegraphics{fig02}
\caption{\label{fig2}Force transmission provides a competitive
advantage. (A)~Monolayer of heterogeneous patient-derived metaplastic
breast cancer cells. E-cadherin in magenta marks the epithelial-like
sub-population. Vimentin in cyan marks the mesenchymal-like
sub-population. (B)~Cluster development over time. The E-cad-expressing
population is expanding. Red line shows the tissue boundaries after
20~h. (C)~Top: cluster development in a mixed tissue consisting of MDCK
WT and E-cad KO cells (green). Bottom: area quantification over time
for different fractions of E-cad KO cells. (D)~Left: map of the
isotropic stress in the cancer tissue corresponding to~(B). Right:
quantification of the average isotropic stress within the two
sub-populations. (E)~Stress map (left) and stress quantification
(right) of the competition between MDCK WT and E-cad KO corresponding
to~(C). Magenta outlines show the initial cluster boundaries.
(D,E)~$P$-values from unpaired $t$-tests. Every datapoint shows a
single measurement in one frame out of $n=5$ (D) or $n=14$ (E) videos
from $N=2$ (D) or $n=4$ (E) independent experiments. Data are presented
as means ${\pm}$~s.d. Scale bars 200~$\upmu$m (A,B,D), 100~$\upmu$m
(C,E). All figures reproduced from \citet{34}.}
\end{figure*}

To investigate this hypothesis more systematically, we generated mixed
cultures of epithelial cell lines, where the cell--cell adhesion
strength of some cells was modulated through modifying the levels of
cadherin expression. We tested various experimental conditions,
including competition between MDCK and MCF10A WT versus E-cad KO cells,
MDCK WT versus E-cad-overexpressing cells and MDCK E-cad KO versus
E-cad/Cadherin-6 double-knockout cells. This experimental approach
further offered the advantage that the ratios of competing cells, like
MDCK WT competing against fluorescently labelled E-cad KO cells, can be
precisely controlled (Figure~\ref{fig2}C). Furthermore, boundary
conditions like the interface geometry or the substrate mechanics can
be modulated.\ We found across this wide range of
\mbox{experimental} conditions that cells with relatively stronger
force transmission mediated by their cell--cell adhesion levels are
exclusively winning. This validated the observations in patient-derived
tumours. It showed that force transmission increases cellular fitness
and provides a crucial advantage in cell \mbox{competition}.

\subsection{Loser cells are eliminated independently of compression at
tissue interfaces}\label{sec1.2}

Compressive stress has been described as the main driver of cell
elimination in mechanical cell competition
\citep{15,16,17,18,19,20,21}. Therefore, we reasoned that increased
cell--cell adhesion might provide a fitness advantage by allowing cells
to collectively exert compressive stresses on their opponents. Unlike
previous studies, our experimental approach enabled direct force and
stress measurements using Bayesian Inversion Stress Microscopy
\citep{36}. This method allows for a time-resolved and local readout of
compression (negative isotropic stress) and tension (positive isotropic
stress) within heterogeneous tissues \citep{37}.

In line with previous reports, we indeed observed in patient-derived
tumour cultures that the winning E-cad positive cells were under
tension, and the losing E-cad negative cells were under compression
(Figure~\ref{fig2}D). Surprisingly, however, we observed the opposite
stress pattern between competing MDCK WT and E-cad KO cells, with the
losing E-cad KO cells being under tension and the winning WT cells
under compression (Figure~\ref{fig2}E). Those measurements demonstrated
that loser cells can be eliminated independently of compression,
challenging the established consensus and suggest the existence of
other, still unknown mechanisms of cell elimination. 

Therefore, we investigated this competition scenario in detail. We
first ruled out other previously conjectured fitness-increasing traits
of winning cells, including cell-autonomous extrusion rates \citep{14},
cell proliferation \citep{19}, crowding \citep{17} and homeostatic
density \citep{20,21}, where we did not observe a difference between
the subpopulations.\ Surprisingly, mechanical \mbox{properties} associated with
increased fitness, namely cell {stiffness} and traction force
generation \citep{16,20}, were even increased in the losing E-cad KO
cells. This suggests that they are overruled by increased cell--cell
adhesion. The spatial distribution of cell extrusion events showed that
losing E-cad KO cells were preferably eliminated close to the tissue
interface (Figure~\ref{fig3}A). Those interfaces were enriched in
phospho-myosin, indicating increased mechanical activity
(Figure~\ref{fig3}B), suggesting a critical role of interface
mechanical activity in cell elimination \citep{38}. 

\begin{figure*}
{\vspace*{-.3pt}}
\includegraphics{fig03}
{\vspace*{-.6pc}}
\caption{\label{fig3}\advance\baselineskip-.1pt Interface stress
fluctuations lead to E-cad KO cell extrusion. (A)~Left: representative
phase contrast image of MDCK WT cells competing against E-cad KO cells
(green). Red circles highlight cell extrusions. Right: probability
distribution of cell extrusion as a function of the distance from the
interface. Cells have a typical diameter of 10--15 cells.
$n=\text{14,729}$ KO and $n=\text{11,031}$ WT extrusions from $N=4$
independent experiments. (B)~Immunostaining of phosphomyosin (red).
E-cad KO cells in cyan. (C)~Example simulation snapshot of the
multiphase field model. (D)~Fluctuations quantified as susceptibility
of the 2D stress, i.e.\ the in-plane isotropic stress field and the
$zz$, or out-of-plane component as a function of distance from the
interface in simulations. (E)~Susceptibility of the in-plane isotropic
stress in experiments. (F)~Example snapshots of a WT cell cluster
undergoing strong area fluctuations. (G)~Mirror situations. Extrusions
are highlighted with red circles. (H)~Schematic of the proposed
mechanism of cell competition. Scale bars $50~\upmu$m (A), $25~\upmu$m
(B). All figures from \citet{34}.}
\end{figure*}

\subsection{Stress fluctuations at active interfaces lead to E-cad KO
cell elimination} \label{sec1.3}

\looseness=-1
To understand the contribution of interface mechanics to governing the
cell competition outcome, we turned to 3D physical modelling using a
multiphase field approach \citep{39} (Figure~\ref{fig3}C), which
provides access to the 3D stress. Simulations of competing model
tissues showed that increased fluctuations, i.e.\ extreme tensile or
compressive values of the in-plane isotropic stress, resulted in
increased out-of-plane stresses leading to cell extrusion. To quantify
those fluctuations, we measured the susceptibility~$X$, and found that
they occurred predominantly at the interface of the competing
populations (Figure~\ref{fig3}D). To test these predictions, we
assessed stress fluctuations in our experimental data and indeed
observed the same striking stress fluctuation increase at the
interface, correlating with the location of cell eliminations
(Figure~\ref{fig3}E). Based on this combination of simulations and
experiments, we proposed a new mode of cell extrusion through increased
stress fluctuations. Importantly, it is not based on compression, and
extruded cells can be under tension.

Since all cells at the tissue interfaces were subjected to those
increased stress fluctuations, we then wondered why only E-cad KO cells
were eliminated. We reasoned that, under such mechanically challenging
conditions, increased force transmission might increase cellular
fitness, as it could allow a more efficient redistribution of stresses
across the tissue. Consequently, forces are shared among many cells,
which improves the mechanical resistance of the collective. Indeed, we
observed that small groups of WT cells could collectively \mbox{sustain}
drastic area fluctuations through cell deformation without being
extruded (Figure~\ref{fig3}F). In mirror {situations} where E-cad KO
islands were surrounded by WT cells, cells could not deform and
stresses were released through cell elimination (Figure~\ref{fig3}G).
Taken together, we showed that increased force transmission increases
mechanical cellular fitness because it allows winning cells to
collectively resist elimination forces caused by stress
fluctuations (Figure~\ref{fig3}H).

\section{Discussion and outlook}

In this study \citep{34}, we discovered that cell--cell adhesion
differences within epithelia can directly lead to cell competition.
Cells with increased intercellular adhesion are exclusively winning,
and we established adhesion-mediated force transmission as a crucial
regulator of cell fitness across different scenarios of cell
competition, including within patient-derived breast cancer xenografts.
This principle could have broad implications in development and
disease. Indeed, recent studies have shown that E-cad deficient cells
are eliminated during zebrafish development to ensure correct pattern
formation \citep{40}. Moreover, increased adhesion allows winning cells
to collectively exert compression on loser populations {in vivo}
and in the context of cancer-associated mutations \citep{41,42}. 

\looseness=-1
Importantly, we demonstrated that the force transmission capacity can
overrule other mechanical cellular properties like stiffness or force
generation, and even allows winners to persist in a compressive
environment, which was before considered the main driver of loser cell
elimination \citep{15,16,17,18,19,20,21}. The surprising observation
that cells under tension were in turn eliminated and became losers led
to the identification of stress fluctuations as a new,
compression-independent cell elimination mechanism. Reports of
cell extrusion mediated by tension in intestinal organoids were
recently published \citep{43}, suggesting that similar mechanisms could
govern cell elimination in physiological contexts. To withstand such
cell fluctuations and to resist elimination, we showed that increased
force transmission provides an inherently collective competitive
advantage, which~\mbox{allows} to dissipate elimination forces more
effectively throughout the tissue. 

The differences in force transmission studied here were particularly
pronounced due to binary expression levels of adhesion molecules.
However, biological and particularly pathological tissues, including
different cancers, can exhibit more heterogeneous protein expression
levels \citep{44}. Thus, future studies need to address competition
dynamics in such scenarios. Furthermore, research needs to address if
and how the principles of mechanical cell competition proposed here
affect the fate of cells on a longer timescale; that is, if invasion
and subsequent metastasis of specific subpopulations could be promoted.
Finally, we speculate that interface stress fluctuations might fulfil
an important function in maintaining tissue boundaries by preventing
the mixing of cell types and by removing undesired cells, which may
have left their fate compartments.

%\section*{Declaration of interests}
%The authors do not work for, advise, own shares in, or receive funds
%from any organization that could benefit from this article, and have
%declared no affiliations other than their research organizations.
\printCOI

\back{}

\printbibliography
\refinput{crbiol20260189-reference.tex}

\end{document}