Available online at www.sciencedirect.

com

ScienceDirect
Physics Procedia 57 (2014) 24 – 28

27th Annual CSP Workshops on “Recent Developments in Computer Simulation Studies in

Condensed Matter Physics”, CSP 2014

Physics of cell adhesion failure and human diseases

Fereydoon Family*
Department of Physics, Emory University, Atlanta, GA 30322

Abstract

Emergent phenomena in living systems, including your ability to read these lines, do not obviously follow as a consequence of
the fundamental laws of physics. Understanding the physics of living systems clearly falls outside the conventional boundaries of
scientific disciplines and requires a collaborative, multidisciplinary approach. Here I will discuss how theoretical and
computational techniques from statistical physics can be used to make progress in explaining the physical mechanisms that
underlie complex biological phenomena, including major diseases. In the specific cases of macular degeneration and cancer that
we have studied recently, we find that the breakdown of the mechanical stability in the local tissue structure caused by weakening
of the cell-cell adhesion plays a key role in the initiation and progression of the disease. This finding can help in the development
of new therapies that would prevent or halt the initiation and progression of these diseases.

© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
© 2014 The Authors. Published by Elsevier B.V.
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the Organizing Committee of CSP 2014 conference.
Peer-review under responsibility of The Organizing Committee of CSP 2014 conference
Keywords: Theory, modeling, and computer simulation of biological systems, Cell adhesion and cell mechanics, Diseases, Cancer
PACS 87.15.A-, 87.17.Pq, 87.17.Rt, 87.19.xj, 87.19.X-)

1. Introduction

Life in all of its forms, including your ability to read these lines, does not follow from fundamental laws of
physics. This fact precludes direct application of physical laws to biological systems. In his 1944 monograph called
“What is Life?” Erwin Schrödinger (1944) proposed that life had two secrets: the passage of an encoded molecule

* Corresponding author. Tel.: +1-404-727-4293; fax: +1-404-727-0873.

E-mail address: [email protected]

1875-3892 © 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of The Organizing Committee of CSP 2014 conference
doi:10.1016/j.phpro.2014.08.126
 Fereydoon Family / Physics Procedia 57 (2014) 24 – 28 25

from parents to offspring that is responsible for heritable characteristics, and the spontaneous emergence of order in
living systems. The first idea has culminated in the unprecedented progress made in genetics and molecular
biophysics. On the other hand, still little is known about the origin of emergent phenomena in living systems, even
though some progress has been made in elucidating collective behavior and self-organization in certain condensed
matter and quantum systems.
There are subtle distinctions that separate many toy models exhibiting emergent behavior in biologically inspired
models from predictive understanding of emergent phenomena in living systems, including the pathogenesis of
human diseases. This lack of progress is clearly evident, for example, in the fact that the physical mechanisms of
only an anecdotally small number of catastrophic ailments have been discovered.
In this paper I will review the results of some recent work, Shirinifard et al. (2011), Gryn and Family (2014), we
have done using computer simulations of realistic models of biological systems to investigate the initiation,
progression and spreading of two types of human diseases: (1) Age-Related Macular Degeneration, and (2) Cancer
Tumor Growth and Invasion. The main question we address here is what is the role of cell-cell adhesion failure in
the transition from a healthy organ to one that promotes initiation and spreading of a disease? Organs in living
systems are made up of tissues that communicate with other cell types via cell–cell interactions at the cell surface.
Our main finding is that pathogenic processes cannot be attributed alone to genetic changes in a single cell, but are
the result of the complex interplay between the failure of cellular regulation mechanisms and the presence of a
permissive microenvironment. In fact our results show that in the cases that we have studied the biological systems
become vulnerable to growth and spreading of diseases once the mechanical and physical stability of a tissue is
compromised due to weakening of cell adhesion. This suggests that therapies based on controlling or eliminating
adhesion failure could be effective approach for preventing and halting the initiation and progression of certain
diseases.

2. Simulating Biological Systems

A computational approach is perhaps the most effective way to deal with the multilevel complexity typical of
biological systems. Such an approach allows us to simplify the biological problem by providing both a concise
description of its essential features and the possibility of incorporating the relevant mechanisms and parameters. A
significant advantage of a computational approach is that it provides a controlled way of determining the
consequences of different experimental conditions and serves as a guide for future experiments
Recent advances in condensed matter and statistical physics have been achieved by modeling disparate types of
materials as collections of large number of identical interacting elements. When viewed as a unique state of matter,
biological organs and tissues can also be considered as a collection of interacting and inter-communicating cells
representing the fundamental elements of the system at a mesoscopic scale. In this type of description, cells are
represented by one or more set of discrete units with rules describing their movements and interactions. One of the
most effective approaches of this type is the Cellular Potts Model, Glazier and Graner (1993), Swat et al. (2012),
(also known as the Glazier-Graner-Hogeweg model, Balter et al. (2009)) in which each element can be represented
by a collection of spatial units that capture the biophysical properties of individual cells, such as shape, movement or
adhesion, and their interactions with the neighboring cells. In this type of approach, the cell-scale elements obey a
set of prescribed rules, that may depend on their type and on the signals they receive from the neighbors and from
the environment. CPM has been successfully used to model many problems in biological and biomedical field,
including morphogenesis, cell sorting, angiogenesis, wound healing, and tumor vasculogeneis, as well as non-
biological applications, including simulations of foam drainage and foam rheology (see Swat et al. (2012), Balter et
al. (2009) for references to various applications).
The Cellular Potts Model (CPM) is a generalization of the Q-state Potts Model from statistical mechanics and
shares with it the idea of modeling dynamics using energy minimization under random fluctuations. The details of
the CPM approach can be found in several publications (see Swat et al. (2012), Balter et al. (2009) and references
therein) as well as at the CompuCell3D website (http://www.compucell3d.org), where extensive tutorials and guides
about CPM are available. Cells in the CPM are represented as spatially extended but internally structureless objects
with complex shapes. Cell behaviors and interactions are described in terms of effective energies. On the lattice,
each site or pixel is associated with a single spin. A collection of pixels with the same spin represents a cell. The
26 Fereydoon Family / Physics Procedia 57 (2014) 24 – 28

change in shape of the cell and/or cell motion happens through mutation of spins of individual pixels. Any physical
process that can be described in terms of an internal energy term can be incorporated in the simulation by adding
that energy term to the effective energy. At each time step a lattice site belonging to a particular cell is randomly
selected and proposed to copy its spin value into a randomly selected nearest-neighbor site in the neighboring cell.
The probability for this exchange depends on the energy cost between the initial and the final configuration. Since
the probability must remain invariant under a constant shift in the energy scale, the only function that satisfies this
condition is the exponential function. Thus, a Metropolis type algorithm is used for spin updates that are accepted
with Boltzmann-like probability.
Cellular adhesion is essential in maintaining multicellular structure, because it plays a key role in the movement
and the mechanical stability of biological tissues. Cell adhesion is the binding of a cell to a surface or substrate, the
extracellular matrix, or another cell. Adhesion occurs from the action of proteins called cell adhesion molecules,
such as selectins, integrins, and cadherins. In addition, cellular adhesion can link the cytoplasm of cells and can be
involved in signal transduction. Consequently, one of the key terms in the system effective energy is the adhesion
energy (Glazier and Graner (1993)). The adhesion energy is based on the Differential Adhesion Hypothesis, first
proposed in 1960’s by Steinberg and others, Foty and Steinberg (2005), which asserts that cellular movement is
controlled by the rearrangement of cells to a more thermodynamically stable structure. This is achieved by
maximizing the amount of energy that is utilized adhering the cells together, which decreases the free energy
available in the system. As cells with similar strengths of surface adhesion bond to one another, bonding energy in
the overall system increases, and interfacial free energy decreases causing the system to be more thermodynamically
stable. As I will discuss below, our results show that cell adhesion failure is a key mechanism in the pathogenesis of
two specific diseases that we have studied.

3. Age-Related Macular Degeneration (AMD)

Age-Related Macular Degeneration (AMD) is the leading cause of blindness in the elderly and is a looming
epidemic in our aging society, Holz et al. (2011). Currently, no effective therapies are available for prevention or
halting of this debilitating disease. Choroidal neovascularization (CNV), Grossniklaus (2004), is the pathogenic
mechanism that leads to more severe form of AMD. In CNV, new blood vessels from the choriocapillaries initially
penetrate the Bruch's membrane (BrM) and then spread into the retina layers and eventually affect the
photoreceptors, causing irreversible vision loss. Our results question the validity of the current hypotheses regarding
the causes of CNV.
CNV involves the interaction of two fundamental structures of the eye, the retina and the choroid. The outer
retina is a layered structure that is situated between the photoreceptors and the choroid. The choroid lies in the
posterior of the eye and consists of the dense choriocapillaris that bring oxygen and nutrients to the eye and dispose
of chemicals and other waste products. The retinal pigment epithelium (RPE) is a monolayer of pigmented epithelial
cells that is situated between the photoreceptors and the choroid. The RPE plays numerous roles in maintenance of
the retina. Between the RPE and the choroid lies Bruch’s membrane (BrM), a strong, multi-layered collageneous
membrane that separates the RPE from the choriocapillaris. The inner part of the retina, adjacent to the vitreous
humor, includes the retinal vasculature and layers of neural cells. The outer retina, adjacent to the RPE, includes the
rod and cone photoreceptors.
Multiple hypotheses have been proposed to explain CNV initiation, growth and patterning (for comprehensive
reviews, see Spade et al. (2003), Zarbin et al. (2004)). These hypotheses are divided into two major groups based on
their proposed mechanism of action: (1) VEGF overexpression, and (2) irregularities in BrM. A key factor that had
not been considered is the role of interactions among the different types of cells and the possibility that their
impairment could have an effect on CNV. To better understand the mechanism for CNV, we have considered the
role and the correlations of adhesion failure in the initiation and progression of CNV.
We have used CPM and developed a three dimensional model, Shirinifard et al. (2012), that allows us to probe
the effect of specific mechanisms without confounding crosstalk and to study the effects of multiple mechanisms
acting simultaneously or sequentially. This approach allowed us to explore many more combinations of bio-
mechanistic hypotheses and parameter choices than we could in experiments. We simulated different adhesion
scenarios by assigning one of three levels: normal, moderately impaired, and severely impaired to each of the five
 Fereydoon Family / Physics Procedia 57 (2014) 24 – 28 27

key adhesion parameters, Shirinifard et al. (2012). For each adhesion scenario we simulated our retina both in the
absence and presence of a tip cell for one simulated year.
We have developed, Shirinifard et al. (2012), a morphometric quantification and classification algorithm to
classify CNV patterns. Our classification scheme is compatible with current histological classifications and can be
applied to images of appropriately labeled sections and 3D images of the retina.
We find that previously neglected RPE-BrM, RPE-RPE and RPE-POS adhesion failures suffice to determine the
loci and progression of choroidal neovascularization. Our simulations confirm all clinically known patterns of CNV.
But, while most studies hypothesize that CNV primarily results from the effects of excess VEGF or holes in BrM,
we find that CNV results from combinations of impairment of RPE-RPE junctional adhesion, adhesion failure in
RPE-BrM complex, and adhesion of the RPE to the photoreceptor outer segments (RPE-POS adhesion). These
findings suggest that defects in adhesion dominate CNV initiation and progression.

4. Cancer Tumor Growth and Invasion

Normal cells in a tissue survive within well-regulated and mechanically stable environmental conditions and are
subject to microenvironmental control. On the other hand, tumor cells must overcome a host of hostile
microenvironment conditions, including hypoxia, growth factor deprivation, and loss of adhesion to the ECM.
Therefore, understanding how cancer tumors grow is made difficult by the complexity that arises from the interplay
of several factors including cancer cell heterogeneity, tumor cell interactions within the tumor and with the local
extra cellular matrix (ECM) and stochastic mutations and adaptations during progression. The question we address
here is: how do tumor cells overcome these obstacles and are able to invade the ECM? Specifically, we have
concentrated on the role of microenvironment modifications and the resulting weakening of ECM adhesion on
cancer tumor growth and invasion.
Invasion is basically a three-step process involving changes in tumor cell adhesion, degradation of the ECM and
finally the ability of the tumor cells to migrate in the modified ECM environment. This is consistent with the idea
that the major function of the ECM is to provide a scaffold to support the organization of cells into specific tissues.
Under normal conditions, the ECM provides a physical barrier against invasion and metastasis. However, the ECM
is linked strongly, through cell adhesion processes, to the tumor cells. Modifications to this interaction that degrade
cell adhesion can facilitate not only tumorigenesis, but also the process of tumor invasion.
We have developed a three-dimensional model using the Cellular Potts Model and have studied the dynamics of
the growth of a cancer tumor cell in a healthy ECM microenvironment. We have used the parameters that have been
used in previous models of tumor growth and angiogenesis, Swat (2012), Balter (2009), and have exclusively
concentrated on the role of adhesion failure in tumor growth. To invade the extracellular matrix environment, a cell
has to adhere to its local environment in order to pull itself through the matrix. Clinical and experimental studies
have long shown that cancer cells modify the local environment through various mechanisms such as changes in the
local H+ concentration, Estrella (2013), and release of enzymes. When the local matrix is too dense, the cell has to
degrade the matrix in order to get through, but when it does not have sufficient friction, the cell produces proteins to
locally increase its traction.
Our results show that changes to the microenvironment and weakening of cell-cell adhesion, due to such
processes as chronic inflammation or upregulation of enzymes, strongly correlate with cancer tumor survival,
growth and invasion. A heterogeneous local environment enables tumor cells to migrate more easily and adapt to
other unfavorable conditions, such as hypoxia by migrating along the gradients of oxygen and nutrient
concentrations, Estrella (2013). By studying the velocity correlation among motile cells, we have found that
collective motion of cells is also enhanced with cell adhesion weakening. Enhanced collective behavior is perhaps
an additional mechanism that promotes cancer invasion. Our results support the longstanding hypothesis, Mantovani
(2013), that persistent inflammatory conditions promote tumor growth by continually upregulating enzymes that
weaken the adhesion properties of the ECM. The results also provide support for another finding that the pH of solid
tumors is acidic and the diffusion of H+ into the tumor microenvironment leads to remodeling of the
microenvironment and therefore cancer invasion, Estrella (2013). This is partly due to the ability of cancer cells to
adapt and thrive in acidic environments.
28 Fereydoon Family / Physics Procedia 57 (2014) 24 – 28

5. Conclusions

In this brief report I have discussed the results of studying the initiation and progression of two different types of
diseases, age-related macular degeneration and cancer tumor growth and invasion. The goal of these studies was
very specific: To determine the possible role of cell adhesion weakening on the initiation and progression of these
diseases. Cellular Potts Model is an effective approach for such a study, as it provides a mathematical and
computational environment that allows for realistic modeling of biological systems from the cellular to the
macroscopic sizes. The main finding of these studies is that adhesion failure can be a key factor in the development
of diseases. This finding supports, for example, the hypothesis that inflammation, which causes mechanical
degradation of the local tissue by weakening of the cell adhesion, plays a key role in promoting certain diseases,
including cancer tumor growth and invasion.

Acknowledgements

I would like to acknowledge and thank James Glazier, Hans Grossniklaus, Scott Gens, Yi Jiang, Abbas
Shirinifard, and Maciej Swat for their collaborations on some of the works discussed here.

References

Balter, A., Merks, R.M.H., Popławski, N.J., Swat, M., Glazier, J.A., 2009. The Glazier-Graner-Hogeweg Model: Extensions, Future directions,
and opportunities for further study. in “Single-Cell-Based Models in Biology and Medicine”, R. A. Anderson, Mark A. J. Chaplain, Katarzyna
A. Rejniak (Ed.), pp. 151-167.
Estrella, V., Chen, T., Lloyd, M., Wojtkowiak, J., Cornnell, H.H., Ibrahim-Hashim, A., Bailey, K., Balagurunathan, Y., Rothberg, J.M., Sloane,
B.F., Johnson, J., Gatenby, R.A., and Gillies, R.J., 2013. Acidity generated by the tumor microenvironment drives local invasion. Cancer
Research, doi:10.1158/0008-5472.CAN-12-2796.
Foty, Ramsey A.; Steinberg, Malcolm S. 2005. The differential adhesion hypothesis: a direct evaluation. Developmental Biology 278, 255–263.
Glazier, J.A, Graner, F., 1993. Simulation of the differential adhesion driven rearrangement of biological cells. Phys. Rev. E 47, 2128.
Grossniklaus, H.E., Green, W.R., 2004. Choroidal neovascularization. Am. J. Ophthalmol. 137, 496–503.
Gryn, O., Family, F., 2014. Effects of adhesion failure on cancer growth and invasion, in progress.
Holz, F.G., Pauleikhoff, D., Spaide, F., Bird, A.C., 2012. “ Age-related Macular Degeneration”, Springer, 2nd Edition.
Mantovani, A., Allavena, P. , Sica, A., Balkwill, F., 2008. Cancer-related inflammation. Nature. 454, 436–444.
Schrödinger, E., 1944. “What is Life?”, Cambridge University Press, Cambridge, UK.
Shirinifard, A., Glazier J.A., Swat, M., Gens, J.S., Family, F., Jiang, Y., Grossniklaus, H.E., 2012. Adhesion failures determine the pattern of
choroidal neovascularization in the eye: a computer simulation study. PLoS Comp. Biology 8, e1002440.
Spaide, R., Armstrong, D., Browne, R., 2003. Choroidal neovascularization in age-related macular degeneration–what is the cause? Retina 23,
595–614.
Swat, M.H., Thomas, G.L., Belmonte, J.M., Shirinifard, A., Hmeljak, D., Glazier, J. A., 2012. Multi-scale modeling of tissues using
CompuCell3D. Methods in Cell Biology 110, 325-366.
Zarbin, M.A., 2004. Current concepts in the pathogenesis of age-related macular degeneration. Arch. Ophthalmol 122, 598–614.