overview mechanotransduction: a major regulator of homeostasis...

Overview

Mechanotransduction: a majorregulator of homeostasis anddevelopmentKevin S. Kolahi and Mohammad R.K. Mofrad∗

In nearly all aspects of biology, forces are a relevant regulator of life’s form andfunction. More recently, science has established that cells are exquisitely sensitiveto forces of varying magnitudes and time scales, and they convert mechanicalstimuli into a chemical response. This phenomenon, termed mechanotransduction,is an integral part of cellular physiology and has a profound impact on thedevelopment of the organism. Furthermore, malfunctioning mechanical propertiesor mechanotransduction often leads to pathology of the organism. In thisreview, we describe mechanotransduction and the theories underlying howforces may be sensed, from the molecular to organism scale. The influence ofmechanotransduction on normal and abnormal development, such as stem celldifferentiation and cancer, is also reviewed. Studies illustrate the diversity ofmechanotransduction, and the major role it has on organism homeostasis. Cellsemploy a variety of mechanisms, which differ depending upon cell type andenvironment, to sense and respond to forces . 2010 John Wiley & Sons, Inc. WIREs SystBiol Med

In general, all animals consist of four general tissuetypes: nervous, skeletal, connective, and epithelial

tissues. These tissues are organized into distinctgeometries and hence underlie the functional aspectsof many organs and organ systems. To survive, anorganism relies on the proper functions of organsand each organ in a body satisfies a specific survivaldemand, such as bodily fluid waste removal (kidneys),nutrient and oxygen perfusion to tissues (heart),absorption and solid excretion (gastrointestinal tract),and gas-exchange (lungs).

The four tissue types can be thought of as havinggeneral roles for each organ. Nervous tissue, forinstance, is responsible for the conduction of signalsto the organ, and can toggle organ function on oroff. Connective tissues form the scaffolding of theorgan and are responsible for forming the matrix ofthe organ’s cells and for the stable anchoring of theorgan in the animal. Physical forces are provided bymuscular tissue. Muscular tissue can force large-scaletransport of substances into and out of an organ, but

∗Correspondence to: [email protected]

Molecular Cell Biomechanics Laboratory, Department of Bioengi-neering, University of California, Berkeley, CA 94720-1762, USA

DOI: 10.1002/wsbm.79

can also support tissue architecture. The fourth tissuetype, epithelial tissue, can be thought of as the tissueunderlying absorption, excretion, secretion, sensation,and protection.

Although these general concepts can be appliedto organs, not all organs necessarily contain all fourtissue types. Evolution has specialized many organs tocarry out specific tasks, such as the central nervoussystem. The central nervous system can be consideredto contain mainly nervous tissue and very little, ifany, muscular tissue. In addition, these tissue typesare intimately associated and rely upon one anotherfor function.

In contemporary medicine, organs are thoughtof as modular in structure and function. However,organs work as systems and cooperatively inphysiology. For example, waste removal requires notonly the kidneys for filtration of bodily fluids, butalso the action of the liver to prepare compoundsfor excretion. To function properly, the kidneys alsorequire the action of the perfusion system, drivenby the heart. The heart is, in turn, dependent uponthe respiratory system, consisting of the lungs andbreathing muscles for its oxygen demand.

Recent advancements in molecular and cellularbiology allowed the discovery that tissue development

2010 John Wiley & Sons, Inc.

Overview www.wiley.com/wires/sysbio

is not absolutely autonomous, but tissues of adeveloping organism coordinate their maturation inan integrated manner.1,2 These studies are astoundingin that they implicate that many tissues of differentfates and origins communicate with one anotherreciprocally.

Another landmark of the 21st century science isthe discovery that cells and tissues communicate andrespond to forces.3–6 The transduction of mechanicalstimuli into a cellular signal and response is termedmechanotransduction.3–7

To achieve homeostasis, feedback exists betweenthe targets of signaling and the origin of the signal.There is thus reciprocity in signaling pathways.The target reciprocally communicates with theoriginator of the signal to modulate future responses.Reciprocity of signaling during development is notlimited to electrochemical signals, but can also occurthrough mechanical means, and this can be termedmechanoreciprocity.8,9 Developing tissues can thusinteract via forces and through physical contact. Thusthe aim is to understand both the biological andthe mechanical interactions of cells and tissues. Byunderstanding how tissues interact to form biologicalorgans of the developing organism, one can hopefullyexploit these in regenerative medicine, for example.

Mechanotransduction research is fundamentalin the underlying concepts of being able to oneday guide cells spatially and temporally to createdefined structures in three dimensions. Cells in vivoare constantly reshaping tissues by varying mechanicaland biochemical properties, both spatially andtemporally.10,11 These changes, on a grander scale,ultimately result in what is termed morphogenesis,differentiation, determination, and are a part ofdevelopment.

Major developmental processes are influencedby mechanotransduction. For instance, what marksthe left and right sides of the organism is determinedby coordinated ciliary beating which is thoughtto influence local fluid dynamics in the cellularenvironment.12 In addition, morphogenesis of organsoften involves a complex execution of forces to resultin proper tissue topology and shaping, such as duringthe formation of the chambers of the heart.13,14

Additionally, mechanotransduction can underliemany abnormal processes in organisms. For exam-ple, one of the first diseases to be correlated withbiomechanics and mechanotransduction is atheroscle-rosis, a disease that is now the leading cause of deathin the United States.15,16 Researchers have shownthat low-oscillatory shear stress correlates with sitesof atherosclerotic plaques, and that the low-shear

stress environment dramatically alters endothelialorganizations, especially their cytoskeleton.17–20

For centuries, orthopedists have known thatbone growth and healing, among the many factorsincluding diet, are also correlated to weight bearingactivities. Since then, many other examples of howmechanics plays a major role in biology have surfaced,but how this process specifically occurs at themolecular and cellular levels remains elusive.

In the remainder, we will review the conceptsthat define the field of mechanotransduction beginningwith the molecular scale and proceeding to theorgan scale. We discuss the current theories as tohow molecules can be influenced by force and thecorresponding reactions of the cells and tissues thatsense these changes. The force and time scales thatlife is sensitive to are also relevant to understandingmechanotransduction theory. Lastly, we discuss howmechanotransduction plays a role in normal andabnormal developmental processes and propose someareas of inquiry for future study.

MOLECULAR CELL BIOMECHANICSUltimately, intracellular propagations of all stimulioccur through biochemical cascades to alter transcrip-tion and cellular activity. Until the origin of mechan-otransduction theory, it was widely assumed thatchemically propagating stimuli also began throughbiochemistry, i.e., binding and oligomerization orchanges in electrical conductance leading to bind-ing or enzymatic conformational changes. However,recent evidence has led to the discovery that cells alsorespond to their physical environment and these aretransduced into intracellular biochemical cascades.Therefore, the term ‘mechanotransduction’ describesthe specific capacity of life to transduce a mechanicalsignal into a biochemical signal. Mechanotransduc-tion is a rapidly growing field since its implication inthe mid-1960s by Y.C. Fung.7,21

Interestingly, although it is known that mechan-ical factors do dramatically alter cellular and eventissue behaviors, their relationship to biochemicalsignals remains elusive. Even further, it is debatedwhether or not cells are actually actively respondingto the mechanical factors, or if the mechanical factorspassively enable other cell behaviors. For example,the observation that cells migrate and behave moremesenchymal-like on stiffer substrates may not bethat cells actually respond to the stiff substrate, butrather that certain expression patterns are enabled inthe presence of greater physical substrate support.22 Acell’s intention is difficult, if not impossible, to justify.Nevertheless, mechanics cannot be neglected if one


WIREs Systems Biology and Medicine Mechanotransduction

were to try to understand cellular development andfunction.

Forces are increasingly being recognized asmajor regulators of cell structure and function.23 Thebalance of forces among cells determines multicellularorganization as much as the expression of genes does.Recent studies in vitro and even in model organisms,such as Drosophila melanogaster or Caenorhabditiselegans, combine physical modeling with experimentalmeasurements to provide insight on the processes thatinfluence cell patterns in vivo.24

MECHANOTRANSDUCTIONA major hallmark of modern molecular and cellbiology is the discovery of the cytoskeleton in bothprokaryotic and eukaryotic organisms. This discoveryprompted the revision of the model that cells arebasically bags of enzymes. The cytoskeleton is a stressbearing structure and governs cell shape. Structurally,the cytoskeleton also serves as the mechanical linkbetween the exterior and internal environments.

During processes like cell movement or shapechange, the cytoskeleton is actively and dynamicallyremodeling. The precise regulation of cell movementis the epitome of mechanotransduction. For example,a cell that is migrating must be able to transducean initial chemical cascade into a physical effect.Furthermore, the cell must also be able to recognizewhat forces it has applied and how the cytoskeletonis organized in order to successfully carry out itsmovement. Thus, a cell must in every day activity beable to both transduce a biochemical stimulus intoa mechanical effect, but also vice-versa, transducea mechanical stimulus into a biochemical effect, thelatter we typically call mechanotransduction.

It seems obvious to some degree that the formercase is entirely possible through the action of forceproducing enzymes, such as motor proteins andcytoskeletal filament polymerizations. However, howthe cell can recognize the force it is applying and theforce it feels from the external environment is not soclear. Biochemical pathways in the cell are under pre-cise regulation, require an input of energy, and are notgenerally reversible. These two pathways, althoughseemingly related, proceed via unique mechanisms.

Recently, the cytoskeleton has also been impli-cated in serving as a scaffold for many biochemicalactivities. Transmembrane chemoreceptors or chan-nels often link to the cytoskeleton, namely actin.25,26

Even classical signal transduction cascades are oftendemonstrated to be intimately linked and localizedto the cytoskeleton.25,26 In fact, some normal andpathogenic processes rely on their anchors to the

cytoskeleton to carry out their activities. For example,the human immunodeficiency virus requires cytoskele-tal binding to endocytose and infects cells.27 One canreadily identify, therefore, that mechanics plays a largerole in biological behaviors at varying scales, from themolecular level to the scale of an organism.

One must take into account that the magnitudesof length and time at the molecular scale may diferfrom the macroscopic scale. The challenge of modernbiomechanics and mechanotransduction is to bridgethe gap between the micro and macro scales. Thiswill depend largely upon the integration of knowledgefrom the molecular to the organism scale. Ideally,one would be able to one day extrapolate frombiomechanics at the molecular scale and predictthe macroscopic effects at not only the cell scale,but also the tissue, organ, and even on the levelof the whole organism (Figure 1). For example,when a gene encoding an extracellular componentprotein in D. melanogaster, bola, is mutated, the eggchamber of the organism develops spherically ratherthan oblong in wild type28–31 (Figure 1). Althoughmultiple mutations in differing extracellular matrix(ECM) encoding genes result in similar phenotypes,the mechanical implications of such mutations at thecellular scale underlying the phenotype, nor the tissuemorphogenetic alterations are known.28–31

Theoretically, one should be able to identifythe cellular changes, which could underlie changesin the tissue morphology (Figure 1). Although theseassertions are highly speculative and the goalambitious, the use of such knowledge could be usefulfor diagnosing the molecular bases of diseases andfuture cures. Such foresight could also be used indesigning future tissues and organs for regenerativemedicine.

MECHANOTRANSDUCTIONINFLUENCES PHYSIOLOGYThe cardiovascular system is markedly sensitive tomechanical stress. It is well established that fluid flowforces regulate the development and physiology of theheart and the vascular network.32–36 At the molecu-lar and cellular scale, vascular endothelia align theirF-actin stress fibers along the primary orientation offluid shear stress.37 In areas of high unidirectionaloscillatory shear stress, vascular endothelium exhibitsa uniform orientation of their cytoskeletal network,which produces a cylindrical supracellular arrange-ment, and is atheroprotective.38,39

In areas of low oscillatory shear stress orrecirculation zones, the vascular endothelium developsa random orientation of stress fibers and does not



S7

I

a p

B Awt C bola D

pa

K

bola S7wt

H

Wildtype Mutant

PA

FIGURE 1 | A single mutation ina gene dramatically alters the localcell shape and also the global tissuemorphology. The organism developsabnormally spherical. In normaldevelopment the epitheliumappears stretched a long a uniformaxis, and the mutation disrupts thisprocess. The mutation results in adecrease in net force within theepithelium. (Adapted withpermission from Ref 30. Copyright2001 Development).

form a supracellular structure. At these sites, there is agreater propensity for vascular wall thinning, resultingin aneurysms, or atherosclerotic plaques.39–42 Someresearchers speculate that wall thinning may be ahomeostatic response to increase tissue stresses in theendothelial wall.

Another example where fluid shear stressesinfluence tissue homeostasis is in bone. In haversian orcortical bone, osteocytes are thought to regulate bonestructure by responding to dynamic fluid shear stressesthat are normally generated during movement.43

Bone cells, osteocytes, extend long membranousprocesses through bone channels, canaliculi, andconnect to share cytoplasmic material via gapjunctions. The osteocytes can thus respond to fluidflow that is induced by squeezing during bending ofstress in bone.44

THE MOLECULAR BASIS OFMECHANOTRANSDUCTION

The signaling cascades that become activated bymechanical stress are actively studied. However, themeans by which this mechanical stress is initiallyconverted into a biochemical signal is much lessunderstood. A myriad of theories exist that mightexplain the process of mechanotransduction, but mostrequire further development and refinement.

An excellent example of mechanotransductionis the sensory system of hair follicles. In hair follicles,the cilia that compose the hair bundle are physically

linked to transmembrane potassium channels.45 Anyforce that is applied to the hair deflects the ciliaand in turn stretches the potassium channels and thepotassium influx initiates the biochemical-signalingcascade.

While many cells are known to contain stretch-sensitive ion channels, such as endothelial vascularepithelium,17,46–49 the atomic structure of only a fewis known, and molecular dynamics simulations provesomewhat inconclusive. Furthermore, only recentlyare efforts being made to identify the downstreameffects of such mechanotransducive events in othercells and tissues.

The cellular response to force can vary frombeing essentially instantaneous to on the order ofdays and even years as with heart disease, forexample.50 Most rapid responses to force occurthrough transient changes in the transmembraneelectric potentials because of ion flux throughtransmembrane channels51 (Figure 2). The completeelectrochemical impulse can be completed on theorder of milliseconds.19,20 These stretch-sensitiveion channels may be mechanically coupled to thecytoskeleton, or may be sensitive to membranetensions (Figure 2).

It has also been demonstrated that changes inmembrane fluidity during cell stretching can directlyinfluence associations of transmembrane receptormolecules.52–54 The time scale for these events is notknown, but could be potentially immediate. Muchresearch is surfacing to support that these eventsare possible due to force activated conformational



Forces via intercellular junctions

Extracellular matrix Forces at extracellular attachments

Forces via glycocalyx Cytoskeletal

forces

Fluid shear stress

Forces to nucleus

Glycocalyx

Actin stress fiber network

FIGURE 2 | A schematic illustration of some candidates that sense force and propagate the signaling cascade. While this illustration is not anexhaustive list of every candidate known to be sensitive to force, it describes some canonical mechanotransduction pathways, such as through theintegrin–cytoskeleton linkage, and through stretch-sensitive ion channels.

changes of proteins, directly influencing their bind-ing affinities. Force could alter the binding affinitiesfor cytoskeletal binding proteins at focal adhesioncomplexes, for instance. Eventually, these conforma-tional changes could potentially lead to activation ofmultiple signal transduction pathways and alter geneexpression patterns.

A major misconenception of mechanosensationput to rest by Davies was that mechanosensation hadto occur at the site of force application.37 Daviesillustrated that stresses, such as fluid shear stress, canbe transmitted to remote parts of the cell throughvarious intracellular structures.37 To elaborate, thecytoskeleton is a highly interconnected web thatconnects cell–cell junctions, focal adhesions, thenuclear membrane, and cytoskeletal binding factors(Figure 2). Thus, any force at one site in the cell canbe immediately propagated through the network andsignal transduction can occur at any of these sitesinstantly. Researchers have even demonstrated thatthis can actually be many orders of magnitude fasterthan intraceullular biochemical cascades.43

SIGNALING THROUGH FOCALADHESIONS

Many biological researchers today appreciate thatother proteins, in addition to transmembrane chan-nels, undergo conformational changes when subjectedto stress. Arguably the most widely known and

studied mechanism of force sensation and trans-duction is through focal adhesion complexes. Focaladhesions bridge the extracellular matrix to theintracellular environment. Integrins are transmem-brane heterodimeric transmembrane receptors thatbind extracellular matrix molecules and link them tointraceullar cytoskeletal binding proteins like talin,vinculin, filamin, paxilin, and focal adhesion kinase(FAK), for example.7,55–57 These cytoskeletal bindingproteins bridge integrins to the F-actin cytoskeleton.

The integrin’s cytoplasmic domain is associatedwith cytoskeletal binding proteins and can alsobe activated by force to associate with a host ofsecond messengers to initiate signaling cascades. Thissignal cascade will not only recruit more cytoskeletalassociations to increase mechanical bracing, butalso increase the expression of focal adhesions andlocalization of more focal adhesions to the area. Theresponse to force at a focal adhesion is logical, whichincreases its strength and the cell’s ability to withstanddeformation. However, there is a limit to this responsewhich will be discussed in greater detail in subsequentsections.

The entire protein–protein interaction map isnot known at focal complexes, but some examples aresummarized in Figure 3. It is important to note thatthe nature and regulation of force-induced activationmay differ drastically between these illustrated factors(Figure 3). One example illustrates how integrinreceptors can activate FAK, either directly or by FAK’sindependently sensing force. A conformational changeof FAK may alter its binding affinity for one of several



FIGURE 3 | Mechanotransductioncan occur in many locationssimultaneously within the cell.Forces can be transmitted from theexterior via the extracellular matrixor cell–cell junctions, or directly intostructures like the glycocalyx. Thesetransmitted forces can effectmembrane fluidity andbiochemistry, or propagate furtherthrough the cytoskeleton and evendeform the nucleus. All of theseevents may occur in concert andtheir contribution need not beequivalent or even cooperative.

Extracellular matrix

Force K+ Ca2+

[K+]

[Ca2+]

Integrin heterodimer

Talin

Vinculin

Rho

Filamin

Transcriptional adjustment

FAKRas

PaxilinPI3-KSrcGrb2, Grb-7

PKCMLCKCamKCalcineurin

GrafCsk,Cas,CalpainHic-5

known binding partners. In addition, FAK activitymay vary after the application of force, and this can inturn lead to the activation of other signaling moleculessuch as Rho, Rac, and Ras, to name a few.58

The experience of force at a focal adhesionnot only leads to recruitment of cytoskeletal bindingproteins, but also increases the number of recruitedsignal transduction molecules. A focal contact thathas progressed to accumulate a number of molecularfactors is termed a focal complex.

NUCLEAR SIGNALING

The nesprins are a family of proteins that linkthe cytoskeleton to the nucleus.59–61 Nesprin 1 and

2 associate with actin filaments and SUN domaincontaining proteins of the nuclear membrane.59–61

Nesprin 3 can also associate with SUN domaincontaining proteins, but links to the intermediatefilament network.59–61 The lamins are proteins thatform the nuclear envelope, or stabilize intranuclearstructure and associate with the SUN domaincontaining proteins and with chromatin associatedfactors.59–61 Therefore, forces at the extracellularmatrix can immediately propagate into the nucleus(Figure 3). This may initiate changes in transcriptionand subsequent changes in translation.

Additionally, forces propagated to the nucleusmay physically alter the rate of nuclear cytoplasmictransport.62 Thus, multiple means exist to sense forces



Seconds

Stretch-sensitive ion channel activation Changes in transcription and translation

Cytoskeletal remodeling Developmental processes, both normal and abnormal (pathological)

Changes in extracellular matrixChanges in cell and tissue morphologies

Changes in cell metabolism i.e. stimulation of endo-and exocytosis Increase in mitosis (due to increases in membrane tension, for example)

Increase in intracellular [Ca2+]GPCR activationMAPK pathway activationPI-3K activationTransient increase of IP3Increased release of NO, PGI2, endothelin (from vascular endothelium, for example)

Changes in chromatin assembly and nucleocytoplasmic transport (speculative)

Minutes Hours Days Years

FIGURE 4 | The varying time and length scales of cellular responses are illustrated in this figure. In general, as time increases so do themacroscopic changes to the organism. The events that characterize short and quick responses, tend to be second messenger cascades orelectrochemical. Other changes, such as gene transcription and cell/tissue morphological changes occur on the order of minutes to hours. Finally, ifthese events continue the final effect is large-scale changes in tissue organizations and morphological development of the organism. (Adapted withpermission from Ref 65. Copyright 2003 BioMed Central).

at the nucleus; the topology of DNA can be directlyaffected, or the transport of transcription factors intoand out of the nucleus may differ.

An increasing amount of research is demonstrat-ing the grand role force has on cellular homeostasisand function. Cells often respond to force by mod-ifying intracellular structure, migration, and modi-fying transcription and translation. More recently,researchers have also revealed another excitingfront in mechanotransduction where forces are eventhought to influence post-translation modifications.Indeed, interdisciplinary researchers employ biochem-ical knowledge of glycosylation to better understandhow the turnover rate of the glycocalyx is influencedby shear stresses, for example.63,64

FORCE AND LENGTH SCALES OFMECHANOTRANSDUCTION

An interesting aspect of mechanotransduction thatcontrasts with other canonical cellular signal trans-duction pathways is the various responses that can beachieved with a seemingly uniform ligand/stimulus:

force. This is surprising given that the nature of thecellular response to force can vary not only betweendiffering cell types, but even within a uniform celltype. This is possible partly because cells and tissueexperience forces of great range in magnitude andtime scale or frequency (Figure 4).

Utilizing an in situ assay for the exposure ofshielded or buried cysteine residues in proteins incells, Johnson et al. were able to demonstrate thatcytosolic proteins do unfold in their normal cellularenvironment.66 Surprisingly, in both red blood cells,cells that regularly experience fluidic shear stress, andmesenchymal stem cells, there existed proteins thatregularly unfolded.

While it is true for the most part that forceswithin tissues and cells are distributed and decreasewith increasing distance from the site of application,cellular structure indicates means by which forcescan be directly transmitted large distances and evenconcentrated. For example, cells are attached to theextracellular environment at discrete locations, focalcontacts being one type. A uniformly applied stress tothe top surface of a cell is inherently concentrated as



a consequence of there being only 1–10% of surfacearea contact with the substrate at the bottom surface.Thus, a stress can be magnified by many orders ofmagnitude.39

Mack et al. have demonstrated such stress focus-ing at focal contacts through cell force applicationwith fibronectin coated magnetic beads.39 In fibrob-last cultures, the effect of focal adhesion stress focusinghas also been reproduced.67

Atomic force microscopy and optical traps areexperimental tools that can apply forces small enoughto slowly extend proteins and probe interactions viaforce. In cases of extracellular adhesions, it has beenshown that integrin–extracellular matrix componentadhesions can be ruptured with forces on the order of30–100 pN.68 Proteins are viscoelastic and hence thestrain rate markedly affects the unfolding pathway ofthe protein as well as the force required for unfolding.

Using molecular dynamics simulation andatomic force microscopy, mechanical linking proteinslike titin in muscle, filamin, and α-actinin in F-actincross linking, and talin in focal adhesion formationhave been actively studied for their ability to senseforce inputs.55–57,69,70 The forces for many of the con-formational changes exhibited by these proteins fallwithin the range of 10–100 pN and occur on theorder of picoseconds to microseconds. The sensationof force by these proteins must be greater than thethermal energy available to the system (kT) which isapproximately 4 pN·nm.

Researchers have quantified the force that cellsactively contract on their substrate, which is measuredto approximately 5.5 nN/µm2.71,72 This measurementhas made possible the extrapolation of the forceexperienced by a single integrin molecule to be on theorder of several piconewtons. Mack et al. have shownthat forces applied via magnetic beads to adherentvascular endothelial cells can elicit a response if onthe order of 1 nN, corresponding to the thresholdof 1 Pa that Davies et al. proposed to approximatethe threshold for stimulation by hemodynamic shearstress.37,39

The body experiences forces on time scales thatare many orders of magnitude greater than whatcells and molecules require to respond to mechanicalstimuli. This can be analogous to having cells, whichtypically have transmembrane electrical potentials onthe order of hundreds of millivolts, to experiencingexternal electrical fields on the order of volts tokilovolts on a daily basis.

Just as chemical physiological processes exist inthe organism to preserve homeostasis, processes mustexist to preserve homeostasis in light of the manymechanical stimuli that an organism experiences on a

daily basis. These effects must be remarkably robust,yet exquisitely sensitive to account for the staggeringranges in force, length, and time scales experienced.

MECHANOTRANSDUCTION INEPITHELIAL TISSUESIn 1824, Etienne Serres put forth the embryologicalparallelism ‘ontogeny recapitulates phylogeny’.73

Consistent with the Meckel–Serres Law, epithelialtissue is considered to be one of the first tissuesto evolve and is the first tissue formed duringdevelopment. Furthermore, epithelial tissue is believedto be one of the most basic tissue types, and thereforecan serve as an excellent model for research in‘developmental biomechanics’.

Epithelium occurs on all inner and outer surfacesof organs. During development, epithelial cells formconfluent monolayers and these epithelial tissue sheetstwist and contort to form various structures togive rise or house developing organs and bodilycompartments.74

Epithelial cells have an apical face and a basalface (Figure 5). The apical face functions in absorp-tion and secretion of matter from or into the luminalspace formed by the epithelial tissue. The basal faceof the epithelium is the site attachment to the ECM,a complex network of post-translationally modifiedprotein filaments, such as collagens, laminins, fibrins,fibrilarins, and many more6,9 (Figure 5). ECM fila-ments are highly charged because of extensive post-translational modifications such as glycosylation andsulfation, and are given the name glycosaminogly-cans. The extensive charge of the ECM underlies itswater retaining properties and hence directly influ-ences the viscoelasticity. In aging the number of gly-cosaminoglycans or overall charge of the ECM tendsto decrease reducing the tissue viscoelasticity.

Epithelial tissue sheets are organized by theirexpression of adhesional proteins on their membranesurface, primarily their lateral membrane surfaces(Figure 5). Many adhesional protein complexes existwithin epithelial cells and they can have varyingdegrees of adhesion strength. Cellular adhesionis highly regulated and is organized in specificorientations and geometries giving rise to manyintegral functional properties of the epithelium.

One adhesional structure that is integral to tissuearchitecture and highly studied is the zonula adherens,or adherens junctions. The proteins responsible forcell–cell adherens junctions are the calcium iondependent adhesion molecules, the cadherins. Manydifferent isoforms or homologs exist, and exhibitdiffering pairing affinities. When two cells adhere



Zonula occludins (tight junctions)

Apical face Zonula adherens (adherens junction)

Apical actomyosin network

Septate junctions

Basal faceBasal extracellular matrix

Epithelial tissue adhesion architecture

FIGURE 5 | A generalized epithelial tissue architecture depicting some characteristic epithelial adhesion structures and geometry. The epithelialtissue is polarized with an apical face and a basal face. The apical face functions mainly to import and export. To increase surface area, the membraneis highly folded forming structures called villi and within the villi are even smaller microvilli. At the basal domain is the extracellular matrix that formsthe tissue scaffold. Three major cell–cell adhesion structures exist each with different accompanying physiological functions, the tight junctions,adherens junctions, and the septate junctions. All of these junctions also serve as intracellular anchor points for the cytoskeleton, commonly actin,microtubules, and intermediate filaments.

to one another via cadherins, the adhesive complexformed is termed a desmosome, or macula adherentes,in Latin meaning ‘adhering spot’.

Interestingly, the desmosomes formed betweentwo cells need not be between identical cadherinisoforms. Cadherins have been shown to form notonly homodimers between cells, but also heterodimersbetween isoforms of markedly differing affinities.75,76

Consequentially, epithelial cells can prefer to adheremore tightly to neighbors expressing isoforms thatmaximize adhesional strength.75,77 This hypothesiswas proposed originally by Steinberg given thisheterodimerization phenomenon, and was named thedifferential adhesion hypothesis (DAH).78,79

The DAH proposes that epithelial tissue mor-phogenesis can be driven by the segregation ofepithelial cells due to expression of differingcadherins.78,80,81 In other words, two differing celltypes can form immiscible, separate clusters solelybased upon their differing adhesion affinities.76 Italso proposes that to interchelate cell types, similaradhesion affinities must exist.75,78 The DAH is oftendescribed similar to the theory of chemical solubility,in which similar interactions, or interaction affinitybetween particles should promote solubility. How-ever, recent work has revealed that the affinities of

cadherins expressed by cells do not correlate with thepredictions of cell lineage segregation.76,82

Nevertheless, the DAH has been demonstratedto be consistent with a number of morphogeneticprocesses,80,81 and this inconsistency has spurreda number of researchers to propose that activeprocesses may be at play.75,82 One of the majoractive processes that is proposed to be at playis mechanotransduction, and many investigationssupport that the cadherins may respond to force,namely β-catenin.82 When cyclical stress is appliedto cells expressing β-catenin, the stiffness of the cellsincreases, due in part to the increased recruitmentof filamentous actin to cadherins.82 Although theseexperimental observations support that β-catenin canfunction as a mechanosensor, it does not directlyreveal why adhesion affinities do not correlateto cell–cell segregation predictions. However, theseworks do demonstrate the dynamic nature betweenthe cytoskeleton and intercellular interactions.

As the cytoskeleton is responsible for cell shape,adhesion therefore also affects the geometry of the cellby interacting with and modifying the cytoskeleton.83

Interestingly, this effect is reciprocal, as the cytoskele-ton can exert stresses on adhesion structures, andpromote their disassembly, for example.81



CELL AND TISSUE MORPHOGENESISRELIES ON MECHANICAL CUES

During development of multicellular organisms,epithelial cell sheets coordinate to achieve a myriadof specific and distinctive shape changes. Invagina-tion, evagination, folding, intercalation (convergentextension), cell flattening (epiboly), ingression, egres-sion, and branching are widespread examples ofepithelial morphogenesis in the animal kingdom.84–86

Remarkably, the epithelial cells within the tissue mustprecisely organize adhesion, actin–myosin contrac-tility, apical–basal and planar cell polarity duringmorphogenesis.

During morphogenesis, each cell must processand physically respond to patterning and polarityinformation in order to effectively take part in tis-sue growth. Cell behaviors during morphogenesisinclude processes such as cell growth, migration,death, division, and shape changes. Biological tis-sues have two mechanical properties that directlycontradict each other. They exhibit a stable struc-ture, which is needed to resist stresses, but theyalso demonstrate a dynamic and plastic behavior,allowing for remodeling.81 The balance of these twoopposing mechanisms can be thought of as leadingto tissue homeostasis.81 The balance between inter-cellular adhesion, an outwardly directed force, andintracellular contractility, an inwardly directed force,can bring about tissue homeostasis. Cells can tilt thebalance of forces to create polarized epithelial layersand control cell sorting.81

The orchestration of such mechanisms has beenmainly studied through the identification of signalingpathways that control cell behavior both in time andspace. However, it is an open area of investigationas to how this information is processed to controlcell shape and dynamics. A range of developmentalphenomena can be explained by the regulation of cellsurface tension through adhesion and cortical actinnetworks.81

Although the interplay between adhesion andcontractility can explain or support a number of devel-opmental phenomena, it is unknown how this can beexploited or exactly how this leads to the observedmorphogenesis, i.e., by identifying molecular candi-dates and their spatiotemporal expression. In addi-tion, describing certain developmental or pathologicalmechanisms solely by adhesion and contractility vari-ations, such as in cellular metastasis or mesenchy-mal cell migrations, may be too oversimplifying tobe useful. Morphogenesis ultimately relies upon themechanical environment for cues.

STEM CELL MORPHOGENESIS

If development and morphogenesis is exquisitelysensitive to mechanical cues, it is not surprisingthat these cues are equally influential in stem celldifferentiation. It has been repeatedly demonstratedthat stem cell differentiation can be reliably controlledsolely by mechanical cues.5,87–90 Although in vivostem cell differentiation likely involves more than justmechanical cues, i.e., biochemical cues as well, thismechanically driven differentiation is robust and canbe reproduced reliably.

In general, research in this field is supportingthat stem cells differentiate to whatever tissuestiffness most resembles that of their substrate.6 Forexample, nervous tissue is comparatively soft to otherhuman tissues, and neurons therefore form fromstem cells when cultured on soft substrates.87,88,90,91

Myocytes and osteoblasts form on substrates that areincreasingly stiff.6 Moreover, these phenomena occurin multiple stem cell lineages, and even in circulatingstem cells.90

Even cells that are not stem cells tend to mimicthe stiffness of their substrate, such as fibroblasts.92

Fibroblasts within a sensible range, will contract ontheir substrate such that their average cellular stiffnesswill reflect that of the substrate.92 Fibroblasts as amature cell line may be unique in this capacity as theygenerally serve a repair or wound healing role in thebody. Therefore, it is conceivable that to form repairtissues, it would be ideal to most closely resemblewhat was previously lost or damaged.

TUMOROGENESIS

Cells can misinterpret mechanical cues and becomemalignant. In some cases, like breast cancer, forinstance, the stiffness of the substrate directlycauses female mouse mammary epithelium to becomemalignant.93,94 Even more astonishing is that themalignant cells can be transferred to soft substrates,resembling normal physiological stiffness, and the cellswill return to normal. It is hypothesized that similarmechanisms exist underlying the progression of oneof the most deleterious forms of cancer, glioblastomamultiforme.95

It is possible that mechanical cues can potentiatean existing or new mutation allowing pathogenecity,but this is unknown. A defining characteristic ofnearly all tumors is the increased expression of matrixmetalloproteases.96,97 Matrix metalloproteases arewidely expressed in metastatic cancers, as a meansto potentially create a greater void for tumorproliferation, or possibly to modify their mechanical



environment for other means. This is an active area ofresearch with much still remaining to be investigated.

CONCLUSIONS: AN INTEGRATION OFMECHANOTRANSDUCTION FROMMOLECULAR, CELLULAR, TISSUE ANDORGAN/ORGANISMAL LEVELS

Mechanotransduction is a relatively newly discoveredphenomenon describing the translation of forces intoa biochemical language the cell can understand. Cellsand tissues are extremely sensitive to changes in theirmechanical environment and respond in ways that arestill explicitly unclear.

Cells employ more than just chemical cues toregulate their physiology, and mechanics plays an inte-gral role in regulating cell movements, morphologies,and even metabolism. During development, a num-ber of complex cell shape changes and populationreorganizations occur to result in the complex tissuegeometries of the organism. These involve coordinatedmovements of cells and massive matrix remodeling toform these tissues. Mechanotransduction undoubtedlyplays a role during these processes.

Just as in other signal transduction pathways,feedback loops exist to promote homeostasis. Amechanical response to the environment, suchas decreasing cellular contractility or cytoskeletalintegrity, or increasing activity matrix degradationor deposition, can significantly adjust the mechanicalenvironment. This can in turn, lead to a reciprocalresponse within that same cell, or lead to a change innearby cell populations.9

The implications of this are astounding, andcan, potentially, lead to new theories in areas of tissueengineering, regenerative medicine, and development.For example, although stem cells can be driven viamechanical changes to differentiate to target cell typesor some tissues, no methods exist to create completebiological organs.98 Currently the major hurdle isculturing large tissues in three dimensions.98

In addition to this major problem, it is alsonecessary to interchelate multiple tissue types andcell types into a defined three-dimensional topologyfor proper organ function. For example, vasculartissue must be present in all target organs.99 Theprocess by which these tissues organize themselvesand differentiate into particular shapes and topologiesare unknown. Cell environments are dynamic duringin vivo development and morphogenesis, and nor cancurrent approaches capture this. A cell’s environmentis constantly being reshaped and therefore tissueengineering in vitro must also be dynamic.89

It is often neglected that many of the forcesand mechanical cues of a developing organ or tissuemay come from tissues not necessarily from the samelineage. In fact, it is entirely possible that mechan-otransduction can occur between tissues, or intertissuemechanotransduction. A fundamental understandingof the in vivo micromechanics of processes involved inmorphogenesis is essential to employ these processesin vitro to differentiate stem cells to a target tissue.

Understanding how a mechanical, dynamic feed-back between interacting tissues can drive concerteddevelopment during reproductive growth is integralfor future prospects in regenerative medicine.89 Forexample, the mechanical identification that neuronsregulate axonal growth by tension to connect to otherneurons, has allowed for the creation of devices thatcan in a user-defined manner, guide neuron axonalgrowth both in vitro and in vivo.11,100 With a singleobservation in the mechanics of development of neuralconnective systems, these devices were fabricated andnow make possible the repair of nerve connections inthe human nervous system in situ.11

It may be advantageous, then, to take atop–down approach rather than a bottom up inregards to tissue and organogenesis. The assumptionthat fully developed organs are modular may not bevalid during development. Organs may require allaccessory tissues with which they develop, despitethese tissues not being intimately involved in organfunction, nor derived from the same germ layer.Research may benefit to test if developing tissues canbe shaped in a controlled manner.

In addition, the magnitude of developmentalfactors omitted during classical in vitro regenerativemedicine applications is largely confounding in ourefforts to understand differentiation. It may serveresearch better, to instead allow development toproceed, but aim to precisely guide cell movementsby varying select parameters. This method, in situdirected tissue engineering, would benefit in allowingthe researcher to make direct cause and effectrelationships, thus enhancing the wealth of knowledgein directed growth. In situ directed tissue engineeringefforts could test if ontogeny can be recapitulated andmodified in vitro.

Many pathologies and abnormalities can surfacewhen mechanotransduction malfunctions, includingbut not limited to tumorogenesis. For an excellentreview on pathological conditions resulting fromimproper mechanotransduction, see Jaalouk andLammerding.34

Physical changes described by mechanics arethe ultimate means tissues communicate, and changesin these mechanics describe changes in spatial and



temporal organizations of tissues, supporting the bio-logical function. Organogenesis thus requires a greater

understanding of the relationships between the bio-chemistry and their net physical effects.

ACKNOWLEDGEMENTS

Financial support through a National Science Foundation CAREER award (CBET#0955291) is gratefullyacknowledged.

REFERENCES

1. Raya A, et al. Notch activity acts as a sensor for extra-cellular calcium during vertebrate left-right determi-nation. Nature 2004, 427:121–128.

2. Goodwin M, Yap AS. Classical cadherin adhesionmolecules: coordinating cell adhesion, signaling andthe cytoskeleton. J Mol Histol 2004, 35:839–44.

3. Iqbal J, Zaidi M. Molecular regulation of mechan-otransduction. Biochem Biophys Res Commun 2005,328:751–755.

4. Ingber DE. Cellular mechanotransduction: putting allthe pieces together again. FASEB J 2006, 20:811–827.

5. Ingber DE. The mechanochemical basis of cell andtissue regulation. Mech Chem Biosyst 2004, 1:53–68.

6. Discher DE, Janmey P, Wang YL. Tissue cells feeland respond to the stiffness of their substrate. Science2005, 310:1139–1143.

7 Mofrad M, Kamm R. Cellular Mechanotransduc-tion: Diverse Perspectives from Molecules to Tissuesy.New York: Cambridge University Press; 2009, 375.

8. Lopez JI, Mouw JK, Weaver VM. Biomechanical reg-ulation of cell orientation and fate. Oncogene 2008,27:6981–6993.

9. Butcher DT, Alliston T, Weaver VM. A tense situ-ation: forcing tumour progression. Nat Rev Cancer2009, 9:108–122.

10. Robertson MJ, Gip P, Schaffer DV. Neural stemcell engineering: directed differentiation of adult andembryonic stem cells into neurons. Front Biosci 2008,13:21–50.

11. Phillips JB, et al. Neural tissue engineering: a self-organizing collagen guidance conduit. Tissue Eng2005, 11:1611–1617.

12. Marshall WF, Kintner C. Cilia orientation and thefluid mechanics of development. Curr Opin Cell Biol2008, 20:48–52.

13. Taber LA. Biophysical mechanisms of cardiac looping.Int J Dev Biol 2006, 50:323–332.

14. Voronov DA, et al. The role of mechanical forces indextral rotation during cardiac looping in the chickembryo. Dev Biol 2004, 272:339–350.

15. Dewey CF Jr, et al. The dynamic response of vascularendothelial cells to fluid shear stress. J Biomech Eng1981, 103:177–185.

16. Hahn C, Schwartz MA. Mechanotransduction in vas-cular physiology and atherogenesis. Nat Rev Mol CellBiol 2009, 10:53–62.

17. Barakat AI. Responsiveness of vascular endotheliumto shear stress: potential role of ion channels andcellular cytoskeleton (review). Int J Mol Med 1999,4:323–332.

18. Barbee KA, et al. Subcellular distribution of shearstress at the surface of flow-aligned and non-aligned endothelial monolayers. Am J Physiol 1995,268:H1765–H1772.

19. Lehoux S, Tedgui A. All strain, no gain: stretch keepsproliferation at bay via the NF-kappaB response geneiex-1. Circ Res 2003, 93:1139–1141.

20. Lehoux S, Tedgui A. Cellular mechanics and geneexpression in blood vessels. J Biomech 2003,36:631–643.

21. Kassab GS. ‘‘Bert’’ Fung: the father of modern biome-chanics. Mech Chem Biosyst 2004, 1:5–22.

22. Kumar S, Weaver VM. Mechanics, malignancy, andmetastasis: the force journey of a tumor cell. CancerMetastasis Rev 2009, 28:113–127.

23. Discher D, et al. Biomechanics: cell research and appli-cations for the next decade. Ann Biomed Eng 2009,37:847–859.

24. Fernandez-Gonzalez R, Zallen JA. Epithelial organi-zation: may the force be with you. Curr Biol 2008,18:R163–R165.

25. Huveneers S, Danen EH. Adhesion signal-ing—crosstalk between integrins, Src and Rho. J CellSci 2009, 122:1059–1069.

26. Lindberg U, et al. The microfilament system andmalignancy. Semin Cancer Biol 2008, 18:2–11.

27. Jimenez-Baranda S, et al. Filamin-A regulates actin-dependent clustering of HIV receptors. Nat Cell Biol2007, 9:838–846.

28. Bateman J, et al. The receptor tyrosine phosphataseDlar and integrins organize actin filaments in the



Drosophila follicular epithelium. Curr Biol 2001,11:1317–1327.

29. Deng WM, et al. Dystroglycan is required for polar-izing the epithelial cells and the oocyte in Drosophila.Development 2003, 130:173–184.

30. Frydman HM, Spradling AC. The receptor-like tyro-sine phosphatase lar is required for epithelial pla-nar polarity and for axis determination withindrosophila ovarian follicles. Development 2001,128:3209–3220.

31. Gutzeit HO. The microfilament pattern in the somaticfollicle cells of mid-vitellogenic ovarian follicles ofDrosophila. Euro J Cell Bio 1990, 53:349–356.

32. Gimbrone MA, et al. Endothelial dysfunction, hemo-dynamic forces, and atherogenesis. Ann N Y Acad Sci2000, 902:230–239; discussion 239–40.

33. Haga JH, Li YS, Chien S. Molecular basis of the effectsof mechanical stretch on vascular smooth muscle cells.J Biomech 2007, 40:947–960.

34. Jaalouk DE, Lammerding J. Mechanotransductiongone awry. Nat Rev Mol Cell Biol 2009, 10:63–73.

35. Lammerding J, Kamm RD, Lee RT. Mechanotrans-duction in cardiac myocytes. Ann N Y Acad Sci 2004,1015:53–70.

36. Li YS, Haga JH, Chien S. Molecular basis of theeffects of shear stress on vascular endothelial cells. JBiomech 2005, 38:1949–1971.

37. Davies PF. Flow-mediated endothelial mechanotrans-duction. Physiol Rev 1995, 75:519–560.

38. Dai G, et al. Distinct endothelial phenotypes evokedby arterial waveforms derived from atherosclerosis-susceptible and -resistant regions of human vas-culature. Proc Natl Acad Sci U S A 2004,101:14871–14876.

39. Mack PJ, et al. Force-induced focal adhesion translo-cation: effects of force amplitude and frequency. AmJ Physiol Cell Physiol 2004, 287:C954–C962.

40. Boussel L, et al. Aneurysm growth occurs at regionof low wall shear stress: patient-specific correlationof hemodynamics and growth in a longitudinal study.Stroke 2008, 39:2997–3002.

41. Mofrad MRK, et al. Characterization of theatherosclerotic carotid bifurcation using MRI, finiteelement modeling, and histology. Ann Biomed Eng2004, 32:932–946.

42. Tzima E, et al. Activation of Rac1 by shear stress inendothelial cells mediates both cytoskeletal reorgani-zation and effects on gene expression. EMBO J 2002,21:6791–6800.

43. Wang N, Tytell JD, Ingber DE. Mechanotransductionat a distance: mechanically coupling the extracellularmatrix with the nucleus. Nat Rev Mol Cell Biol 2009,10:75–82.

44. Santos A, Bakker AD, Klein-Nulend J. The role ofosteocytes in bone mechanotransduction. OsteoporosInt 2009, 20:1027–1031.

45. Holt JR, Corey DP. Two mechanisms for transduceradaptation in vertebrate hair cells. Proc Natl Acad SciU S A 2000, 97:11730–11735.

46. Hamill OP. Twenty odd years of stretch-sensitivechannels. Pflugers Arch 2006, 453:333–351.

47. Ninio DM, Saint DA. The role of stretch-activatedchannels in atrial fibrillation and the impact ofintracellular acidosis. Prog Biophys Mol Biol 2008,97:401–416.

48. Suchyna TM, Sachs F. Mechanosensitive channelproperties and membrane mechanics in mouse dys-trophic myotubes. J Physiol 2007, 581:369–387.

49. Zagorodnyuk VP, et al. Mechanotransduction andchemosensitivity of two major classes of bladder affer-ents with endings in the vicinity to the urothelium. JPhysiol 2009, 587:3523–3538.

50. Weinberg EJ, Schoen FJ, Mofrad MR. A computa-tional model of aging and calcification in the aorticheart valve. PLoS One 2009, 4:e5960.

51. Ingber DE. Tensegrity-based mechanosensing frommacro to micro. Prog Biophys Mol Biol 2008,97:163–179.

52. Li S, et al. The role of the dynamics of focal adhesionkinase in the mechanotaxis of endothelial cells. ProcNatl Acad Sci U S A 2002, 99:3546–3551.

53. Sawada Y, Sheetz MP. Force transduction by Tritoncytoskeletons. J Cell Biol 2002, 156:609–615.

54. Zhang YL, Frangos JA, Chachisvilis M. Mechanicalstimulus alters conformation of type 1 parathyroidhormone receptor in bone cells. Am J Physiol CellPhysiol 2009, 296:C1391–C1399.

55. Golji J, Collins R, Mofrad MR. Molecular mechanicsof the alpha-actinin rod domain: bending, torsional,and extensional behavior. PLoS Comput Biol 2009,5:e1000389.

56. Kolahi KS, Mofrad MR. Molecular mechanics of fil-amin’s rod domain. Biophys J 2008, 94:1075–1083.

57. Lee SE, et al. Molecular dynamics study of talin-vinculin binding. Biophys J 2008, 95:2027–2036.

58. Katsumi A, et al. Integrins in mechanotransduction. JBiol Chem 2004, 279:12001–12004.

59. Crisp M, et al. Coupling of the nucleus and cyto-plasm: role of the LINC complex. J Cell Biol 2006,172:41–53.

60. Haque F, et al. SUN1 interacts with nuclear laminA and cytoplasmic nesprins to provide a physi-cal connection between the nuclear lamina and thecytoskeleton. Mol Cell Biol 2006, 26:3738–3751.

61. Warren DT, et al. Nesprins: intracellular scaffoldsthat maintain cell architecture and coordinate cellfunction? Expert Rev Mol Med 2005, 7:1–15.



62. Wolf C, Mofrad MR. On the octagonal structureof the nuclear pore complex: insights from coarse-grained models. Biophys J 2008, 95:2073–2085.

63. Lopez-Quintero SV, et al. The endothelial glycocalyxmediates shear-induced changes in hydraulic con-ductivity. Am J Physiol Heart Circ Physiol 2009,296:H1451–H1456.

64. Tarbell JM, Ebong EE. The endothelial glycocalyx:a mechano-sensor and -transducer. Sci Signal 2008,1:pt8.

65. Yan L, et al. Knockout of ERK5 causes multipledefects in placental and embryonic development.BMC Dev Biol 2003, 3:11.

66. Johnson CP, et al. Forced unfolding of proteins withincells. Science 2007, 317:663–666.

67. Hu S, et al. Intracellular stress tomography revealsstress focusing and structural anisotropy in cytoskele-ton of living cells. Am J Physiol Cell Physiol 2003,285:C1082–C1090.

68. Lehenkari PP, Horton MA. Single integrin moleculeadhesion forces in intact cells measured by atomicforce microscopy. Biochem Biophy Res Commun1999, 259:645–650.

69. Oberhauser AF, et al. Stepwise unfolding of titinunder force-clamp atomic force microscopy. Proc NatlAcad Sci U S A 2001, 98:468–472.

70. Schwaiger I, et al. A mechanical unfolding intermedi-ate in an actin-crosslinking protein. Nat Struct MolBiol 2004, 11:81–85.

71. Balaban NQ, et al. Force and focal adhesion assembly:a close relationship studied using elastic micropat-terned substrates. Nat Cell Biol 2001, 3:466–472.

72. Geiger B, et al. Transmembrane crosstalk between theextracellular matrix–cytoskeleton crosstalk. Nat RevMol Cell Biol 2001, 2:793–805.

73. Montgomery W. The politics of evolution. Morphol-ogy, medicine, and reform in radical London. AdrianDesmond. University of Chicago Press, Chicago,1990. x, 503 pp., illus. $34.95. Science and its con-ceptual foundations. Science 1990, 248:883–884.

74. Marieb EN, Hoehn K. Anatomy and Physiologyy.Benjamin Cummings; 2007, 1048.

75. Leckband D. Beyond structure: mechanism anddynamics of intercellular adhesion. Biochem SocTrans 2008, 36:213–220.

76. Shi Q, Chien YH, Leckband D. Biophysical propertiesof cadherin bonds do not predict cell sorting. J BiolChem 2008, 283:28454–28463.

77. Lecuit T. Adhesion remodeling underlying tissue mor-phogenesis. Trends Cell Biol 2005, 15:34–42.

78. Foty RA, Steinberg MS. The differential adhesionhypothesis: a direct evaluation. Dev Biol 2005,278:255–263.

79. Steinberg MS, Poole TJ. Strategies for specifying formand pattern: adhesion-guided multicellular assem-bly. Philos Trans R Soc Lond B Biol Sci 1981,295:451–460.

80. Lecuit T, Le Goff L. Orchestrating size and shapeduring morphogenesis. Nature 2007, 450:189–192.

81. Lecuit T, Lenne PF. Cell surface mechanics and thecontrol of cell shape, tissue patterns and morphogen-esis. Nat Rev Mol Cell Biol 2007, 8:633–644.

82. Chien YH, et al. Two stage cadherin kinetics requiremultiple extracellular domains but not the cytoplas-mic region. J Biol Chem 2008, 283:1848–1856.

83. Ingber DE. Mechanobiology and diseases of mechan-otransduction. Ann Med 2003, 35:564–577.

84. Fristrom D. The cellular basis of epithelial morpho-genesis. A review. Tissue Cell 1988, 20:645–690.

85. Pilot F, Lecuit T. Compartmentalized morphogenesisin epithelia: from cell to tissue shape. Dev Dyn 2005,232:685–694.

86. Quintin S, Gally C, Labouesse M. Epithelial morpho-genesis in embryos: asymmetries, motors and brakes.Trends Genet 2008, 24:221–230.

87. Engler AJ, et al. Matrix elasticity directs stem celllineage specification. Cell 2006, 126:677–689.

88. Engler AJ, et al. Extracellular matrix elasticity directsstem cell differentiation. J Musculoskelet NeuronalInteract 2007, 7:335.

89. Ghosh K, Ingber DE. Micromechanical control of celland tissue development: implications for tissue engi-neering. Adv Drug Deliv Rev 2007, 59:1306–1318.

90. Zajac AL, Discher DE. Cell differentiation throughtissue elasticity-coupled, myosin-driven remodeling.Curr Opin Cell Biol 2008, 20:609–615.

91. Thakar RG, et al. Contractility-dependent modula-tion of cell proliferation and adhesion by microscaletopographical cues. Small 2008, 4:1416–1424.

92. Solon J, et al. Fibroblast adaptation and stiffnessmatching to soft elastic substrates. Biophys J 2007,93:4453–4461.

93. Novaro V, Roskelley CD, Bissell MJ. Collagen-IV andlaminin-1 regulate estrogen receptor alpha expressionand function in mouse mammary epithelial cells. JCell Sci 2003, 116:2975–2986.

94. Rønnov-Jessen L, Bissell MJ. Breast cancer by proxy:can the microenvironment be both the cause andconsequence? Trends Mol Med 2009, 15:5–13.

95. Ulrich TA, de Juan Pardo EM, Kumar S. The mechan-ical rigidity of the extracellular matrix regulates thestructure, motility, and proliferation of glioma cells.Cancer Res 2009, 69:4167–4174.

96 Tallant C, Marrero A, Gomis-Ruth FX. Matrix met-alloproteinases: fold and function of their catalyticdomains. Biochim Biophys Acta 2009, 15.



97. Wolf K, Friedl P. Mapping proteolytic cancer cell-extracellular matrix interfaces. Clin Exp Metastasis2009, 26:289–298.

98. Tsang VL, Bhatia SN. Fabrication of three-dimensional tissues. Adv Biochem Eng Biotechnol2007, 103:189–205.

99. Borenstein JT, et al. Microfabrication of three-dimensional engineered scaffolds. Tissue Eng 2007,13:1837–1844.

100. Heidemann SR, Buxbaum RE. Tension as a regu-lator and integrator of axonal growth. Cell MotilCytoskeleton 1990, 17:6–10.


overview mechanotransduction: a major regulator of homeostasis...

Documents