Stem Cell Differentiation is Regulated byExtracellular Matrix Mechanics

Stem cells mechanosense the stiffness of their microenvironment, which im-

pacts differentiation. Although tissue hydration anti-correlates with stiffness,

extracellular matrix (ECM) stiffness is clearly transduced into gene expression

via adhesion and cytoskeleton proteins that tune fates. Cytoskeletal reorga-

nization of ECM can create heterogeneity and influence fates, with fibrosis

being one extreme.

Lucas R. Smith,Sangkyun Cho, and Dennis E. Discher

Molecular & Cell Biophysics Lab, Physical Sciences OncologyCenter, University of Pennsylvania, Philadelphia, Pennsylvania

Introduction to Mechanosensing

Brain tissue is very soft and bone is rigid, but docells sense and respond to such physical differ-ences? Understanding most physiological func-tions has invariably relied on physics as well aschemistry, but a test-tube vision of well-stirredbiology ignores the mechanical properties of mosttissues as well as any collective effects on cells.New opportunities to address such issues are pro-vided by the sensitivity of stem cells in differenti-ation and the many types of stem cells that havebeen identified or generated de novo.

Embryonic stem cells and the related inducedpluripotent stem cells (iPS cells) give rise to hun-dreds of specialized cell types in the human body.Beyond functionally committed cells such asneurons, muscle cells, or bone cells, pluripotentstem cells also generate adult stem cells that aremulti-potent. Adult stem cells reside in specifictissues to provide for more restricted regenera-tion throughout life. To drive stem cells to anappropriate fate, coordination of inductive cuesis necessary, and various soluble factors in“cocktails” are certainly potent in this respect.However, physical factors—specifically the soft-ness or stiffness of the microenvironment— canalso contribute to differentiation (23).

The ease of deforming tissue or a cell is de-scribed by its mechanical properties, and to firstorder excludes changes in volume since we are, ofcourse, mostly incompressible water. Biologicaltissues deform when a mechanical stress (force perunit area) is applied, and the mechanical proper-ties of solid and semi-solid tissues are often sim-plified to an elastic modulus (mechanical stress perstrain) that varies widely across tissues (20). Braintissue requires very little stress to extend or shear itand has a low elastic modulus (E � 1 kPa), makingthe tissue “soft,” whereas “rigid” calcified bone hasan elastic modulus orders of magnitude higher(E � 1 GPa); all other solid tissues fall betweenthese two extremes (31, 67, 97) (FIGURE 1A). Water

content also decreases with tissue stiffness(FIGURE 1B) as various (non-fat) constituents in-crease in weight fraction, particularly ECM pro-teins such as collagens that are the most abundantproteins in the body; tissue softness and tissuewater content are thus “colligative” properties.Most reductionist studies with stem-cell culturesnonetheless use rigid and hydrophobic tissue cul-ture plastic, even though cultures of committedcells on soft hydrogels has been known since Pel-ham and Wang (71) to dramatically limit cellspreading and adhesive signaling relative to stiffsubstrates. Control over both adhesive ligands (i.e.,surface biochemistry) and gel mechanics(FIGURE 1C) was essential to proving this point,and much earlier work might be interpreted asimplying such matrix mechanosensitivity (5, 90).However, none of these early studies related me-chanical properties of tissues to culture sub-strates, likely because the needed instrumentsare rare in physiology and cell biology laborato-ries. Micro-scale tools such as atomic force mi-croscopes (AFM) have indeed been essential forthe mechanical characterization not only of tis-sues and stem-cell niches on cellular and subcel-lular scales but also the gels used to mimic them(49). AFM remains a workhorse for measuringsubstrate mechanics on the cellular scale, and avariety of techniques are now available to alsomeasure cellular forces and displacements (76).

Adhesive ligands are of course crucial for cells tomolecularly engage their surroundings, and suchligands would appear abundantly displayed on ex-tracellular matrix (ECM) molecules. Synthetic gelscan likewise be modified to display sufficient li-gand, such that adhesion is not limiting, althoughthis must always be verified when working withsynthetic gels. Using polyacrylamide gels (PA) thatwere first covalently modified with constant colla-gen (PA is inert and thus widely used for proteinseparations), Pelham and Wang varied substratestiffness through cross-linking and then demon-strated that 3T3 fibroblasts spread less on softer gels

REVIEWPHYSIOLOGY 33: 16–25, 2018. Published December 6, 2017; doi:10.1152/physiol.00026.2017

1548-9213/18 Copyright © 2018 Int. Union Physiol. Sci./Am. Physiol. Soc.16

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

(71). Stiff substrates, including rigid plastic and glass,promote spreading of most cells, whereas soft sub-strates limit spreading and produce more roundedcell morphologies (FIGURE 1D). Although PA hydro-gels are widely used to control substrate stiffness,other synthetic polymers, such as polydimethylsilox-ane (PDMS) elastomers, in addition to modifiedbiopolymers, such as alginate and hyaluronic acid(HA), have also been used (35). In general, gel stiff-ness is controlled by adjusting the concentration ofpolymer and/or cross-linker, and the gels are func-tionalized with collagen, fibronectin, or other ECMprotein to provide the biochemistry needed for celladhesion.

In tissues, ECM proteins also confer structureand mechanics. Most cells interact with ECM typ-ically by adhering, but ECM also acts as a depot forsemi-soluble growth factors, some of which de-pend on ECM mechanical properties for release(39). Fibrillar collagens (collagen-1, 2, 3, 5, 6 . . .)are the primary shear-resistant proteins of ECM,and the wide range of stiffnesses reported for tis-sues indeed correlate well with the amounts ofcollagen, over 4 –5 logs in collagen level (84). Tis-sues that sustain high mechanical stress (muscle,bone) have the most collagen and are the stiffest,whereas tissues that are relatively protected frommechanical stress (brain, marrow) have low colla-gen and are soft. Collagenase addition to such tis-sues will soften them in tens of minutes (84), whichis of course consistent with methods for isolatingprimary cells. Purified collagen self-assembles intoa fibrous gel with a concentration-dependent elasticmodulus, for which collagen fiber elastic modulus,bending modulus, lattice spacing, and cross-linkingcan all modulate gel stiffness (79). Fibers also tend toalign with the direction of strain and generally stiffenunder tension (64). Non-fibrillar collagens, pro-teoglycans, and other matrix components also mod-ulate the mechanical properties of ECM, either ontheir own or through their interactions with collagenfibrous network.

For cells to sense their mechanical microenvi-ronment, they must apply a force on that environ-ment, just as we need to push and tug on amaterial surface to probe its mechanics. Such cellforces are generated at least in part by the myosinII mini-filaments that pull on the actin cytoskele-ton and adhesions to drive cell and ECM contrac-tion (FIGURE 1E). Stiff substrates provide amechanical resistance or load that favors acto-my-osin assembly and cell spreading. Inhibition of my-osin II (or F-actin) pharmacologically indeedleads to cell rounding regardless of the stiffnessof the substrate, and so cells lose all mechano-sensitivity to matrix. Cell spreading while con-tracting on a stiff substrate requires stableadhesions to the ECM. Integrins coalesce into

focal adhesion complexes while binding specificECM ligands (83), and focal adhesions grow withtension to license cells to spread while pulling onthe matrix (8).

Although stiffness-dependent adhesion andspreading occur within short time scales of 1–2 h

FIGURE 1. Universal scale of micro-stiffness for tissuesA: stem cells derive the tissues across that body that vary in stiffness of wide scales,from fluid like in the marrow at �1 kPa to rigid bone in the GPa range. The stiffnessmeasured as microelasticity correlates with expression of collagen across the rangeof tissues but is generally much softer than the rigid plastic typically used in cell cul-ture (21). B: hydration level of several human tissues after extraction of fat from a46-yr-old male (26). Cartilage hydration state is age dependent and approximatedfor a 46-yr-old male (1). Bone matrix hydration is determined as a percentage of wa-ter and organic bone matrix (55, 56). The hydration state of tissues is inversely pro-portional to the tissue microelasticity (E) and collagen content. C: AFM is used toprobe tissue or gel stiffness on the scale of the cell. The microelasticity is deter-mined by measuring the restoring force relative to the indentation distance and de-pends on the probe tip (88). D: various cell types spread out when placed on asubstrate with collagen coating of a stiff underlying gel. The spread-out cells containa robust cytoskeleton with abundant actin stress fibers. E: collagen of the ECM pro-vides adhesion sites for transmembrane integrins of the cell that form the basis offocal adhesions. Focal adhesions anchor the actin cytoskeleton at the membrane,whereas the LINC complex anchors the cytoskeleton at the nuclear membrane. LINCcomplexes also interact with the nuclear stiffness, determining lamins just inside thenuclear membrane, which provides a direct link to chromatin and DNA.

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org 17

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

(71), more profound and lasting changes inphenotype, such as lineage specification of stemcells, requires specific transcriptional programsto be activated and/or repressed in a process ofmechanotransduction from the ECM to thenucleus. Among the mechanosensitive signalingpathways that transmit ECM stiffness signals totranscriptional machinery are the integrin-basedfocal adhesions with focal adhesion kinase (FAK)that initiates multiple mechanosensitive path-ways (32). These include ERK, JNK, Wnt-�-catenin, and Hippo pathways (45). Many of thesepathways regulate transcription factors thattranslocate into the nucleus in response to mi-croenvironment mechanics. Yes-associated pro-tein (YAP) and transcriptional coactivator withPDZ-binding motif (TAZ) are examples: they areusually in the cytosol of cells cultured on softmatrices, but when cell contractility increases ona stiff matrix, YAP/TAZ translocate into the nu-cleus, seemingly independent of the canonicalHippo signaling pathway (22). YAP/TAZ influ-ences differentiation of stem cells by such path-ways or others, and illustrate the broad range ofmechanisms by which differentiation of stemcells can be manipulated by their mechanicalenvironment.

Mechanosensitive signaling pathways with solu-ble cytoplasmic factors undoubtedly contribute(45), but mechanotransduction is also likelythrough direct physical linkage from ECM all theway to DNA. Filamentous actin that engages integ-rins together with microtubules and intermediatefilaments connect to the nucleus through nesprinsthat span the outer nuclear membrane. Nesprinscan in turn engage SUN proteins that span theinner nuclear membrane and bind to the underly-ing lamina proteins (17). A- and B-type laminsform the primary structural meshwork of the nu-clear envelope and determine the mechanicalproperties of the nucleus (70). A-type lamin levelsvary widely between tissues, but the ratio of A-typeto B-type lamins scales systematically with tissuecollagen levels and tissue stiffness over severalorders of magnitude, so that nuclear stiffnessincreases with ECM stiffness (84). Importantly,the nuclear lamins directly regulate transcriptionfactors with broad roles in differentiation such asserum response factor (SRF) and retinoic acidreceptors (RAR). The lamins also interact withand control integral membrane proteins, such aslamin-B receptor (LBR), that bind and regulatechromatin factors to complete the linkage fromECM to DNA (68). These linkages have beenshown to influence the differentiation of adultstem cells (84).

Matrix Stiffness DirectedDifferentiation

Human bone marrow-derived mesenchymal stemcells (MSCs) likely differentiate in situ into bone,cartilage, and fat (86); unlike mouse marrow,human adult marrow in long bones is filled withfat. Mouse marrow MSCs clearly express nestin insitu, which suggests some neuronal origins orcharacter (65). Human MSCs were the first stemcells used to demonstrate the in vitro influence ofmatrix stiffness on stem cell differentiation (23). Totest mechanosensitive lineage specification ofMSCs, PA gel formulations to mimic the tissuestiffness of brain, muscle, and osteoid (bone sur-face) were used. Importantly, the gels were coatedwith equal concentrations of collagen so that all ofthe MSCs cultured atop the gel interacted with thesame ligand. Based on protein and transcript expres-sion, MSCs differentiated over several days or moretoward neural, muscle, and bone cells on the tissue-matched substrates (FIGURE 2A). This mechano-transduction was lost when cells were treated withblebbistatin, which inhibits myosin II and therebyblocks the ability of MSCs to pull on their substrateand “feel” matrix stiffness. Importantly, phenotypicchanges in cell morphology occur in hours: neuron-like dendritic branching on soft gels, myoblast-likeelongation on intermediate stiffness gels, and osteo-blastic spreading on stiff gels. Although it is likely thatsuch human marrow-derived MSCs do not becomeneurons or muscle cells in vivo (13), multiple meth-ods of controlling substrate stiffness have demon-strated clear effects on MSC differentiation. Forexample, arrays of micropillars can be made thatallow cells to adhere to the tops with short and un-bendable pillars or else with long and flexible pillars(28), and MSCs on the short pillars responded to thehigh stiffness by undergoing osteogenesis, whereasMSCs on the longer pillars responded to the softsubstrate and underwent adipogenesis. Soft gels thatare made very thin on top of rigid glass likewise favorosteogenesis, with high lamin-A and low LBR,whereas thick versions of the same soft gels coordi-nately repress lamin-A and upregulate LBR, which isinteresting for its role in lipid biopsynthesis (8a). Theuse of 2D gels may be appropriate to mimic theprocess of MSCs adhering to the bone surface anddifferentiating into osteoblasts, but adipogenesislikely involves tissue building more in 3D. Embed-ding cells within 3D hydrogels presents some tech-nical challenges, but alginate modified for celladhesion showed that soft gels favored MSC adipo-genesis in 3D, whereas stiffer gels favored osteogen-esis (43).

Differentiation to multiple lineages is character-istic of stem cells, of course, but an ability to

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org18

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

self-renew without differentiation is also a definingproperty of stem cells that has shown to be down-stream of matrix mechanosensing. Skeletal musclemaintains a resident quiescent population of mus-cle stem cells (MuSC) that becomes activated andundergoes 1) symmetric division into muscle pro-genitors, 2) symmetric division that maintainsstemness, or 3) asymmetric division to produceone daughter cell of each cell type. Culture systemsthat constrain mouse MuSCs on asymmetric fi-bronectin-coated substrates leads to these asym-metric divisions (96). Symmetric divisions thatmaintain stemness are important to preserve thestem-cell population for future injuries. Using a gelstiffness to mimic skeletal muscle stiffness, MuSCunderwent significant symmetric stem-cell divisionthat was not observed on stiffer substrates (96), aswell as being able to maintain the quiescent state onsofter gels (62) (FIGURE 2B). MuSCs grown on gelswith the stiffness of healthy muscle were even able torepopulate the MuSC niche when injected back intoa mouse muscle as opposed to cells grown on stiffersubstrates (33). To repair muscle, MuSCs eventu-ally differentiate into contractile myotubes andmyofibers. In vitro, this is typically done on plasticand results in myotubes that are stuck in an im-mature state and do not progress to form highlyaligned and packed sarcomeres that are observedin mature muscle. However, when C2C12 cells, aMuSC-like cell line, were differentiated on PA gelswith an intermediate stiffness, as in healthy mus-cle, sarcomeric striation was maximized comparedwith softer gels and stiffer gels (16) (FIGURE 2C).Quiescent MuSCs are present in a complex 3Denvironment sandwiched between a mature mus-cle fiber and the basement membrane. Recent ef-forts have been made to engineer synthetic musclefibers as a substrate for MuSCs, again showing that,when the stiffness of these synthetic fibers ismatched to that of real fibers, the MuSCs main-tained a quiescent stem cell pool and an enhancedability to engraft in vivo (74).

Cardiac muscle, on the other hand, has a morelimited adult stem-cell progenitor pool. Along withits obviously critical function, this makes derivingcardiomyocytes from iPSCs a major regenerativemedicine target. Many protocols seek to use iPSCsto generate functional cardiomyocytes relying ontranscriptional induction on rigid plastic, with lim-ited success or else unknown contributions fromECM made by the cells (78). However, the ECMstiffness that a cardiomyocyte experiences in anembryo and through development is much softerthan the rigid plastic used in such studies (59). It isinteresting that, transplanting human PSC-derivedcardiomyocytes into a neonatal rat heart with asoft microenvironment facilitates nearly full matu-ration of the cardiomyocytes, which is not

observed when the cardiomyocytes are trans-planted into a stiffer adult rat heart (14). Embry-onic cardiomyocytes also tune their beating to thestiffness of their substrate. On very soft gels, they

FIGURE 2. Stiffness-based differentiation of stem cellsA: MSCs plated on collagen functionalized gels of the indicated stiffness differen-tiate over the course of a week into cells from tissues corresponding to the stiff-ness of the gel. Soft gels direct adipogenic differentiation, whereas the more stiffgels induce osteoblast differentiation. B: substrate stiffness drives the type ofproliferation among MuSC, where substrates with tissue-like stiffness often main-tain their stem-like properties (defined by expression of Pax7), stiff substrates de-plete the stem cell pool by inducing differentiation of both daughter cells(defined by expression of MyoD). C: muscle progenitors eventually differentiateand fuse into multinucleated muscle fibers filled with contractile sarcomeres. Theformation of sarcomeres is most prevalent on tissue-like stiffness, whereas stiff-ness above or below that level leads to impaired sarcomerogenesis. D: immaturecardiomyocytes are derived along a series of steps from embryonic stem cells inthe soft embryo. As the heart matures, it becomes more stiff, and the cardiomyo-cytes become more aligned and contractile. However, in a stiffer fibrotic heart,contraction of cardiomyocytes is impaired. Alternatively, iPSCs can be induced toform cardiomyocytes on plastic; however, they are not able to produce the con-tractile force of a healthy mature cardiomyocyte.

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org 19

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

disassemble the contractile machinery of the sar-comere that allows them to contract and producemechanical work on stiff gels (59) (FIGURE 2D).However, when the gel substrate becomes too stiff,mimicking more fibrotic post-infarct scars, sar-comeres are disrupted, and beating is inhibited.In addition to sarcomeric organization and con-tractility being downstream of matrix stiffness,the transcriptional mechanosensitivity of hearthas also been shown across many disease statesto have collagen I expression—a gauge for ECMstiffness— coupled to lamin A/C expression inthe nucleus (15). Although the precise mecha-nisms remain to be elucidated, on stiff matrices,YAP/TAZ shuttles into the nucleus (22, 34),whereas NKX-2.5, a repressor of smooth muscleactin (ACTA2), shuttles out of the nucleus (19).These systems highlight how stem cells and pro-genitors rely on mechanical cues from the extra-cellular matrix to modulate the transcriptionalprogram used to drive differentiation and matu-ration of cells.

Osmotic Regulation of MatrixMechanosensing

Soft matrix suppresses myosin II activity and cyto-skeletal assembly, but hypotonic media causessimilar cell rounding with disruption of the actincytoskeleton and changes in ion channel activity(24, 61). Changes in osmotic pressure well below orabove isotonic pressure cause proportional volumechanges of cells and nuclei, with hypotonic con-ditions causing chromatin decondensation andhypertonic conditions resulting in chromatincondensation (46). Physiological differences inosmotic pressure are not evident for tissues thatotherwise vary in stiffness by orders of magni-tude (FIGURE 1A), but regulation of MSC differ-entiation on soft and stiff gels by osmoticpressure (36) is nonetheless understandable,based at least on the long-established effects ofosmotic pressure on the cytoskeleton (61). In-deed, given the higher water content in soft tis-sues and the lower water content in stiff tissues(FIGURE 1B), the pleiotropic effects of osmoticpressure changes and hydration motivate furtherstem-cell differentiation experiments that com-bine osmotic-hydration changes with drugs thattarget specific molecules of the cytoskeleton. Itwill be interesting to clarify, for example,whether osteogenesis on soft matrix that can beincreased by ~50% using hypertonic media (36)(which dehydrates cells and drives actomyosinassembly) is suppressed or not by simultaneousinhibition of myosin II.

Stress or Strain Relaxation andMatrix Heterogeneity

Many stem cells express ECM components in amechanosensitive manner. Using void-forming al-ginate hydrogels, murine MSCs express the mostcollagen with an (osteoid-like) alginate stiffness of20–60 kPa (44). Cells also degrade ECM using matrixmetalloproteinases (MMPs). Primary mouse fibro-blasts transduced with the four reprogramming fac-tors (“Yamanaka” factors) and cultured in 3Dhydrogels of different initial stiffness, MMP-degrad-ability, and adhesive ligand illustrate the importanceof matrix malleability (9). The reprogramming effi-ciency to iPSCs proved highest with highly de-gradable gels of initial stiffness of 0.6 kPa; such astiffness is similar to that of embryos, which arealso likely to yield and flow under stress similarto brain tissue (59).

Cell forces can make ECM flow as occurs mac-roscopically with soft brain tissue under strain, butECM in some tissues can also fully recover itsshape and exhibit elasticity, as is a requirement ofcardiac tissue (59). Not all tissues flow under cell-generated stresses or strains (or else our tissueswould have no shape when isolated), but the abil-ity to confer gels with relaxation properties hasrecently been accomplished with relaxation timescales that are tunable from 1 to 40 min, whichaffects differentiation (12). MSCs in 9-kPa gels un-dergo more adipogenesis with slow-relaxing gelscompared with fast-relaxing gels, whereas MSCs in17-kPa gels exhibit maximal osteogenesis with fast-relaxing gels (98). Importantly, MSCs pull on thefast-relaxing gel and surround themselves with moreECM and more ligand. As a general rule in polymerphysics, a high concentration of polymer will have ahigher stiffness. Synthetic fiber matrices with finecontrol of individual fiber stiffness, size, and cross-linking have also been made, although sparsemeshworks limit the presentation of ligand forcell adhesion (2). MSCs spread more on soft-fibermeshes compared with stiff fiber meshes, butonce again the soft meshes condense and likelystiffen beneath the cell (FIGURE 3A). Both ofthese examples demonstrate the importance oftaking care to distinguish the initial bulk me-chanics from the final heterogeneous mechanics(and ligand) of a microenvironment.

Some tissues (such as liver) might have relativelyisotropic mechanics at the level of the cell, butmany musculoskeletal tissues (such as tendon) ex-perience forces more in one direction and exhibitanisotropic ECM with aligned collagen fibers. Un-derstandably, the orientation of ECM stiffness canalso impact stem-cell differentiation. Collagen fi-ber nano-films on mica exhibit very long-range

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org20

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

fiber orientation, and MSCs extend along such fi-bers in a more myogenic-like morphology; how-ever, when the fibers are cross-linked, the MSCsshow no orientation preference and undergo os-teogenesis on the stiffer films (47). Skeletal muscleprogenitors (MuSCs) also exhibit enhanced differ-entiation and maturation on stiff fiber substratescompared with the typical isotropic gels orculture plastic, even generating functional neu-romuscular junctions in cocultures with mo-toneurons (38). Mechanical anisotropy of ECM canaffect morphology, maturation, and/or differentia-tion, similar to experimentally controlled cellshapes, by patterning ECM ligand on a substrate;with cardiomyocytes, for example, rod-shapepatterns produced more forceful contractionsrelative to circular patterns (7). Printing a rod-shaped pattern of ligand on gels with physiolog-ical stiffness understandably maximized thematuration of iPSC-derived cardiomyocytes,which exhibited greater mechanical work andcalcium handling as well as more physiologicalintracellular structure (77).

Mechanical anisotropy in 3D has also been in-vestigated with gel overlays or “sandwich” gels thatpresent distinct mechanical properties above andbelow the cell. Some niches in tissues are similarlystratified: MSCs undergo osteogenesis on top ofbone with soft marrow above, and MuSCs typi-cally have semi-stiff muscle below and semi-stiffbasement membrane above. Bio-compatible gelsfor overlays include HA that can be syntheticallymodified to control elasticity on either side ofsandwiched cells (FIGURE 3B). Bonding betweenthe gels should be minimized to allow cells tofreely change morphology at the gel-gel inter-face. MSC morphology indeed responds to thestiffer gel, regardless of being above or below thecells (75); the stiffest medium thus dominatescell responses. Hepatic stellate cells (HSCs) arealso mechanosensitive, maintaining quiescenceon soft substrates and activating toward a myo-fibroblast-like state on stiff substrates, but HSCson a soft gel plus are also activated by a rigidoverlay (69).

Cells are only able to mechanically sense ap-proximately one cell length away on gels such asPA gels, but within a purely fibrous network, cells,including MSCs, are able to increase their sensingradius 10-fold (87). By mixing soluble collagen Iinto a soft PA hydrogel, collagen-I fibers bundleand segregate into a heterogeneous distribution ofbranched fibers that mimic fibrotic scar tissue (19).The heterogeneously soft- and stiff-gel regions arethen uniformly functionalized with collagen for ad-hesion, and cultures of MSCs respond as if culturedon a uniformly stiff hydrogel rather than a softhydrogel in terms of multiple mechanosensitive

markers, including �-SMA (FIGURE 3C), which isan important cellular maker of contractility as wellas scarring. �-SMA is often observed in fibrotictissues and even precedes the overexpression ofcollagen in chemically induced models of liver fi-brosis (69). The stiffest components of a heteroge-nous ECM thus dominate mechanosensitivedifferentiation.

Fibrosis, the Pathologic Matrix,and Differentiation

Fibrosis is the pathologic accumulation of ECMwithin a tissue and is generally associated withaberrant wound-healing mechanisms that result inscarring. Dysregulated repair can initiate feed-for-ward systems that result in progressive fibrotic tis-sue accumulation that undermines stem cell-basedrepair (37). Given the association of fibrosis with awide array of tissue injuries, it contributes to nearlyhalf of all deaths in the developed world (95). Fi-brosis reflects an imbalance between synthesis and

FIGURE 3. ECM malleability and heterogeneityA: MSCs placed in soft fibrous networks are able to pull thefibers and pack them near the cell, creating heterogeneity withhigh stiffness and ligand presentation local to the cell. How-ever, in fibrous networks with stiff fibers, the cell is unable todeform the network and change the local stiffness. B: simplesandwich gels demonstrate that stiff regions dominate matrixmechanosensing. When a soft gel is placed on top of cells oneither soft or stiff lower substrates, the cells maintain theirrounded or spread morphology, respectively. However, whenrounded cells are on a soft substrate or overlaid with a stiffgel, they actively spread to adopt the morphology expectedfrom a stiff substrate. C: gels can also be made with heteroge-nous underlying collagen fibers that also introduce heteroge-nous stiffness within the 2D gel. Reminiscent of a fibroticenvironment, these gels are termed “scar in a dish.” AlthoughMSCs on a soft or stiff substrate adopt a rounded or spreadmorphology, respectively, the mechanical heterogeneity cre-ates a near homogenous population of MSCs with a spreadmorphology. This spread morphology includes robust �SMA,which is a hallmark of fibrotic cells.

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org 21

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

degradation of ECM, both of which can be influ-enced by mechanical stress (FIGURE 4). Since ECMis the major determinant of tissue mechanicalproperties, the disrupted ECM architecture in fi-brosis has mechanical consequences. For example,healthy lung tissue with an elasticity of 2 kPa stiff-ens to �15 kPa in idiopathic pulmonary fibrosis(6). In vivo elastography has become a reliable waywith ultrasound to assess such trends in situ; forexample, healthy liver has an elasticity of �5 kPa,and cirrhotic liver has an elasticity of �12 kPa (18).Similar increases in tissue stiffness and collagencontent are also observed in fibrotic striated mus-cle (81) and kidney (51). Many studies show alteredmechanosensing in fibrotic ECM is responsible forboth the impaired ability of healthy cells to regen-erate functioning tissue as well as promoting fibro-blast activation—a type of differentiation—tomyofibroblasts that secrete even more ECM (10).

Mechanical properties likely stimulate the acti-vation of myofibroblasts to produce ECM in skin(40), heart (29), lung (42), liver (69), and kidney(57). Mechanosensing of matrix seems critical tothe positive feedback in fibro-proliferative disor-ders (92). Unfortunately, myofibroblasts do noteasily revert to their previous phenotype upon ma-trix softening (3); reprogramming is feasible undersome conditions (30, 60), but soft and malleablematrix remains best for reprogramming any cell toan iPSC, as cited above. The increased contractilityof myofibroblasts and other resident cells in the

fibrotic tissue can mechanically impact nearbycells. In the liver, activation of the HSCs aroundsinusoidal endothelial cells restricts blood flow inthe damaged region (82). Given the known role ofhypoxia signaling in collagen expression, this fur-ther propagates fibrosis (58). Macrophages fromthe circulation or already resident in tissue are alsorecognized as mediators of fibrosis in many tissuesby secretion of pro-inflammatory signals andwound-healing signals. Macrophages are mecha-nosensitive, but whether stiff matrices cause themto differentiate into pro-inflammatory (73) or anti-inflammatory (27) phenotypes remains unclear.

Although many cell types with differentiationpotential are mechanosensitive in culture, the spe-cific cell type responsible for mechanosensing thefibrotic environment in vivo appears tissue-depen-dent. The resident fibroblast population in idio-pathic pulmonary fibrosis senses stiffness andbecomes activated into a myofibroblast phenotypein response to stiffness (63), whereas in liver theHSCs are primarily responsible for the fibrotic re-sponse and require a stiff environment for activa-tion (69). In skeletal muscle, fibro-adipogenicprogenitors fulfill a similar role (48), although theirmatrix mechanosensing has not been investigated.Tissue-resident MSCs are a major contributor tomyofibroblast populations (52) and likely overlapwith the populations of hepatic stellate cells in liver(50) or fibro-adipogenic progenitors in muscle (85).In mice, when Gli1� resident MSCs are geneticallyablated, fibrosis is ameliorated in a range of tis-sues, and cardiac function is preserved (52).

The process of epithelial-mesenchymal transi-tion (EMT) involves the transdifferentiation offunctional epithelial cells into mesenchymal cells,which can be important during development andwound healing, but may also go awry in fibrosis(41, 53). Epithelial cells from lung (11) and kidney(54) have been shown to undergo EMT and prog-ress into fibrogenic myofibroblasts, but the role ofEMT in the liver has been called into question (91).EMT critically involves not only the recruitment ofadditional matrix-modifying cells, but also the lossof cells fulfilling the tissue’s primary function.TGF-� signaling has been described as the primarydriver of EMT; however, multiple reports haveshown that both TGF-� signaling and matrix rigid-ity are necessary to elicit EMT (66, 89). In additionto EMT, lineage-tracing experiments have alsodemonstrated the possibility of endothelial-to-mesenchymal transitions (EndoMT) in fibrosis ofthe heart, lung, and kidney (72); however, the roleof mechanosensing in (EndoMT) is less wellknown. Importantly, mechanosensing can also dis-rupt proper functioning of resident cells even with-out undergoing transdifferentiation processes.

FIGURE 4. Schematic of stem-cell mechanosensing in fibrosisFibrosis is the pathologic accumulation of ECM that is driven by myofibroblast inmany tissues. A host of resident stem cells are known to differentiate into myofi-broblast, which in many cases is known to be mechanosensitive. Myofibroblastssecrete matrix components that incorporate into the matrix to increase stiffness.Myofibroblasts are also highly contractile and stress the matrix, which leads toeven higher levels of stiffness and also limits the ability of MMPs to degrade thematrix. Mechanical stress on the matrix has also been demonstrated to releaseactive TGFB from the matrix, a critical soluble factor in the differentiation of myo-fibroblasts and other fibrotic programs. These processes demonstrate the criticalnature of ECM mechanics in the positive feedback loop that results in progressivefibrosis.

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org22

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

In addition to the mechanosensitive differentia-tion of many cell types, ECM itself is mechanosen-sitive, with fibrils under load being protected fromMMP-based degradation (25). This “use it or loseit” mechanism applied in a fibrotic context makesexisting scar tissues more difficult to remodel. Themechanical signal produced by myofibroblast con-traction is propagated further through alignmentof collagen fibrils between contractile elementsthat enable long-range force transmission (4, 87).In vitro experiments show collagen is compactedand well aligned between contractile groups ofcells (80), creating an ECM architecture reminis-cent of bridging fibrosis in the liver. This propaga-tion of strain through the fibrotic ECM can alsoimpact signaling by soluble factors. Myofibroblastcontraction on stiff matrices directly activates la-tent TGF-�, which further cranks the wheel of pos-itive feedback in fibrosis (94). This brings intofocus the interplay between soluble, cellular, andmechanical signals in fibrotic disease (93). Al-though soluble factors have long been a focus ofscientific inquiry in fibrosis, the impact of mecha-nosensing in fibrosis underscores the broader im-portance of matrix mechanosensing in cell andtissue differentiation.

Concluding Prospectus

Various types of stem and progenitor cells have ademonstrated ability to sense the stiffness of ECM.The morphology, cytoskeleton, and migration ofthese cells are all responsive within hours to ECMmechanics, whereas effects on proliferation and/ordifferentiation follow over days. Stem cells can alsoreorganize and sometimes synthesize their ECM,which creates a local niche and adds mechanicalheterogeneity. This is important because thestiffest ECM dominates the cell response. In fi-brosis, this can feed forward with stiffer ECMleading to increased fibrogenic states of cells andfurther stiffening. Various engineered gels sys-tems have clearly demonstrated matrix mecha-nosensitivity, but a myriad of factors in vivocontinue to make it challenging to isolate theprecise role of matrix mechanosensing in tissuedifferentiation. Although MSCs have been usefulfor documenting many aspects of mechanosen-sitive differentiation, further investigation ofhow iPSCs and other stem and progenitor cellsrespond to ECM mechanics—in combinationwith potent soluble factors—will likely provecritical to unlocking the therapeutic potential ofsuch cells. �

The authors were supported by the U.S. National Insti-tutes of Health National Cancer Institute under PSOCAward No. U54 CA-193417, by the National Heart Lungand Blood Institute under Award Nos. R01 HL-124106

and R21 HL-128187, by the National Institute of Arthritisand Musculoskeletal and Skin under Award No. K99AR-067867 (L.S.), and also by the U.S. National ScienceFoundation Materials Science and Engineering Center(MRSEC) and Science & Technology Center (CMMI-1548571) grants to Penn. Additional support was pro-vided by the U.S.-Israel Binational Science Foundation(BSF) and the Human Frontiers Sciences Program(HFSP). The content is solely the responsibility of theauthors and does not necessarily represent the officialviews of the National Institutes of Health or other grant-ing agencies.

No conflicts of interest, financial or otherwise, are de-clared by the author(s).

Author contributions: L.R.S., S.C., and D.E.D. interpretedresults of experiments; L.R.S. and D.E.D. prepared figures;L.R.S. and D.E.D. drafted manuscript; L.R.S., S.C., andD.E.D. edited and revised manuscript; L.R.S., S.C., andD.E.D. approved final version of manuscript.

References1. Amadò R, Werner G, Neukom H. Water content of human

articular cartilage and its determination by gas chromatog-raphy. Biochem Med 16: 169–172, 1976. doi:10.1016/0006-2944(76)90020-X.

2. Baker BM, Trappmann B, Wang WY, Sakar MS, Kim IL, ShenoyVB, Burdick JA, Chen CS. Cell-mediated fibre recruitmentdrives extracellular matrix mechanosensing in engineered fibril-lar microenvironments. Nat Mater 14: 1262–1268, 2015. doi:10.1038/nmat4444.

3. Balestrini JL, Chaudhry S, Sarrazy V, Koehler A, Hinz B. Themechanical memory of lung myofibroblasts. Integr Biol 4:410–421, 2012. doi:10.1039/c2ib00149g.

4. Bates JH, Davis GS, Majumdar A, Butnor KJ, Suki B. Linkingparenchymal disease progression to changes in lung mechan-ical function by percolation. Am J Respir Crit Care Med 176:617–623, 2007. doi:10.1164/rccm.200611-1739OC.

5. Ben-Ze’ev A, Farmer SR, Penman S. Protein synthesis re-quires cell-surface contact while nuclear events respond tocell shape in anchorage-dependent fibroblasts. Cell 21: 365–372, 1980. doi:10.1016/0092-8674(80)90473-0.

6. Booth AJ, Hadley R, Cornett AM, Dreffs AA, Matthes SA,Tsui JL, Weiss K, Horowitz JC, Fiore VF, Barker TH, MooreBB, Martinez FJ, Niklason LE, White ES. Acellular normal andfibrotic human lung matrices as a culture system for in vitroinvestigation. Am J Respir Crit Care Med 186: 866 – 876,2012. doi:10.1164/rccm.201204-0754OC.

7. Bray MA, Sheehy SP, Parker KK. Sarcomere alignment isregulated by myocyte shape. Cell Motil Cytoskeleton 65:641–651, 2008. doi:10.1002/cm.20290.

8. Burridge K, Guilluy C. Focal adhesions, stress fibers andmechanical tension. Exp Cell Res 343: 14–20, 2016. doi:10.1016/j.yexcr.2015.10.029.

8a. Buxboim A, Irianto J, Swift J, Athirasala A, Shin JW, RehfeldtF, Discher DE. Coordinated increase of nuclear tension andlamin-A with matrix stiffness outcompetes lamin-B receptorthat favors soft tissue phenotypes. Mol Biol Cell 28: 3333–3348, 2017. doi:10.1091/mbc.E17-06-0393.

9. Caiazzo M, Okawa Y, Ranga A, Piersigilli A, Tabata Y, LutolfMP. Defined three-dimensional microenvironments boost in-duction of pluripotency. Nat Mater 15: 344–352, 2016. doi:10.1038/nmat4536.

10. Carver W, Goldsmith EC. Regulation of tissue fibrosis by thebiomechanical environment. BioMed Res Int 2013: 101979,2013. doi:10.1155/2013/101979.

11. Chapman HA. Epithelial-mesenchymal interactions in pulmo-nary fibrosis. Annu Rev Physiol 73: 413–435, 2011. doi:10.1146/annurev-physiol-012110-142225.

12. Chaudhuri O, Gu L, Klumpers D, Darnell M, Bencherif SA,Weaver JC, Huebsch N, Lee HP, Lippens E, Duda GN,Mooney DJ. Hydrogels with tunable stress relaxation regu-late stem cell fate and activity. Nat Mater 15: 326–334, 2016.doi:10.1038/nmat4489.

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org 23

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

13. Chen Q, Shou P, Zheng C, Jiang M, Cao G, YangQ, Cao J, Xie N, Velletri T, Zhang X, Xu C, ZhangL, Yang H, Hou J, Wang Y, Shi Y. Fate decision ofmesenchymal stem cells: adipocytes or osteo-blasts? Cell Death Differ 23: 1128–1139, 2016.doi:10.1038/cdd.2015.168.

14. Cho GS, Lee DI, Tampakakis E, Murphy S, Ander-sen P, Uosaki H, Chelko S, Chakir K, Hong I, SeoK, Chen HV, Chen X, Basso C, Houser SR, Toma-selli GF, O’Rourke B, Judge DP, Kass DA, KwonC. Neonatal transplantation confers maturationof PSC-derived cardiomyocytes conducive tomodeling cardiomyopathy. Cell Reports 18: 571–582, 2017. doi:10.1016/j.celrep.2016.12.040.

15. Cho S, Irianto J, Discher DE. Mechanosensing bythe nucleus: from pathways to scaling relation-ships. J Cell Biol 216: 305–315, 2017. doi:10.1083/jcb.201610042.

16. Cooper ST, Maxwell AL, Kizana E, Ghoddusi M,Hardeman EC, Alexander IE, Allen DG, NorthKN. C2C12 co-culture on a fibroblast substra-tum enables sustained survival of contractile,highly differentiated myotubes with peripheralnuclei and adult fast myosin expression. CellMotil Cytoskeleton 58: 200 –211, 2004. doi:10.1002/cm.20010.

17. Crisp M, Liu Q, Roux K, Rattner JB, Shanahan C,Burke B, Stahl PD, Hodzic D. Coupling of thenucleus and cytoplasm: role of the LINC com-plex. J Cell Biol 172: 41–53, 2006. doi:10.1083/jcb.200509124.

18. Degos F, Perez P, Roche B, Mahmoudi A, Asse-lineau J, Voitot H, Bedossa P; FIBROSTIC studygroup. Diagnostic accuracy of FibroScan andcomparison to liver fibrosis biomarkers in chronicviral hepatitis: a multicenter prospective study(the FIBROSTIC study). J Hepatol 53: 1013–1021,2010. doi:10.1016/j.jhep.2010.05.035.

19. Dingal PC, Bradshaw AM, Cho S, Raab M, Bux-boim A, Swift J, Discher DE. Fractal heterogene-ity in minimal matrix models of scars modulatesstiff-niche stem-cell responses via nuclear exit ofa mechanorepressor. Nat Mater 14: 951–960,2015. doi:10.1038/nmat4350.

20. Discher DE, Janmey P, Wang YL. Tissue cells feeland respond to the stiffness of their substrate.Science 310: 1139–1143, 2005. doi:10.1126/science.1116995.

21. Discher DE, Mooney DJ, Zandstra PW. Growthfactors, matrices, and forces combine and con-trol stem cells. Science 324: 1673–1677, 2009.doi:10.1126/science.1171643.

22. Dupont S, Morsut L, Aragona M, Enzo E, Giulitti S,Cordenonsi M, Zanconato F, Le Digabel J, ForcatoM, Bicciato S, Elvassore N, Piccolo S. Role of YAP/TAZ in mechanotransduction. Nature 474: 179–183, 2011. doi:10.1038/nature10137.

23. Engler AJ, Sen S, Sweeney HL, Discher DE. Ma-trix elasticity directs stem cell lineage specifica-tion. Cell 126: 677–689, 2006. doi:10.1016/j.cell.2006.06.044.

24. Fernández P, Pullarkat PA. The role of the cyto-skeleton in volume regulation and beading tran-sitions in PC12 neurites. Biophys J 99: 3571–3579, 2010. doi:10.1016/j.bpj.2010.10.027.

25. Flynn BP, Bhole AP, Saeidi N, Liles M, Dimarzio CA,Ruberti JW. Mechanical strain stabilizes reconstitutedcollagen fibrils against enzymatic degradation bymammalian collagenase matrix metalloproteinase 8(MMP-8). PLoS One 5: e12337, 2010. doi:10.1371/journal.pone.0012337.

26. Forbes RM, Cooper AR, Mitchell HH. The com-position of the adult human body as determinedby chemical analysis. J Biol Chem 203: 359–366,1953.

27. Friedemann M, Kalbitzer L, Franz S, Moeller S,Schnabelrauch M, Simon JC, Pompe T, Franke K.Instructing Human Macrophage Polarization byStiffness and Glycosaminoglycan Functionaliza-tion in 3D Collagen Networks. Adv Healthc Ma-ter 6: 1600967, 2017. doi:10.1002/adhm.201600967.

28. Fu J, Wang YK, Yang MT, Desai RA, Yu X, Liu Z,Chen CS. Mechanical regulation of cell functionwith geometrically modulated elastomeric sub-strates. Nat Methods 7: 733–736, 2010. doi:10.1038/nmeth.1487.

29. Galie PA, Westfall MV, Stegemann JP. Reducedserum content and increased matrix stiffnesspromote the cardiac myofibroblast transition in3D collagen matrices. Cardiovasc Pathol 20: 325–333, 2011. doi:10.1016/j.carpath.2010.10.001.

30. Gaça MD, Zhou X, Issa R, Kiriella K, Iredale JP,Benyon RC. Basement membrane-like matrix in-hibits proliferation and collagen synthesis by ac-tivated rat hepatic stellate cells: evidence formatrix-dependent deactivation of stellate cells.Matrix Biol 22: 229–239, 2003. doi:10.1016/S0945-053X(03)00017-9.

31. Gefen A, Margulies SS. Are in vivo and in situbrain tissues mechanically similar? J Biomech 37:1339–1352, 2004. doi:10.1016/j.jbiomech.2003.12.032.

32. Geiger B, Spatz JP, Bershadsky AD. Environmen-tal sensing through focal adhesions. Nat Rev MolCell Biol 10: 21–33, 2009. doi:10.1038/nrm2593.

33. Gilbert PM, Havenstrite KL, Magnusson KE,Sacco A, Leonardi NA, Kraft P, Nguyen NK,Thrun S, Lutolf MP, Blau HM. Substrate elasticityregulates skeletal muscle stem cell self-renewalin culture. Science 329: 1078–1081, 2010. doi:10.1126/science.1191035.

34. Graham DM, Burridge K. Mechanotransductionand nuclear function. Curr Opin Cell Biol 40: 98–105, 2016. doi:10.1016/j.ceb.2016.03.006.

35. Gribova V, Crouzier T, Picart C. A material’s pointof view on recent developments of polymericbiomaterials: control of mechanical and biochem-ical properties. J Mater Chem 21: 14354–14366,2011. doi:10.1039/c1jm11372k.

36. Guo M, Pegoraro AF, Mao A, Zhou EH, Arany PR,Han Y, Burnette DT, Jensen MH, Kasza KE,Moore JR, Mackintosh FC, Fredberg JJ, MooneyDJ, Lippincott-Schwartz J, Weitz DA. Cell volumechange through water efflux impacts cell stiff-ness and stem cell fate. Proc Natl Acad Sci U S A114: E8618–E8627, 2017. doi:10.1073/pnas.1705179114.

37. Gurtner GC, Werner S, Barrandon Y, LongakerMT. Wound repair and regeneration. Nature 453:314–321, 2008. doi:10.1038/nature07039.

38. Happe CL, Tenerelli KP, Gromova AK, Kolb F,Engler AJ. Mechanically patterned neuromuscu-lar junctions-in-a-dish have improved functionalmaturation. Mol Biol Cell 28: 1950–1958, 2017.doi:10.1091/mbc.E17-01-0046.

39. Hinz B. The extracellular matrix and transforminggrowth factor-�1: Tale of a strained relationship.Matrix Biol 47: 54–65, 2015. doi:10.1016/j.matbio.2015.05.006.

40. Hinz B, Mastrangelo D, Iselin CE, Chaponnier C,Gabbiani G. Mechanical tension controls granula-tion tissue contractile activity and myofibroblastdifferentiation. Am J Pathol 159: 1009–1020, 2001.doi:10.1016/S0002-9440(10)61776-2.

41. Hinz B, Phan SH, Thannickal VJ, Galli A, Bochaton-Piallat ML, Gabbiani G. The myofibroblast: one func-tion, multiple origins. Am J Pathol 170: 1807–1816,2007. doi:10.2353/ajpath.2007.070112.

42. Huang X, Yang N, Fiore VF, Barker TH, Sun Y,Morris SW, Ding Q, Thannickal VJ, Zhou Y. Matrixstiffness-induced myofibroblast differentiation ismediated by intrinsic mechanotransduction. AmJ Respir Cell Mol Biol 47: 340–348, 2012. doi:10.1165/rcmb.2012-0050OC.

43. Huebsch N, Arany PR, Mao AS, Shvartsman D, AliOA, Bencherif SA, Rivera-Feliciano J, Mooney DJ.Harnessing traction-mediated manipulation ofthe cell/matrix interface to control stem-cell fate.Nat Mater 9: 518–526, 2010. doi:10.1038/nmat2732.

44. Huebsch N, Lippens E, Lee K, Mehta M, KoshyST, Darnell MC, Desai RM, Madl CM, Xu M, ZhaoX, Chaudhuri O, Verbeke C, Kim WS, Alim K,Mammoto A, Ingber DE, Duda GN, Mooney DJ.Matrix elasticity of void-forming hydrogels con-trols transplanted-stem-cell-mediated bone for-mation. Nat Mater 14: 1269–1277, 2015. doi:10.1038/nmat4407.

45. Humphrey JD, Dufresne ER, Schwartz MA.Mechanotransduction and extracellular matrixhomeostasis. Nat Rev Mol Cell Biol 15: 802–812,2014. doi:10.1038/nrm3896.

46. Irianto J, Swift J, Martins RP, McPhail GD, KnightMM, Discher DE, Lee DA. Osmotic challengedrives rapid and reversible chromatin condensa-tion in chondrocytes. Biophys J 104: 759–769,2013. doi:10.1016/j.bpj.2013.01.006.

47. Ivanovska IL, Swift J, Spinler K, Dingal D, Cho S,Discher DE. Cross-linked matrix rigidity and sol-uble retinoids synergize in nuclear lamina regu-lation of stem cell differentiation. Mol Biol Cell28: 2010–2022, 2017. doi:10.1091/mbc.E17-01-0010.

48. Joe AW, Yi L, Natarajan A, Le Grand F, So L,Wang J, Rudnicki MA, Rossi FM. Muscle injuryactivates resident fibro/adipogenic progenitorsthat facilitate myogenesis. Nat Cell Biol 12: 153–163, 2010. doi:10.1038/ncb2015.

49. Jorba I, Uriarte JJ, Campillo N, Farré R, NavajasD. Probing micromechanical properties of theextracellular matrix of soft tissues by atomicforce microscopy. J Cell Physiol 232: 19–26,2017. doi:10.1002/jcp.25420.

50. Kordes C, Sawitza I, Götze S, Herebian D,Häussinger D. Stellate cells are mesenchymalstem cells. Eur J Med Res 19, Suppl 1: S6, 2014.doi:10.1186/2047-783X-19-S1-S6.

51. Korsmo MJ, Ebrahimi B, Eirin A, Woollard JR,Krier JD, Crane JA, Warner L, Glaser K, Grimm R,Ehman RL, Lerman LO. Magnetic resonance elas-tography noninvasively detects in vivo renalmedullary fibrosis secondary to swine renal ar-tery stenosis. Invest Radiol 48: 61–68, 2013. doi:10.1097/RLI.0b013e31827a4990.

52. Kramann R, Schneider RK, DiRocco DP, MachadoF, Fleig S, Bondzie PA, Henderson JM, Ebert BL,Humphreys BD. Perivascular Gli1� progenitorsare key contributors to injury-induced organ fi-brosis. Cell Stem Cell 16: 51–66, 2015. doi:10.1016/j.stem.2014.11.004.

53. Lamouille S, Xu J, Derynck R. Molecular mecha-nisms of epithelial-mesenchymal transition. NatRev Mol Cell Biol 15: 178–196, 2014. doi:10.1038/nrm3758.

54. LeBleu VS, Taduri G, O’Connell J, Teng Y, CookeVG, Woda C, Sugimoto H, Kalluri R. Origin andfunction of myofibroblasts in kidney fibrosis. NatMed 19: 1047–1053, 2013. doi:10.1038/nm.3218.

55. Lees S. Considerations regarding the structure ofthe mammalian mineralized osteoid from view-point of the generalized packing model. ConnectTissue Res 16: 281–303, 1987. doi:10.3109/03008208709005616.

56. Lees S, Escoubes M. Vapor pressure isotherms, com-position and density of hyperdense bones of horse,whale and porpoise. Connect Tissue Res 16: 305–322,1987. doi:10.3109/03008208709005617.

57. Leight JL, Wozniak MA, Chen S, Lynch ML, ChenCS. Matrix rigidity regulates a switch betweenTGF-�1-induced apoptosis and epithelial-mesen-chymal transition. Mol Biol Cell 23: 781–791,2012. doi:10.1091/mbc.E11-06-0537.

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org24

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.

58. Lokmic Z, Musyoka J, Hewitson TD, Darby IA.Hypoxia and hypoxia signaling in tissue repairand fibrosis. Int Rev Cell Mol Biol 296: 139–185,2012. doi:10.1016/B978-0-12-394307-1.00003-5.

59. Majkut S, Idema T, Swift J, Krieger C, Liu A,Discher DE. Heart-specific stiffening in early em-bryos parallels matrix and myosin expression tooptimize beating. Curr Biol 23: 2434–2439, 2013.doi:10.1016/j.cub.2013.10.057.

60. Marinkovic A, Liu F, Tschumperlin DJ. Matrices ofphysiologic stiffness potently inactivate idio-pathic pulmonary fibrosis fibroblasts. Am J Re-spir Cell Mol Biol 48: 422–430, 2013. doi:10.1165/rcmb.2012-0335OC.

61. Mills JW, Schwiebert EM, Stanton BA. Evidencefor the role of actin filaments in regulating cellswelling. J Exp Zool 268: 111–120, 1994. doi:10.1002/jez.1402680207.

62. Monge C, DiStasio N, Rossi T, Sébastien M, SakaiH, Kalman B, Boudou T, Tajbakhsh S, Marty I,Bigot A, Mouly V, Picart C. Quiescence of humanmuscle stem cells is favored by culture on naturalbiopolymeric films. Stem Cell Res Ther 8: 104,2017. doi:10.1186/s13287-017-0556-8.

63. Moore MW, Herzog EL. Regulation and rele-vance of myofibroblast responses in idiopathicpulmonary fibrosis. Curr Pathobiol Rep 1: 199–208, 2013. doi:10.1007/s40139-013-0017-8.

64. Motte S, Kaufman LJ. Strain stiffening in collagenI networks. Biopolymers 99: 35–46, 2013. doi:10.1002/bip.22133.

65. Méndez-Ferrer S, Michurina TV, Ferraro F, Ma-zloom AR, Macarthur BD, Lira SA, Scadden DT,Ma’ayan A, Enikolopov GN, Frenette PS. Mesen-chymal and haematopoietic stem cells form aunique bone marrow niche. Nature 466: 829–834, 2010. doi:10.1038/nature09262.

66. O’Connor JW, Gomez EW. Biomechanics ofTGF�-induced epithelial-mesenchymal transition:implications for fibrosis and cancer. Clin TranslMed 3: 23, 2014. doi:10.1186/2001-1326-3-23.

67. Oftadeh R, Perez-Viloria M, Villa-Camacho JC,Vaziri A, Nazarian A. Biomechanics and mecha-nobiology of trabecular bone: a review. J Bio-mech Eng 137: 010802115829–010802, 2015.doi:10.1115/1.4029176.

68. Olins AL, Rhodes G, Welch DB, Zwerger M, OlinsDE. Lamin B receptor: multi-tasking at the nu-clear envelope. Nucleus 1: 53–70, 2010. doi:10.4161/nucl.1.1.10515.

69. Olsen AL, Bloomer SA, Chan EP, Gaça MD,Georges PC, Sackey B, Uemura M, Janmey PA,Wells RG. Hepatic stellate cells require a stiffenvironment for myofibroblastic differentia-tion. Am J Physiol Gastrointest Liver Physiol301: G110 –G118, 2011. doi:10.1152/ajpgi.00412.2010.

70. Pajerowski JD, Dahl KN, Zhong FL, Sammak PJ,Discher DE. Physical plasticity of the nucleus instem cell differentiation. Proc Natl Acad Sci USA104: 15619–15624, 2007. doi:10.1073/pnas.0702576104.

71. Pelham RJ, Jr, Wang Y. Cell locomotion and focaladhesions are regulated by substrate flexibility.Proc Natl Acad Sci USA 94: 13661–13665, 1997.doi:10.1073/pnas.94.25.13661.

72. Piera-Velazquez S, Li Z, Jimenez SA. Role of en-dothelial-mesenchymal transition (EndoMT) inthe pathogenesis of fibrotic disorders. Am JPathol 179: 1074–1080, 2011. doi:10.1016/j.ajpath.2011.06.001.

73. Previtera ML, Sengupta A. Substrate stiffnessregulates proinflammatory mediator productionthrough TLR4 activity in macrophages. PLoS One10: e0145813, 2015. doi:10.1371/journal.pone.0145813.

74. Quarta M, Brett JO, DiMarco R, De Morree A,Boutet SC, Chacon R, Gibbons MC, Garcia VA, SuJ, Shrager JB, Heilshorn S, Rando TA. An artificialniche preserves the quiescence of muscle stemcells and enhances their therapeutic efficacy. NatBiotechnol 34: 752–759, 2016. doi:10.1038/nbt.3576.

75. Rehfeldt F, Brown AE, Raab M, Cai S, Zajac AL,Zemel A, Discher DE. Hyaluronic acid matricesshow matrix stiffness in 2D and 3D dictates cy-toskeletal order and myosin-II phosphorylationwithin stem cells. Integr Biol 4: 422–430, 2012.doi:10.1039/c2ib00150k.

76. Rehfeldt F, Schmidt CF. Physical probing of cells.J Phys D Appl Phys 50: 463001, 2017. doi:10.1088/1361-6463/aa8aa6.

77. Ribeiro AJ, Ang YS, Fu JD, Rivas RN, MohamedTM, Higgs GC, Srivastava D, Pruitt BL. Contrac-tility of single cardiomyocytes differentiatedfrom pluripotent stem cells depends on physio-logical shape and substrate stiffness. Proc NatlAcad Sci USA 112: 12705–12710, 2015. doi:10.1073/pnas.1508073112.

78. Robertson C, Tran DD, George SC. Concise re-view: maturation phases of human pluripotentstem cell-derived cardiomyocytes. Stem Cells 31:829–837, 2013. doi:10.1002/stem.1331.

79. Sharma A, Licup AJ, Jansen KA, Rens R, Shei-nman M, Koenderink GH, MacKintosh FC. Strain-controlled criticality governs the nonlinearmechanics of fibre networks. Nat Phys 12: 584–587, 2016. doi:10.1038/nphys3628.

80. Shi Q, Ghosh RP, Engelke H, Rycroft CH,Cassereau L, Sethian JA, Weaver VM, LiphardtJT. Rapid disorganization of mechanically inter-acting systems of mammary acini. Proc Natl AcadSci USA 111: 658–663, 2014. doi:10.1073/pnas.1311312110.

81. Smith LR, Lee KS, Ward SR, Chambers HG,Lieber RL. Hamstring contractures in childrenwith spastic cerebral palsy result from a stifferECM and increased in vivo sarcomere length. JPhysiol 589: 2625–2639, 2011. doi:10.1113/jphysiol.2010.203364.

82. Soon RK, Jr, Yee HF Jr. Stellate cell contraction:role, regulation, and potential therapeutic target.Clin Liver Dis 12: 791–803, 2008. doi:10.1016/j.cld.2008.07.004.

83. Sun Z, Guo SS, Fässler R. Integrin-mediatedmechanotransduction. J Cell Biol 215: 445–456,2016. doi:10.1083/jcb.201609037.

84. Swift J, Ivanovska IL, Buxboim A, Harada T, Din-gal PC, Pinter J, Pajerowski JD, Spinler KR, ShinJW, Tewari M, Rehfeldt F, Speicher DW, DischerDE. Nuclear lamin-A scales with tissue stiffnessand enhances matrix-directed differentiation.Science 341: 1240104, 2013. doi:10.1126/science.1240104.

85. Uezumi A, Ikemoto-Uezumi M, Tsuchida K. Rolesof nonmyogenic mesenchymal progenitors inpathogenesis and regeneration of skeletal mus-cle. Front Physiol 5: 68, 2014. doi:10.3389/fphys.2014.00068.

86. Ullah I, Subbarao RB, Rho GJ. Human mesenchy-mal stem cells - current trends and future pro-spective. Biosci Rep 35: e00191, 2015. doi:10.1042/BSR20150025.

87. Wang H, Abhilash AS, Chen CS, Wells RG, She-noy VB. Long-range force transmission in fibrousmatrices enabled by tension-driven alignment offibers. Biophys J 107: 2592–2603, 2014. doi:10.1016/j.bpj.2014.09.044.

88. Wang Y-L, Discher DE. Cell Mechanics, Volume83 (Methods in Cell Biology). Cambridge, MA:Academic Press, 2007.

89. Wei SC, Fattet L, Tsai JH, Guo Y, Pai VH, MajeskiHE, Chen AC, Sah RL, Taylor SS, Engler AJ, YangJ. Matrix stiffness drives epithelial-mesenchymaltransition and tumour metastasis through aTWIST1-G3BP2 mechanotransduction pathway.Nat Cell Biol 17: 678–688, 2015. doi:10.1038/ncb3157.

90. Weiss P, Garber B. Shape and movement of mes-enchyme cells as functions of the physical struc-ture of the medium: contributions to aquantitative morphology. Proc Natl Acad SciUSA 38: 264–280, 1952. doi:10.1073/pnas.38.3.264.

91. Wells RG. The epithelial-to-mesenchymal transi-tion in liver fibrosis: here today, gone tomorrow?Hepatology 51: 737–740, 2010.

92. Wells RG. Tissue mechanics and fibrosis. BiochimBiophys Acta 1832: 884–890, 2013. doi:10.1016/j.bbadis.2013.02.007.

93. Wells RG, Discher DE. Matrix elasticity, cytoskel-etal tension, and TGF-beta: the insoluble andsoluble meet. Sci Signal 1: pe13, 2008. doi:10.1126/stke.110pe13.

94. Wipff PJ, Rifkin DB, Meister JJ, Hinz B. Myofibro-blast contraction activates latent TGF-beta1 fromthe extracellular matrix. J Cell Biol 179: 1311–1323, 2007. doi:10.1083/jcb.200704042.

95. Wynn TA. Common and unique mechanisms reg-ulate fibrosis in various fibroproliferative dis-eases. J Clin Invest 117: 524–529, 2007. doi:10.1172/JCI31487.

96. Yennek S, Burute M, Théry M, Tajbakhsh S. Celladhesion geometry regulates non-random DNAsegregation and asymmetric cell fates in mouseskeletal muscle stem cells. Cell Reports 7: 961–970, 2014. doi:10.1016/j.celrep.2014.04.016.

97. Yoshikawa Y, Yasuike T, Yagi A, Yamada T. Trans-verse elasticity of myofibrils of rabbit skeletalmuscle studied by atomic force microscopy.Biochem Biophys Res Commun 256: 13–19, 1999.doi:10.1006/bbrc.1999.0279.

98. Zemel A, Rehfeldt F, Brown AE, Discher DE, Sa-fran SA. Optimal matrix rigidity for stress fiberpolarization in stem cells. Nat Phys 6: 468–473,2010. doi:10.1038/nphys1613.

REVIEW

PHYSIOLOGY • Volume 33 • January 2018 • www.physiologyonline.org 25

Downloaded from www.physiology.org/journal/physiologyonline by Sean Boyer (096.068.246.177) on December 14, 2017.Copyright © 2018 American Physiological Society. All rights reserved.