Intrinsic Mechanical Cues and Their Impact on Stem Cells and Embryogenesis

Abstract

Although understanding how soluble cues direct cellular processes revolutionised the study of cell biology in the second half of the 20th century, over the last two decades, new insights into how mechanical cues similarly impact cell fate decisions has gained momentum. During development, extrinsic cues such as fluid flow, shear stress and compressive forces are essential for normal embryogenesis to proceed. Indeed, both adult and embryonic stem cells can respond to applied forces, but they can also detect intrinsic mechanical cues from their surrounding environment, such as the stiffness of the extracellular matrix, which impacts differentiation and morphogenesis. Cells can detect changes in their mechanical environment using cell surface receptors such as integrins and focal adhesions. Moreover, dynamic rearrangements of the cytoskeleton have been identified as a key means by which forces are transmitted from the extracellular matrix to the cell and vice versa. Although we have some understanding of the downstream mechanisms whereby mechanical cues are translated into changes in cell behaviour, many of the signalling pathways remain to be defined. This review discusses the importance of intrinsic mechanical cues on adult cell fate decisions, the emerging roles of cell surface mechano-sensors and the cytoskeleton in enabling cells to sense its microenvironment, and the role of intracellular signalling in translating mechanical cues into transcriptional outputs. In addition, the contribution of mechanical cues to fundamental processes during embryogenesis such as apical constriction and convergent extension is discussed. The continued development of tools to measure the biomechanical properties of soft tissues in vivo is likely to uncover currently underestimated contributions of these cues to adult stem cell fate decisions and embryogenesis, and may inform on regenerative strategies for tissue repair.

Keywords: development; embryogenesis; mechanotransduction; stem cell; stiffness.

FIGURE 1

Extrinsic and intrinsic cues in mechanotransduction. (A) Cell differentiation has been shown to be affected by mechanical forces external to the cell (extrinsic) such as shear stress from fluid flow and more local mechanical cues (intrinsic) such as cell density, shape and elasticity of the surrounding extracellular matrix (ECM). (B) As a general concept, mechano-transduction involves the transfer of mechanical cues from the cell surface to the nucleus via the cytoskeleton. This activates downstream cell signalling cascades, which can influence cell fate decisions. In addition, a transcriptional feedback loop allows cells to maintain a cytoskeletal equilibrium that is responsive to changes in their mechano-environment. This is particularly important for processes like cell migration, in which continual cytoskeletal remodelling is required for persistent cell motility.

FIGURE 2

Summary of cell responses to ECM elasticity, topography and micropillars. (A) Fluorescence microscopy images show typical cell response on soft (0.5 kPa) and stiff (40 kPa) substrates. In general, cells (in this case embryonic neural crest cells) on soft (0.5 kPa) substrates remain rounded, whilst those on stiff (40 kPa) ECM spread and have organised F-actin fibres, as seen by the phalloidin staining (PHAL, green). (B) Typically, MSC cultured on long flexible micropillars respond similarly as they would on soft ECM and have a rounded morphology, whilst those on short inflexible micropillars behave as they would on stiff ECM and spread. (C) In general, MSC cultured on wider microgrooves show enhanced adipogenesis, whilst those on stiff substrates have an elongated morphology which promotes osteogenesis (Fu et al., 2010; Nikkhah et al., 2012; Abagnale et al., 2015). Scale bar 100 μm.

FIGURE 3

Schematic representation of mechanotransduction pathways. Mechanical stimuli are perceived by mechano-sensors at the cellular-ECM interface, such as integrin-FA complexes, GPCR, AJ and stretch-activated ion channels. This activates several cellular signalling pathways involving kinases or transcription factors (MAPK, ERK, JNK, PKC, AP-1), as well as Rho small GTPases (RhoA). RhoA-GTP regulates actin structure by (1) activating mDia to promote actin polymerisation (2) activating ROCK, which enhances actin contractility by activating NMM II phosphorylation, and (3) preventing actin disassembly by inhibiting the actin-severing protein COF. The remodelling of F-actin and increased cytoskeletal tension also regulates YAP/TAZ, which translocate to the nucleus in response to mechanical strain. At AJ, cadherin-actomyosin connections form via α-cat and ß-cat. An increase in tension at cell-cell contacts induces unfolding of α-cat, which promotes recruitment of AJ-stabilisation proteins such as vinculin. In response to a loss of cell-cell adhesion or mechanical stimulation, ß-cat can translocate to the nucleus, to activate mechanosensitive genes. Nuclear mechano-transduction occurs via the LINC complex, which directly couples the nuclear envelope to the cytoskeleton. NES 1/2 form a link between actin and SUN 1/2 proteins in the perinuclear space, which interact with the nuclear lamina via EM and lamin A. Nesprin proteins also connect the nuclear lamina with intermediate filaments and microtubules (not depicted here). JNK, c-Jun N-terminal kinase; PKC, protein kinase C; AP-1, activator protein 1; FAC, focal adhesion complex; GCPR, G-protein coupled receptor; IC, ion channel; ECM, extracellular matrix; AJ, adherens junction; α-cat, alpha-catenin; ß-cat, beta-catenin; YAP, yes associated protein; TAZ, WW domain-containing transcription regulator protein 1 NES 1/2, nesprin-1/2; SUN 1/2, sun-domain containing protein 1/2; EM, emerin; AP-1, activator protein 1; ERK, extracellular-receptor kinase; ROCK, rho-associated protein kinase; RhoA, ras homolog family member A; COF, cofillin; NMM II, non-muscle myosin II. Created using BioRender.com.

FIGURE 4

Schematic representation of focal adhesion kinase signalling. FA signalling: FAK is recruited to integrin clusters at the cell-ECM boundary in response to changes to ECM stiffness, or other physical cues. This initiates formation of the FA complex by recruitment of various proteins such as TLN and VCL and CAS, which transduce mechanical stimuli from the ECM to the cellular cytoskeleton. VASP, Zyx and α-actinin directly regulate actin assembly. Three general FA layers are depicted, including the integrin signalling layer, force transduction layer and actin regulatory layer. FA, focal adhesion; ECM, extracellular matrix; ITα;ITß, integrin subunit α and ß; FAK, focal adhesion kinase; TLN, talin; VCL, vinculin; Zyz, zyxin; NMM II, non-muscle myosin II; VASP, vasodilator-stimulated phosphoprotein.

FIGURE 5

YAP/TAZ mechanism of action. Schematic showing mechanical regulation of YAP/TAZ activity and modulation of cell behaviour by YAP/TAZ in MSC. Osteogenesis and skeletal muscle fates are promoted by stiff ECM and a low cell density, allowing MSC to spread and generate cytoskeletal tension via F-actin stress fibres. The stiff matrix promotes stress fibre formation and YAP/TAZ nuclear translocation. Conversely, adipogenic fates are promoted by soft ECM and high cell-cell contact. The soft matrix prevents stress fibre formation, thus MSC cannot generate tension and display only cortical actin. As such, YAP/TAZ are retained in the cytoplasm, undergo proteasomal degradation and are rendered inactive, promoting adipogenesis. Created using Biorender.com.

FIGURE 6

Hippo-dependent and Hippo-independent regulation of YAP/TAZ. YAP/TAZ are known to be regulated via the HIPPO signalling pathway and by a mechanically regulated HIPPO-independent mechanism. (Left) HIPPO control of YAP and TAZ. The HIPPO pathway regulates organ growth as well as cell proliferation, migration and differentiation. In tightly packed tissues, proliferation is regulated by contact inhibition via the HIPPO pathway. Tight junctions and adherens junctions between cells interact with and activate MST1/2, which recruit SAV1, and subsequently phosphorylate LATS1/2. This phosphorylation is facilitated by the scaffold proteins MOB1 A/B and NF2. In turn, LATS1/2 phosphorylate YAP/TAZ, leading to cytoplasmic sequestering of these proteins, and their eventual ubiquitination/degradation. F-actin has been proposed to regulate YAP/TAZ localisation via the HIPPO pathway by inhibiting LATS1/2 and/or upstream factors, thus preventing phosphorylation and cytoplasmic retention of YAP/TAZ. (Right) ECM stiffness also regulates YAP/TAZ in a HIPPO-independent mechanism. Cells interact with their ECM via integrins; in stiff environments, focal adhesion assembly is promoted, which activates Rho-ROCK signalling, which in turn activates F-actin stress fibre formation and translocation of YAP/TAZ to the nucleus, where these proteins regulate gene expression via activation of TEAD1-4. External application of force to the nucleus has also been shown to open nuclear pores and allow increased YAP/TAZ entry into the nucleus. Cytoskeletal inhibitors affect different parts of the mechanotransduction pathway; Y-27632 inhibits ROCK, Latrunculin A inhibits F-actin polymerisation and Blebbistatin inhibits Myosin (all depicted in red). MST1/2, mammalian ste-20-like kinases 1/2; SAV1, salvador family WW domain containing protein 1; LATS1/2, large tumour suppressors 1/2; MOB1 A/B, MOB kinase activator 1A; NF2, neurofibromatosis type 2; TEAD1-4, TEA domain family member 1-4; FAC, focal adhesion kinase. Created using Biorender.com.