Metabolic and Mechanical Cues Regulating Pluripotent Stem Cell Fate

TICB 1462 No. of Pages 16

Special Issue: Stem Cell Biology

Review

Metabolic and Mechanical Cues Regulating Pluripotent Stem Cell Fate Tânia Perestrelo,1,2,3,4,8 Marcelo Correia,1,2,8,9 João Ramalho-Santos,2,5,* and Denis Wirtz4,6,7,* The ability to shift between metabolic states and to tightly regulate cellular mechanical properties have been described as crucial events in the achievement of correct embryonic development. Indeed, metabolic and mechanical manipulations in vitro have led to the discovery of new methods to control cell fate. As these two modulators are usually studied separately, in this review article, we describe how cellular mechanics and metabolic characteristics regulate embryonic development in vivo and describe the role of these cues in the regulation of pluripotency and differentiation in vitro. We also pinpoint possible connections between metabolism and mechanotransduction, highlighting recent findings in the Yes-associated protein, phosphoinositide 3-kinase, and AMP-activated protein kinase signaling pathways, and how they may be relevant in modulating cell fate in other contexts.

Highlights Metabolic and mechanical cues control stem cell fate in vivo and in vitro. There is a synergistic intertwining of metabolism, stem cell fate, differentiation, and reprogramming. The mechanical properties of embryonic cells are crucial for their precise development. Recent knowledge of the mechanical properties at different cell stages of embryonic development has brought to light the role of mechanosensing in cell fate decision. In vitro, biophysical cues are key modulators in ESC fate and cellular reprogramming.

Embryonic Development Embryogenesis is a highly complex and dynamic, yet precise process. It begins with fertilization of the oocyte with a sperm cell, which gives rise to a totipotent cell, the zygote. This cell has the potential to produce all embryonic and extraembryonic lineages [1]. Importantly, embryogenesis is a continuous process. As development takes place cells face important fate decisions, which culminate in a new individual organism (Figure 1). The need for a better understanding of early mammalian development has led scientists to look for in vitro models that can recapitulate cell pluripotency, commitment, and differentiation. Thus, different cell populations have been successfully isolated in vitro including mouse embryonic stem cells (mESCs) [4,5], human ESCs (hESCs) [6], induced pluripotent stem cells (iPSCs) reprogrammed from somatic mouse and human cells [7,8], and epiblast (EPI)-derived stem cells (EpiSCs) [9,10] (Figure 1). While ESCs are derived from preimplantation blastocysts, EpiSCs are obtained from the post-implantation EPI and iPSCs are reprogrammed from somatic cells (Figure 1). These cell models have led to the recognition that maintenance of each developmental state is achieved through a balance of transcription factor networks (Figure 2). How these different transcription factors are maintained or modulated is still not completely understood. Biochemical modulators, such as leukemia inhibitory factor (LIF) and inhibitors of glycogen synthase kinase-3 (Gsk3) and mitogen-activated protein kinase kinase (MEK), maintain mESCs in a pluripotent state, while their withdrawal leads to differentiation [11]. Moreover, the successful isolation of EpiSCs and iPSCs in culture led to the ability to compare different cell states and to the discovery that metabolism (see Glossary) is a keystone in stem cell fate, responsible not only for the production of energy but also for the synthesis of precursors, which are crucial for sensing the energetic and redox status of the cell in the developmental process, chromatin remodeling, and gene transcription regulation. Finally, the discovery that the physical characteristics of culture conditions, notably that substrate

Trends in Cell Biology, Month Year, Vol. xx, No. yy

The connection between metabolic and mechanotransduction signaling pathways is emerging as key in determining cell fate but remains largely unexplored. Evidence suggests that YAP, MAPK, and PI3K signaling pathways act as molecular crosstalk between metabolism and mechanotransduction.

1

PhD Program in Experimental Biology and Biomedicine (PDBEB), Institute for Interdisciplinary Research (IIIUC), University of Coimbra, Coimbra, 3030789, Portugal 2 Center for Neuroscience and Cell Biology (CNC), University of Coimbra, Coimbra, 3004-504, Portugal 3 Johns Hopkins Institute for NanoBioTechnology, Baltimore, MD 21218, USA 4 Department of Chemical and Biomolecular Engineering, The Johns Hopkins University, Baltimore, MD 21218, USA 5 Department of Life Sciences, University of Coimbra, Coimbra, 3000-

https://doi.org/10.1016/j.tcb.2018.09.005 © 2018 Published by Elsevier Ltd.

1

TICB 1462 No. of Pages 16

stiffness, can regulate the fate of mesenchymal stem cells (MSCs) [12] opened a new research window that led to the discovery that physical modulators, such as stiffness and topography, are able to control ESC fate [13,14]. In this review article, we highlight the importance of metabolism and mechanical cues in cell fate regulation.

Metabolism: An Important Player in Stem Cell Fate Cells rely on two processes to produce energy in the form of ATP: glycolysis and oxidative phosphorylation (OXPHOS; Box 1). In addition to its role in energy production, glycolysis and the Krebs cycle provide intermediates for other molecular pathways. Cells have the plasticity to shift between metabolic pathways to match their requirements for ATP and cell precursors. Nonetheless, most differentiated cells rely on OXPHOS to achieve their energetic demands. Glycolysis and OXPHOS in Embryonic Development During embryonic development, metabolic pathways are tightly regulated to satisfy demands for both energy and precursor molecules, which are essential for cellular replication and specialization (Figure 1). It all starts before fertilization, during oogenesis, where a boost in mitochondria biogenesis results in a two- to threefold increase in mitochondrial content [15]. Mitofusin-2 regulates oocyte development as its downregulation reduces mitochondrial membrane potential as well as fertilization and cleavage rate [16]. Since mitochondria do not replicate between fertilization and the blastocyst stage, the increase in mitochondrial content before fertilization is crucial to ensure sufficient mitochondrial number and to meet energetic and precursor demands in all embryonic cells [17]. After fertilization, the embryo relies on OXPHOS energy metabolism (Figure 1) [18,19]. In the early embryonic stages, mitochondria are small, globular shaped, and contain poorly developed cristae; nevertheless, they are still active. After several embryonic divisions, the mitochondrial content in each cell declines [20], and at the morula state, there is an increase in glucose transporters in each cell, which gradually shifts the embryo to a glycolytic phenotype with the preimplantation EPI combining an OXPHOS and glycolytic metabolism (Figure 1) [19– 21]. After implantation, glycolysis is the major metabolic pathway in the EPI, where glucose is metabolized to lactate, presumably due to the hypoxic uterine wall environment. OXPHOS is reestablished and mitochondria biogenesis and maturation occur again after blood flow begins in the embryo [22] (Figure 1). Importantly, when comparing the different in vivo embryonic states with in vitro-derived ESCs and EpiSCs, equivalent mitochondrial morphologies and metabolism requirements are seen: ESCs seem to retain similar characteristics to the preimplantation inner cell mass (ICM), while EpiSCs are similar to the post-implantation EPI [23] (Figure 1). Glycolysis and OXPHOS in ESCs and EpiSCs Naïve mESCs have small and spherical mitochondria, with immature cristae and reduced mitochondrial DNA (mtDNA) copy number, equivalent to what is found in the preimplantation ICM cells in the blastocyst. Nevertheless, these cells have a bivalent metabolism and can switch between glycolysis and OXPHOS [23]. Importantly, mitochondrial function is dependent not only on the expression of mtDNA genes but also on the expression of nuclear genes. For instance, using mESCs as a model system, signal transducer and activator of transcription 3 (Stat3), a downstream effector of the LIF pathway, bound mtDNA and increased the expression of mitochondrial transcripts, enhancing OXPHOS metabolism [24]. These results suggest a synergetic role between the mitochondria respiratory chain and the nuclear transcription of pluripotent genes mediated by Stat3. Similar to the post-implantation EPI, mEpiSCs have a more mature mitochondrial network, but still less mature than those found in differentiated cells [23]. mEpiSCs are more glycolytic; 2

Trends in Cell Biology, Month Year, Vol. xx, No. yy

456, Portugal 6 Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, MD 21231, USA 7 Johns Hopkins Physical SciencesOncology Center, The Johns Hopkins University, Baltimore, MD 21218, USA 8 Co-first authors 9 Current address: i3S (Instituto de Investigação e Inovação em Saúde) and IPATIMUP (Institute of Molecular Pathology and Immunology, University of Porto), University of Porto, Porto 4200-135, Portugal

*Correspondence: [email protected] (J. Ramalho-Santos) and [email protected] (D. Wirtz).

TICB 1462 No. of Pages 16

consequently, their basal and reserve levels of respiration are lower than that of naïve mESCs [23]. These features have been related to the low expression of multiple nuclear-encoded proteins of the respiratory chain complexes in EpiSCs, namely, cytochrome c oxidase (from complex IV), and nicotinamide adenine dinucleotide (NADH) dehydrogenase (from complex I), compared to mESCs [23]. In addition, hypoxia-inducible factor 1 a (Hif1a) levels are increased in mEpiSCs compared to mESCs [23]. Hif1a is a transcription factor that acts as an oxygen sensor inside the cell. Low levels of oxygen promote the expression of Hif1a, which controls the expression of essential genes that promote glycolysis, such as lactate dehydrogenase A, responsible for the conversion of pyruvate to lactate; and pyruvate dehydrogenase kinase 1 (PDK1), responsible for the phosphorylation of pyruvate dehydrogenase (PDH) and, consequently, its inactivation [25]. Interestingly, ectopic expression of Hif1a or reduced oxygen concentrations promote the transition of mESCs from the naïve state to the primed state, which is the first step towards pluripotent cell differentiation [23,26]. In addition, dichloroacetate, an inhibitor of PDK1 and therefore an activator of the PDH complex, promotes differentiation of mESCs, suggesting PDK as a metabolic gatekeeper of pluripotency [27]. Also, 3bromopyruvate, an inhibitor of hexokinase II and glyceraldehyde-3-phosphate dehydrogenase, which are crucial enzymes in the glycolytic pathway, led to a metabolic switch and loss of pluripotency, even in the presence of pluripotent cell culture conditions [28]. Stressing the notion that reduction of OXPHOS in the primed state is essential for maintenance of pluripotency, antimycin A (AA), an inhibitor of complex III of the electron transport chain (ETC), is able to maintain pluripotency in hESCs and inhibit differentiation of mESCs [29–32]. Epigenetic Modulators and Reprogramming Besides its role in energy production, metabolism is also implicated in the generation of cellular building blocks involved in epigenetic regulation. Glycolysis and the Krebs cycle provide precursors for the synthesis of nucleotides, non-essential amino acids and lipids, used to produce macromolecules essential for cell division, as well as intermediates used in enzymatic reactions, such as acetylation and methylation (Figure 3, Key Figure) [33,34]. For instance, acetyl-coenzyme A (CoA) contributes not only to the Krebs cycle but also to the synthesis of lipids and acetylation of amino acid residues, such as lysine, on histones and other proteins [35,36], while a-ketoglutarate, a Krebs cycle intermediate, contributes to the demethylation of DNA and histones [37]. The acetylation/deacetylation of histones regulates global chromatin architecture and, consequently, the control of gene transcription and stem cell fate. There are two main protein families involved in the control of this process: histone acetyltransferases (HATs) and histone deacetylases (HDACs). Acetylation of histones is associated with open chromatin structures and plays a critical role in the maintenance of stem cell pluripotency [38]. Inhibition of glycolytic enzymes decreases histone acetylation, connecting glycolytic production of acetyl-CoA with histone acetylation and the maintenance of the pluripotent state [36]. By contrast, HDAC promotes histone deacetylation, which is one of the primary features upon ESC differentiation [36,39]. HDACs are divided in two main groups: ‘classical’ HDACs, which are Zn2+ dependent, and sirtuins, which are dependent of the NAD+ pool. Shifts from anaerobic (glycolysis) to aerobic (OXPHOS) metabolism and vice versa are linked to changes in the NAD+/NADH ratio, which make sirtuins important cellular metabolic sensors [40]. Methylation is another epigenetic mark that plays a crucial role in the maintenance of stem cell pluripotency. Methylation/demethylation promotes a modulation of chromatin structure, and this process is tightly interconnected with the metabolic status of the cell [41]. DNA and histone methylation are regulated by DNA methyltransferases (DNMTs) and histone

Glossary Anabolic pathways: pathways responsible for the production of macromolecules, such as nucleotides, amino acids, and lipids. Anabolic pathways require energy from ATP or nicotinamide adenine dinucleotide (NADH). Blastoid: an in vitro 3D combination of ESCs and trophoblast stem cells that resembles the transcription profile and the structure of the blastocyst. Blastomere compaction: process in which blastomeres flatten against each other, their cell-cell contact is maximized, and they can no longer be distinguished morphologically. This takes place before the morula stage of embryo development. Catabolic pathways: pathways that break down molecules into smaller units to produce energy or building blocks. Gastruloid: an in vitro multicellular model of gastrulation. Glycolytic flux: the rate at which molecules progress through glycolysis. Metabolism: combination of anabolic and catabolic mechanisms that occur inside a cell. Naïve state: ESCs capable of unlimited self-renewal capacity and able to differentiate into all three germ layers in vitro and to form teratomas and chimeras in vivo. The earliest stage of pluripotency. Organoid: an in vitro multicellular structure that contains many cell types resembling a specific adult organ. Pluripotent cells: cells that can differentiate in vitro as well as in vivo, giving rise to all three embryonic lineages and germ cells, as well as being capable of self-renewal. Primed state: ESCs that are not capable of forming chimeras in vivo, although they are still able to form teratomas and differentiate into all three germ layers in vitro. A later stage of pluripotency. Self-renewal: cells that can proliferate indefinitely, while maintaining pluripotency. Stiffness: reported in Pascal (Pa) units, it is the ability of a material to resist to a force. It is formally defined as Young’s elastic modulus (E) and results from the relationship between the force applied to a defined area

Trends in Cell Biology, Month Year, Vol. xx, No. yy

3

TICB 1462 No. of Pages 16

methyltransferases, respectively. In ESCs, the lack of DNMTs, for example, promotes maintenance of pluripotency and self-renewal, while these cells fail to differentiate [42,43]. Histone demethylases can be lysine-specific demethylases, which target histone 3-lysine 4 residues and histone 3-lysine 9 residues and require flavin adenine dinucleotide to catalyze this process, or lysine and arginine demethylases, which require a-ketoglutarate to promote demethylation [41]. a-Ketoglutarate is a cofactor of a-ketoglutarate/Fe2+ dioxygenases (enzymes responsible for histone and DNA demethylation), whereas succinate is a competitive inhibitor of these enzymes. Thus, in ESCs, an elevated a-ketoglutarate-to-succinate ratio, maintained by glucose or glutamine metabolism, promotes pluripotency through histone and DNA demethylation [37]. Furthermore, this demethylation pattern matches the epigenetic markers characteristic of pluripotency in ESCs that are responsible for the suppression of differentiation and expression of pluripotent genes. Epigenetic regulation is also crucial during somatic cell reprogramming to iPSCs. Upon induction of pluripotency, cells must remodel their chromatin to reactivate pluripotency-associated genes, while silencing differentiation-related genes. Metabolically, the increase of PDK and hexokinase II or the inhibition of the ETC through AA or rotenone increases reprogramming efficiency [44,45]. Moreover, reprogramming factors, with the exception of Oct4, can be replaced by stimulating glycolytic gene expression through the activation of key metabolic enzymes [46], suggesting that metabolism and the cell state are also interconnected during cellular reprogramming [47]. In addition to metabolism, biophysical cues influence somatic cellular reprogramming to iPSCs (for a detailed review, see [48]). For example, 2D soft matrices as well as 3D matrices of defined stiffness increase the reprogramming efficacy of fibroblasts through the facilitation of mesenchymal-to-epithelial transition (a crucial step for the success of reprogramming) [14,49]. Accordingly, substrate topography such as microgrooves (10 mm in width) or aligned nanofibers significantly improve reprogramming efficiency by promoting chromatin structure rearrangement [13]. These conditions induce the reorganization of the cytoskeleton and actin-myosin contractility, which results in the alteration of key epigenetic regulators of expression, including a decrease in HDAC activity and an increase in methyltransferase activity (WD repeat domain 5) and consequently increases in histone H3 acetylation and di- and trimethylation of lysine 4, respectively [13]. iPSCs formed by the combination of gene transfer and soft substrates or specific substrate topography are able to generate teratomas in vivo and embryoid bodies in vitro, both of which are hallmarks of pluripotent cells [13,14]. The inhibition of actomyosin contractility with blebbistatin results in the reduction of epigenetic changes and reprograming efficiency, which suggests that the cell epigenetic state is also regulated by mechanotransduction processes [13]. Interestingly, studies focused on metabolism and iPSC formation suggest that the glycolytic shift occurs before expression of pluripotency genes, reinforcing metabolism as a key player in cell fate [50,51]. In addition, studies focused on the role of mechanotransduction and iPSC formation propose that gene transfer and biophysical cues synergistically stimulate pluripotent gene expression, and the increase of reprogramming efficacy. It will be interesting to address whether these two aspects are related, for example, if biophysical cues play a role in the glycolytic shift, and whether there is any synergy between metabolism and the mechanotransduction signal that could be tapped to increase cellular reprogramming efficiency.

Mechanical Cues: A Forgotten Piece in Regulating Stem Cell Fate Mechanical cues have recently been recognized as important regulators of cell fate. Changes in extracellular matrix (ECM) mechanics, substrate stiffness, topography, and physical forces such as shear stress, tension, and compression are examples of external mechanical cues that 4

Trends in Cell Biology, Month Year, Vol. xx, No. yy

(stress) and the resultant deformation of the material (strain). A stiff matrix is more resistant to deformation than a soft matrix. Teratoma: a multicellular type of rare tumor used as an in vivo pluripotency assay, the only one that can be performed with human cells. The teratoma assay assesses the ability of cells to spontaneously differentiate into tissues from the three germ layers when injected into immunocompromised mice. If they are pluripotent cells should always form teratomas.

TICB 1462 No. of Pages 16

OXPHOS

OXPHOS Glycolysis

OXPHOS Glycolysis

OXPHOS Glycolysis

(A)

Parietal endoderm (PE)

FerƟlizaƟon

Zygote

2 cells

4 cells

8 cells

Morula CompacƟon and polarizaƟon

Oocyte Zona pellucida

Blastomeres

Blastocyst

PrimiƟve endoderm

Trophectoderm

Inner cell mass

ImplantaƟon

PrimiƟve endoderm (PrE)

Visceral endoderm (VE)

Mesoderm

Epiblast Blastocoel cavity

Epiblast (Epi)

DefiniƟve endoderm Ectoderm

(B)

ESC

OXPHOS Glycolysis

EpiSC

OXPHOS Glycolysis

iPSC OXPHOS Glycolysis

In vitro

Figure 1. Embryonic Development in Mammals and In Vitro-Derived Cells: From Preimplantation to Gastrulation. (A) Schematic of cell fate decisions that lead to the formation of the three embryonic germ layers and the extraembryonic lineages. After fertilization, the zygote is formed. The zygote gives rise to two cells, four cells, and then eight cells that are morphologically distinct from each other. These cells rely mostly on oxidative phosphorylation (OXPHOS) for ATP. At the eight-cell stage, the cells undergo compaction, polarization, and asymmetric cell division, giving rise to a compact morula, and the first cell-fate decision occurs: trophectoderm (TE) (green) and inner cell mass (ICM) (purple) are formed. At the 32-cell stage, the blastocoel cavity starts to form inside the embryo, giving rise to the blastocyst, where the second cell-fate decision occurs: ICM gives rise to the epiblast (EPI) (orange) and the primitive endoderm (PrE) (blue) [2]. PrE is positioned facing the blastocoel cavity, while EPI is positioned between the PrE and the TE. At this stage, cells rely both on glycolysis and OXPHOS. After Epi, PrE, and TE specification, the blastocyst is ready to be activated and to implant. During implantation, glycolysis is the main source of ATP. After implantation, PrE gives rise to parietal endoderm (PE) and visceral endoderm (VE), and EPI will give rise to mesoderm, definitive endoderm, and ectoderm [3]. These cells have a shift in their metabolic requirements and rely mostly on OXHPOS. (B) Different pluripotent cell populations have been isolated and propagated under in vitro conditions. Embryonic stem cells (ESCs) are derived from the ICM of the preimplanted blastocyst, whereas epiblast-derived stem cells (EpiSCs) are derived from the post-implantation EPI. Similar to the ICM, ESCs rely both on glycolysis and OXPHOS, while EpiSCs shift their metabolism to glycolysis. Induced pluripotent stem cells (iPSCs) are a result of the molecular reprogramming of somatic cells in vitro by the forced exogenous expression of a combination of key pluripotent transcription factors such as Oct4, Klf4, Sox2, and c-myc. These cells shut down OXPHOS and rely on glycolysis as their main source of ATP.

influence cell fate through a mechanotransduction process (Box 2). It is now accepted that mechanical cues work together with biochemical cues in the regulation of embryogenesis and organogenesis, while cell mechanical properties are associated with a readout of physiological functions or pathological features. Mechanical Cues in Embryonic Development Recently, mechanical characteristics have been taken into consideration and different methods have been developed to analyze oocyte and zygote quality. During fertilization, the sperm cell first penetrates the zona pellucida, a specialized extracellular matrix made of a network of glycoproteins that surrounds and protects the oocyte [52,53]. After gamete fusion, there is a triggering of biochemical signals that induces a 3D architectural change of the zona pellucida known as zona reaction or ‘hardening’ [53]. Mechanically, different studies have been conducted to analyze zona pellucida elasticity before and after fertilization [54–59]. Although different techniques were used, all studies show that the zona pellucida becomes stiffer after fertilization [54–59]. Moreover, the viscoelasticity properties of the oocyte after fertilization were recently associated with embryo viability [58,59]. These results suggest that the mechanical properties of the zona pellucida can function, in an early stage, as a predictive factor of embryonic development. Trends in Cell Biology, Month Year, Vol. xx, No. yy

5

TICB 1462 No. of Pages 16

ESC

Na ve pluripotency EpiSC

Tfcp2l1

LIF

ERK2

Stat3

Klf5 Klf4

Sox2

Nanog

β-Catenin

Tcf3

Esrrb

Primed pluripotency

Oct4

Foxa2

Klf2 Fgf5 Dppa3

Rex1

Sox2

Nanog

T

Oct4

Le y2

Sox17 Foxa2

Le y2 T Oct4

Zeb2 Nes n

Sox17

Soma

Fgf5

Sox2

(i) (ii)

c linea

So

ge diff e

ren a on

S ff

Figure 2. Role of Transcription Factors and Stiffness in Naïve Pluripotency, Primed Pluripotency, and Somatic Lineage Differentiation. The main transcription factors involved in naïve and primed mouse embryonic stem cell (mESC) pluripotency, as well as in lineage specification (mesendoderm and ectoderm), are represented. The core pluripotency transcription factors Oct4, Sox2, and Nanog are represented in yellow. When together, Oct4, Sox2, and Nanog maintain ESCs in a pluripotent state. In blue are the transcription factors specific for the naïve state of pluripotency, with black arrows representing some of the pathways that activate these genes. Specific primed transcription factors responsible for maintaining cells in a primed pluripotency state are represented in green. Upon differentiation, specific transcription genes are maintained in each lineage: (i) mesendoderm [Oct4, Sox17, Lefty2, Foxa2, T (T-brachyury)] and (ii) ectoderm (Sox2, Fgf5, Nestin, and Zeb2). While soft substrates (represented in light red) have been shown to induce naïve pluripotency, the increase of stiffness (represented in dark red) has been correlated with somatic lineage differentiation.

The mechanical properties of embryonic cells are also crucial for their precise development and recent knowledge of the changes in mechanical properties during different stages of embryo development have brought to light the role of mechanosensing in cell fate decisions. For example, the evaluation of surface tension in the eight-cell stage blastomere, using a micropipette aspiration technique, demonstrated that the increase of surface tension, promoted by

Box 1. Summary of Glycolysis and OXPHOS Glycolysis is the pathway responsible for the breakdown of one molecule of glucose into two molecules of pyruvate, and the production of two net molecules of ATP and two net molecules of NADH [21]. Pyruvate, the end product of glycolysis, can usually then go one of two ways: under anaerobic conditions, pyruvate is reduced to lactate that is excreted to the extracellular space; whereas under aerobic conditions, pyruvate is oxidatively decarboxylated to acetylCoA by the pyruvate dehydrogenase complex (PDH), in the mitochondrial matrix, with production of one NADH molecule. Acetyl-CoA is then the fuel of the Krebs cycle, located inside the mitochondria. In the Krebs cycle, three molecules of NADH, one moiety of flavin adenine dinucleotide (FADH2) of the succinate dehydrogenase enzyme, and one molecule of guanosine triphosphate (GTP) are produced from each acetyl-CoA that gets into the Krebs cycle [21]. At the end, from each glucose molecule, four molecules of ATP, ten molecules of NADH, and two reduced FADH2 (part of the enzyme succinate dehydrogenase) are generated. The NADH and succinate (FADH2) produced thus far are oxidized, and, on average, 34 additional ATP molecules per molecule of glucose are generated [21]. Therefore, mitochondria are known as the powerhouse of the cell. It should however be noted that many cells, including some types of stem cells, can carry out aerobic glycolysis, using the Krebs cycle mostly for precursor production, not ATP generation.

6

Trends in Cell Biology, Month Year, Vol. xx, No. yy

TICB 1462 No. of Pages 16

Key Figure

The Interplay between Mechanotransduction and Metabolism Signaling Pathways

E- c a

Cell 2

dhe

[1]

rin

Cell 1

E-cad

HIPPO signaling on

herin

Glu

e co s

LKB1 p120 Abl

P AMPK

Glu

cos

e

LATS

B-cat α-cat

Vinculin

Me Me H isto ne Me

ROCK

ATP

ATP

Fo rce

Myosin

ATP

Fo

rc e

Aldolase

PIP3

Ac Rac

e Ɵv

Aldolase AcƟn

AC

PI3K

His ton e

F-ac Ɵn

Akt

Cholesterol biosynthesis

AC

P

Glycolysis

Aldolase Cytoskeleton turnover

Inac Ɵve

YAP TAZ

Aldolase Aldolase

ECM

AC

P

YAP TAZ TEAD

Cytoskeleton reorganizaƟon

ATP

Glycolysis

FAK

Me

P

RhoA

Glycolysis

[2]

Me

Nucleus

[4] es

bul

u rot

Mic

F-acƟn Aldolase

c

Intermediate filaments ATP

Citrate ATP ATP ATP ATP

a-KG

f on o n ulaƟ Reg hylaƟo met

f on o ulaƟ n Reg tylaƟo ace

[3]

ria

ond

och Mit

Figure 3. (1) The transmission of mechanical forces between adjacent cells. Cadherins are responsible for the force transmission between adjacent cells. Inside the cell, cadherins recruit catenins, which have associated cytoskeleton-binding proteins responsible for signaling transmission. The energy sensor AMP-activated protein kinase (AMPK) is also associated to the cadhesome (a pool of structural and regulatory protein components of the adherens junctions), and its activation promotes glucose uptake and consequently an increase in glycolysis and ATP production, which are important for cytoskeleton rearrangement. (2) The transmission of (See figure legend on the bottom of the next page.) Trends in Cell Biology, Month Year, Vol. xx, No. yy

7

TICB 1462 No. of Pages 16

Box 2. Mechanotransduction Inside the cell, mechanical stimuli are converted to biochemical signals, in a process called mechanotransduction. This process enables cell adaptation to the surrounding environment and can be divided into three stages: mechanotransmission, mechanosensing, and mechanoresponse [123]. Mechanotransmission is the transmission of mechanical forces, through protein–protein interactions, between the ECM and the cell or between neighboring cells. Transmembrane integrins are the bridge between the cell and the ECM and work as both force transmitters and mechanosensors. The mechanical linkage between the ECM and the actin cytoskeleton is mediated by focal adhesion complexes [124]. Inside the cell, structural scaffolds bind to the cytoplasmic domain of integrins and continue the transduction pathway mechanically, with the transduction of force to the actin cytoskeleton, with consequent regulation of actomyosin structure, and transmission of force to the nucleus through the LINC (linker of nucleoskeleton and cytoskeleton) complex [123,125]. These forces can be transmitted for long distances and are faster than biochemical signals since the propagation of the mechanical force is dependent on the resistance of the protein–protein interactions. Cells can also receive mechanical signals through their neighboring cells via cell-cell adhesions [126]. Cadherins are the transmembrane proteins responsible for the transmission of the force between adjacent cells, and similar to integrins, they act as both force transmitters and mechanosensors [126]. The cytoplasmic domain of cadherins recruits catenins, which have associated cytoskeleton-binding proteins, whereas the cadherin extracellular domain forms intercellular bonds with cadherins of neighboring cells. Mechanosensing is the ability to convert the mechanical force into a biochemical output through molecular changes of focal adhesions and conformational changes of their individual components. Inside the cell, there are also signaling scaffolds that bind to the cytoplasmic domain of integrins and continue the transduction pathway biochemically [75,123]. Finally, the mechanoresponse, as implied by its name, is the downstream effect of the applied force that in the short term promotes the modulation of cell-ECM binding of focal adhesions stability and of cytoskeleton strength. In the long term, it induces the activation of signaling pathways and changes of gene expression [75,123].

actomyosin contractility, is essential to induce blastomere compaction [60]. Blastomere compaction has long been known to be related to an increase in adhesion molecules such as Ecadherin [61], but the molecular pathways that regulate this event are poorly understood. Recently, E-cadherin was shown to be present in filopodia-like protrusions that occur at the eight-cell stage, which have been proposed to be responsible for maintaining tension to control cell shape and promote compaction [62]. Although this study proposes that E-cadherindependent filopodia generate the mechanical forces to promote compaction, an independent study proposes that compaction is generated by actomyosin cortex tension, where E-cadherin prevents an increase in contractility at cell–cell contacts [60]. One possible explanation for these opposing results is that the actomyosin cytoskeleton can also be disrupted upon filopodia laser ablation, causing blastomeres to decompact [63]. In addition, the authors show that tension at cell–cell contacts is smaller than at the embryo surface, suggesting that adhesive forces, such as filopodia, cannot be the main driver of compaction, as they operate by pulling on neighboring cells [63]. During blastomere compaction, an apico-basal cell polarity in each blastomere is established, and blastomeres undergo two rounds of symmetric or asymmetric division [64]. In the first case, both daughter cells inherit the apical domain resulting in two polar cells, both maintaining their mechanical forces between the extracellular matrix (ECM) and a cell. Broken arrows indicate a presumable interaction between mechanotransduction and metabolism upon ECM–cell interaction. Cells sense the substrate through ECM–integrin interactions. Inside the cell, interactions between focal adhesions lead to the signaling cascade initiation by focal adhesion kinase (FAK), which activates the phosphoinositide 3-kinase (PIK3) pathway. The force activates Rac, resulting in actin cytoskeleton turnover and release of aldolase to the cytoplasm, which increases glycolysis flux. PIP3, phosphatidylinositol (3,4,5)-trisphosphate. (3) Metabolism as a source of intermediates used in enzymatic reactions, such as acetylation and methylation. Citrate and a-ketoglutarate (a-KG) produced in the mitochondria through the Krebs cycle contribute to DNA and histone acetylation and methylation, respectively. (4) Mechanotransduction and metabolism as regulators of Yes-associated protein (YAP)/ transcriptional co-activator with postsynaptic density 95/disc-large/zona occludens (PDZ)-binding motif (TAZ) location. Stiff substrates induce F-actin polymerization and consequently YAP and transcriptional co-activator with PDZ-binding motif (TAZ) translocation to the nucleus, whereas cell-cell adhesions are responsible for activating the Hippo signaling pathway and induce YAP/TAZ degradation signaled by large tumor suppressor kinase (LATS). By contrast, cholesterol biosynthesis and glycolysis are responsible for the translocation of YAP/TAZ to the nucleus and their binding to the TEAD family of transcription factors.

8

Trends in Cell Biology, Month Year, Vol. xx, No. yy

TICB 1462 No. of Pages 16

outer position. In the second case, only one daughter inherits the apical domain, resulting in one polar (outside cell) and one apolar cell (inside cell). This eventually leads to the outside layer of cells forming the trophoblast or trophectoderm (TE) that will participate in placental development and an interior cluster of ICM cells that contributes to the fetal lineage. However, how a blastomere adopts an internal or external position within the embryo has long been a topic of debate, with cell polarity, cell position, and contractility models described as independent regulators of TE/ICM fate [65,66]. Recently, mechanosensing was shown to be crucial for the correct positioning of cells during TE/ICM specification, acting as the bridge between polarity, position, and cell fate [67]. This model suggests that upon asymmetric division, the two daughter cells acquire different levels of contractility; apolar blastomeres have low levels of atypical protein kinase C (aPKC) and increased levels of cortical myosin, which increases cell contractility and regulates Yes-associated protein (YAP) localization in the cytoplasm, in opposition to polar blastomeres [67]. In addition to the mechanical properties of the embryo, the uterus has been suggested as an external mechanical force involved in embryonic anterior–posterior axis formation, which is critical for post-implantation development [68,69]. With the use of microfabricated cavities to recapitulate the external constraints promoted by the uterus during implantation, one group reported that mouse embryo egg-cylinder shape acquisition and distal visceral endoderm (DVE) specification are dependent on the external spatial restriction, with the width and stiffness of the cavity as vital players in the correct DVE development and anterior-posterior axis acquisition [68]. However, doubts about this model have emerged due to the ability to induce the DVE in embryos cultured in a hanging drop method or in a dish without a physical constraint, raising the importance of the self-organization properties of these cells [70,71]. Nevertheless, mechanical forces derived from cell–cell contacts or ECM-cell adhesion are increasingly recognized as key regulators in gastrulation and organ formation, where cells undergo changes in their motility and shape, which are either affected by mechanical forces, or generate mechanical forces that are crucial for correct embryo development [72–75]. Indeed, ESCs cultured in different confinement experiments in 2D patterns or within 3D suspension or matrices have led to the development of blastoids, gastruloids, and organoids, cells on a spatial organized shape and signaling environments that try to recapitulate the different states of in vivo development [76]. Many different organoid models have been generated to study the chemical and physical requirements for stem cells to differentiate into different tissue types [77,78]. Specific mechanical and ECM environments were recently shown to be crucial to intestinal stem cell maintenance and differentiation [79]. Fibronectin-based adhesion and high matrix stiffness were described as being responsible for stem cell survival and proliferation, through activation of YAP; whereas laminin-based adhesion and soft substrates were required to promote intestinal stem cell differentiation [79]. Considering the importance of cells sensing the correct force, geometric confinement of ESCs was shown to promote a self-organized pattern and induce trophectoderm, ectoderm, and mesendoderm primitive-streak markers in embryoid bodies with a similar cell type distribution as the gastrula during embryogenesis [80]. By contrast, blastoids were just recently established using ESCs and trophoblast stem cells cultured in 3D [81,82]. This model recapitulates the transcription and morphological patterns seen in the blastocyst state 3.5 of the embryo and opens the door for new approaches to study genetic and mechanical principles in blastocyst formation and implantation. All things considered, there is a clear pattern showing a closer in vitro recapitulation when the native microenvironment of cells in the different tissues and development stages is considered. Trends in Cell Biology, Month Year, Vol. xx, No. yy

9

TICB 1462 No. of Pages 16

Stiffness in ESC Fate ESCs fate in vitro has also been shown to be regulated through biophysical cues in addition to biochemical cues. Generally, ESCs are cultured in tissue culture polystyrene, and their fate is regulated only using biochemical factors (i.e., liquid media manipulation of growth factors or inhibitors). However, reports have shown that ESCs sense mechanical properties of the culture dishes. When mESCs and hESCs are cultured on dishes of different stiffness ranges, their fate changes accordingly (Figure 2). This effect was first reported in MSCs, where the commitment to a cell lineage was shown to be dependent on the similarity of the substrate elasticity to tissues in vivo [12]. Soft substrates (1 kPa) mimic the preimplantation embryo, the uterine epithelium, and the brain; substrates of intermediate stiffness (10 kPa) mimic the muscle; and stiffer substrates (100 kPa) mimic the bone [12,56,83]. mESCs cultured on very soft substrates (0.2-, 0.6-, and 2-kPa polyacrylamide gels) maintain pluripotency for many passages, even in the absence of the pluripotency growth factor LIF [84,85]. Under these conditions, mESC cultures are homogeneous, with round and compact colonies. These cells can develop teratomas, thus showing pluripotent properties [84]. An increase in substrate stiffness promotes an increase in basal traction forces in mESC colonies and affects cell spreading, but not cell attachment [84,86]. Upon induction of spontaneous mESC differentiation, substrate stiffness also plays an important role. mESCs cultured in the absence of LIF on 7.5-kPa substrates tend to differentiate into mesoendoderm [85]. Embryoid bodies derived from mESCs show an increase in cardiomyogenic differentiation when cultured on substrates with an elasticity of approximately 6 kPa [87]. Nevertheless, the use of substrates of high stiffness (40–2700 kPa) to culture mESCs in the absence of LIF leads to a positive correlation between genes expressed in the primitive streak during gastrulation and substrate stiffness [86]. The differentiation-related genes T-brachyury, Eomes, Foxa2, and N-cadherin increase their expression as stiffness increases, an effect that is independent of cell density, suggesting that substrate stiffness has a direct rather than an indirect effect on cell differentiation [86]. In the presence of osteogenic supplements, stiffer substrates promote an increase in osteogenic differentiation, compared to mESCs cultured on soft substrates [86]. Similar results have been reported in hESCs triggered to differentiate, where cells were cultured in 3D scaffolds of different stiffness to model germ layer specification upon gastrulation. The culture of hESCs in scaffolds of high elasticity (1.5–6 MPa) induces a transition from primitive streak expression to a mesoderm lineage, scaffolds of intermediate elasticity (0.1–1 MPa) induce endoderm-specific gene expression, and softer scaffolds (<0.1 MPa) increase the expression of ectoderm-specific genes [88]. In addition, hESCs cultured on soft flat (2D) substrates (0.1 and 0.7 kPa) form neural ectoderm more efficiently that hESCs cultured on stiffer substrates (75 kPa) but, contrary to mESCs, hESCs do not maintain their pluripotency on soft substrates in the absence of biochemical pluripotency cues [89]. Interestingly, mESCs cultured on 2-kPa polyacrylamide gels in the presence of the pluripotency-inducing 2i medium (combination of an MEK inhibitor, PD0325901, and a GSK3 inhibitor, CHIR99021), but in the absence of LIF, have higher expression of ectoderm lineage markers, indicating that these cells tend to differentiate even in the presence of specific pluripotency-maintaining inhibitors [85]. However, the presence of a Src inhibitor (GP77675), in combination with GSK3 in serum-free medium, designated as ‘alternative 2i medium’, was able to inhibit mESC differentiation and sustain mESC self-renewal and pluripotency, with these cells able to form teratomas and produce chimeric mice [85]. Therefore, Src is proposed to be an early mechanotransduction signal that induces differentiation when mESCs are cultured on substrates with specific elasticities. This mechanism was shown to involve the Src-Src homology/collagen A mitogen-activated protein kinase (MAPK) pathway [85]. In addition, cell stiffness also dictates cell sensitivity in response to mechanical stimuli. Upon the induction of a 10

Trends in Cell Biology, Month Year, Vol. xx, No. yy

TICB 1462 No. of Pages 16

local small cyclic stress, soft cells such as mESCs are more prone to spread and to lose Oct4 expression than stiffer cells, with Src, myosin II contractility, and cdc42 involved in the induction of this focal adhesion cell spreading [90]. These results suggest that small forces might have an important role during embryonic development, where cells are known to have soft properties. In addition, the initiation of force-induced spreading response and consequently differentiation are induced by activation of Src through the integrin cytoskeleton pathway.

The Connection between Mechanotransduction and Metabolism: What Do We Know? The simultaneous deregulation of both mechanotransduction and metabolic pathways has been described as a cause (or consequence) of diseases, such as osteoarthritis, cardiovascular diseases, and cancer [91–93]. These two pathways seem to be responsible for the cell energy imbalance and the deregulation of post-translational modifications in proteins responsible for the mechanosensing and the mechanoresponse in these diseases. Thus, in a simplistic way, mechanotransduction induces a rearrangement of the cytoskeleton, and this rearrangement requires energy. Furthermore, mitochondria need to interact with the cytoskeleton to maintain their normal function and proper localization, which is crucial for cell metabolism. However, the molecular crosstalk between these two signaling pathways is only now starting to be revealed, and the role of this crosstalk in pluripotency remains to be established. Nevertheless, many interesting aspects are being discovered, notably the differentiation of cardiac tissue and fate commitment in cancer biology, which may also be relevant given the metabolic similarities between pluripotent stem cells and many cancer cell types [21]. For example, the number and morphology of mitochondria are dependent on fission and fusion processes, which are carefully balanced in the cell as they are required to maintain a functional mitochondrial population. The cytoskeleton has been shown to be crucial in mitochondrial dynamics, and new research has found that mitochondria fission is triggered by mechanical forces [94]. Moreover, stretch, which mimics cell contractility, was shown to induce a structural organization of actin, microtubules, and mitochondria, promoting an increase of mitochondrial membrane potential in cardiac differentiation [95]. Nevertheless, while microtubules and vimentin were shown to have a crucial role in mechanotransduction and consequently mitochondrial membrane potential, non-muscle myosin II, the microtubule motor kinesin, and the mitochondrial fission protein dynamin-related protein 1 (Drp1) were crucial to the maintenance of the mitochondrial membrane potential during stretch fluctuations [95]. Taken together, mitochondria dynamics and activity seem to be clearly dependent on the mechanical context of the cell, with the cytoskeleton playing a major role. YAP is a very promising candidate in bridging metabolism and mechanotransduction. YAP and transcriptional co-activator with postsynaptic density 95/disc-large/zona occludens (PDZ)binding motif (TAZ) are localized in the nucleus when transcriptionally active and are known to interact with the TEAD family of transcription factors. YAP/TAZ have a role in cellular polarization, migration, organ growth, and induction of stem cell self-renewal and inhibition of differentiation and apoptosis [96]. YAP/TAZ are also well known cell mechanosensors and mechanoresponders and have been shown to be regulated by the Hippo signaling pathway, Wnt signaling pathway, and actin cytoskeleton integrity [96,97]. In addition, metabolism has recently been suggested to also regulate YAP/TAZ activity [98]. The mevalonate pathway, involved in cholesterol biosynthesis, was shown to regulate YAP/TAZ activity through geranylation of Rho GTPAse [99,100]. Through a different mechanism, glycolysis was also shown to regulate YAP/TAZ activity, with active phosphofructokinase, an enzyme that mediates the first committed step of glycolysis, shown to interact with TEAD and stabilize its interaction with YAP/ Trends in Cell Biology, Month Year, Vol. xx, No. yy

11

TICB 1462 No. of Pages 16

TAZ [101]. Moreover, the AMP-activated protein kinase (AMPK), which is activated upon glucose starvation or energy stress, was shown to induce the phosphorylation of YAP and consequently promote its cytoplasmic retention and inactivation [102–104]. It is also well known that YAP/TAZ activation are related to tumor malignancy, and these results suggest that metabolism is directly linked to cellular proliferation through YAP/TAZ regulation. Indeed, in breast cancer cells both glucose metabolism and the mevalonate pathway are strongly associated with YAP/TAZ activity and with the progression of these tumors towards more advanced and malignant stages, with phosphofructokinase acting downstream of the mevalonate pathway [99–101]. YAP was also shown to promote transcription of the GLUT3 glucose transport receptor and to control the transcription of enzymes involved in nucleotide biosynthesis [102,105,106]. Taking into account that YAP has also been described to be activated upon tumor ECM stiffening [107,108], YAP might perhaps contribute to a positive feedback mechanism between cellular stiffening, DNA replication, and the increase of energy requirements characteristic of tumor cells. As noted previously, this is particularly relevant given that the metabolic characteristics of proliferating tumor cells are similar to pluripotent stem cells [25]. Although multiple studies have recently shown that metabolic alterations are correlated with mechanotransduction changes in disease, the opposite is also true. Sarcomere mutations (known to induce cardiomyopathies) have been described to change the energetic cost of cardiac contraction and therefore induce metabolic changes in these cells [109]. In addition, proteome studies of a mouse model of collagen VI deficiency, which is associated with a myopathic phenotype, demonstrated that ECM architectural alterations induce mechanotransduction changes through cytoskeleton tension [specifically focal adhesion kinase (FAK) and Rho-associated, coiled-coil-containing protein kinase 1 increase] and consequently cause changes in muscle metabolism, such as a glycolysis decrease [110]. Indeed, cardiomyocytes cultured on soft substrates have an extended lifetime and retain a greater respiratory capacity compared to cardiomyocytes cultured on stiff substrates [111]. Furthermore, the same group recently showed that ECM elasticity is directly correlated with mitochondria basal respiration, ATP production, and maximum respiration capability [112]. Interestingly, the authors coupled ECM elasticity with substrate alignment and showed that there is a pronounced trend in basal respiration, ATP production, and maximum respiration capability compared to isotropic substrates, suggesting that both ECM and tissue architecture better regulate the adaptability of mitochondria to stress and the demand to higher cellular ATP levels [112]. In addition to mitochondrial activity, substrate stiffness was also shown to regulate the mitochondrial network, the nature of which is also linked to function [113]. However, the molecular crosstalk between ECM–cell adhesion, cell–cell adhesion, cytoskeleton, and mitochondria metabolism has just started to be unveiled. The phosphoinositide 3-kinase (PI3K) pathway is activated in response to ECM-cell adhesion (Figure 3) [114]. The PI3K signaling pathway regulates cell metabolism through Akt and induces cytoskeleton rearrangements through the Rac pathway [115]. Recently, PI3K was described to regulate glycolysis via the direct regulation of cytoskeleton dynamics, independently of Akt. One group showed that PI3K induces the disruption of the actin cytoskeleton via Rac, and this cytoskeleton disruption induces the release of aldolase (a glycolysis pathway enzyme) from Factin to the cytoplasm, increasing its activity and consequently the glycolytic flux [116]. Interestingly, the decrease of mechanistic target of rapamycin kinase (mTOR) signaling, a downstream effector of the PI3K/Akt pathway, was shown to induce the endocytosis of b-integrin and its ECM substrate in epithelial cells [117,118]. This ECM internalization is used as a source of energy and building blocks to produce amino acids, which enhances mTOR 12

Trends in Cell Biology, Month Year, Vol. xx, No. yy

TICB 1462 No. of Pages 16

activity and prevents the death of starved cells [118]. These results suggest that the PI3K pathway is responsible for the coordination between cell energy and cell mechanotransduction. By contrast, the already mentioned cell energy sensor AMPK is activated by a low ATP/AMP ratio, and it induces the increase of catabolic pathways while switching off anabolic pathways [119]. Recently, AMPK and one of its activators, liver kinase B1 (LKB1), have been shown to colocalize with the cadherin adhesion complex in a force-dependent manner in epithelial cells (Figure 3) [120]. This colocalization is crucial for the initiation of mechanotransduction signals. Indeed, one study elegantly showed that AMPK is upstream of Rho-Amediated contractility, acting as a mechanoresponsive protein [120]. In addition, AMPK was shown to increase glucose uptake and oxidation upon induction of force on E-cadherin. Thus, AMPK appears to connect mechanotransduction and energy homeostasis in cell–cell adhesion.

Concluding Remarks During embryonic development, cells are in contact with extracellular matrices of varying stiffness, which trigger different cell fates. However, to progress to different fates, cells need to tightly regulate their metabolism to achieve the correct balance of energy and production of precursors for distinct biosynthetic purposes. Recent in vivo studies have highlighted the importance of mitochondrial dynamics, metabolic switches, and mechanical cues on embryo development. The currently accepted models suggest that metabolism and mechanotransduction are important contributors to cell fate specifications from fertilization to organogenesis; however, their interaction has not yet been investigated in detail. It is known that cells that give rise to the fetus rely first on OXPHOS and then switch to a glycolysis-based metabolism upon implantation, and finally rely again predominantly on OXPHOS upon tissue differentiation. By contrast, cell tension and contraction have emerged as important regulators of the first cell fate decisions during embryonic development. Moreover, the recent studies involving blastoids, gastruloids, and organoids bring to light the importance of cells sensing the appropriate microenvironment to recapitulate in vitro the structural and biochemical development that occurs in vivo. Thus, these models open the door to the study of possible interactions between metabolic and mechanical cues, which remains a largely unexplored research area.

Outstanding Questions Does the effect of biophysical modulators overlap with biochemical effects in determining cell fate? What is the impact of biophysical modulators on metabolism? What is the impact of metabolic modulators on cell biophysical properties? How do mechanical signals affect the epigenetic state during cell fate commitment? Which signaling pathways link mechanotransduction and histone methylation and acetylation patterns? Are there synergistic effects between metabolism and mechanotransduction that could be tapped to increase cell reprogramming or differentiation efficiency? Can we recapitulate and modulate the metabolic and mechanical requirements in blastoids or gastruloids? Is YAP a bridging molecule between mechanotransduction and metabolism in stem cells? Are AMPK and PI3K signaling pathways the connection between cell-cell and ECM-cell adhesions and metabolism also in stem cells?

Deciphering the signaling pathways in vitro that mechanical or biochemical cues induce in ESCs, EpiSCs, and iPSCs highlights the importance of these cues on chromatin remodeling and consequently gene transcription regulation. In addition, the findings that independent modulation of metabolism, substrate stiffness, and substrate patterning are crucial to maintain pluripotency, to promote differentiation, or to induce somatic cell reprogramming stress the importance of these two cues working in synergy to induce different cell fates. It will be interesting to decouple the effect of each modulator on stem cell fate and evaluate the impact of biophysical modulators on metabolism and vice versa. Indeed, the latest research is now focusing on the link between mechanotransduction and metabolism in differentiated and cancer cells, which are metabolically similar to pluripotent cells. Key signaling pathways such as YAP, PI3K, and MAPK seem to connect both energy and precursor requirements for cell proliferation and maintenance of state. YAP/TAZ expression was recently shown to reprogram and induce tissue-specific stem cells from differentiated cells, with an independent report showing that YAP is able to bind to promoters of stem cell genes and stimulate their expression [121,122]. Considering that the ectopic expression of YAP prevents ESC differentiation and that during embryonic stem cell differentiation cellular mechanics is altered and cells go through a metabolic shift, it will be interesting in the future Trends in Cell Biology, Month Year, Vol. xx, No. yy

13

TICB 1462 No. of Pages 16

to analyze whether YAP connects the effects of metabolism and mechanotransduction during embryonic stem cell differentiation. Moreover, the recent results suggest AMPK and PI3K as common signaling transducers of metabolism and of cell–cell and of ECM-cell adhesion signaling, respectively. These new results emphasize the importance of understanding the robust integration of these different stimuli in determining cell fate and open the door to a panoply of questions (see Outstanding Questions) that can lead to different cellular reprograming models and novel solutions for tissue engineering. Acknowledgment The authors thank Fundação para a Ciência e Tecnologia (FCT) Portugal, for PhD scholarships attributed to T.P. (SFRH/ BD/51684/2011) and M.C. (SFRH/BD/51681/2011). This work is partly funded by FCT Project CENTRO-01-0145FEDER-028871, HEALTHYAGING2020 CENTRO-45-2015-01, and FEDER funds through the Operational Programme Competitiveness Factors – COMPETE and national funds by FCT-Foundation for Science and Technology under the strategic project UID/NEU/04539/2013 (POCI-01-0145-FEDER-007440).

References 1.

Chazaud, C. and Yamanaka, Y. (2016) Lineage specification in the mouse preimplantation embryo. Development 143, 1063– 1074

17. Udagawa, O. et al. (2014) Mitochondrial fission factor Drp1 maintains oocyte quality via dynamic rearrangement of multiple organelles. Curr. Biol. 24, 2451–2458

2.

Schrode, N. et al. (2013) Anatomy of a blastocyst: cell behaviors driving cell fate choice and morphogenesis in the early mouse embryo. Genesis 51, 219–233

18. Ramalho-Santos, J. and Amaral, S. (2013) Mitochondria and mammalian reproduction. Mol. Cell. Endocrinol. 379, 74–84

3.

Tam, P.P. and Loebel, D.A. (2007) Gene function in mouse embryogenesis: get set for gastrulation. Nat. Rev. Genet. 8, 368–381

4.

Martin, G.R. (1981) Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc. Natl. Acad. Sci. U. S. A. 78, 7634– 7638

5.

Evans, M.J. and Kaufman, M.H. (1981) Establishment in culture of pluripotential cells from mouse embryos. Nature 292, 154– 156

19. Ramalho-Santos, J. et al. (2009) Mitochondrial functionality in reproduction: from gonads and gametes to embryos and embryonic stem cells. Hum. Reprod. Update 15, 553–572 20. Hashimoto, S. et al. (2017) Quantitative and qualitative changes of mitochondria in human preimplantation embryos. J. Assist. Reprod. Genet. 34, 573–580 21. Pereira, S.L. et al. (2014) From gametogenesis and stem cells to cancer: common metabolic themes. Hum. Reprod. Update 20, 924–943 22. Folmes, C.D.L. and Terzic, A. (2015) Metabolic determinants of embryonic development and stem cell fate. Reprod. Fertil. Dev. 27, 82–88

6.

Thomson, J.A. et al. (1998) Embryonic stem cell lines derived from human blastocysts. Science 282, 1145–1147

7.

Takahashi, K. and Yamanaka, S. (2006) Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663–676

8.

Takahashi, K. et al. (2007) Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131, 861– 872

9.

Brons, I.G.M. et al. (2007) Derivation of pluripotent epiblast stem cells from mammalian embryos. Nature 448, 191–195

25. Seagroves, T.N. et al. (2001) Transcription factor HIF-1 is a necessary mediator of the pasteur effect in mammalian cells. Mol. Cell. Biol. 21, 3436–3444

10. Tesar, P.J. et al. (2007) New cell lines from mouse epiblast share defining features with human embryonic stem cells. Nature 448, 196–199

26. Takehara, T. et al. (2012) Reduced oxygen concentration enhances conversion of embryonic stem cells to epiblast stem cells. Stem Cells Dev. 21, 1239–1249

11. Dejosez, M. and Zwaka, T.P. (2012) Pluripotency and nuclear reprogramming. Annu. Rev. Biochem. 81, 737–765

27. Rodrigues, A.S. et al. (2015) Dichloroacetate, the pyruvate dehydrogenase complex and the modulation of mESC pluripotency. PLoS One 10, e0131663

12. Engler, A.J. et al. (2006) Matrix elasticity directs stem cell lineage specification. Cell 126, 677–689 13. Downing, T.L. et al. (2013) Biophysical regulation of epigenetic state and cell reprogramming. Nat. Mater. 12, 1154–1162 14. Choi, B. et al. (2016) Stiffness of hydrogels regulates cellular reprogramming efficiency through mesenchymal-to-epithelial transition and stemness markers. Macromol. Biosci. 16, 199– 206 15. Ramalho-Santos, J. and Rodrigues, A.S. (2013) From oocytes and pluripotent stem cells to fully differentiated fates: (also) a mitochondrial odyssey. In Mitochondrial DNA, Mitochondria, Disease and Stem Cells (St. John, J.C., ed.), pp. 69–86, Humana Press 16. Liu, Q. et al. (2016) Mitofusin 2 regulates the oocytes development and quality by modulating meiosis and mitochondrial function. Sci. Rep. 6, 30561

14

Trends in Cell Biology, Month Year, Vol. xx, No. yy

23. Zhou, W. et al. (2012) HIF1a induced switch from bivalent to exclusively glycolytic metabolism during ESC-to-EpiSC/hESC transition. EMBO J. 31, 2103–2116 24. Carbognin, E. et al. (2016) Stat3 promotes mitochondrial transcription and oxidative respiration during maintenance and induction of naive pluripotency. EMBO J. 35, 618–634

28. Rodrigues, A.S. et al. (2015) Differentiate or die: 3-bromopyruvate and pluripotency in mouse embryonic stem cells. PLoS One 10, e0135617 29. Varum, S. et al. (2009) Enhancement of human embryonic stem cell pluripotency through inhibition of the mitochondrial respiratory chain. Stem Cell Res. 3, 142–156 30. Pereira, S.L. et al. (2013) Inhibition of mitochondrial complex III blocks neuronal differentiation and maintains embryonic stem cell pluripotency. PLoS One 8, 1–16 31. Spitkovsky, D. et al. (2004) Activity of complex III of the mitochondrial electron transport chain is essential for early heart muscle cell differentiation. FASEB J. 18, 1300–1302 32. Chung, S. et al. (2007) Mitochondrial oxidative metabolism is required for the cardiac differentiation of stem cells. Nat Clin Pract Cardiovasc Med 4, 1–12

TICB 1462 No. of Pages 16

33. Lunt, S.Y. and Vander Heiden, M.G. (2011) Aerobic glycolysis: meeting the metabolic requirements of cell proliferation. Annu. Rev. Cell Dev. Biol. 27, 441–464

58. Yanez, L.Z. et al. (2016) Human oocyte developmental potential is predicted by mechanical properties within hours after fertilization. Nat. Commun. 7, 10809

34. Moussaieff, A. et al. (2015) Concise review: energy metabolites: key mediators of the epigenetic state of pluripotency. Stem Cells 33, 2374–2380

59. Andolfi, L. et al. (2016) Investigating the mechanical properties of zona pellucida of whole human oocytes by atomic force spectroscopy. Integr. Biol. 8, 886–893

35. Wellen, K.E. et al. (2009) ATP-citrate lyase links cellular metabolism to histone acetylation. Science 324, 1076–1080

60. Maître, J.-L. et al. (2015) Pulsatile cell-autonomous contractility drives compaction in the mouse embryo. Nat. Cell Biol. 17, 849– 855

36. Moussaieff, A. et al. (2015) Glycolysis-mediated changes in acetyl-CoA and histone acetylation control the early differentiation of embryonic stem cells. Cell Metab. 21, 392–402 37. Carey, B.W. et al. (2015) Intracellular a-ketoglutarate maintains the pluripotency of embryonic stem cells. Nature 518, 413–416 38. Ware, C.B. et al. (2009) Histone deacetylase inhibition elicits an evolutionarily conserved self-renewal program in embryonic stem cells. Cell Stem Cell 4, 359–369 39. Lee, J.-H. et al. (2004) Histone deacetylase activity is required for embryonic stem cell differentiation. Genesis 38, 32–38 40. Correia, M. et al. (2017) Sirtuins in metabolism, stemness and differentiation. Biochim. Biophys. Acta Gen. Subj. 1861 (1 Pt A), 3444–3455 41. Ryall, J.G. et al. (2015) Metabolic reprogramming of stem cell epigenetics. Cell Stem Cell 17, 651–662 42. Tsumura, A. et al. (2006) Maintenance of self-renewal ability of mouse embryonic stem cells in the absence of DNA methyltransferases Dnmt1, Dnmt3a and Dnmt3b. Genes Cells 11, 805–814 43. Jackson, M. et al. (2004) Severe global DNA hypomethylation blocks differentiation and induces histone hyperacetylation in embryonic stem cells. Mol. Cell. Biol. 24, 8862–8871

61. Larue, L. et al. (1994) E-cadherin null mutant embryos fail to form a trophectoderm epithelium. Proc. Natl. Acad. Sci. U. S. A. 91, 8263–8267 62. Fierro-González, J.C. et al. (2013) Cadherin-dependent filopodia control preimplantation embryo compaction. Nat. Cell Biol. 15, 1424–1433 63. Maître, J.-L. (2017) Mechanics of blastocyst morphogenesis. Biol. Cell 109, 323–338 64. Ducibella, T. et al. (1977) Changes in cell surface and cortical cytoplasmic organization during early embryogenesis in the preimplantation mouse embryo. J. Cell Biol. 74, 153–167 65. Anani, S. et al. (2014) Initiation of Hippo signaling is linked to polarity rather than to cell position in the pre-implantation mouse embryo. Development 141, 2813–2824 66. Samarage, C.R. et al. (2015) Cortical tension allocates the first inner cells of the mammalian embryo. Dev. Cell 34, 435–447 67. Maître, J.-L. et al. (2016) Asymmetric division of contractile domains couples cell positioning and fate specification. Nature 536, 344–348 68. Hiramatsu, R. et al. (2013) External mechanical cues trigger the establishment of the anterior-posterior axis in early mouse embryos. Dev. Cell 27, 131–144

44. Son, M.J. et al. (2013) Interference with the mitochondrial bioenergetics fuels reprogramming to pluripotency via facilitation of the glycolytic transition. Int. J. Biochem. Cell Biol. 45, 2512– 2518

69. Matsuo, I. and Hiramatsu, R. (2016) Mechanical perspectives on the anterior-posterior axis polarization of mouse implanted embryos. Mech. Dev. 144 (Pt A), 62–70

45. Varum, S. et al. (2011) Energy metabolism in human pluripotent stem cells and their differentiated counterparts. PLoS One 6, e20914

70. Bedzhov, I. et al. (2015) Development of the anterior-posterior axis is a self-organizing process in the absence of maternal cues in the mouse embryo. Cell Res. 25, 1368–1371

46. Zhu, S. et al. (2010) Reprogramming of human primary somatic cells by OCT4 and chemical compounds. Cell Stem Cell 7, 651–655

71. Bedzhov, I. and Zernicka-Goetz, M. (2014) Self-organizing properties of mouse pluripotent cells initiate morphogenesis upon implantation. Cell 156, 1032–1044

47. Sousa, M.I. et al. (2015) Mitochondrial mechanisms of metabolic reprogramming in proliferating cells. Curr. Med. Chem. 22, 2493–2504

72. Ferreira, R.R. and Vermot, J. (2016) The balancing roles of mechanical forces during left-right patterning and asymmetric morphogenesis. Mech. Dev. 144 (Pt A), 71–80

48. Chen, X. et al. (2018) Insoluble microenvironment facilitating the generation and maintenance of pluripotency. Tissue Eng. B Rev. 24, 267–278

73. Sutherland, A.E. (2016) Tissue morphodynamics shaping the early mouse embryo. Semin. Cell Dev. Biol. 55, 89–98

49. Caiazzo, M. et al. (2016) Defined three-dimensional microenvironments boost induction of pluripotency. Nat. Mater. 15, 344– 352

74. Ingber, D.E. (2006) Mechanical control of tissue morphogenesis during embryological development. Int. J. Dev. Biol. 50, 255– 266

50. Folmes, C.D.L. et al. (2011) Somatic oxidative bioenergetics transitions into pluripotency-dependent glycolysis to facilitate nuclear reprogramming. Cell Metab. 14, 264–271

75. Mammoto, T. et al. (2013) Mechanobiology and developmental control. Annu. Rev. Cell Dev. Biol. 29, 27–61

51. Prigione, A. et al. (2014) HIF1a modulates cell fate reprogramming through early glycolytic shift and upregulation of PDK1-3 and PKM2. Stem Cells 32, 364–376 52. Gupta, S.K. (2015) Role of zona pellucida glycoproteins during fertilization in humans. J. Reprod. Immunol. 108, 90–97 53. Clift, D. and Schuh, M. (2013) Restarting life: fertilization and the transition from meiosis to mitosis. Nat. Rev. Mol. Cell Biol. 14, 549–562 54. Drobnis, E.Z. et al. (1988) Biophysical properties of the zona pellucida measured by capillary suction: is zona hardening a mechanical phenomenon? J. Exp. Zool. 245, 206–219 55. Sun, Y. et al. (2003) Mechanical property characterization of mouse zona pellucida. IEEE Trans. Nanobiosci. 2, 279–286 56. Murayama, Y. et al. (2006) Mouse zona pellucida dynamically changes its elasticity during oocyte maturation, fertilization and early embryo development. Hum. Cell 19, 119–125 57. Khalilian, M. et al. (2010) Estimating Young’s modulus of zona pellucida by micropipette aspiration in combination with theoretical models of ovum. J. R. Soc. Interface 7, 687–694

76. Shahbazi, M.N. and Zernicka-Goetz, M. (2018) Deconstructing and reconstructing the mouse and human early embryo. Nat. Cell Biol. 20, 878–887 77. McCauley, H.A. and Wells, J.M. (2017) Pluripotent stem cellderived organoids: using principles of developmental biology to grow human tissues in a dish. Development 144, 958–962 78. Dahl-Jensen, S. and Grapin-Botton, A. (2017) The physics of organoids: a biophysical approach to understanding organogenesis. Development 144, 946–951 79. Gjorevski, N. et al. (2016) Designer matrices for intestinal stem cell and organoid culture. Nature 539, 560–564 80. Warmflash, A. et al. (2014) A method to recapitulate early embryonic spatial patterning in human embryonic stem cells. Nat. Methods 11, 847–854 81. Rivron, N. (2018) Formation of blastoids from mouse embryonic and trophoblast stem cells. Protoc. Exch. Published online May 14, 2018. http://dx.doi.org/10.1038/protex.2018.051 82. Rivron, N.C. et al. (2018) Blastocyst-like structures generated solely from stem cells. Nature 557, 106–111

Trends in Cell Biology, Month Year, Vol. xx, No. yy

15

TICB 1462 No. of Pages 16

83. Kolahi, K.S. et al. (2012) Effect of substrate stiffness on early mouse embryo development. PLoS One 7, e41717 84. Chowdhury, F. et al. (2010) Soft substrates promote homogeneous self-renewal of embryonic stem cells via downregulating cell-matrix tractions. PLoS One 5, e15655 85. Shimizu, T. et al. (2012) Dual inhibition of Src and GSK3 maintains mouse embryonic stem cells, whose differentiation is mechanically regulated by Src signaling. Stem Cells 30, 1394–1404 86. Evans, N.D. et al. (2009) Substrate stiffness affects early differentiation events in embryonic stem cells. Eur. Cell Mater. 18, 1–14 87. Shkumatov, A. et al. (2014) Matrix rigidity-modulated cardiovascular organoid formation from embryoid bodies. PLoS One 9, e94764 88. Zoldan, J. et al. (2011) The influence of scaffold elasticity on germ layer specification of human embryonic stem cells. Biomaterials 32, 9612–9621 89. Keung, A.J. et al. (2012) Soft microenvironments promote the early neurogenic differentiation but not self-renewal of human pluripotent stem cells. Integr. Biol. 100, 130–134 90. Chowdhury, F. et al. (2010) Material properties of the cell dictate stress-induced spreading and differentiation in embryonic stem cells. Nat. Mater. 9, 82–88 91. Zignego, D.L. et al. (2015) Mechanotransduction in primary human osteoarthritic chondrocytes is mediated by metabolism of energy, lipids, and amino acids. J. Biomech. 48, 4253–4261 92. Pasqualini, F.S. et al. (2016) Mechanotransduction and metabolism in cardiomyocyte microdomains. Biomed Res. Int. 2016, 4081638 93. Tung, J.C. et al. (2015) Tumor mechanics and metabolic dysfunction. Free Radic. Biol. Med. 79, 269–280 94. Helle, S.C.J. et al. (2017) Mechanical force induces mitochondrial fission. eLife 6, e30292 95. Bartolak-Suki, E. et al. (2015) Fluctuation-driven mechanotransduction regulates mitochondrial-network structure and function. Nat. Mater. 14, 1049–1057 96. Halder, G. et al. (2012) Transduction of mechanical and cytoskeletal cues by YAP and TAZ. Nat. Rev. Mol. Cell Biol. 13, 591–600 97. Low, B.C. et al. (2014) YAP/TAZ as mechanosensors and mechanotransducers in regulating organ size and tumor growth. FEBS Lett. 588, 2663–2670 98. Santinon, G. et al. (2016) Control of YAP/TAZ activity by metabolic and nutrient-sensing pathways. Trends Cell Biol. 26, 289–299

106. Cox, A.G. et al. (2016) Yap reprograms glutamine metabolism to increase nucleotide biosynthesis and enable liver growth. Nat. Cell Biol. 18, 886–896 107. Rice, A.J. et al. (2017) Matrix stiffness induces epithelial-mesenchymal transition and promotes chemoresistance in pancreatic cancer cells. Oncogenesis 6, e352 108. Nukuda, A. et al. (2015) Stiff substrates increase YAP-signalingmediated matrix metalloproteinase-7 expression. Oncogenesis 4, e165 109. Crilley, J.G. et al. (2003) Hypertrophic cardiomyopathy due to sarcomeric gene mutations is characterized by impaired energy metabolism irrespective of the degree of hypertrophy. J. Am. Coll. Cardiol. 41, 1776–1782 110. De Palma, S. et al. (2013) Changes in muscle cell metabolism and mechanotransduction are associated with myopathic phenotype in a mouse model of collagen VI deficiency. PLoS One 8, e56716 111. McCain, M.L. et al. (2014) Micromolded gelatin hydrogels for extended culture of engineered cardiac tissues. Biomaterials 35, 5462–5471 112. Lyra-Leite, D.M. et al. (2017) Mitochondrial function in engineered cardiac tissues is regulated by extracellular matrix elasticity and tissue alignment. Am. J. Physiol. Heart Circ. Physiol. 313, H757–H767 113. Bartolak-Suki, E. et al. (2017) Regulation of mitochondrial structure and dynamics by the cytoskeleton and mechanical factors. Int. J. Mol. Sci. 18, 1812 114. Zeller, K.S. et al. (2010) PI3-kinase p110alpha mediates beta1 integrin-induced Akt activation and membrane protrusion during cell attachment and initial spreading. Cell Signal. 22, 1838–1848 115. Cantley, L.C. (2002) The phosphoinositide 3-kinase pathway. Science 296, 1655–1657 116. Hu, H. et al. (2016) Phosphoinositide 3-kinase regulates glycolysis through mobilization of aldolase from the actin cytoskeleton. Cell 164, 433–446 117. Rainero, E. et al. (2015) Ligand-occupied integrin internalization links nutrient signaling to invasive migration. Cell Rep. 10, 398–413 118. Muranen, T. et al. (2017) Starved epithelial cells uptake extracellular matrix for survival. Nat. Commun. 8, 13989 119. Herzig, S. and Shaw, R.J. (2018) AMPK: guardian of metabolism and mitochondrial homeostasis. Nat. Rev. Mol. Cell Biol. 19, 121–135

99. Sorrentino, G. et al. (2014) Metabolic control of YAP and TAZ by the mevalonate pathway. Nat. Cell Biol. 16, 357–366

120. Bays, J.L. et al. (2017) Linking E-cadherin mechanotransduction to cell metabolism through force-mediated activation of AMPK. Nat. Cell Biol. 19, 724–731

100. Wang, Z. et al. (2014) Interplay of mevalonate and Hippo pathways regulates RHAMM transcription via YAP to modulate breast cancer cell motility. Proc. Natl. Acad. Sci. U. S. A. 111, E89–E98

121. Panciera, T. et al. (2016) Induction of expandable tissue-specific stem/progenitor cells through transient expression of YAP/TAZ. Cell Stem Cell 19, 725–737

101. Enzo, E. et al. (2015) Aerobic glycolysis tunes YAP/TAZ transcriptional activity. EMBO J. 34, 1349–1370

122. Lian, I. et al. (2010) The role of YAP transcription coactivator in regulating stem cell self-renewal and differentiation. Genes Dev. 24, 1106–1118

102. Wang, W. et al. (2015) AMPK modulates Hippo pathway activity to regulate energy homeostasis. Nat. Cell Biol. 17, 490–499

123. LaCroix, A.S. et al. (2015) Molecular-scale tools for studying mechanotransduction. Annu. Rev. Biomed. Eng. 17, 287–316

103. DeRan, M. et al. (2014) Energy stress regulates hippo-YAP signaling involving AMPK-mediated regulation of angiomotinlike 1 protein. Cell Rep. 9, 495–503

124. Wang, N. et al. (1993) Mechanotransduction across the cell surface and through the cytoskeleton. Science 260, 1124–1127

104. Mo, J.-S. et al. (2015) Cellular energy stress induces AMPKmediated regulation of YAP and the Hippo pathway. Nat. Cell Biol. 17, 500–510

125. Sheehy, S.P. and Parker, K.K. (2011) The role of mechanical forces in guiding tissue differentiation. In Tissue Engineering in Regenerative Medicine (Bernstein, H.S., ed.), pp. 77–97, Humana Press

105. Santinon, G. et al. (2018) dNTP metabolism links mechanical cues and YAP/TAZ to cell growth and oncogene-induced senescence. EMBO J. 37, e97780

126. Leckband, D.E. et al. (2011) Mechanotransduction at cadherinmediated adhesions. Curr. Opin. Cell Biol. 23, 523–530

16

Trends in Cell Biology, Month Year, Vol. xx, No. yy