Crosstalk between Signaling Pathways and Energy Metabolism in Pluripotency

Abstract

The sequential change from totipotency to multipotency occurs during early mammalian embryo development. However, due to the lack of cellular models to recapitulate the distinct potency of stem cells at each stage, their molecular and cellular characteristics remain ambiguous. The establishment of isogenic naïve and primed pluripotent stem cells to represent the pluripotency in the inner cell mass of the pre-implantation blastocyst and in the epiblast from the post-implantation embryo allows the understanding of the distinctive characteristics of two different states of pluripotent stem cells. This review discusses the prominent disparities between naïve and primed pluripotency, including signaling pathways, metabolism, and epigenetic status, ultimately facilitating a comprehensive understanding of their significance during early mammalian embryonic development.

Introduction

Cell signaling pathways governing pluripotency, marker gene expressions, and glucose metabolism are cellular and molecular features that differ between mouse and human embryonic stem cells (ESCs). These discrepancies arise from the unique characteristics exhibited by the cellular states of pre- and post-implantation embryos rather than variations among species (1). ESCs and epiblast stem cells (EpiSCs) are pluripotent stem cells (PSCs) derived from the pre-implantation inner cell mass (ICM) and post-implantation epiblasts of embryos, respectively. Naïve and primed pluripotency represent two distinct PSC states, mirroring the distinct pluripotency observed in pre- and post-implanted embryos (1, 2). Given the significant environmental changes and outcomes after implantation, glycolytic metabolism preference (3-6), epigenetic specification (7, 8), loss of chimeric potential (9), and other marked differences in cellular features in primed PSCs reflect that EpiSCs in post-implantation embryos are prone to differentiation (i.e., ‘primed’ to differentiation).

Unlike EpiSCs, which undergo drastic germ-line differentiation, ESCs from the ICM of pre-implantation blastocysts (residing in naïve pluripotency) maintain pluripotency to resist differentiation. This resilience of naïve PSCs is attributed to capacitation, acting as a barrier that exits pluripotency, impeding progression toward an intermediate state (10, 11). The discernible variations in propensity are particularly evident when considering cultural conditions. Notably, the use of fetal bovine serum (FBS) to sustain pluripotency exemplifies the contrasting behaviors of mouse ESCs (naïve-like) as opposed to human ESCs (primed-like). Thus, exploring the molecular mechanisms of how naïve and primed PSC models harbor each molecular feature akin to ESCs and EpiSCs can expand our understanding of early embryo development in mammals, including humans. As such, this review specifically focuses on signaling and metabolic pathway differences.

Main Text

Signaling pathways regulating pluripotency

Naïve and primed PSCs display unique molecular signatures when cultured in vitro (12, 13). In general, leukemia inhibitory factor (LIF) combined with bone morphogenetic protein 4 (BMP4) predominantly sustains naïve pluripotency (14). Additionally, chemically inhibiting mitogen-activated protein kinase kinase 1 (MEK1) and glycogen synthase kinase 3β (GSK3β) by two inhibitors (i.e., PD0325901 for inhibitor of MEK1 [iMEK1] and CHIR99021 for inhibitor of GSK3β [iGSK3β]) (hereafter LIF/2i) is a standardized culture protocol to stabilize the naïve state and prevent differentiation (Fig. 1A) (1, 15, 16). The requirement of simultaneous inhibition of MEK1 and GSK3β for naïve pluripotency maintenance recapitulates the role of Netrin-1, family of laminin-related secreted proteins, in inhibiting these signaling pathways in the pre-implanted embryos (15). On the other hand, primed pluripotency relies on fibroblast growth factor 2 (FGF2) and Activin signaling pathways, which induce ESC differentiation upon activation (Fig. 1B) (17, 18). Similarly, the distinctive characteristics of each pluripotent state are upheld through critical signaling pathways. This section explores key signals that regulate naïve and primed pluripotency, primarily focusing on Janus kinase/signal transducer and activator of transcription (JAK/STAT), MEK1/extracellular signal-regulated kinase 1 (ERK1) and ERK2, and Wingless and INT-1 (WNT)/β-catenin pathways.

Fig. 1.

Graphical illustration of main signaling pathways in naïve (A) and primed (B) pluripotent stem cells (PSCs). (A) (i) BMP4 mediates active phosphorylation of Smad1/5/8 to form protein complex with Smad4 to induce Dusp9, encoding Mkp4 (a dual phosphatase for MAPK) to inactivate Erk1/2. (ii) LIF binds to gp130/LIFR and activates JAK/STAT3 pathway to induce naïve pluripotency. Upon LIF mediated LIFR phosphorylation recruits Zap70 and Shp2 serving as both negative feedback of JAK/STAT3 signaling and a signaling transducer to Ras/Raf toward Erk1/2 activation. PKC and SRC serve as upstream kinase transducing signals toward Erk1/2. (iii) Wnt binding to LRP and Fz receptor inhibits β-catenin destruction complex (i.e., Gsk3β/Axin1/APC) and stabilizes β-catenin for Tcf3 inhibition. Plasma membrane localization of β-catenin/Oct4 in a complex with E-cadherin is found in naïve PSCs. CHIR99021 as a Gsk3β inhibitor; PD0325901 as a Mek1 inhibitor. (B) (i) Fgf2 binding to Fgf2 receptor recruits Shp2 to the plasma membrane, to serve a binding site for Grb2. Sos activation by Grb2 binding activates Ras and the downstream signaling which leads to lineage specification. PKC and SRC serve as upstream kinase transducing signals toward Erk1/2. Activin binding to Activin Receptor (ActR) induces Smad2/3 phosphorylation. Gray-colored proteins refer to inhibited proteins.

JAK/STAT signaling pathway: Since the initial establishment of mouse ESCs (mESCs), LIF supplementation is often used to maintain mESCs in vitro (1, 19, 20). JAK/STAT3 signaling, downstream LIF signaling, has been widely recognized as a vital signaling pathway for maintaining and inducing naïve pluripotency (21, 22). Notably, the zeta-chain-associated protein kinase 70 (ZAP70), a T cell-specific non-receptor tyrosine kinase, is highly expressed in mESCs (19). Since ZAP70 negatively modulates JAK/STAT signaling upon T-cell receptor (TCR) signaling (23), the loss of ZAP70 and subsequent uncoupling of Src homology 2 domain-containing phosphatase 1 (SHP1) or SHP2 protein phosphatase from signaling complex, disrupts mESCs’ differentiation potential by stimulating JAK/STAT3 signaling (19, 24). Similarly, the loss of SHP2, directly dephosphorylating active JAK phosphorylation (25), encourages JAK/STAT3 signaling to favor pluripotency maintenance of mESCs (26) and naïve ESCs (but not primed ESCs) in vitro (27).

MEK/ERK1 and 2 signaling pathway: Other than activating JAK/STAT3 signaling, protein kinase C (PKC) (28), SRC (29), and ERK1/2 (28) inhibition enhances mESC self-renewal. As PKC and SRC serve as primary kinase transducing signals toward ERK1/2, these studies highlight the prerequisite of constant inhibition of ERK1/2 activity for naïve pluripotency. Likewise, the BMP4-dependent effects on mESC self-renewal result from induction of DUSP9 (a dual specific phosphatase for ERK1/2), negatively modulating ERK1/2 (30). SHP2, whose genetic perturbation favors mESCs self-renewal by activating JAK/STAT3 as a tyrosine phosphatase (26), also functions as a fundamental adapter protein to transmit the receptor signal toward MEK1/ERK1 and 2 (31). Thus, SHP2 loss decouples the signal from LIF receptor activation toward MEK/ERK, positively affecting naïve pluripotency (27). ERK1/2-dependent octamer-binding transcription factor 4 (OCT4) (32), NANOG (33), and krüppel-like factor 4 (KLF4) (34) phosphorylation interferes with these transcription factors for pluripotency and promotes endoderm specification (35). These findings demonstrate that suppressing ERK1/2 signaling is critical for promoting naïve pluripotency (36). Thus, diverse MEK1 inhibitors, such as SU5402, PD184352, and PD0325901, are currently used to maintain naïve pluripotency in vitro.

On the other hand, mouse EpiSCs (primed-like) from the implanted epiblasts are established in vitro through FGF and Activin supplementation (1, 17). The FGF4 requirement is underscored by the multiple evidence that the absence of key ERK1/2 signaling mediators from FGF4, such as GRB2 (37), FGF2 receptor (38), FGF4 (39), and SHP2 (40), promotes embryonic lethality after implantation and induces defective trophoblast differentiation (41). Based on the live reporter system for ERK activity, this recent study reveals that FGF to ERK1/2 signaling occurs in a spatial and temporal lineage-specific manner in pre-implanted embryos. Unlike constant primitive endoderm and trophectoderm activation, sporadic ERK1/2 pulse activation is observed in ESCs in ICM (42). Given the vital functions of mitogen-activated protein kinase (MAPK), including ERK1/2, in lineage priming or determination (43), the failure of primed convertsion from naïve pluripotency in mESCs lacking SHP2, indicates that ERK1/2 activity is crucial for not only lineage determination and but also maintaining primed pluripotency after implantation (27).

Wnt/β-Catenin pathway: WNT, a mammalian Drosophila wingless (wg) homolog that severely impairs fly embryo development if mutated, exhibits similar signaling behavior between vertebrates and invertebrates. Furthermore, WNT is a critical molecule for tissue-specific stem cells during mammalian development (44). As previously mentioned, Netrin-1 mediates GSK3β inhibition to stabilize β-catenin in blastocysts (15). This chemical inhibition of GSK3β, which eventually enhances WNT signaling through β-catenin-dependent gene expressions, is a standardized protocol for inducing and maintaining naïve pluripotency along with iMEK1 (9, 45, 46). The subsequent cytoplasmic β-catenin stabilization promotes pluripotency through TCF3-dependent gene expressions (47-50) or an Oct4 and E-cadherin complex in the plasma membrane that modulates the pluripotency network (51). Nonetheless, Wnt/β-catenin pathway activation is essential for maintaining self-renewal and pluripotency in naïve PSCs and pre-implanted embryos.

Energy metabolism-regulating pluripotency

The embryo undergoes drastic metabolic changes throughout early development, particularly during peri-implantation periods. During these stages, the embryo relies on internal energy reserves until it establishes a continuous supply of nutrients from maternal sources post-implantation (52, 53). Essentially, the cellular machinery responsible for energy production in the embryo adapts based on the primary substrates consumed at each developmental phase (54). This adaptation predicates how embryos utilize various energy sources, including glucose, glutamine, and lipids. Furthermore, it also influences unique mitochondrial functions and their diverse pathways for energy production. This section focuses on the metabolic rearrangements in glucose, lipid, and mitochondrial metabolisms of the epiblast during embryo implantation and the drastic transition from naïve to primed pluripotency (Fig. 2).

Fig. 2.

Graphical illustration of main metabolism pathways in naïve (A) and primed (B) pluripotent stem cells (PSCs). (A) Bivalent metabolism with active glycolysis and OxPHOS contributes to ATP production. Excess glucose is converted and stored as glycogen by activation of Gys1 upon Gsk3β inhibition. Subsequent inhibition of AMPK by glycogen activates de novo fatty acid synthesis in naïve PSCs. (B) The attenuated mitochondrial metabolism in primed pluripotency is highlighted. Primed PSCs predominantly utilize glycolysis to generate ATP. The activation of AMPK due to depletion of glycogen inhibits overall fatty acid synthesis. However, certain fatty acids (e.g., MUFAs), are recognized for their crucial function in sustaining primed human PSCs. Decreased fatty acid oxidation by epigenetic suppression of CPT1 is a typical feature of primed PSCs. Gray-colored proteins refer to inhibited proteins. The unique mitochondrial structure distinguishing naïve from primed PSCs is also underlined.

Glucose metabolism: Upon implantation, there is a metabolic adaptation from a glucose-poor to a glucose-rich state (55). The profound metabolic shift is recapitulated in naïve and primed ESCs, wherein a transition from a bivalent metabolism in glycolysis and oxidative phosphorylation (OxPHOS) to a preference for glycolysis occurs (1, 3, 4, 6, 18, 45). Due to the high glucose demand of rapidly proliferating cells and the subsequent elevated glycolysis, this metabolic characteristic provides an advantage in generating ample biomass for amino acid, lipid, and nucleotide synthesis (56). This activity aligns with the higher proliferation rate observed in primed PSCs compared to naïve PSCs (57, 58), which is associated with the metabolic changes during the transition from naïve to primed pluripotency states (Fig. 2). The core transcription factors for pluripotency, such as Pou5f1 (encoding OCT4) (59, 60) and Esrrb (61), determine naïve specific glucose metabolism. Comparatively, Lin28A expression induces glycolytic dependence in primed PSCs (62).

Intracellular glycogen accumulation, a cellular storage of excess glucose, has been studied since the 1960s (63). During embryonic development, glycogen levels rise from the 2-cell stage to the early blastocyst stage but sharply decline after the late blastocyst stage (63). GSK3β, the inhibitory kinase for glycogen synthase, suppresses glycogen synthesis by inhibiting glycogen synthase 1 (GYS1) phosphorylation. Thus, iGSK3β’s consistent inhibition of GSK3β in naïve ESCs increases glycogen levels, unlike primed ESCs (4). Considering the active OxPHOS for energy production, the excess glucose due to the bivalent glucose metabolism in naïve ESCs may account for the naïve specific glycogen accumulation in undifferentiated cells (4). The subsequent study reveals that intracellular glycogen levels are not just a consequence of GSK3β inhibition. These levels also actively serve as a signaling molecule to inhibit AMP-activated protein kinase (AMPK) activation, a key kinase for intracellular energy sensors through direct binding to carbohydrate binding module (64), and the subsequent catabolism of fatty acids in naïve ESCs (5). Of note, the constant inhibition of AMPK by a high intracellular glycogen activates acetyl-CoA carboxylase (ACC), a prominent enzyme for de novo fatty acid synthesis. This occurs alongside an increased glucose demand in naïve ESCs, facilitating fatty acid anaplerosis, which is necessary for maintaining naïve pluripotency (Fig. 2A). Given the significance of fatty acids in ESCs and early embryo development (65-68), glycogen regulating fatty acid levels by inhibiting AMPK in naïve pluripotency suggests regular crosstalk between signaling pathways and energy metabolism (5).

Lipid metabolism: Lipid metabolism is intricately regulated, balancing catabolism (fatty acid oxidation, FAO) and anabolism (de novo fatty acid synthesis), tailored to the cell’s specific metabolic demands. Thus, comprehending this dynamic lipid metabolism can offer valuable insights into the cellular state, particularly during the transition from naïve to primed pluripotency (69, 70). The distinct lipid metabolism of ESCs compared to differentiated somatic cells, abundant metabolites derived from unsaturated fatty acids, is determined by mass spectrometry-based metabolomic profiles (71). The notable rise of long-carbon-chain lipids in primed ESCs or post-implanted embryos results from decreased FAO activity due to epigenetically suppressed carnitine acyltransferase 1 (CPT1), the initial step for mitochondrial FAO (72).

On the other hand, de novo fatty acid synthesis for pluripotency has been demonstrated through multiple examples. The activity of fatty acid synthase (FASN) (67) and stearoyl-CoA desaturate (SCD1), a key enzyme in monounsaturated fatty acid (MUFA) biosynthesis (73, 74), is crucial for human ESC (hESC) survival as chemically inhibiting either of these enzymes prompts massive cell death (Fig. 2B). Similarly, supplemental oleic acids (belonging to MUFA) or de novo fatty acid synthesis of MUFAs promotes cellular reprogramming (75) and pluripotency maintenance (66). Consistently, L-carnitine-induced FAO decreases fatty acid levels in embryos, improving pregnancy rates following embryo transfer in human (76) and bovine models (77). A few animals, especially cows, pigs, and cats, that require longer implantation times compared to mice have higher fatty acid levels in embryos (78, 79). These findings imply that intracellular fatty acids or lipids in embryos are not only an energy reservoir during peri-implantation but actively determine embryo development.

Interplay between signaling and metabolism for pluripotency: The metabolic transition from glycolysis in primed cells to bivalent metabolism in naïve ESCs, achieved through LIF/2i treatment, allows naïve ESCs to be independent of glutamine due to an active TCA cycle (80). In contrast, primed hESCs critically depend on glutamine for survival (81). Additionally, LIF/2i treatment alters glucose metabolism to produce α-ketoglutarate, a vital co-factor that supports active demethylation in naïve ESCs (82). Mitochondrial metabolism restoration, a prominent feature for naïve pluripotency, is primarily induced by Essrb expression and subsequent OxPHOS gene transcription (83) and activation (61).

Notably, Esrrb induction from GSK3β inhibition (84) and OxPHOS gene transcription through STAT3 activation (46) exemplifies the intricate signaling and metabolic interplay in naïve pluripotency. The diminished JAK/STAT3 signaling in ESCs from Oct4-null embryos, coupled with the downregulation of vital glycolysis enzymes like Hk2 and Pkm (85), indicates molecular crosstalk between signaling and cellular energy metabolisms in pluripotency initiation and maintenance. The activation of enzymes in glycolysis through an OCT4-dependent induction of SIRT2, a deacetylase (3), increases glycolysis cellular reprogramming for human induced PSCs (iPSCs). This highlights the essential roles of acetate or acetyl-CoA in glycolysis during histone acetylation of pluripotent genes (86).

Conclusion

Our understanding of cellular and molecular events during peri-implantation in embryos has been significantly augmented through multiple isogenic cellular models for naïve and primed PSCs in mice and humans. In addition, expanded potential stem cells (EPSCs) (87), totipotent stem cells (88), or 2-cell-like stem cells (89) from PSCs successfully recapitulated the early embryonic stage. This potentiated further induction of extraembryonic lineages (e.g., trophoblast and primitive endoderm) from PSCs that compose the entire embryo to support epiblasts during embryo development (90-92). Recent advancements in 3D embryo-like structures, such as blastoids (93), gastruloids (94), or peri-gastrulation embryo models (95-98) run parallel to our efforts to comprehend the unrevealed biology of very early life stages. These innovations are crucial for overcoming the challenges associated with the limited availability of mice or human embryo resources.

Funding Statement

Funding This research was supported by Korean Fund for Regenerative Medicine funded by Ministry of Science and ICT, and Ministry of Health and Welfare (Grant number RS-2022-00070316 and RS-2023-00218543), Republic of Korea.