The development, use, and challenges of electromechanical tissue stimulation systems

BACKGROUND

Tissue stimulations greatly affect cell growth, phenotype, and function, and they play an important role in modeling tissue physiology. With the goal of understanding the cellular mechanisms underlying the response of tissues to external stimulations, in vitro models of tissue stimulation have been developed in hopes of recapitulating in vivo tissue function.

METHODS

Herein we review the efforts to create and validate tissue stimulators responsive to electrical or mechanical stimulation including tensile, compression, torsion, and shear.

RESULTS

Engineered tissue platforms have been designed to allow tissues to be subjected to selected types of mechanical stimulation from simple uniaxial to humanoid robotic stain through equal-biaxial strain. Similarly, electrical stimulators have been developed to apply selected electrical signal shapes, amplitudes, and load cycles to tissues, lending to usage in stem cell-derived tissue development, tissue maturation, and tissue functional regeneration. Some stimulators also allow for the observation of tissue morphology in real-time while cells undergo stimulation. Discussion on the challenges and limitations of tissue simulator development is provided.

CONCLUSIONS

Despite advances in the development of useful tissue stimulators, opportunities for improvement remain to better reproduce physiological functions by accounting for complex loading cycles, electrical and mechanical induction coupled with biological stimuli, and changes in strain affected by applied inputs.

Inhibition of the ubiquitin-proteasome system reduces the abundance of pyruvate dehydrogenase kinase 1 in cultured myotubes

Pyruvate dehydrogenase kinase (PDK), which phosphorylates the pyruvate dehydrogenase complex, regulates glucose metabolism in skeletal muscle. PDK1, an isozyme whose expression is controlled by hypoxia-inducible factor-1α (HIF-1α), is thought to play a role in muscle adaptation to hypoxia. While transcriptional upregulation of PDK1 by HIF-1α is well characterised, mechanisms controlling proteolysis of PDK1 in skeletal muscle have not been thoroughly investigated. Proteasome inhibitor MG132 paradoxically reduced the abundance of PDK1 in human cancer cells and rat L6 myotubes, suggesting that MG132 might direct PDK1 towards autophagic degradation. The objectives of our current study were to determine (1) whether MG132 suppresses PDK1 levels in primary human myotubes, (2) whether chloroquine, an inhibitor of autophagy, prevents MG132-induced suppression of PDK1 in L6 myotubes, and (3) whether PYR-41, an inhibitor of ubiquitination, suppresses PDK1 in L6 myotubes. Using qPCR and/or immunoblotting, we found that despite markedly upregulating HIF-1α protein, MG132 did not alter the PDK1 expression in cultured primary human myotubes, while it suppressed both PDK1 mRNA and protein in L6 myotubes. The PDK1 levels in L6 myotubes were suppressed also during co-treatment with chloroquine and MG132. PYR-41 markedly increased the abundance of HIF-1α in primary human and L6 myotubes, while reducing the abundance of PDK1. In L6 myotubes treated with PYR-41, chloroquine increased the abundance of the epidermal growth factor receptor, but did not prevent the suppression of PDK1. Collectively, our results suggest that cultured myotubes degrade PDK1 via a pathway that cannot be inhibited by MG132, PYR-41, and/or chloroquine.

Aberrant evoked calcium signaling and nAChR cluster morphology in a SOD1 D90A hiPSC-derived neuromuscular model

Familial amyotrophic lateral sclerosis (ALS) is a progressive neuromuscular disorder that is due to mutations in one of several target genes, including SOD1. So far, clinical records, rodent studies, and in vitro models have yielded arguments for either a primary motor neuron disease, or a pleiotropic pathogenesis of ALS. While mouse models lack the human origin, in vitro models using human induced pluripotent stem cells (hiPSC) have been recently developed for addressing ALS pathogenesis. In spite of improvements regarding the generation of muscle cells from hiPSC, the degree of maturation of muscle cells resulting from these protocols has remained limited. To fill these shortcomings, we here present a new protocol for an enhanced myotube differentiation from hiPSC with the option of further maturation upon coculture with hiPSC-derived motor neurons. The described model is the first to yield a combination of key myogenic maturation features that are consistent sarcomeric organization in association with complex nAChR clusters in myotubes derived from control hiPSC. In this model, myotubes derived from hiPSC carrying the SOD1 D90A mutation had reduced expression of myogenic markers, lack of sarcomeres, morphologically different nAChR clusters, and an altered nAChR-dependent Ca2+ response compared to control myotubes. Notably, trophic support provided by control hiPSC-derived motor neurons reduced nAChR cluster differences between control and SOD1 D90A myotubes. In summary, a novel hiPSC-derived neuromuscular model yields evidence for both muscle-intrinsic and nerve-dependent aspects of neuromuscular dysfunction in SOD1-based ALS.

Importance of the electrophoresis and pulse energy for siRNA-mediated gene silencing by electroporation in differentiated primary human myotubes

BACKGROUND

Electrotransfection is based on application of high-voltage pulses that transiently increase membrane permeability, which enables delivery of DNA and RNA in vitro and in vivo. Its advantage in applications such as gene therapy and vaccination is that it does not use viral vectors. Skeletal muscles are among the most commonly used target tissues. While siRNA delivery into undifferentiated myoblasts is very efficient, electrotransfection of siRNA into differentiated myotubes presents a challenge. Our aim was to develop efficient protocol for electroporation-based siRNA delivery in cultured primary human myotubes and to identify crucial mechanisms and parameters that would enable faster optimization of electrotransfection in various cell lines.

RESULTS

We established optimal electroporation parameters for efficient siRNA delivery in cultured myotubes and achieved efficient knock-down of HIF-1α while preserving cells viability. The results show that electropermeabilization is a crucial step for siRNA electrotransfection in myotubes. Decrease in viability was observed for higher electric energy of the pulses, conversely lower pulse energy enabled higher electrotransfection silencing yield. Experimental data together with the theoretical analysis demonstrate that siRNA electrotransfer is a complex process where electropermeabilization, electrophoresis, siRNA translocation, and viability are all functions of pulsing parameters. However, despite this complexity, we demonstrated that pulse parameters for efficient delivery of small molecule such as PI, can be used as a starting point for optimization of electroporation parameters for siRNA delivery into cells in vitro if viability is preserved.

CONCLUSIONS

The optimized experimental protocol provides the basis for application of electrotransfer for silencing of various target genes in cultured human myotubes and more broadly for electrotransfection of various primary cell and cell lines. Together with the theoretical analysis our data offer new insights into mechanisms that underlie electroporation-based delivery of short RNA molecules, which can aid to faster optimisation of the pulse parameters in vitro and in vivo.

Assessment of Cell Viability in Electrically Excitable Muscle Cells Through Intact Twitch Stimulation

Muscle cells (i.e. skeletal muscle fibers) are fully viable and functional when their excitation-contraction (EC) coupling machinery is intact. This involves intact membrane integrity with polarized membrane, functional ion channels for action potential generation and conduction, an intact electro-chemical interface at the level of the fiber's triad, followed by sarcoplasmic reticulum Ca²⁺ release, and subsequent activation of the chemico-mechanical interface at the level of the contractile apparatus. The ultimate end result is then a visible twitch contraction upon a brief electrical pulse stimulation. For many biomedical studies involving single muscle cells, intact and viable myofibers are of utmost importance. Thus, a simple global screening method that involves a brief electrical stimulus applied to single muscle fibers and assessment of visible contraction would be of high value. In this chapter, we describe step-by-step protocols to (i) obtain intact single muscle fibers from freshly dissected muscle tissue using an enzymatic digestion procedure and (ii) provide a workflow for the assessment of twitch response of single fibers that can be ultimately classified as viable. For this, we have prepared a unique stimulation pen for which we provide the fabrication guide for do-it-yourself rapid prototyping to eliminate the need for expensive specialized commercial equipment.

Insulin, dibutyryl-cAMP, and glucose modulate expression of patatin-like domain containing protein 7 in cultured human myotubes

Expression of patatin-like phospholipase domain containing protein 7 (PNPLA7), also known as neuropathy target esterase-related esterase (NRE), a lysophospholipase, increases with fasting and decreases with feeding in mouse skeletal muscle, indicating it is regulated by insulin, counterregulatory hormones, such as glucocorticoids and catecholamines, and/or nutrients. In cultured mouse adipocytes insulin reduces Pnpla7 expression, underscoring the possibility that insulin regulates PNPLA7 in skeletal muscle. The first aim of this study was to establish whether PNPLA7 is functionally expressed in cultured human skeletal muscle cells. The second aim was to determine whether PNPLA7 is regulated by insulin, glucocorticoids, cAMP/protein kinase A pathway, and/or glucose. Cultured human skeletal muscle cells expressed PNPLA7 mRNA and protein. Gene silencing of PNPLA7 in myoblasts reduced the phosphorylation of 70 kDa ribosomal protein S6 kinase and ribosomal protein S6 as well as the abundance of α1-subunit of Na⁺,K⁺-ATPase and acetyl-CoA carboxylase, indirectly suggesting that PNPLA7 is functionally important. In myotubes, insulin suppressed PNPLA7 mRNA at 1 g/L glucose, but not at low (0.5 g/L) or high (4.5 g/L) concentrations. Treatment with synthetic glucocorticoid dexamethasone and activator of adenylyl cyclase forskolin had no effect on PNPLA7 regardless of glucose concentration, while dibutyryl-cAMP, a cell-permeable cAMP analogue, suppressed PNPLA7 mRNA at 4.5 g/L glucose. The abundance of PNPLA7 protein correlated inversely with the glucose concentrations. Collectively, our results highlight that PNPLA7 in human myotubes is regulated by metabolic signals, implicating a role for PNPLA7 in skeletal muscle energy metabolism.

Energy metabolism in skeletal muscle cells from donors with different body mass index

Obesity and physical inactivity have a profound impact on skeletal muscle metabolism. In the present work, we have investigated differences in protein expression and energy metabolism in primary human skeletal muscle cells established from lean donors (BMI<25 kg/m2) and individuals with obesity (BMI>30 kg/m2). Furthermore, we have studied the effect of fatty acid pretreatment on energy metabolism in myotubes from these donor groups. Alterations in protein expression were investigated using proteomic analysis, and energy metabolism was studied using radiolabeled substrates. Gene Ontology enrichment analysis showed that glycolytic, apoptotic, and hypoxia pathways were upregulated, whereas the pentose phosphate pathway was downregulated in myotubes from donors with obesity compared to myotubes from lean donors. Moreover, fatty acid, glucose, and amino acid uptake were increased in myotubes from individuals with obesity. However, fatty acid oxidation was reduced, glucose oxidation was increased in myotubes from subjects with obesity compared to cells from lean. Pretreatment of myotubes with palmitic acid (PA) or eicosapentaenoic acid (EPA) for 24 h increased glucose oxidation and oleic acid uptake. EPA pretreatment increased the glucose and fatty acid uptake and reduced leucine fractional oxidation in myotubes from donors with obesity. In conclusion, these results suggest that myotubes from individuals with obesity showed increased fatty acid, glucose, and amino acid uptake compared to cells from lean donors. Furthermore, myotubes from individuals with obesity had reduced fatty acid oxidative capacity, increased glucose oxidation, and a higher glycolytic reserve capacity compared to cells from lean donors. Fatty acid pretreatment enhances glucose metabolism, and EPA reduces oleic acid and leucine fractional oxidation in myotubes from donor with obesity, suggesting increased metabolic flexibility after EPA treatment.

Insight Into the Metabolic Adaptations of Electrically Pulse-Stimulated Human Myotubes Using Global Analysis of the Transcriptome and Proteome

Electrical pulse stimulation (EPS) has proven to be a useful tool to interrogate cell-specific responses to muscle contraction. In the present study, we aimed to uncover networks of signaling pathways and regulatory molecules responsible for the metabolic effects of exercise in human skeletal muscle cells exposed to chronic EPS. Differentiated myotubes from young male subjects were exposed to EPS protocol 1 (i.e. 2 ms, 10 V, and 0.1 Hz for 24 h), whereas myotubes from middle-aged women and men were exposed to protocol 2 (i.e. 2 ms, 30 V, and 1 Hz for 48 h). Fuel handling as well as the transcriptome, cellular proteome, and secreted proteins of EPS-treated myotubes from young male subjects were analyzed using a combination of high-throughput RNA sequencing, high-resolution liquid chromatography-tandem mass spectrometry, oxidation assay, and immunoblotting. The data showed that oxidative metabolism was enhanced in EPS-exposed myotubes from young male subjects. Moreover, a total of 81 differentially regulated proteins and 952 differentially expressed genes (DEGs) were observed in these cells after EPS protocol 1. We also found 61 overlapping genes while comparing the DEGs to mRNA expression in myotubes from the middle-aged group exposed to protocol 2, assessed by microarray. Gene ontology (GO) analysis indicated that significantly regulated proteins and genes were enriched in biological processes related to glycolytic pathways, positive regulation of fatty acid oxidation, and oxidative phosphorylation, as well as muscle contraction, autophagy/mitophagy, and oxidative stress. Additionally, proteomic identification of secreted proteins revealed extracellular levels of 137 proteins were changed in myotubes from young male subjects exposed to EPS protocol 1. Selected putative myokines were measured using ELISA or multiplex assay to validate the results. Collectively, our data provides new insight into the transcriptome, proteome and secreted proteins alterations following in vitro exercise and is a valuable resource for understanding the molecular mechanisms and regulatory molecules mediating the beneficial metabolic effects of exercise.

Microphysiological system for studying contractile differences in young, active, and old, sedentary adult derived skeletal muscle cells

Microphysiological systems (MPS), also referred to as tissue chips, incorporating 3D skeletal myobundles are a novel approach for physiological and pharmacological studies to uncover new medical treatments for sarcopenia. We characterize a MPS in which engineered skeletal muscle myobundles derived from donor-specific satellite cells that model aged phenotypes are encapsulated in a perfused tissue chip platform containing platinum electrodes. Our myobundles were derived from CD56+ myogenic cells obtained via percutaneous biopsy of the vastus lateralis from adults phenotyped by age and physical activity. Following 17 days differentiation including 5 days of a 3 V, 2 Hz electrical stimulation regime, the myobundles exhibited fused myotube alignment and upregulation of myogenic, myofiber assembly, signaling and contractile genes as demonstrated by gene array profiling and localization of key components of the sarcomere. Our results demonstrate that myobundles derived from the young, active (YA) group showed high intensity immunofluorescent staining of α-actinin proteins and responded to electrical stimuli with a ~1 μm displacement magnitude compared with non-stimulated myobundles. Myobundles derived from older sedentary group (OS) did not display a synchronous contraction response. Hypertrophic potential is increased in YA-derived myobundles in response to stimulation as shown by upregulation of insulin growth factor (IGF-1), α-actinin (ACTN3, ACTA1) and fast twitch troponin protein (TNNI2) compared with OS-derived myobundles. Our MPS mimics disease states of muscle decline and thus provides an aged system and experimental platform to investigate electrical stimulation mimicking exercise regimes and may be adapted to long duration studies of compound efficacy and toxicity for therapeutic evaluation against sarcopenia.

Novel integrated workflow allows production and in-depth quality assessment of multifactorial reprogrammed skeletal muscle cells from human stem cells

Skeletal muscle tissue engineering aims at generating biological substitutes that restore, maintain or improve normal muscle function; however, the quality of cells produced by current protocols remains insufficient. Here, we developed a multifactor-based protocol that combines adenovector (AdV)-mediated MYOD expression, small molecule inhibitor and growth factor treatment, and electrical pulse stimulation (EPS) to efficiently reprogram different types of human-derived multipotent stem cells into physiologically functional skeletal muscle cells (SMCs). The protocol was complemented through a novel in silico workflow that allows for in-depth estimation and potentially optimization of the quality of generated muscle tissue, based on the transcriptomes of transdifferentiated cells. We additionally patch-clamped phenotypic SMCs to associate their bioelectrical characteristics with their transcriptome reprogramming. Overall, we set up a comprehensive and dynamic approach at the nexus of viral vector-based technology, bioinformatics, and electrophysiology that facilitates production of high-quality skeletal muscle cells and can guide iterative cycles to improve myo-differentiation protocols.

Branched-Chain Amino Acid Deprivation Decreases Lipid Oxidation and Lipogenesis in C2C12 Myotubes

Impaired lipid metabolism is a common risk factor underlying several metabolic diseases such as metabolic syndrome and type 2 diabetes. Branched-chain amino acids (BCAAs) that include valine, leucine and isoleucine have been proven to share a role in lipid metabolism and hence in maintaining metabolic health. We have previously introduced a hypothesis suggesting that BCAA degradation mechanistically connects to lipid oxidation and storage in skeletal muscle. To test our hypothesis, the present study examined the effects of BCAA deprivation and supplementation on lipid oxidation, lipogenesis and lipid droplet characteristics in murine C2C12 myotubes. In addition, the role of myotube contractions on cell metabolism was studied by utilizing in vitro skeletal-muscle-specific exercise-like electrical pulse stimulation (EPS). Our results showed that the deprivation of BCAAs decreased both lipid oxidation and lipogenesis in C2C12 myotubes. BCAA deprivation further diminished the number of lipid droplets in the EPS-treated myotubes. EPS decreased lipid oxidation especially when combined with high BCAA supplementation. Similar to BCAA deprivation, high BCAA supplementation also decreased lipid oxidation. The present results highlight the role of an adequate level of BCAAs in healthy lipid metabolism.

Phosphorylation of Na+,K+-ATPase at Tyr10 of the α1-Subunit is Suppressed by AMPK and Enhanced by Ouabain in Cultured Kidney Cells

Na⁺,K⁺-ATPase (NKA) is essential for maintenance of cellular and whole-body water and ion homeostasis. In the kidney, a major site of ion transport, NKA consumes ~ 50% of ATP, indicating a tight coordination of NKA and energy metabolism. AMP-activated protein kinase (AMPK), a cellular energy sensor, regulates NKA by modulating serine phosphorylation of the α1-subunit, but whether it modulates other important regulatory phosphosites, such as Tyr10, is unknown. Using human kidney (HK-2) cells, we determined that the phosphorylation of Tyr10 was stimulated by the epidermal growth factor (EGF), which was opposed by inhibitors of Src kinases (PP2), tyrosine kinases (genistein), and EGF receptor (EGFR, gefitinib). AMPK activators AICAR and A-769662 suppressed the EGF-stimulated phosphorylation of EGFR (Tyr1173) and NKAα1 at Tyr10. The phosphorylation of Src (Tyr416) was unaltered by AICAR and increased by A-769662. Conversely, ouabain (100 nM), a pharmacological NKA inhibitor and a putative adrenocortical hormone, enhanced the EGF-stimulated Tyr10 phosphorylation without altering the phosphorylation of EGFR (Tyr1173) or Src (Tyr416). Ouabain (100–1000 nM) increased the ADP:ATP ratio, while it suppressed the lactate production and the oxygen consumption rate in a dose-dependent manner. Treatment with ouabain or gene silencing of NKAα1 or NKAα3 subunit did not activate AMPK. In summary, AMPK activators and ouabain had antagonistic effects on the phosphorylation of NKAα1 at Tyr10 in cultured HK-2 cells, which implicates a role for Tyr10 in coordinated regulation of NKA-mediated ion transport and energy metabolism.

Higher glucose availability augments the metabolic responses of the C2C12 myotubes to exercise-like electrical pulse stimulation

The application of exercise-like electrical pulse simulation (EL-EPS) has become a widely used exercise mimetic in vitro. EL-EPS produces similar physiological responses as in vivo exercise, while less is known about the detailed metabolic effects. Routinely, the C2C12 myotubes are cultured in high-glucose medium (4.5 g/L), which may alter EL-EPS responses. In this study, we evaluate the metabolic effects of EL-EPS under the high- and low-glucose (1.0 g/L) conditions to understand how substrate availability affects the myotube response to EL-EPS. The C2C12 myotube, media, and cell-free media metabolites were analyzed using untargeted nuclear magnetic resonance (NMR)-based metabolomics. Furthermore, translational and metabolic changes and possible exerkine effects were analyzed. EL-EPS enhanced substrate utilization as well as production and secretion of lactate, acetate, 3-hydroxybutyrate, and branched-chain fatty acids (BCFAs). The increase in BCFAs correlated with branched-chain amino acids (BCAAs) and BCFAs were strongly decreased when myotubes were cultured without BCAAs suggesting the action of acyl-CoA thioesterases on BCAA catabolites. Notably, not all EL-EPS responses were augmented by high glucose because EL-EPS increased phosphorylated c-Jun N-terminal kinase and interleukin-6 secretion independent of glucose availability. Administration of acetate and EL-EPS conditioned media on HepG2 hepatocytes had no adverse effects on lipolysis or triacylglycerol content. Our results demonstrate that unlike in cell-free media, the C2C12 myotube and media metabolites were affected by EL-EPS, particularly under high-glucose condition suggesting that media composition should be considered in future EL-EPS studies. Furthermore, acetate and BCFAs were identified as putative exerkines warranting more research.NEW & NOTEWORTHY The present study examined for the first time the metabolome of 1) C2C12 myotubes, 2) their growth media, and 3) cell-free media after exercise-like electrical pulse stimulation under distinct nutritional loads. We report that myotubes grown under high-glucose conditions had greater responsiveness to EL-EPS when compared with lower glucose availability conditions and increased media content of acetate and branched-chain fatty acids suggests they might act as putative exerkines warranting further research.

Activation of (un)regulated cell death as a new perspective for bispyridinium and imidazolium oximes

Oximes, investigated as antidotes against organophosphates (OP) poisoning, are known to display toxic effects on a cellular level, which could be explained beyond action on acetylcholinesterase as their main target. To investigate this further, we performed an in vitro cell-based evaluation of effects of two structurally diverse oxime groups at concentrations of up to 800 μM, on several cell models: skeletal muscle, kidney, liver, and neural cells. As indicated by our results, compounds with an imidazolium core induced necrosis, unregulated cell death characterized by a cell burst, increased formation of reactive oxygen species, and activation of antioxidant scavenging. On the other hand, oximes with a pyridinium core activated apoptosis through specific caspases 3, 8, and/or 9. Interestingly, some of the compounds exhibited a synergistic effect. Moreover, we generated a pharmacophore model for each oxime series and identified ligands from public databases that map to generated pharmacophores. Several interesting hits were obtained including chemotherapeutics and specific inhibitors. We were able to define the possible structural features of tested oximes triggering toxic effects: chlorine atoms in combination with but-2(E)-en-1,4-diyl linker and adding a second benzene ring with substituents such as chlorine and/or methyl on the imidazolium core. Such oximes could not be used in further OP antidote development research, but could be introduced in other research studies on new specific targets. This could undoubtedly result in an overall improved wider use of unexplored oxime database created so far in OP antidotes field of research in a completely new perspective.

Effect of differentiation, de novo innervation, and electrical pulse stimulation on mRNA and protein expression of Na+,K+-ATPase, FXYD1, and FXYD5 in cultured human skeletal muscle cells

Denervation reduces the abundance of Na⁺,K⁺-ATPase (NKA) in skeletal muscle, while reinnervation increases it. Primary human skeletal muscle cells, the most widely used model to study human skeletal muscle in vitro, are usually cultured as myoblasts or myotubes without neurons and typically do not contract spontaneously, which might affect their ability to express and regulate NKA. We determined how differentiation, de novo innervation, and electrical pulse stimulation affect expression of NKA (α and β) subunits and NKA regulators FXYD1 (phospholemman) and FXYD5 (dysadherin). Differentiation of myoblasts into myotubes under low serum conditions increased expression of myogenic markers CD56 (NCAM1), desmin, myosin heavy chains, dihydropyridine receptor subunit α_1S, and SERCA2 as well as NKAα2 and FXYD1, while it decreased expression of FXYD5 mRNA. Myotubes, which were innervated de novo by motor neurons in co-culture with the embryonic rat spinal cord explants, started to contract spontaneously within 7–10 days. A short-term co-culture (10–11 days) promoted mRNA expression of myokines, such as IL-6, IL-7, IL-8, and IL-15, but did not affect mRNA expression of NKA, FXYDs, or myokines, such as musclin, cathepsin B, meteorin-like protein, or SPARC. A long-term co-culture (21 days) increased the protein abundance of NKAα1, NKAα2, FXYD1, and phospho-FXYD1^Ser68 without attendant changes in mRNA levels. Suppression of neuromuscular transmission with α-bungarotoxin or tubocurarine for 24 h did not alter NKA or FXYD mRNA expression. Electrical pulse stimulation (48 h) of non-innervated myotubes promoted mRNA expression of NKAβ2, NKAβ3, FXYD1, and FXYD5. In conclusion, low serum concentration promotes NKAα2 and FXYD1 expression, while de novo innervation is not essential for upregulation of NKAα2 and FXYD1 mRNA in cultured myotubes. Finally, although innervation and EPS both stimulate contractions of myotubes, they exert distinct effects on the expression of NKA and FXYDs.