Skip to content
naulon

No retraction, correction or concern found when last checked on Oct 21, 2026.

The emerging antioxidant paradigm of mesenchymal stem cell therapy

Rhian Stavely, Kulmira Nurgali

Stem Cells Translational Medicine · 2020 · Review · CC BY

doi:10.1002/sctm.19-0446

Abstract

Mesenchymal stem cells (multipotent stromal cells; MSCs) have been under investigation for the treatment of diverse diseases, with many promising outcomes achieved in animal models and clinical trials. The biological activity of MSC therapies has not been fully resolved which is critical to rationalizing their use and developing strategies to enhance treatment efficacy. Different paradigms have been constructed to explain their mechanism of action, including tissue regeneration, trophic/anti-inflammatory secretion, and immunomodulation. MSCs rarely engraft and differentiate into other cell types after in vivo administration. Furthermore, it is equivocal whether MSCs function via the secretion of many peptide/protein ligands as their therapeutic properties are observed across xenogeneic barriers, which is suggestive of mechanisms involving mediators conserved between species. Oxidative stress is concomitant with cellular injury, inflammation, and dysregulated metabolism which are involved in many pathologies. Growing evidence supports that MSCs exert antioxidant properties in a variety of animal models of disease, which may explain their cytoprotective and anti-inflammatory properties. In this review, evidence of the antioxidant effects of MSCs in in vivo and in vitro models is explored and potential mechanisms of these effects are discussed. These include direct scavenging of free radicals, promoting endogenous antioxidant defenses, immunomodulation via reactive oxygen species suppression, altering mitochondrial bioenergetics, and donating functional mitochondria to damaged cells. Modulation of the redox environment and oxidative stress by MSCs can mediate their anti-inflammatory and cytoprotective properties and may offer an explanation to the diversity in disease models treatable by MSCs and how these mechanisms may be conserved between species. Significance statement The role of mesenchymal stem cells (MSCs) in ameliorating oxidative and nitrosative injury has received considerable attention in recent years. The reduction-oxidation (redox) environment regulates many physiological and pathophysiological mechanisms in cellular biology. Oxidative stress and redox imbalance are mediated by molecular constituents that are present in all living cells and share similar functions. The ability of MSCs to regulate these processes may offer an explanation to the diversity of disease models treatable by MSCs and to the effects of MSCs conserved between species. In this review, evidence of direct and indirect antioxidant mechanisms of MSC therapies is explored.

Full text

From PubMed Central, U.S. National Library of Medicine, as current on Oct 21, 2026. Paragraph numbers are ours, so an answer can point at one.

The role of mesenchymal stem cells (MSCs) in ameliorating oxidative and nitrosative injury has received considerable attention in recent years. The reduction‐oxidation (redox) environment regulates many physiological and pathophysiological mechanisms in cellular biology. Oxidative stress and redox imbalance are mediated by molecular constituents that are present in all living cells and share similar functions. The ability of MSCs to regulate these processes may offer an explanation to the diversity of disease models treatable by MSCs and to the effects of MSCs conserved between species. In this review, evidence of direct and indirect antioxidant mechanisms of MSC therapies is explored.

The role of mesenchymal stem cells (MSCs) in ameliorating oxidative and nitrosative injury has received considerable attention in recent years. The reduction‐oxidation (redox) environment regulates many physiological and pathophysiological mechanisms in cellular biology. Oxidative stress and redox imbalance are mediated by molecular constituents that are present in all living cells and share similar functions. The ability of MSCs to regulate these processes may offer an explanation to the diversity of disease models treatable by MSCs and to the effects of MSCs conserved between species. In this review, evidence of direct and indirect antioxidant mechanisms of MSC therapies is explored.

INTRODUCTION

Mesenchymal stem cells (multipotent stromal cells; MSCs) have been used as tools to treat a broad range of diseases in animal models due to their unique characteristics such as host immune evasion, rapid expansion, and their affluence in adult bone marrow and adipose tissue. The positive outcomes of these studies have driven hundreds of clinical trials into their application for diabetes, inflammatory disorders and various liver, kidney, lung, cardiovascular, musculoskeletal, neurological, and gastrointestinal diseases. 1 While several trials have demonstrated the therapeutic potential of MSCs, the failure to incorporate MSCs into current treatment regimens can be, in part, attributed to the lack of understanding pertaining to their biological mechanisms of action.

Initially, MSCs were explored as tools of regenerative medicine to replace damaged tissue. 2 However, administered MSCs were rarely observed to differentiate and effectively engraft into host tissues despite demonstrating favorable effects in many disease models. 3 Furthermore, the secretome of MSCs was identified to be therapeutic in many disease models in vitro and in vivo. Together, this resulted in a paradigm shift in recognition of the trophic actions of MSCs. 4 Despite extensive research investigating the anti‐inflammatory and trophic constituents of the MSC‐derived secretome, the therapeutic mechanisms of MSCs remain incompletely resolved. 5 MSCs demonstrate therapeutic attributes across xenogeneic barriers and, therefore, the therapeutic mechanisms of MSCs may be similar between species. There is strong evidence that the effects of MSCs are mediated via the secretion of protein/peptide ligands; however, it is equivocal whether these ligands are effective across xenogeneic barriers.

Recently, the role of MSCs in ameliorating oxidative and nitrosative injury has received considerable attention. The reduction‐oxidation (redox) environment regulates many physiological and pathophysiological mechanisms in cellular biology. Antioxidant effects of MSC therapy have been observed in various disease models such as diabetic injuries to the kidney, retina, sensory neurons, brain, and bone formation; chemotherapy‐ or radiation‐induced injury to the lungs, gonads, aorta, and brain; ischemic injury of the brain, heart, kidney, and liver; and traumatic injury to the spine and testis, cognitive disorders, gastrointestinal inflammation, septic injuries, and aging (Figure 1; Table 1). MSCs can directly reduce oxidative stress‐related injury in vitro in glial cells, neurons, cardiomyocytes, renal cells, endothelial cells, immune cells, hepatocytes, islet cells, fibroblasts, skeletal muscle, and other cells (Table 2). Oxidative stress is concomitant with cellular injury, inflammation, and dysregulated metabolism and, therefore, is a key pathophysiological mechanism of many diseases. Oxidative stress and redox imbalance are mediated by molecular constituents that are present in all living cells and share similar functions. Thus, the ability of MSCs to regulate these processes may offer an explanation to the diversity of disease models treatable by MSCs and to the effects of MSCs conserved between species.

Oxidative stress refers to a deviation from the physiological redox state and an increase in pro‐oxidants, or free radicals, that structurally change lipids, proteins, and DNA in a way that causes pathology or damage to a cell or tissue. 6 The most widely studied free radicals are reactive oxygen species (ROS), which can also include reactive molecules that have a stable charge. The three major endogenous ROS include the superoxide anion (O2.−), hydroxyl radical (•OH), and hydrogen peroxide (H2O2). 7, 8 O2.− is predominantly generated by nicotinamide adenine dinucleotide phosphate reduced (NADPH)‐oxidase (NOX) family enzymes or, by the mitochondria, as a by‐product of oxidative phosphorylation. 9 The level of mitochondria‐derived O2.− depends on metabolic substrates, cytosolic Ca2+ levels, pH, and oxygen tension. 10 O2.− generated from complexes of the electron transport chain (ETC) are highly reactive and can damage the mitochondrion. 11 The detoxification of O2.− into H2O2 is mediated by superoxide dismutase (SOD). 9 However, H2O2 can also be generated in various metabolic processes and by dual oxidases (DUOX). 12 While H2O2 is more stable than O2.−, its detoxification is crucial as it possesses a weak peroxide bond that makes it susceptible to reacting with metals, such as Fe2+, to generate reactive •OH through the Fenton reaction. 13 Both, H2O2 and O2.−, are diffusible across cell membranes and can promote cell death and inflammatory signaling. 14, 15 Several studies have demonstrated that MSCs can reduce ROS and biomarkers of oxidative stress. In this review, evidence of direct and indirect antioxidant mechanisms of MSC therapies is explored.

MSCs ARE RESISTANT AND RESPOND TO OXIDATIVE STRESS

The therapeutic properties of MSCs have been explored in many models of disease associated with high levels of ROS and biomarkers of oxidative injury. MSCs must survive these volatile environments to exert their therapeutic effects, which can present as a challenge for their engraftment after administration. Nonetheless, several studies have demonstrated that MSCs are highly resistant to oxidative insult. The oxidative effects of ionizing radiation are limited on MSCs which have been attributed to their ability to directly scavenge free radicals. 16 It has been demonstrated that MSCs are resistant to oxidative and nitrosative stimuli in vitro which is associated with constitutively expressed antioxidant enzymes SOD1, SOD2, catalase (CAT), and glutathione peroxidase (GPx), in addition to high levels of the antioxidant glutathione (GSH). 17 Depletion of GSH results in a loss of tolerance to oxidative stress. MSCs also constitutively express heat‐shock protein 70 (HSP70) and sirtuin (SIRT)3, 18 which may also play a role in the resistance of MSCs to oxidative/nitrosative injury. SIRT1 is also required for MSC survival against H2O2 and its overexpression has a protective effect. 19 Likewise, SIRT6 has been suggested to confer resistance to oxidative insult and basal ROS production in MSCs via downstream production of antioxidants including heme oxygenase‐1 (HO‐1). 20 Overexpression of HO‐1 ameliorates elevations in ROS and cellular senescence in SIRT6‐null MSCs and, therefore, appears to be a critical component of the survival mechanism of MSCs in oxidative environment. 20

In addition to wielding constitutive antioxidants, MSCs are also capable of significant adaptions in response to redox stress. MSCs exposed to lipopolysaccharide (LPS) produce oxidative and nitrosative free radicals. 18 In parallel, several adaptive processes are observed including the upregulation and/or nuclear translocation of redox‐sensitive factors (nuclear factor kappa‐B [NFκB], thioredoxin [TRX1], apurinic/apyrimidinic endonuclease redox effector factor‐1 [APE1/Ref‐1], nuclear factor erythroid 2‐related factor 2 [NRF2], forkhead box O3 [FOXO3]. and HO‐1), as well as mitochondrial remodeling and autophagy. Similarly, MSCs exposed to hypoxic conditions (1.5%‐2% O2) exhibit increased intracellular ROS and cells respond by upregulating the expression of hypoxia‐inducible factor 1 alpha (HIF‐1α), erythropoietin receptor, CAT, HO‐1, and antiapoptotic Bcl‐2 family of proteins. 21 Exposure of MSCs to hypoxia during in vitro culture can enhance their anti‐inflammatory, antioxidative, and cytoprotective properties. 21, 22, 23 This suggests that the ability of MSCs to tolerate and respond to oxidative environment may be critical to their engraftment and therapeutic efficacy at sites of tissue injury.

EFFECT OF MSCs ON OXIDATIVE STRESS BIOMARKERS AND ROS

In disease models, oxidative stress is typically quantified via biomarkers of oxidation to DNA and proteins or lipid peroxidation. Administration of MSCs has been demonstrated to reduce levels of one or more of these markers in a variety of animal models associated with oxidative stress (Table 1). Injection of MSCs themselves may not be critical to their antioxidant effects as administration of their conditioned medium (CM) also reduced lipid peroxidation in a model of ureteral obstruction‐induced kidney injury. 24 Administration of MSC‐derived exosomes was also effective to rescue protein oxidation and lipid peroxidation in animal models of septic and hyperglycemic brain injury and cognitive impairment. 25, 26 Likewise, DNA oxidation and lipid peroxidation caused by cisplatin‐induced kidney damage are alleviated by exosomes from human umbilical cord‐derived MSCs (UC‐MSCs); these results were confirmed in vitro with renal proximal tubular cells. 27 Treatments with MSCs were also demonstrated to reduce levels of ROS in animal models of diabetic retinopathy and nephropathy, severe acute pancreatitis, ureteral obstruction‐induced kidney damage, Alzheimer's disease, and metabolic renovascular disease. 24, 28, 29, 30, 31, 32 Specifically, MSC treatments have been shown to reduce levels of H2O2 in intestinal inflammation and several organs in a model of premature aging. 33, 34 MSCs also reduced levels of O2.− in colitis and spontaneous stroke. 34, 35

Typically, the reduction of oxidative stress markers by MSC treatments is associated with functional recovery and positive outcomes in animal models. The exception to this is the diversity of responses at various stages of hepatocarcinoma whereby antioxidant effects of MSCs reduce tumor burden at the early stages of disease by protecting the integrity of DNA but increase tumor progression at the late stages of the disease possibly by reducing ROS‐associated cell death. 36 The timing of treatments may also affect MSCs ability to attenuate oxidative stress as pretreatment with MSCs is more effective to prevent oxidative stress in septic lung injury and acute liver failure. 37, 38

Several studies have also investigated the therapeutic properties of MSCs on nitrosative stress which is particularly of interest in neural diseases. MSCs reduced the volatile peroxynitrite (ONOO−) in a model of intracerebral hemorrhage and nitrite levels in diabetic sensory neuropathy. 39, 40 In a model of radiation‐induced neurological complications, intranasal delivery of MSCs reduced inducible nitric oxide synthase (iNOS) expression and oxidative stress biomarkers, which are associated with improved cognitive performance and neuronal survival. 41 In a model of spinal cord injury, adipose tissue‐derived MSCs (AT‐MSCs) also demonstrated antioxidant activity with a reduction in lipid peroxidation and protein oxidation; however, no significant effects were observed for nitrosylation. 42 Furthermore, MSCs have been associated with increased nitric oxide (NO) in a model of Alzheimer's disease which may have been driven by preventing the loss of neuronal nitric oxide synthase (nNOS). 43 Alternatively, several studies observed a reduction in inflammation‐induced iNOS and nNOS expression after MSC treatment. 37, 44, 45, 46, 47 Overall, while the majority of studies support antioxidative effects of MSCs, their antinitrosative effects are unclear and likely to be disease and tissue specific. Furthermore, NO can be produced by nonhuman MSCs which is thought to be critical to their immunomodulatory function which may also explain these inconsistent effects on nitrosative stress. 42, 48

Although studies in cells and animal models unequivocally demonstrate that MSC treatments reduce levels of oxidative stress, albeit limited data exist from human studies. Nonetheless, favorable outcomes in a case study utilizing MSCs to treat the lungs of a subject previously exposed to sulfur mustard gas were attributed to the antioxidant properties of MSCs as evidenced by reduced lipid peroxidation levels in the sputum. 49

The antioxidant effects of MSC treatments are likely to be a specific property of these cells as they are more efficacious than hematopoietic stem cells and fibroblast at reducing oxidative stress in carbon tetrachloride‐induced‐liver injury and sepsis, respectively. 50, 51 Likewise, fibroblast exosomes have no effect on kidney injury‐induced by ischemia and chemotherapy. 27, 52

The alleviation of oxidative stress in animal models is associated with decreased pro‐inflammatory cytokines and markers of cellular death highlighting the close association between these processes. The precise mechanisms of in vivo MSC treatments are difficult to determine as cell death, inflammation, and oxidative stress occur concomitantly and perpetuate each other. However, a growing body of evidence suggests that MSCs have a direct role on suppressing oxidative stress and ROS production which may mediate their antiapoptotic and anti‐inflammatory effects.

ANTIOXIDATIVE MECHANISMS OF MSCs

The potential for MSCs to attenuate oxidative injury is unequivocally demonstrated by the reduction in ROS and biomarkers of oxidative stress in many disease models. Evidence form in vitro models suggests that MSCs directly protect cells from oxidative stimuli (Table 2). This is often associated with a reduction in ROS suggesting that MSCs avert the negative effects of oxidative stress by reducing the oxidative stimuli. The antioxidative effects of MSCs often occur in a paracrine manner in vitro and the administration of MSC‐conditioned medium (CM) can also reduce oxidative stress in vivo suggesting a paracrine component to their mechanism. 24 Nevertheless, others report that the antioxidant effects of MSCs can be cell contact dependent 53; albeit, these mechanisms could be disease and tissue dependent. Currently, MSCs have been proposed to reduce oxidative injury via scavenging free radicals, enhancing host antioxidant defenses, modulating the inflammatory response, augmenting cellular respiration and mitochondrial functions, or donating their mitochondria to protect damaged cells (Figure 1). 37, 50, 51, 53, 54

Antioxidant defense and scavenging

To maintain redox homeostasis and prevent excessive production of free radicals, cells rely on a complement of enzymatic antioxidants, including SODs, CAT, GPx, and small nonenzymatic antioxidants, such as GSH. After MSC treatments, the total antioxidant capacity of tissues are enhanced as observed in models of chemotherapy‐induced neuropathy, hyperglycemia‐induced cardiac damage, systemic sclerosis, hepatocarcinoma, and acute liver injury. 36, 55, 56, 57, 58 MSCs are receptive to oxidative stimuli and exhibit all necessary machinery to efficiently process ROS. 18, 59 Furthermore, media conditioned by MSCs have potent antioxidant capacity indicating that MSCs actively secrete antioxidants. 60 MSC‐CM has more effective antioxidant properties than CM from lung fibroblasts. 61 In several models of disease, MSC treatments upregulate the expression of antioxidant defense enzymes in vivo (Table 1). Therefore, the antioxidant effects of MSCs may be explained by their ability to directly scavenge free radicals and by enhancing antioxidant defenses in host tissues through upregulation of antioxidant enzymes.

Volatile O2 − is produced during cellular respiration by the mitochondria and NOX enzymes during tissue inflammation; O2 − is eliminated by SOD which catalyzes its conversion to H2O2. Antioxidant effects of MSC treatments have been associated with enhanced SOD activity or the expression of SODs in models of aging, age‐related erectile dysfunction, chemotherapy‐induced pulmonary fibrosis or gonadotoxicity colitis, pancreatitis, septic lung injury, arthritis, hepatotoxicity and hepatic ischemia reperfusion injury, Alzheimer's disease, and ovarian autografts. 33, 34, 37, 38, 43, 50, 62, 63, 64, 65, 66, 67, 68, 69, 70 MSCs secrete all isoforms of SOD including SOD1 and SOD2, which are archetypically not released extracellularly. 33, 59, 71, 72 Thus, it is difficult to interpret whether enhanced SOD activity and/or SOD expression is due to MSCs or host‐tissue‐derived SOD. In in vitro studies, MSC‐CM or MSCs in transwell cocultures promote SOD activity in tert‐Butyl hydroperoxide or UV‐exposed fibroblasts, H2O2‐treated neural stem cells or retinal ganglion cells, and dexamethasone‐induced muscle atrophy model. 60, 73, 74, 75, 76 These studies suggest that MSCs enhance SOD activity in cells exposed to oxidative stimuli. Likewise, MSC‐CM increases SOD1 expression in islet cells exposed to pro‐inflammatory cytokines and SOD2 in tert‐Butyl hydroperoxide‐treated umbilical endothelial cells. 77, 78 In endothelial cells, increased SOD2 expression was regulated by signal transducer and activator of transcription (STAT3) signaling and knockdown of either SOD2 or STAT3 decreased the antiapoptotic effects of the MSC‐CM. 77 These findings suggest that MSC upregulation of SOD in host tissues may be critical to their antioxidant effects. Alternatively, MSCs stimulated with TNF‐α and IFN‐γ were found to secrete high levels of SOD3, which was a major contributor to the antioxidant properties of MSCs in the amelioration of NO‐induced neuronal death in vitro. 71 Likewise, SOD3 expression in MSCs was necessary to suppress neutrophil respiratory burst and the accumulation in immune complex‐mediated dermal vasculitis. 79 Similarly, silencing of SOD2 in MSCs inhibits the antioxidant properties and therapeutic efficacy of their exosomes in hepatic I/R injury in vivo and H2O2‐treated human fetal hepatocytes which can be recovered by the addition of a SOD2 mimetic. 80

CAT and GPx are responsible for detoxifying H2O2 by its conversion to oxygen and water. MSCs secrete CAT and upregulation of CAT expression is associated with the therapeutic properties of MSCS in models of Friedreich's ataxia, radiation‐induced aortic injury, septic lung injury, and hepatoxicity. 33, 37, 50, 68, 81, 82 Likewise, upregulation of CAT associates with the antioxidant properties of MSCs in aging and colitis which parallels measurable reductions in H2O2. 33, 34 Exosomes derived from MSCs also express functional CAT. 31, 83 Inhibition of CAT suppresses their protective effects on amyloid‐β oligomer‐induced damage to hippocampal neurons indicating that this enzyme may also mediate antioxidant effects of MSCs. 83 MSC treatments have also been demonstrated to upregulate the expression of GPx in septic lung injury, severe acute pancreatitis, small bowel ischemia/reperfusion (I/R) injury, and Friedreich's ataxia. 37, 67, 81, 84 MSC‐CM can also enhance GPx activity in fibroblasts under oxidative stimuli. 60 Therefore, both CAT and GPx may play a role in MSC treatments in reducing oxidative stress by the elimination of H2O2 which was commonly used as a cell death inducing oxidative stimulus in in vitro studies.

The glutathione system is critical to scavenging ROS in animals, plants and fungi. GSH exerts its antioxidant effects by reducing free radicals and peroxides. During this process, glutathione disulfide (GSSG) is generated which is converted back to GSH by NADPH and is catalyzed by glutathione reductase. Increased levels of GSSG to GSH are suggestive of an oxidative redox state. Conversely, higher GSH‐to‐GSSG ratio is observed after MSC treatment in models of type 2 diabetes, in the hippocampus after chronic ethanol intake and hypoxia‐induced hypertension. 85, 86, 87 Likewise, MSCs improve GSH‐to‐GSSG ratio in sepsis which is not achieved by fibroblast. 51 Similarly, MSCs were found to increase GSH levels in diabetic models, septic lung injury, arthritis, hepatotoxicity, and epilepsy. 37, 38, 66, 68, 88, 89, 90 These effects were also observed by MSC‐CM in a model of unilateral ureteral obstruction which may suggest GSH was host‐tissue‐derived. 24 The effects of MSCs on GSH levels may be mediated by their ability to upregulate glutathione reductase which was observed in septic lung injury, acute pancreatitis, and I/R injury of the small bowel. 37, 67, 84 Notably, MSCs have also been shown to upregulate the expression of glutathione S‐transferases (GSTs) in cells and tissues which detoxify many damaging molecules that arise from redox imbalance such as peroxidized lipids. 88, 91 Glutathione reductase and GSTs are secreted by MSCs via exosomes. 31 While GSH has been the best studied scavenger in MSC treatments, they have also been demonstrated to upregulate the expression of the enzyme NAD(P)H quinone oxidoreductase 1 (NQO1) which detoxify quinones that contribute to the generation of ROS in septic lung injury, small bowel I/R injury in vivo, and cytokine‐exposed islet cell in vitro. 78, 84, 92

The multifunctional antioxidant HO‐1 has also been implicated in the therapeutic effects of MSC treatments. HO‐1's antioxidant and therapeutic mechanisms have been attributed to its ability to degrade heme, which is pro‐oxidative and the scavenging abilities of its products, biliverdin and bilirubin. HO‐1 expression is inducible in response to oxidative stress via nuclear factor erythroid 2‐related factor 2 (NRF2) which appears to be important in the adaptive responses of MSCs to inflammation and ROS. 18 Downregulation of NRF2 and HO‐1 has been observed after MSC treatment of chemotherapy‐induced pulmonary fibrosis and diabetic sensory neuropathy which may be explained by a reduction in oxidative stimuli. 40, 63 Conversely, the antioxidant effects of MSCs in several other models associate with upregulation of HO‐1 such as radiation‐induced aortic injury, septic lung injury, pancreatitis, and renal injuries caused by altered oxygen tensions and cisplatin. 30, 82, 92, 93, 94 HO‐1 was determined to partially contribute to the effects of MSCs in pancreatitis as its inhibition with zinc protoporphyrin negated some of the effects of MSC treatments, including upregulation of CAT and increased SOD activity, which may indicate that these processes are downstream of HO‐1 activity. 30 After MSC treatment of small bowel I/R injury, a larger number of HO‐1 expressing cells were observed which did not appear to completely colocalize with engrafted MSCs suggesting that treatments may increase HO‐1 expression in cells of the host. 84 It has been demonstrated that overexpression of HO‐1 in MSCs enhances their therapeutic activity in septic lung injury which was attributed to its pro‐survival properties. 95 Others have reported that MSCs still respond efficiently to oxidative insult with silenced HO‐1 by upregulating GSH pathway enzymes. 96 Nevertheless, the CM of MSCs derived from HO‐1−/− mice are unable to attenuate cisplatin‐induced renal injury and therefore HO‐1 appears to have an important role in the antioxidant properties of the MSC secretome. 94

Together, these studies demonstrate that antioxidants secreted by MSCs and their ability to upregulate host antioxidant defenses contribute to the suppression of oxidative stress. The exosomes derived from MSCs appear to be particularly rich in machinery to process ROS and can include, but not limited to, GPx, GSTs, SOD1‐3, peroxiredoxin 1‐6, CAT, cytoglobin, prostaglandin‐endoperoxide synthase 1, peroxidasin, albumin, apolipoprotein E, glutathione‐disulfide reductase, and thioredoxin reductase 1‐2. 31, 80 Notwithstanding, recombinant application of factors secreted by MSCs, such as hepatocyte growth factor (HGF) and basic fibroblast growth factors (bFGF), has been demonstrated to upregulate Gpx1, CAT, and SOD activity via SIRT1 and FOXO1 during age‐related loss of ovarian function. 97 Mechanisms of antioxidant defense mediated by scavenging of ROS by MSCs and the host could occur independently or simultaneously and are likely to be disease‐specific.

Antioxidant effects on inflammation

Immune function is regulated by free radicals and the redox system; leukocytes and pro‐inflammatory mediators enhance the formation of free radicals and perturb the redox environment creating a positive feedback cycle. 98 The immunomodulatory action of MSCs is a well‐documented phenomenon; however, their role in the interactions between the immune system and oxidative stress is not fully understood. Oxidative and/or nitrosative free radicals unequivocally play a role in all grades of acute and chronic inflammation. At physiological levels, they act as cellular signals modifying function and initiating necessary cell death programs. However, excessive generation of free radicals and/or inadequate scavenging results in protein oxidation, lipid peroxidation, and DNA damage that can be detrimental both intrinsically to the cell and the surrounding microenvironment. The ROS and reactive nitrogen species involved can take many forms and be generated from a variety of sources. Large amounts of the highly reactive O2 − anion are generated from NOX expressed by innate leukocytes. 99 SOD catalyzes the conversion of O2 − to H2O2 which phagocytes and neutrophils use to generate hypochlorous acid (HOCl) via myeloperoxidase (MPO). 99 Collectively, this is referred to as respiratory burst, a crucial element of the bactericidal response and inflammatory signaling. Nonetheless, MPO activity and O2 − are also associated with various inflammatory diseases. In inflammatory bouts, there is often a parallel increase in the expression of iNOS in leukocytes and, thus, subsequent generation of the free radical NO. Nonimmune cells such as epithelial cells are also capable of expressing iNOS and NOX to generate NO and O2 −.

MSC treatments reduce inflammation and oxidative stress in colitis, pancreatitis, arthritis, sepsis, vasculitis, stroke, myocardial infarction, hyperoxic lung injury, and I/R injury of the kidneys and bowel (Table 1). These effects have included a reduction in inflammatory cytokines TNFα, IFNγ, interleukin (IL)‐1β, IL‐6, IL‐9, and IL‐4; decreased expression of ROS producing enzymes NOX, MPO, and iNOS; as well as a net reduction in the infiltration of immune cells such as neutrophils (Table 1). The anti‐inflammatory effects of MSCs in pancreatitis were partially dependent on their expression of the antioxidant pathway enzyme HO‐1. 30 Previously, it was demonstrated in a model of sepsis that MSC treatments can reduce pro‐inflammatory cytokines in the serum and normalize thiol/disulfide redox pairings responsible for free radical scavenging. 51 Decreased levels of IL‐1β and TNFα superseded restoration of redox homeostasis. This suggests that the aversion of oxidative injury was secondary to the immunomodulation of pro‐inflammatory signaling, at least in acute septic inflammation. Conversely, MSCs can directly reduce oxidative injury in many cell types in vitro; thus, it is likely that MSCs may also reduce oxidative stress in tissues by mechanism other than suppressing the immune system. 27, 54, 60, 75, 77, 100 This is highlighted in in vivo and organotypic ex vivo models of myocardial I/R injury where MSC‐derived exosomes ameliorated infarction injury without altering leukocyte recruitment. 101 In in vivo experiments, protein oxidation was reduced by MSC‐derived exosomes after 1 hour; neutrophils were yet to infiltrate into the tissue. After 24 hours, MSCs reduced peripheral blood leukocyte numbers and neutrophil infiltration into the myocardium; thus, the antioxidative activity of MSCs preceded signals recruiting leukocytes. 101 This suggests that MSCs can attenuate oxidative stress‐induced tissue injury first, which can limit the recruitment of immune cells and subsequent inflammation in this model. This may be mediated by their ability to suppress NOX1 and 2 on resident cells which are downregulated by MSC treatments in acute myocardial infarction, sepsis‐induced brain injury, acute ischemic stroke, I/R injury to kidneys, and small bowel which were all associated with reduced inflammation. 25, 84, 102, 103, 104

Neutrophils appear to be key mediators of oxidative stress in inflammation. These cells harbor an abundance of MPO, a major catalyst for hypochlorite and NO‐derived oxidants. 105, 106 MSCs attenuate the infiltration of neutrophils and reduce MPO levels in several disease models. 50, 107, 108 MSCs can also directly dampen the respiratory burst in neutrophils and suppress MPO activity required to produce free radical required for their pro‐inflammatory function which was dependent on SOD3 and occurs in a paracrine manner. 79, 80, 109 Likewise, MSCs can also directly decrease ROS and MPO in stimulated monocytes and macrophages which suppress their pro‐inflammatory phenotype. 110, 111 These data suggest that MSCs not only suppress the immune system to prevent oxidative injury, but also that their mechanism of immunosuppression is reliant on their antioxidant properties.

Cellular bioenergetics

Free radicals are produced by several metabolic processes and the mitochondria during cellular respiration. Dysfunction in mitochondria can cause cellular injury which is mediated through the generation of O2.− and proteins that initiate cellular apoptosis. Depolarization of the mitochondrial membrane potential (Δψm) is a hallmark of mitochondrial dysfunction leading to cell death. Hyperglycemia can also cause oxidative stress via several mechanisms including the formation of free radical as by‐products of glucose auto‐oxidation that deplete antioxidant defense and advanced glycation end products that induce cellular stress. In models of hyperglycemia, MSC treatments can reduce the expression of glucose and fatty acid transports in kidney and cardiac tissue cells, which prevents glucose transport and ROS generation. 29, 56 Therefore, the antioxidant effects of MSC treatments in models of diabetes may be downstream of glycemic control. Conversely, regulation of mitochondrial function and oxidative phosphorylation by MSCs has been implicated in several disease models. MSC treatments improve chemotherapy‐induced cognitive impairment, which associates with enhanced respiratory capacity of the mitochondria. 112 The effects of MSCs on the mitochondria appear to occur in a paracrine manner as MSC‐CM increases the oxygen consumption rate of hypoxia‐exposed neonatal porcine islet cells and LPS‐treated macrophages. 113, 114 Similarly, MSC‐derived extracellular vesicles suppress mitochondrial O2.− levels in H2O2‐treated human fetal hepatocytes, which is associated with a reduction in apoptosis. 80

The potential for MSCs to directly attenuate mitochondrial dysfunction has been demonstrated in an in vitro model of I/R injury in mouse ventricular myocytes. 54 Within 5 minutes of reperfusion, cells exhibited an exaggerated Δψm hyperpolarization, which was reduced by conditioning the reperfusion solution with MSCs. The exaggerated hyperpolarization was followed by a continuous depolarization in controls after 15 minutes which was also attenuated by the paracrine secretion of MSCs. Decay of the Δψm was likely a result of the mitochondrial permeability transition pore opening. The exaggerated hyperpolarization of the Δψm was also averted by a mitochondrial ROS scavenger which simultaneously decreased mitochondrial O2 − generation demonstrating the close relationship between these events. Similarly, MSC secretion decreased mitochondrial O2 −, which led to the suggestion that MSCs may also attenuate Δψm dysfunction via scavenging of O2 −.

Depolarization of the Δψm in cisplatin‐treated renal proximal tubular cells has also been reportedly attenuated using exosomes derived from UC‐MSCs. 27 In vitro, BM‐MSCs were demonstrated to upregulate uncoupling protein 2 (UCP2) transcription in H2O2‐treated alveolar basal epithelial adenocarcinoma cells which reduces the formation of mitochondria‐derived O2 − by lowering the proton‐motive force across the mitochondrial membrane and provides another potential mechanism for the alleviation of mitochondrial dysfunction. 100 This was regulated by the paracrine secretion of stanniocalcin‐1 by MSCs, which enhanced UCP2, correlating with cell survival and decreased ROS generation. MSCs secreted stanniocalcin‐1 may also attenuate inflammation as it decreases mitochondrial ROS and subsequent activation of the nucleotide‐binding oligomerization domain (NOD)‐like receptor protein 3 (NLRP3) inflammasome. 111 BM‐MSCs inhibited the activity of the NLRP3 inflammasome in primed macrophages which is responsible for recognizing damage‐associated molecular patterns and initiating the inflammatory cascade through activation and secretion of IL‐1β. 115, 116 MSCs have also been demonstrated to secrete the redox‐sensitive protein DJ‐1, which has established roles in maintaining mitochondrial biogenesis and respiratory chain efficiency and could potentially mediate the neuroprotective effects of the MSC secretome as shown in Parkinson's disease models. 117 Collectively, these studies demonstrate that MSCs can ameliorate mitochondrial dysfunction in a paracrine manner with diverse therapeutic outcomes.

Mitochondrial donation

Recently, a concept has emerged that MSCs may be able to alter oxidative phosphorylation and ROS generation in cells through donation of mitochondria themselves. Islam et al 118 observed mitochondrial transfer from human BM‐MSCs to alveolar epithelium in a mouse model of LPS‐induced lung injury. BM‐MSC administration attenuated decreased intracellular ATP in the alveoli caused by lung injury; notably, ATP (visualized by a molecular probe) was predominantly restored at the site of mitochondrial transfer and immediately surrounding alveoli. MSCs with a mutation in connexin 43, a protein involved in the formation of gap junctions, were unable to transfer mitochondria despite being functionally competent and subsequently did not restore ATP, surfactant secretion, or reduce leukocyte infiltration. This phenomenon only occurred in LPS‐exposed lungs indicating that mitochondrial transfer is dependent on stimulants from damaged tissues.

MSCs cultured in hyperoxic (21% O2: normoxic atmosphere) conditions produce high levels of mitochondrial O2.−, depolarize Δψm, and induce mitophagy. 119 Mitochondria are loaded into phagosomes and shuttled to the plasma membrane. 119 These effects were reduced by culturing MSCs closer to a normoxic oxygen concentration (5% O2: hypoxic atmosphere). Macrophages have been observed to phagocytose these vesicles containing the partially depolarized mitochondria, which can fuse with endogenous mitochondria in macrophages. This protects silica‐exposed macrophages by increasing their oxygen consumption rate and decreasing mitochondrial O2.− production. These effects could not be elicited when MSCs were substituted by human fibroblasts. 119 This suggests that mitochondrial transfer may be stimulated by oxidative stress in MSCs. Conversely, MSCs have also been demonstrated to engulf mitochondria from other somatic cells exposed to H2O2. MSCs degraded the engulfed mitochondria which stimulated HO‐1 expression, mitochondrial biogenesis in MSCs, and the transfer of functional MSCs to damaged cells. 53 Inhibition of mitophagy negated the cytoprotective effects of MSCs in other somatic cells, which suggests that MSC sensing of damaged mitochondria may mediate their therapeutic responses. 53 Supporting this, the cell contact‐dependent transfer of mitochondria from chemotherapy‐treated T lymphocytes to MSCs was also determined to be critical to their ability to decrease mitochondrial O2.− production and cell death in T lymphocytes. 120 Nonetheless, in this study, mitochondrial transfer appeared to be predominately unidirectional and very few MSC‐derived mitochondria were observed in T cells. This suggests that MSC sensing of mitochondria can promote therapeutic mechanisms other than mitochondrial donation. 120

Although extracellular vesicles can contain whole mitochondria, several studies suggest that the donation of mitochondria may be contact‐dependent. MSCs have been found to transfer mitochondria via tunneling nanotubes (TNT) to glucose‐deprived and hypoxia‐reoxygenated cardiomyocytes which prevented Δψm depolarization and cell apoptosis. Albeit inhibition of TNT formation only partially reversed the effects of MSCs indicating other cytoprotective mechanisms were still active. 121 Similarly, MSCs transfer mitochondria to chemotherapy‐treated endothelial cells, which appears to occur in a unidirectional manner, unlike in T lymphocytes. 120, 122 Miro1 is important to TNT formation and its overexpression in MSCs can enhance mitochondrial transfer. 123 The effects of MSC mitochondrial donation are sufficient to rescue cellular respiration, proliferation, and motility in mitochondria‐depleted osteosarcoma cells. 124 These effects can be maintained for 45 passages, which highlights the therapeutic potential of MSC‐derived mitochondria. 124 Exogenous application of mitochondria isolated from MSCs may also offer therapeutic benefit and are able to protect dexamethasone‐treated muscle cells form oxidative stress in vitro. 125 Albeit, delivery of MSC‐derived mitochondria in in vivo poses a challenge. While several studies have demonstrated contact‐dependent transfer of mitochondria between MSCs and other cells, MSCs have also been reported to donate mitochondria to LPS‐treated macrophages via secreted extracellular vesicles. 114 Exposure of macrophages to MSC‐derived exosomes promoted their indication to the type 2 phenotype, which exerted anti‐inflammatory effects after adoptive transfer in septic lung injury. 114 The effects of MSCs were dependent on enhancing mitochondrial function in macrophages and were inhibited by damaging mitochondria in MSCs and blocking extracellular vesicles. 114 The therapeutic use of MSCs to deliver functional mitochondria to damaged tissue is an intriguing concept and warrants further study; however, another recent advancement reported by Panfoli et al 126 suggests that the exosomes of MSCs are capable of oxidative phosphorylation independent of the mitochondria. Subsets of MSC‐derived exosomes isolated from the umbilical cord of term newborns were discovered to contain complexes of the ETC embedded in the membrane. These exosomes possessed an electrochemical membrane potential, consumed O2, and produced ATP. The therapeutic application of these exosomes is yet to be investigated; nonetheless, this may present a viable tool to restore dysfunctional oxidative phosphorylation and ATP synthesis in damaged cells.

CONCLUSION

The presented studies evidently demonstrate that MSCs exhibit antioxidant potential either directly via scavenging of ROS and donating mitochondria or indirectly by upregulation antioxidant defenses in other cells and altering cellular bioenergetics. These effects can occur in combination with the previously recognized trophic and vesicular components of the MSC secretome acting directly on regenerative pathways. Likewise, antioxidant and trophic pathways appear to mediate the cytoprotective effect of MSC treatments which are ROS dependent. MSCs have frequently been utilized in inflammatory diseases to modulate the immune response. In this context, immunosuppression can avert ROS generation which is generated by MPO and NOX enzymes as a part of the inflammatory response. However, MSCs have now been shown to exert immunosuppressive effects by dampening ROS production and enhancing mitochondrial function in macrophages and neutrophils. Therefore, the antioxidant properties confer a role in the trophic and anti‐inflammatory mechanisms of MSC therapy. Considering that oxidative stress is implicated in almost every disease, these antioxidant properties, along with regenerative capacity of MSC seretome, may explain why MSC treatments are useful for such a spectrum of seemingly unlinked pathologies (Figure 1). Future studies should seek to clarify disease‐specific nuances of the antioxidative mechanisms of MSCs and MSC‐derived products. Likewise, improving the antioxidant effects of MSCs by enhancing the expression of antioxidant enzymes or promoting mitochondrial donation may be useful to optimize MSC‐based therapies and improve outcomes.

CONFLICT OF INTEREST

The authors declared no potential conflicts of interest.

AUTHOR CONTRIBUTIONS

R.S.: conception and design, manuscript writing, final approval of manuscript. K.N.: conception and design, manuscript writing, financial support, final approval of manuscript.

References 148

  1. 1.

    Squillaro T, Peluso G, Galderisi U. Clinical trials with mesenchymal stem cells: an update. Cell Transplant. 2016;25(5):829‐848.

    Not in our library10.3727/096368915x689622

  2. 2.

    Minguell JJ, Erices A, Conget P. Mesenchymal stem cells. Exp Biol Med. 2001;226(6):507‐520.

    Not in our library10.1177/153537020122600603

  3. 3.

    Caplan AI, Dennis JE. Mesenchymal stem cells as trophic mediators. J Cell Biochem. 2006;98(5):1076‐1084.

    Not in our library10.1002/jcb.20886

  4. 4.

    van Poll D, Parekkadan B, Rinkes IB, et al. Mesenchymal stem cell therapy for protection and repair of injured vital organs. Cell Mol Bioeng. 2008;1(1):42‐50.

    Not in our library

  5. 5.
  6. 6.
  7. 7.

    Birben E, Sahiner UM, Sackesen C, Erzurum S, Kalayci O. Oxidative stress and antioxidant defense. World Allergy Organ J. 2012;5(1):9‐19.

    Not in our library10.1097/wox.0b013e3182439613

  8. 8.

    Schieber M, Chandel NS. ROS function in redox signaling and oxidative stress. Curr Biol. 2014;24(10):R453‐R462.

    Not in our library10.1016/j.cub.2014.03.034

  9. 9.

    Dan Dunn J, Alvarez LAJ, Zhang X, Soldati T. Reactive oxygen species and mitochondria: a nexus of cellular homeostasis. Redox Biol. 2015;6:472‐485.

    Not in our library10.1016/j.redox.2015.09.005

  10. 10.

    Aon MA, Cortassa S, O'Rourke B. Redox‐optimized ROS balance: a unifying hypothesis. Biochim Biophys Acta. 2010;1797(6–7):865‐877.

    Not in our library10.1016/j.bbabio.2010.02.016

  11. 11.

    Chen Y, Azad MB, Gibson SB. Superoxide is the major reactive oxygen species regulating autophagy. Cell Death Differ. 2009;16(7):1040‐1052.

    Not in our library10.1038/cdd.2009.49

  12. 12.

    De Deken X, Corvilain B, Dumont JE, et al. Roles of DUOX‐mediated hydrogen peroxide in metabolism, host defense, and signaling. Antioxid Redox Signal. 2014;20(17):2776‐2793.

    Not in our library10.1089/ars.2013.5602

  13. 13.

    Winterbourn CC. Toxicity of iron and hydrogen peroxide: the Fenton reaction. Toxicol Lett. 1995;82:969‐974.

    Not in our library10.1016/0378-4274(95)03532-x

  14. 14.

    Fisher AB. Redox signaling across cell membranes. Antioxid Redox Signal. 2009;11(6):1349‐1356.

    Not in our library10.1089/ars.2008.2378

  15. 15.

    Bienert GP, Schjoerring JK, Jahn TP. Membrane transport of hydrogen peroxide. Biochim Biophys Acta (BBA)—Biomembr. 2006;1758(8):994‐1003.

    Not in our library10.1016/j.bbamem.2006.02.015

  16. 16.

    Chen M‐F, Lin C‐T, Chen W‐C, et al. The sensitivity of human mesenchymal stem cells to ionizing radiation. Intern J Radiat Oncol Biol Phys. 2006;66(1):244‐253.

    Not in our library10.1016/j.ijrobp.2006.03.062

  17. 17.

    Valle‐Prieto A, Conget PA. Human mesenchymal stem cells efficiently manage oxidative stress. Stem Cells Dev. 2010;19(12):1885‐1893.

    Not in our library10.1089/scd.2010.0093

  18. 18.

    Gorbunov NV, Garrison BR, McDaniel DP, et al. Adaptive redox response of mesenchymal stromal cells to stimulation with lipopolysaccharide inflammagen: mechanisms of remodeling of tissue barriers in sepsis. Oxid Med Cell Longev. 2013;2013:1‐16.

    Not in our library10.1155/2013/186795

  19. 19.

    Liu T, Ma X, Ouyang T, et al. SIRT1 reverses senescence via enhancing autophagy and attenuates oxidative stress‐induced apoptosis through promoting p53 degradation. Int J Biol Macromol. 2018;117:225‐234.

    Not in our library10.1016/j.ijbiomac.2018.05.174

  20. 20.
  21. 21.
  22. 22.

    Qin HH, Filippi C, Sun S, Lehec S, Dhawan A, Hughes RD. Hypoxic preconditioning potentiates the trophic effects of mesenchymal stem cells on co‐cultured human primary hepatocytes. Stem Cell Res Ther. 2015;6(1):237.

    Not in our library10.1186/s13287-015-0218-7

  23. 23.

    Kim Y, Jin HJ, Heo J, et al. Small hypoxia‐primed mesenchymal stem cells attenuate graft‐versus‐host disease. Leukemia. 2018;32(12):2672‐2684.

    Not in our library10.1038/s41375-018-0151-8

  24. 24.

    Liu B, Ding F‐X, Liu Y, et al. Human umbilical cord‐derived mesenchymal stem cells conditioned medium attenuate interstitial fibrosis and stimulate the repair of tubular epithelial cells in an irreversible model of unilateral ureteral obstruction. Nephrol Ther. 2018;23(8):728‐736.

    Not in our library10.1111/nep.13099

  25. 25.

    Chang C‐L, Chen H‐H, Chen K‐H, et al. Adipose‐derived mesenchymal stem cell‐derived exosomes markedly protected the brain against sepsis syndrome induced injury in rat. Am J Transl Res. 2019;11(7):3955‐3971.

    Not in our library

  26. 26.
  27. 27.
  28. 28.

    Ezquer M, Urzua CA, Montecino S, Leal K, Conget P, Ezquer F. Intravitreal administration of multipotent mesenchymal stromal cells triggers a cytoprotective microenvironment in the retina of diabetic mice. Stem Cell Res Ther. 2016;7:42‐42.

    Not in our library10.1186/s13287-016-0299-y

  29. 29.

    Lv S, Cheng J, Sun A, et al. Mesenchymal stem cells transplantation ameliorates glomerular injury in streptozotocin‐induced diabetic nephropathy in rats via inhibiting oxidative stress. Diabetes Res Clin Pract. 2014;104(1):143‐154.

    Not in our library10.1016/j.diabres.2014.01.011

  30. 30.

    Ma Z, Song G, Zhao D, et al. Bone marrow‐derived mesenchymal stromal cells ameliorate severe acute pancreatitis in rats via hemeoxygenase‐1‐mediated anti‐oxidant and anti‐inflammatory effects. Cytotherapy. 2019;21(2):162‐174.

    Not in our library10.1016/j.jcyt.2018.11.013

  31. 31.
  32. 32.

    Yan Y, Ma T, Gong K, Ao Q, Zhang X, Gong Y. Adipose‐derived mesenchymal stem cell transplantation promotes adult neurogenesis in the brains of Alzheimer's disease mice. Neural Regen Res. 2014;9(8):798‐805.

    Not in our library10.4103/1673-5374.131596

  33. 33.

    Xie C, Jin J, Lv X, Tao J, Wang R, Miao D. Anti‐aging effect of transplanted amniotic membrane mesenchymal stem cells in a premature aging model of Bmi‐1 deficiency. Sci Rep. 2015;5(1):13975.

    Not in our library10.1038/srep13975

  34. 34.

    Jung K‐J, Lee KW, Park CH, et al. Mesenchymal stem cells decrease oxidative stress in the bowels of Interleukin‐10 knockout mice. Gut Liver. 2020;14(1):100‐107.

    Not in our library10.5009/gnl18438

  35. 35.

    Calió ML, Marinho DS, Ko GM, et al. Transplantation of bone marrow mesenchymal stem cells decreases oxidative stress, apoptosis, and hippocampal damage in brain of a spontaneous stroke model. Free Radic Biol Med. 2014;70:141‐154.

    Not in our library10.1016/j.freeradbiomed.2014.01.024

  36. 36.

    Zong C, Zhang H, Yang X, et al. The distinct roles of mesenchymal stem cells in the initial and progressive stage of hepatocarcinoma. Cell Death Dis. 2018;9(3):345.

    Not in our library10.1038/s41419-018-0366-7

  37. 37.

    Shalaby SM, Amal S, Abd‐Allah SH, et al. Mesenchymal stromal cell injection protects against oxidative stress in Escherichia coli‐induced acute lung injury in mice. Cytotherapy. 2014;16(6):764‐775.

    Not in our library10.1016/j.jcyt.2013.12.006

  38. 38.

    Liu Z, Meng F, Li C, et al. Human umbilical cord mesenchymal stromal cells rescue mice from acetaminophen‐induced acute liver failure. Cytotherapy. 2014;16(9):1207‐1219.

    Not in our library10.1016/j.jcyt.2014.05.018

  39. 39.

    Chen M, Li X, Zhang X, et al. The inhibitory effect of mesenchymal stem cell on blood‐brain barrier disruption following intracerebral hemorrhage in rats: contribution of TSG‐6. J Neuroinflammation. 2015;12:61‐61.

    Not in our library10.1186/s12974-015-0284-x

  40. 40.

    Evangelista AF, Vannier‐Santos MA, de Assis Silva GS, et al. Bone marrow‐derived mesenchymal stem/stromal cells reverse the sensorial diabetic neuropathy via modulation of spinal neuroinflammatory cascades. J Neuroinflammation. 2018;15(1):189.

    Not in our library10.1186/s12974-018-1224-3

All 148 references

Cited by papers we hold 5

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.