Sepsis-Induced Myopathy and Gut Microbiome Dysbiosis: Mechanistic Links and Therapeutic Targets

Abstract

Sepsis is currently defined as life-threatening organ dysfunction caused by a dysregulated host response to infection. The skeletal muscle system is among the host organ systems compromised by sepsis. The resulting neuromuscular dysfunction and impaired regenerative capacity defines sepsis-induced myopathy and manifests as atrophy, loss of strength, and hindered regeneration after injury. These outcomes delay recovery from critical illness and confer increased vulnerability to morbidity and mortality. The mechanisms underlying sepsis-induced myopathy, including the potential contribution of peripheral organs, remain largely unexplored. The gut microbiome is an immunological and homeostatic entity that interacts with and controls end-organ function, including the skeletal muscle system. Sepsis induces alterations in the gut microbiota composition, which is globally termed a state of “dysbiosis” for the host compared to baseline microbiota composition. In this review, we critically evaluate existing evidence and potential mechanisms linking sepsis-induced myopathy with gut microbiota dysbiosis.

1. INTRODUCTION

Sepsis is currently defined as life-threatening organ dysfunction caused by a dysregulated host response to infection [1]. According to the Center for Disease Control and Prevention, around 1.7 million adults in the United States develop sepsis each year [2]. Early diagnosis and aggressive treatments have resulted in improved in-hospital mortality and overall enhanced survival after sepsis. [3–4]. However, still a significant portion of sepsis survivors can enter chronic critical illness (CCI), which has been defined in the literature as patients who remain in the intensive care unit for greater than 14 days with continued organ failure/insufficiency, for which this cohort suffers dismal long-term outcomes [3]. One proposed underlying pathobiology of CCI is the Persistent Inflammation, Immunosuppression and Catabolism Syndrome (PICS), which has been illustrated to result in long-term organ dysfunction and functional limitations resulting in a higher 5-year morbidity and mortality post-hospitalization [5–6]. While many organs are affected, skeletal muscle dysfunction has long-term repercussions that are often underappreciated after sepsis [7–8].

Skeletal muscle atrophy, characterized by reduction in functional cross-sectional area, occurs early and rapidly during the first week of sepsis and appears to be more severe among patients with multi-organ failure compared to those with single-organ failure [3, 7, 9–10]. Histological evidence suggests infiltration of inflammatory cells such as macrophages within the skeletal muscle that attract monocytes from the blood stream by secreting cytokines and chemokines [7, 11]. Furthermore, one model for the origin of skeletal muscle dysfunction in sepsis involves circulating pathogens and cytokines that activate skeletal muscle pathways associated with decreased protein synthesis (due to overproduction of reactive oxygen species) and accelerated protein degradation (due to enhanced proteasome proteolytic degradation and autophagy pathways) [12]. Although in the present review our focus is on the functional aspects of skeletal muscle and its regenerative potential, it is important to recognize that the myopathy is also accompanied by dysregulation of neural inputs at the neuromuscular junction which has been covered in-depth in other reviews[13–15].

Myogenic skeletal muscle stem cells (satellite cells), which are required for regeneration and repair upon injury, have also been shown to be altered by sepsis and to participate in myopathy both in septic humans [9] and preclinical models of sepsis [16]. In sepsis, the satellite cells display a diminished number, attenuated proliferative capacity and an inability to differentiate into functional myotubes, therefore impairing regeneration [8–9]. This is important for the sepsis survivor, since diminished muscle satellite cell number or function results in the inability to respond to exercise stimuli and potentiates muscle wasting [9, 17]. The mechanisms leading to this dysfunctional satellite cell phenotype in sepsis remain largely unknown.

Sepsis-induced myopathy may also have implications beyond impaired locomotion. Indeed, observations that myofibers possess secretory function demonstrate that skeletal muscle physiology extends beyond locomotion [18–19]. Myofibers secrete cytokines (also referred to as myokines) such as interleukin-6 in response to several stressors including catecholamines [20], heat shock [21] and lipopolysaccharide [22], contributing to endocrine and paracrine signaling and supporting immune cell proliferation [23]. Therefore, skeletal muscles, which constitute approximately 40% of total body mass in humans, can regulate, and be regulated, by other organs via cross-talk [24–25]. Although the regulation of skeletal muscle by distant organs has been studied in exercise settings [26–27], the mechanisms involved in their regulation in critical illness remain unclear and may involve cross-talk with other distant organ systems and immunologically active microbial environments.

The microbiota is composed of trillions of bacteria, viruses, fungi, Protozoa and Archaea that live in a symbiotic and pathobiont state with the host [28]. Bacteria comprising the gut microbiota are the largest and most studied microbial community. In healthy conditions, the gut microbiome, the collective metagenome of the microbiota, supports the immune system by many processes [29]. As occurs during certain disease states, the gut microbiome becomes altered, and if not beneficial to the host, is referred to a dysbiosis [30–32]. The dysbiosis can induce host immune dysregulation and the perpetuation of systemic inflammation [33–37]. Given the limited information regarding regulation of organ dysfunction during sepsis in conjunction with sepsis-induced myopathy, the microbiome may serve as a potential link (Figure 1) given known interactions with the immune system and a variety of diseases [33, 38–40].

Figure 1. Potential links among sepsis, bed rest, gut dysbiosis and muscle dysfunction.

Sepsis and bed rest induce an increase in pathogen- and damage-associated molecular patterns (PAMPs and DAMPs), cytokines, and reactive oxygen species, which leads to both skeletal muscle dysfunction characterized by decreased protein synthesis and increased catabolism. Sepsis and bed rest also promote gut dysbiosis, which in turn exacerbates muscle dysfunction potentially via circulating short chain fatty acids (SCFA), extracellular vesicles (EV) and bacterial metabolites.

As stated, PICS may be an underlying pathobiological mechanism for poor in-hospital and post-discharge outcomes such as physical and cognitive function loss [41–42]. Recent preclinical evidence demonstrates a cross-talk between the gut microbial environment and the skeletal muscle [43–44]. As will be discussed further in this review, in sepsis, impaired skeletal muscle function and regenerative capacity may be influenced by the gut microbiota [45]; however, the mechanism of this crosstalk remains undefined. The goal of this review is to critically evaluate current evidence on the interaction between sepsis-induced myopathy and the gut microbiome, and to identify potential targets for future research on the guts-keletal muscle axis in sepsis.

2. LITERATURE SEARCH AND TERMINOLOGY DEFINITION

A literature search was conducted via the PubMed database for English-language preclinical and clinical studies using the terms “sepsis”, “muscle”, “gut microbiome”, and “cross-talk”. Throughout this review we adopted the term “sepsis-induced myopathy” to describe a persistent manifestation characterized by reductions in neuromuscular force-generating capacity (weakness), atrophy (loss of muscle mass) and compromised skeletal muscle regenerative capacity (satellite cell dysfunction) in sepsis survivors. Other nomenclatures have been proposed to describe similar manifestation including: “ICU acquired weakness” [46], “sepsis-induced muscle wasting” [47], “critical illness myopathy” [48].

3. GUT MICROBIOME DYSBIOSIS IN SEPSIS

Sepsis has been shown to alter diversity of the intestinal microbiota within the first six hours after admission [32, 49]. Evidence exists for changes in the microbiota during sepsis that associates with end-organ dysfunction seen in these patients. For example, patients who develop acute respiratory distress syndrome (ARDS) harbor a pulmonary microbiome that has increased abundance of microbes typically found in the intestinal tract and this can be predictive of survival[50–51]. Whether the presence of these microbiomes in the lung promote a pro-inflammatory state and the process of ARDS is unknown. Furthermore, in a murine model, the microbiota was able to facilitate liver injury in sepsis [52]. In contrast, however, microbes have also been demonstrated to have a beneficial impact by the production of short-chain fatty acids (SCFAs) that protect against sepsis-induced acute kidney injury [53]. While this has not been demonstrated in humans, it raises the question of ameliorating the impact of sepsis on organs through microbial manipulation. One obvious suggestion would be to manipulate the microbiota through the use of antibiotics. However, antibiotic treatment plays a major role in the decreased diversity in critical illness, given that 75% of patients receive antibiotics at some point during their treatment [54]. Altered microbial diversity and function will change microbial metabolite production [55]. For example, disappearance of microbes that represent an important part of the microbiota of healthy individuals such as Faecalibacterium prausnitzii that has anti-inflammatory properties, as well as the genera Faecalibacterium, Prevotella, Blautia, and family Ruminococcaceae, which are known producers of SCFAs [56–57]. These bacterial species metabolize dietary fibers to produce SCFAs in the gut lumen that are absorbed into the portal circulation, mainly acetate, propionate, and butyrate [58]. SCFAs, such as butyrate, provide immunologic defense, regulating the host mucosal immune system and serving as the primary energy source for intestinal epithelial cells, which are a primary defense line against pathogens [59]. SCFAs promote regulatory T cells (Treg) by inhibiting histone deacetylase (HDAC) activity and the forkhead box protein P3 (FOXp3) expression in the colon. Exposure of other cells, such as neutrophils, macrophages, and dendritic cells to SCFAs inhibits production of pro-inflammatory cytokines [60–61]. Butyrate is produced by a number of gut genera, mostly by the Faecalibacterium genera, and is highly abundant in healthy subjects but reduced in older individuals and sepsis patients [58, 62]. Levels of SCFAs are lower in septic patients compared with healthy controls, which could potentially be linked to impaired systemic immunity in sepsis [45]. For example, butyrate regulates and increases frequency of CD4⁺ Treg cells, which produce anti-inflammatory interleukin 10 (IL-10) through FOXp3⁺ pathways for containment of inflammatory responses in mucosal tissues, including the colon [57]. Early stimulation of T reg cells (CD4⁺ T cells), directly links the adaptive and innate immunity systems, and thus these cells are important in modulating an inflammatory response during sepsis. Reduction in T reg cells is associated with decreased effector cytokine production, proliferative capacity and apoptosis [63]. Treg cells play a major role in infection and natural Treg cells are present in the host before the pathogen exposure and the inducible regulatory Treg cells are those that acquire regulatory function in response to infection and contain IL-10. In another study, sepsis decreased the frequency of interleukin 2 (IL-2)⁺ and increased tumor necrosis factor (TNF)⁺ cells within CD4⁺ T cells [64]. Manipulating the capacity of the intestinal microbiota to produce SCFAs would be just one way to translate laboratory findings into the clinical realm. This could be through supplementation of SCFAs directly, reintroducing microbes that are metabolizers of dietary fiber, or even bacteriophage therapy to target harmful bacteria that potentiate a septic response. Identification of microbial metabolites or their products offers yet another entry point for clinical translation. Well-controlled clinical trials are thus needed so that these interventions can be tested in a setting taking into account confounding variable such as medication usage and diet.

4. CROSS-TALK BETWEEN THE GUT MICROBIOME AND SKELETAL MUSCLE

The connection between the gut and the skeletal muscle has been demonstrated in sarcopenia research as well as studies regarding aging-related muscle loss. Sarcopenia, a condition defined as a gradual loss of muscle mass, strength and function, is among the most important factors for developing frailty, [65]. From a clinical perspective, skeletal muscle loss may also result from several conditions that are frequently found in geriatric patients, including malnutrition, low dietary protein intake, impaired intestinal malabsorption and altered digestion [66–67]. The pathological processes leading to sarcopenia are represented by reduced muscle capillarity, reduced insulin sensitivity, and increased subclinical inflammation, resulting in reduced oxygen and nutrient delivery, and altered anabolic/catabolic balance of muscle protein synthesis [68–70]. The intestinal microbiome composition with reduced abundance of several microbial taxa has a role in the development of sarcopenia, demonstrating reduced microbiota diversity related to impaired muscle strength [27, 43]. For example, Fielding et al. have demonstrated that transplanted microbiota from high-functioning older adults to germ-free mice improved hand-grip strength compared to mice transplanted with microbiota from low-functioning older adults. High-functioning older adults had higher percentage of lean mass and were better physically functioning than low-functioning older adults. Microbiota of high-functioning older adults contained greater amounts of family-level Prevotellaceae, genus-level Prevotella and Barnesiella, and species-level Barnesiella intestinihominis, which take part in SCFAs production. These results support the notion that changes in the gut microbiome may have a causative role in the age-related muscle loss [27].

Sepsis-induced reduction of diversity and underproduction of important metabolites such as SCFAs may contribute substantially to the gut-muscle axis and muscle dysfunction process aside from the gut-immune system-maintained inflammation. Additionally, SCFAs also stimulate neural and hormonal signals regulating mitochondrial function [71]. Diminished production of SCFAs in the gut could impair the humoral system and contribute to skeletal muscle wasting driven by the mitochondrial dysfunction. However, based on the current evidence by Rocheteau et al., mitochondria may not be the main driver of muscle wasting in sepsis, but rather profound satellite cell dysfunction, which could severely impair the regenerative capacity of skeletal muscle in sepsis patients [8]. One important element in the self-renewal and plasticity of skeletal muscles to injury is the regenerative capacity of satellite cells [72–75], which are unipotent muscle-resident myogenic stem cells localized between the basal lamina and sarcolemma of myofibers [75]. For instance, the gut microbiome has the ability to regulate cytokines and chemokines that are secreted by the adipose tissue and skeletal muscle [76–77]. Therefore, sepsis-induced gut microbiome dysbiosis can presumably alter the skeletal muscle production of these bioactive molecules towards a more pro-inflammatory environment, which can result in a dysfunctional phenotype, affecting satellite cell biology and blunting regeneration. This gut-satellite cell axis may be a different communication channel connecting the gut, systemic inflammatory response and muscle dysfunction.

Few reports have investigated the relationship between the gut microbiome and the skeletal muscle system. Lahiri et al. demonstrated a connection between the gut microbiota and skeletal muscle mass as well as function. The authors used germ-free and pathogen-free mice to study the communication between the gut microbiota and skeletal muscle by disrupting the gut microbiota in the pathogen-free mice with penicillin and transplanting the germ-free mice with the gut microbiota of pathogen-free mice. The communication with skeletal muscle was rescued by SCFA by 4-week treatment (cocktail of sodium acetate, sodium butyrate, and sodium propionate in their drinking water). In particular, germ-free mice demonstrated higher gene expression related to atrophy (Atrogin-1, the muscle RINGfinger protein-1 [Murf-1] and Forkhead box O3 [FOXO-3]) and impaired regenerative capacity (myoblast determination protein [MyoD] and Myogenin). Germ-free mice supplemented with SCFA demonstrated higher gene expression of mitochondrial biogenesis proteins (peroxisome proliferator-activated receptor gamma coactivator 1-alpha [PGC-1α] and transcription factor A, mitochondrial [TFAM]) and reduced gene expression of atrophy proteins [78]. They also demonstrated lower mitochondrial density by decreased mitochondrial DNA content and gene expression of mitochondrial respiration complexes, which relate to lower energetic capacity of the skeletal muscle. Importantly, transplanting feces of germ-free mice with pathogen-free mouse feces reversed these detrimental changes of muscle degeneration [78].

Gut microbes produce a number of metabolites during digestion of nutrients, many of which play important systemic roles. The SCFAs and various metabolites transport information between the microbes and other organs [79–81]. Extracellular vesicles (EVs) released into the extracellular space may be a key component of cross-talk between the gut microbiome and other organs [82]. Bacteria-derived EVs are spherical bi-layered phospholipids with diameters ranging from 20 to 100 nm and are produced ubiquitously by all Gram-negative and some Gram-positive bacteria [83]. EVs derived from Gram-negative bacteria are composed of outer membrane proteins, lipopolysaccharide (LPS), outer membrane lipids, periplasmic proteins, DNA, and RNA [84]. Without the presence of living bacteria, nano-sized EVs derived from Gram-negative bacteria, such as Escherichia coli, have been shown to induce severe immune responses and septic shock [85]. Therefore, other bacterial components and information carried in EVs can be more potent in inducing a systemic response such as pro-inflammatory processes in peripheral organs [85].

Changes in microbial diversity and function, illness-induced hyperpermeability of the gut barrier, and the translocation of the gut-derived microbial metabolites and toxins traveling to skeletal muscle, may be mechanisms that facilitate gut-muscle cross-talk (Figure 2). However, the mechanistic link between gut dysbiosis and altered microbe-derived metabolite production (e.g. SCFA) and inflammation-induced muscle dysfunction (e.g. satellite cells) is still missing and warrants further experimentation.

Figure 2. Conceptual model of the gut microbiome dysbiosis and impaired proliferation of satellite cells, and thus impaired muscle regeneration.

(A) Healthy gut microbiome in a state of “eubiosis” contributes to the maintenance of an anti-inflammatory microenvironment favoring satellite cells to enter the cell cycle and regenerate skeletal muscle fibers. (B) Sepsis induces gut-microbiome dysbiosis and results in increased EVs, gram-negative and gram-positive bacteria, which favors a pro-inflammatory microenvironment status within the satellite cell niche which halts differentiation.

5. THE GUT MICROBIOME MAY ALTER THE SATELLITE CELL NICHE AND IMPAIR REGENERATION IN SEPSIS

Skeletal muscles display high plasticity as they undergo continuous micro-damage and regeneration in response to physical exercise, disuse, and pathological conditions like trauma and sepsis [86–87]. One important element in the self-renewal and plasticity of skeletal muscles to injury is the regenerative capacity of satellite cells [72–75], which are unipotent muscle-resident myogenic stem cells [75] localized between the basal lamina and sarcolemma of myofibers. These cells are required for skeletal muscle repair and regeneration following injury. Satellite cells are characterized by the expression of transcription factor Pax7, which is required for their self-renewal [88]. Satellite cells reside in a quiescent state and become activated in response to injury or damage, generating proliferating MyoD-positive progenitors (myoblasts). The localization of satellite cells and its surroundings beneath the surrounding basal lamina and outside the myofiber plasma membrane are referred to as the niche or microenvironment [89]. Activation, proliferation and differentiation of this cell population into myotubes requires a regulated sequence of signaling pathways, within the satellite cell niche [89]. Importantly, in sepsis, satellite cells display an attenuated proliferative phenotype with limited to no differentiation capacity [8]. Among the hypothetical regulatory pathways likely to mediate the interactions between the gut microbiome and the satellite cell niche is the Notch signaling pathway.

The Notch signaling pathway is highly conserved among mammals and plays important roles in determining the fate of satellite cells. This pathway defines whether cells remain quiescent or enter the cell cycle to differentiate into myotubes. Dormant satellite cells express a marked abundance of Notch effector genes, suggesting that Notch signaling is active in this quiescent cellular state [90]. Ligands for the Notch receptor include the jagged protein family – Jagged Canonical Notch Ligand 1 and 2 (JAG1 and JAG2) and the delta-like protein family of DLL1, DLL3, and DLL4 [91]. Interestingly, these transmembrane protein ligands to the Notch receptor are present in Gram-negative and Gram-positive bacteria [92], and also on the surface of EVs [93]. Therefore, they could participate in the mechanism linking the gut microbiome with satellite cell regulation (Figure 2) [94–95]. The Notch pathway (Figure 3) is negatively regulated by cytoplasmic Numb proteins [96]. Interruption of Notch activity by Numb is associated with progression down the myogenic pathway, while sustained Notch signaling is associated with maintenance of the undifferentiated state. The Notch activity inhibits the differentiation of specific cell types. Activation of Notch signaling following muscle injury promotes the expansion of satellite cells but prevents differentiation. In contrast, termination of Notch activity allows for subsequent progression down the myogenic lineage. For instance, inhibition of Notch signaling with the soluble JAG protein was shown to impair regeneration, while enhanced Notch activation with a specific antibody facilitated the repair process [97]. Interruption of the Notch signaling pathway is required to permit regeneration of a damaged skeletal muscle fiber [90]. Over-expression of the Notch ligands, in particular DLL1 [94], on the EVs and bacteria arising from a dysbiotic gut microbiome in sepsis, could lead to an imbalance between the Notch signaling activation and inactivation by Numb proteins, which would explain why satellite cells do not differentiate well in sepsis. This hypothesis requires further investigation, as this mechanism has not yet been systematically tested. Further mechanistic studies are needed to elucidate how EVs [82], Gram-negative and Gram-positive bacteria that may originate from the gut, travel to skeletal muscle and mediate hypertrophic or atrophic processes, dependent on the EV type in sepsis. Although intriguing, it is important to emphasize that, to the best of our knowledge, studies exploring the role of EVs in sepsis-induced myopathy are non-existent. Thus, caution must be taken while interpreting the potential link we set forth in this section of the review. This is an area of research that warrants further investigation.

Figure 3. Proposed mechanism of the cross-talk between the gut and the satellite cells.

Sending cell carries a ligand that requires ubiquitination of the Delta protein (DLL) by an ubiquitin ligase Mind bomb (MIB) for activation [107]. Once ubiquitinated, DLL signals to Notch, in the receiving cell (e.g. satellite cell) and triggers a two-step receptor proteolysis event [108]. The first proteolytic event involves the metalloproteinase ADAM cleavage to the proximal membrane and sheds the Notch extracellular domain [108]. The second proteolytic cleavage, mediated by γ-secretase, releases the Notch intracellular domain (NICD) from the inner plasma membrane and transports to the cytosol where it reacts with the DNA-binding proteins, including RBP-J (also known as CBF1 or suppressor Hairless), MAM, CSL, and p300, and forms DNA-binding complex that activates Notch target gene expression [108] including c-Myc, P21 and cyclin D3. EVs=extracellular vesicles.

Immunological mediators of satellite cell dysfunction in sepsis

Recent studies have shown the mechanistic link between the inflammation-driven satellite cell dysfunction and muscle wasting in cecal ligation and puncture (CLP) mice, however, the mechanism responsible for driving the persistent systemic inflammation that contributes to the skeletal muscle pathology is emerging [98–99]. Resident and infiltrated macrophages, localized in the satellite cell microenvironment, are important regulators of the activation process. They undergo alteration in their phenotype from M1 (pro-inflammatory) to M2 (anti-inflammatory). This switch is mediated by necrotic and apoptotic cells undergoing phagocytosis [100]. The M1 macrophages not only secrete cytokines to trigger inflammation and clear cell debris, but also activate satellite cells to enter the cell cycle [101]. Later, the M2 subtype of macrophages sustain tissue healing and promote satellite cell differentiation into myotubes [100]. In the case of sepsis, where the microbiome is altered from baseline, the satellite cell microenvironment may be inadequate to induce the proper macrophage phenotype shift, causing muscle repair impairment. The maintenance of M1 activated macrophages can lead to fibrosis, fat deposition and impairments to the stem cell pool [102].

The evidence for gut-skeletal muscle axis in sepsis is only beginning to emerge. Although more research is needed to identify mechanistic links that may lead to effective therapeutic targets, it is possible to conjecture future therapeutic strategies. The first obvious approach involves prevention of microbiome dysbiosis, which would include fecal microbiota transplantation from healthy subjects given evidence in that stabilization of the microbiota in a young versus old mice in a murine model of sepsis may help to explain improved outcomes in young patients with sepsis compared to old [103]. Additionally, improving the microbial environment by supplementing beneficial bacteria by fecal microbiota transplantation (FMT) that can abrogate the pro-inflammatory process and myopathy may provide clinical improvement in sepsis patients [104]. Further studies on therapies such as FMT should aim to answer question such as timing, dosing, administration, treatment duration and the lasting effects of a single FMT treatment. For example, single FMT treatments were effective in patients with ulcerative colitis resembling the gut microbial environments of healthy donors and may provide a simple clinical intervention [105]. Given that FMT has resulted in death in the recipient [106], transfer of a deleterious microbiota or one that is itself able to incite an inflammatory reaction will need to be taken into consideration. As intestinal microbiota sequencing of patient stool enters the clinical arena, administration of FMTs likely will need to take into account the nuances of the recipient microbiota. The systemic and immunologic repercussions of creating a new complex microbial community is unknown and is an area of needed research in itself.

7. CONCLUSIONS

Muscle dysfunction in sepsis affects patients acutely and chronically due to an uncontrolled inflammatory response that may be perpetuated by sepsis-induced gut microbiome dysbiosis. The change in the microbial diversity due to sepsis and production of metabolites as well as cell information carriers such as EVs may explain cross-talk between the gut microbiome and skeletal muscle regenerative process. Further studies are needed to elucidate the mechanistic action of microbe-derived products on skeletal muscle regenerative properties in sepsis.