The gut microbiome as a target for adjuvant therapy in obstructive sleep apnea

Abstract

Introduction:

Gut dysbiosis is assumed to play a role in obstructive sleep apnea (OSA)-associated morbidities. Pre- and probiotics, short chain fatty acids (SCFA) and fecal matter transfer (FMT) may offer potential as novel therapeutic strategies that target this gut dysbiosis. As more mechanisms of OSA-induced dysbiosis are being elucidated, these novel approaches are being tested in preclinical and clinical development.

Areas covered:

We examine the evidence linking OSA to gut dysbiosis and discuss the effects of pre- and probiotics on associated cardiometabolic, neurobehavioral and gastrointestinal disorders. The therapeutic potential of short chain fatty acids (SCFA) and fecal matter transplantation (FMT) are also discussed. We reviewed the National Center for Biotechnology Information database, including PubMed and PubMed Central between 2000 -2020.

Expert opinion:

To date, there are no clinical trials and only limited evidence from animal studies describing the beneficial effects of pre- and probiotic supplementation on OSA-mediated dysbiosis. Thus, more work is necessary to assess whether prebiotics, probiotics and SCFA are promising future novel strategies for targeting OSA-mediated dysbiosis.

1. Introduction

Obstructive sleep apnea (OSA) is a sleep disorder characterized by narrowing of the pharyngeal airways leading to repetitive episodes of airway collapse or increased upper airway resistance. These recurring events are usually associated with intermittent hypoxia (IH) and micro-arousals with concomitant sleep fragmentation. Although the total nocturnal sleep quantity might not be necessarily affected, sleep quality is usually poor in patients with OSA, which explains many of the symptoms seen in patients with OSA such as excessive daytime sleepiness and fatigue, poor concentration, mood alterations, and memory impairments. These symptoms can affect work productivity, increase the risk for motor vehicle and on the job accidents, as well as adversely impact social interactions, academic performance, and many other neuro-cognitive consequences [1]. IH, such as occurs in OSA, has significant impact on a variety of organ systems. Indeed, there is a strong body of literature linking IH to different diseases such as systemic hypertension, coronary artery disease, obesity, diabetes mellitus, pulmonary hypertension, cerebrovascular diseases, Alzheimer disease, mood disorders, erectile dysfunction and even cancer [2–5].

IH can directly affect cells of multiple types, or it can induce alterations in susceptible cells through the effects induced by the IH-mediated release of different mediators. It is assumed nowadays that IH plays a major role in the activation and propagation of oxidative stress and systemic inflammation by acting as a potent stimulant for the release of inflammatory mediators which contribute to many of the co-morbid disorders associated with OSA [6–9]. In this manuscript, we will briefly review the evidence linking IH, gut dysbiosis and systemic inflammation, and will focus on therapeutic interventions targeting gut dysbiosis (namely, pre- and probiotics, short chain fatty acids (SCFA) and fecal matter transplantation (FMT)), and their potential uses as novel adjuvant therapies for patients with OSA. We reviewed the National Center for Biotechnology Information database, including PubMed and PubMed Central between 2000 -2020 using the keywords: “obstructive sleep apnea, gut dysbiosis, probiotics, prebiotics, short-chain fatty acids, fecal matter transfer, cardiovascular, metabolic, cognitive, anxiety, depression, gastrointestinal” in different combinations and permutations.

2. The gut microbiome and gut dysbiosis

Microbiota is a term used to include all microbes that live in and colonize different organs (such as gastrointestinal tract, lung, upper airways, and skin). These microbes include bacteria, viruses, fungi, archaea, and eukarya [10]. The gut harbors most of the microbiota with more than 100 trillion bacteria from more than 1,000 species [11]. The mutual interactions between the host and the microbiota maintains a neutral, yet actively regulated and responsive ecosystem through which the host benefits from the microbiota and vice versa [12]. Any perturbations in the balance of the system can lead to a state of dysbiosis. The healthy gut microbiota provides the gut with many benefits, such as maintaining the integrity of the gastrointestinal epithelial cells, regulating the gut and the overall host immune system, protecting against a variety of opportunistic pathogens, and providing energy to the colonocytes, a function which plays a critical role in maintaining the gut integrity and permeability [13–16]. Gut dysbiosis invokes the presence of harmful species of bacteria that can potentially generate excessive amounts of specific compounds that are detrimental to the gut or directly produce toxic substances that can induce leakiness of the gut or traverse the leaky epithelial tight junctions into the systemic circulation. Some of these toxins (such as lipopolysaccharides produced by some gram-negative bacteria) can induce a state of local and systemic low-grade inflammation that has been shown in some studies [17] (Figure. 1). Indeed, systemic inflammation is one of the main underlying pathological mechanisms linking OSA to CVD [18].

Figure 1: Mechanisms and treatments of OSA-induce dysbiosis.

BA: bile acid, IH: intermittent hypoxia, LPS: lipopolysaccharides, OSA: obstructive sleep apnea, ROS: reactive oxygen species, TMA: trimethylamine, TMAO: trimethylamine oxide, WBCs: white blood cells

The five most common bacterial phyla that reside in the colon are Bacteroidetes, Firmicutes, Actinobacteria, Proteobacteria, and Cerrucomicrobia [19]. Some of the health promoting bacterial species include Lactobacillaceae, Bifidobacteriaceae, Erysipelotrichaceae, Ruminococcaceae, and Clostridiales which belong to the phylum Bacteroidetes. Bacteroidetes (Gram-negative) inhabit the gastrointestine (mainly the colon) and play a critical role in carbohydrate and fiber fermentation. This process produces short chain fatty acids (SCFA) such as butyrate, acetate, and propionate. SCFAs provide a major source of nutrients and energy to colonocytes [3, 20, 21]. Induction of gut dysbiosis with unhealthy diets or following IH will consume many of these SCFAs and lead to colonocyte dysfunction, manifesting as weakened intercellular tight junctions that connect the epithelial cells and preserve the gut epithelial barrier. Consequently, the gut becomes leakier. At the same time, the process of hypoxia/re-oxygenation per-se is directly toxic to tight junctions and increases the risk of leaky gut [21].

On the other hand, Prevotella, Paraprevotella, Desulfovibrio, and Lachnospiraceae species belong to the phylum Firmicutes. More than 90% of the colon bacteria consist of Bacteroidetes and Firmicutes [22]. Increased presence of Desulfovibrio induces mucin degradation and subsequently increases gut permeability [23, 24]. There is some evidence that augmentation of Lachnospiraceae in the gut can increase the risk of obesity [25]. Similarly, Prevotella is strongly linked to systemic inflammation through the production of lipopolysaccharides [26].

3. Intermittent hypoxia and gut dysbiosis (animal and human models)

The episodic nature IH associated with apnea and hypopnea events leads to fluctuation in the oxygen partial pressure (PaO₂) in the gut lumen across a centrifugal gradient. These changes in PaO₂ shift the gut microbiome toward a dominance of anaerobic and facultative anaerobic bacteria at the expense of obligate aerobic bacteria [27]. Of note, these findings are derived from studies conducted in this field using animal models (mice and rats) with very few human trials. Most of the animal studies used environmental hypoxic chambers, where mice are exposed to cyclical changes in the fraction of oxygen inspired (FIO₂) eliciting corresponding targeted changes in blood PaO₂ for a specific number of hours during the sleep period of the animals and followed by normal air at 21% FIO₂ for the remained of the circadian period mimicking periods of wakefulness during which normal PaO₂ is present in OSA patients. The duration of exposures varies between 1-20 weeks depending on the study design. In addition, a wide variety of IH profiles have been used, which differ in both the number of cycles of hypoxia-reoxygenation occurring every hour (simulating a specific apnea-hypopnea index) and the severity of hypoxemia targeted (usual SpO₂ achieved between 60-90%) [28]. Another model used inflatable devices implanted in the mice trachea or externally applied which through periodic inflation and deflation mimic obstructive respiratory events [29–31]. The variability in study design and results using these two models has not been evaluated.

It is worth mentioning that IH is not the only sleep disorder that affects the gut microbiota. Sleep deprivation for a short period of time (48 hours) was associated with subtle changes in the gut microbiota [32]. Another study showed that chronic sleep deprivation (mean total sleep time 5.58 hours/night) who were given the chance to extend their sleep for two weeks (mean total sleep time 6.6 hours/night) did not results in changes in the gut microbiota [33]. Another example is circadian rhythm disorders. Circadian misalignment has been shown to have a bidirectional effect on the gut microbiome, where both central and peripheral clocks affect and are affected by changes in the gut microbiome [34, 35]. A common example in this regard in shift workers who exhibit alterations in the gut microbiome that can be partially contributing to the associated higher risk of cardiovascular and metabolic diseases in this group [36].

The fecal sample analysis to characterize the microbiota starts with DNA and RNA isolation followed by 16S r RNA sequencing. Nowadays, in the era of advanced technology in biomedical engineering and bioinformatic analysis, more detailed and accurate methods have emerged in the field of metagenomics. The mainstay amplicon sequencing is 16S rRNA [37, 38]. However, more specialized and unbiased databases (e.g. gunshot approaches used in whole metagenome sequencing) are being utilized nowadays [39, 40]. Another critically important tool that is used to further identify a group of bacteria is metabolomics, which analyzes different molecules and metabolites produced by different bacterial species to identify these species [41]. Ultimately, the combination of metagenomics and metabolomics provides a more comprehensive and accurate analysis of the gut microbiome (Figure 2).

Figure 2:

Pipeline for processing fecal samples for microbiome characterization.

Most of the stool samples obtained from IH-exposed animal models showed an increase in the Firmicutes/Bacteroidetes ratio, which is the hallmark of gut dysbiosis [24]. Specific bacterial species under the phylum Firmicutes include Prevotella and Desulfovibrio and both are increased with IH exposures [17, 24]. In rats, IH combined with high salt diet increased the severity of systemic hypertension likely via increasing levels of blood trimethylamine N-oxide (TMAO), enhanced release of Th1-related interferon-γ (IFN-γ) and reduced levels of the anti-inflammatory cytokine (TGF-β1). These changes coincided with changes in the gut microbiome of the exposed rats, particularly reflected by major decreases in Lactobacillus abundance [42]. Similar studies have confirmed these findings across multiple rodent species [43–48]. Xue et al [49]. suggested that it is not only IH, but rather IH and hypercapnia that plays a role in inflammation and atheroma formation in the pulmonary artery and aortic arch in mice. Although the impact of hypercapnia on the gut microbiome is still controversial, the hypothesis of hypercapnia as a common denominator between OSA and gut dysbiosis deserves expanded exploration through future experimental studies.

One of the pathological conditions that is linked to IH is inflammation. Gut dysbiosis is associated with increased abundance of Prevotella and Klebsiella which are gram negative bacteria that produce toxins such as lipopolysaccharide (LPS). When these toxins leak from the gut into the systemic circulation, they stimulate the release of other inflammatory mediators, such as interleukin-6 (IL-6) and tumor necrosis factor- α (TNF-α), that lead to systemic inflammation [17]. Inflammation plays a critical role in the pathophysiology of many co-morbid diseases (such as hypertension, coronary artery disease, obesity, and diabetes mellitus), and has been implicated in the vast majority of the morbidities associated with OSA as well [50–54].

In humans, few studies examined the association between IH and dysbiosis. The first study on these issues identified increased LPS-binding protein levels in children with OSA, and particularly among obese children with OSA, suggesting increased translocation of LPS to the circulation from the gut [55]. In a trial included pediatric age group (2 years) with history of snoring, there was evidence of increase Firmicutes/Bacteroidetes (F/B) ratio and decrease Actinobacteria/Proteobacteria ratio suggestive of inflammation [56]. In adults, one study showed an increase in the abundance of Proteobacteria, Bacteroidetes, Firmicutes, Fusobacteria, and Acidobacteria in the lungs suggestive of lung dysbiosis [57]. Another study examined the nasal microbiota in patients with OSA and found an increase in some inflammatory mediators (IL-8 and IL-6) [58]. In a small cohort of patients with OSA, three enterotypes, namely predominance of Bacteroides, Ruminococcus, and Prevotella, respectively were identified. Interestingly, arousal-related parameters and distribution of sleep stages were significantly altered among OSA patients with an apnea-hypopnea index (AHI) ≥15/hour total sleep duration among those patients in whom Prevotella predominated [59]. In another preliminary study [60], functional analysis in the fecal microbiomes of 93 OSA patients and their corresponding controls revealed significant differences in the gut microbiota profile between the two study groups. Additionally, decreased in short-chain fatty acid (SCFA)-producing bacteria and increased pathogens, accompanied by elevated levels of IL-6, while the abundance of Lactobacillus was significantly associated with homocysteine levels.

4. Pre- and probiotics as therapeutic interventions in gut dysbiosis

Probiotics refer to species of live bacteria that confer beneficial health effects on the host when ingested in adequate amounts [61]. Health benefits have been demonstrated for specific probiotic strains mainly of the following genera: Lactobacillus (e.g. prevent and decrease the severity of some allergic diseases such as atopic dermatitis), Bifidobacterium (e.g. decrease the severity of diarrhea, reduce symptoms of necrotizing enterocolitis) , Saccharomyces (e.g. reduce duration of traveler’s diarrhea and irritable bowel syndrome), Enterococcus (e.g. treatment of antibiotic-associated diarrhea), Streptococcus (e.g. reduce irritable bowel syndrome symptoms), Pediococcus, Leuconostoc, Bacillus and Escherichia coli (e.g. decrease the severity of inflammatory bowel diseases and decrease the severity of constipation) [62]. Probiotics can exert a wide range of effects such as improving metabolism and immunity through interactions with the host microbiota and also directly with the intestinal epithelium [63, 64]. They can improve mucosal defenses of the GI tract by: i) Blocking the colonization of pathogenic bacteria through decreasing luminal pH, inhibiting adhesion and invasion to epithelial cells and producing antimicrobial compounds. ii) Reinforcement of the mucosal barrier against pathogenic bacteria through enhancement of mucous production and tight junction protein phosphorylation. iii) Influencing both innate and acquired immune system locally in the gut wall [65].

Prebiotics can be defined as “a substrate utilized selectively by host microorganisms conferring a health benefit” [66]. Dietary fiber is one of the main classes of prebiotics that includes inulin, fructooligosaccharides (FOS), galactooligosaccharides (GOS) and resistant starch [67]. Prebiotics can prevent intestinal invasion of pathogens due to their anti-adhesive properties [68] and their ability to stimulate the growth of favorable probiotic bacteria such as Lactobacillus and Bifidobacterium [69]. Byproducts produced through prebiotic digestion and fermentation by gut microbiota (such as SCFAs) can positively modulate local and systemic effects of the immune system [70]. In addition to the role of dietary fibers as a source of SCFA and probiotics, they were found to be negatively correlated with the severity of OSA. Stelmach-Mardas et al [71]. concluded that alcohol consumption is positively correlated with the severity of OSA and high fiber diet is negatively correlated with OSA severity.

The past few decades have seen a surge in experimental and clinical studies on gut dysbiosis and its role in a wide spectrum of diseases. This surge was accompanied by multiple interventions that includes a large variety of pre- and probiotics. Although, many of the beneficial effects of pre- and probiotics can be seen in general population, in the next sections, we will discuss the beneficial effects of pre- and probiotics in cardiometabolic, neurobehavioral and gastrointestinal diseases in the context of sleep related breathing disorders.

4.1. Cardiometabolic Disease

Cardiovascular disease (CVD) continues to be the leading cause of death worldwide with increasing prevalence of common metabolic risk factors such as obesity, type 2 diabetes and the metabolic syndrome [72]. Inadequate effectiveness of pharmaceutical therapies for the management of atherosclerosis, a hallmark of CVD, combined with the recent developments in understanding of the gut microbiome - host interactions have fueled the interest in pre- and probiotics as a potential add-ons therapy for CVD [73]. In gut dysbiosis, compromised intestinal barrier can lead to LPS and other bacterial components leakage triggering inflammatory responses through monocyte recruitment and activation of Toll-like receptors [74]. Immune cell recruitment, production of pro-inflammatory factors and oxidative stress are the main underlying mechanisms promoting the initiation and propagation of endothelial dysfunction and atherosclerosis [75]. Furthermore, increased TMAO, an oxidized form of the bacterial metabolite trimethylamine from dietary choline and L-carnitine, can also promote inflammation, thrombosis, macrophage cell formation and disrupt lipid homeostasis [76]. Primary bile acids synthesized from cholesterol can be deconjugated by certain microbiota that have bile salt hydroxylase (BSH) activity to secondary bile acid salts, which are less soluble and less efficiently reabsorbed from the intestine, resulting in more cholesterol demand to synthesize more bile acids. This cholesterol lowering-effect can be greatly affected by dysbiosis [77]. It is also well documented that gut dysbiosis can reduce SCFA-producing bacteria [78]. Taken together, dysbiosis-mediated atherosclerosis is a promising target for treatment of CVD. Indeed, OSA induces systemic inflammation, oxidative stress and increased sympathetic activity leading to endothelial dysfunction and atherosclerosis [18, 79], and is considered as an independent risk factor for CVD [80]. It is becoming an irrefutable fact that OSA-mediated gut dysbiosis contributes to CVD.

A multitude of recent randomized control trials (RCTs) have been conducted to examine the effects of pre-, pro-, and symbiotic supplements in patients with cardiovascular and metabolic diseases (Tables 1 and 2). Meta-analyses of such interventional studies revealed beneficial effects on cardiometabolic parameters. Zheng et al. identified 13 trials that included 671 patients for analysis and showed that pre-, pro- and synbiotic therapies significantly reduced C- reactive protein (CRP), malondialdehyde (MDA), total cholesterol (TC) and low-density lipoprotein (LDL) levels; they also increased glutathione (GSH), total antioxidant capacity (TAC) and high-density lipoprotein (HDL) in patients with chronic kidney disease (CKD) compared to placebo-treated groups. Subgroup analyses showed that duration of intervention, body mass index (BMI) and age had an effect of therapy on outcomes [81]. Another meta-analysis conducted by the same group on 16 RCTs with 1,060 patients reported that microbial treatment improved biomarkers of inflammation and oxidative stress in diabetic patients [82]. A total of 19 RCTs that included 967 participants with hypercholesterolemia were analyzed by Mo and colleagues and examined the effects of probiotics on serum lipid concentrations. The study found that probiotic interventions reduced TC and LDL levels compared to control with no significant effects on either TG or HDL concentrations. The lipid lowering effects were greater for longer intervention times, specific probiotic strains and were also more beneficial among younger participants [83]. Zhang et al. included 10 RCTs with myocardial infarction (n=1,392), coronary artery disease (n=4,490), stroke (n=7,078), and uncategorized CVD (n=51,707) that examined the effects of fermented dairy food intake on CVD risk in a recent meta-analysis. There was a significant decrease in CVD risk associated with fermented dairy food intake. In a subgroup analysis, cheese and yogurt were associated with decreased CVD risk [84]. Although clinical and observational studies suggest that natural pre- and probiotics have potential benefits on cardiometabolic disease, there is still not enough evidence to routinely and conclusively recommend them at this time, due to the heterogeneity of the approaches used and the lack of specific well circumscribed therapeutic targets and risk groups [85].

Table 2: Pre- and probiotic treatment effects on metabolic parameters.

FBG: fasting blood glucose, HbA1C: hemoglobin A1C, HDL: high-density lipoprotein, HOMA-IR: Homeostatic model of assessment of insulin resistance, hs-CRP: high sensitivity C-reactive protein, INF-λ: interferon-λ, LBP: lipopolysaccharide binding protein, LDL: low-density lipoprotein, NAFLD: non-alcoholic fatty liver disease, TC: total cholesterol, TG: triglycerides, TNF-α: tumor necrosis factor-α.

Pre-/Probiotic	Condition	Dose	Effects	Ref
Ecologic® Barrier (Bifidobacterium and Lactobacillus strains)	obese postmenopausal women	2 g sachets, 2 sachets/ day (1010/day) for 12 weeks	↓ Homocysteine, TNF-α, TC, LDL, and TG	Majewska et al.[182] 2020
UB316 (contains Bifidobacterium and Lactobacillus strains)	type 2 diabetics	5 x 109 CFU twice a day for 12 weeks	↓ HbA1c ↓ Weight ↔ FBG, HOMA-IR, insulin, TC, TG, HDL, and LDL levels. ↑ Quality of life	Madempudi et al.[183] 2019
whole-grain pasta containing barley β-glucans and Bacillus coagulans BC30, 6086	healthy sedentary overweight and obese subjects	1 serving/day of innovative pasta for 12 weeks	↓ hs-CRP and plasma LDL/HDL cholesterol ratio ↓ Plasma resistin	Angelino et al.[184] 2019
Symbiter Omega (multistrain probiotic with flax and wheat germ oil)	type-2 diabetics with NAFLD	1 sachet (10g) of (250 mg omega-3 fatty acids 1-5%) with probiotics for 8 weeks	↓ Fatty liver index ↔ Liver stiffness ↓ Serum gamma-glutamyl transpeptidase, TG, and TC, IL-1β, TNF-α, IL-8, IL-6, and INF-γ.	Kobyliak et al.[185] 2018
Bifidobacterium animalis subsp. lactis CECT 8145 (Ba8145) and heat-killed form (h-k Ba8145)	Abdominally obese subjects	1 capsule/day (1010 CFU of Ba8145, 1010 CFU of h-k Ba8145 for 3 months.	Ba8145 and h-k Ba8145: ↓ Waist circumference, waist circumference/height ratio, Conicity index and BMI h-k Ba8145: ↓ Visceral fat area, diastolic blood pressure and HOMA index.	Pedret et al.[186] 2019
Symbiotic supplement (contains Bifidobacterium, Lactobacillus strains) + oat β-glucan	Specific-pathogen free mice fed HFD	Oral gavage (108 CFU/day) + 1g/kg body weight/day of oat β-glucan for 12 weeks	↓ Body weight gain ↑ HOMA-IR score ↓ Fasting insulin, TC and LBP ↓ Hepatic steatosis and adipocyte size	Ke et al.[187] 2019
Lactobacillus plantarum TAR4, Lactobacillus acidophilus ATCC 4356	Sprague-Dawley rats fed high-cholesterol diet	Oral gavage (2 ml (109 CFU/ml) for 4 weeks	↓ TC, TG, liver TG	Lim et al.[188] 2020
Microencapsulated Lactobacillus plantarum LIP-1	Wistar rats fed HFD	Intragastric administration (20 mL supplement/kg BW (2 × 109 CFU/ml) for 4 weeks	↓ TC ↑ cholesterol excretion	Yao et al.[189] 2020
Bifidobacterium adolescentis, B. bifidum or Lactobacillus rhamnosus	C57BL/6J mice fed HFD and sucrose	intragastric administration (5 × 109 CFU/mL) for 12 weeks	↓Fasting and postprandial glucose levels ↑ Glucose tolerance ↓ Pancreatic damage L. rhamnosus >> Bifidobacterium strains in the regulation of blood lipid levels	Wang et al.[190] 2020

In animal models (Tables 1 and 2), probiotic supplementation (VSL#3) reduced high fat diet (HFD)-induced lesion development in apolipoprotein-E knockout (ApoE^−/−) mice in addition to reduced vascular inflammation, adhesion molecules, oxidized LDL, TNF-α, plasma TMAO, and TMAO-induced atherosclerosis [86]. A similar study in ApoE ^−/− mice examined the effects of two Lactobacillus strains (Lactobacillus acidophilus 4356 and 4962) on atherosclerosis. L.4356 significantly reduced atherosclerotic lesion, plasma TC and LDL via inhibition of intestinal cholesterol absorption, but no effects were observed in the L.4962 group.[87] In another study, plasma TMAO levels and TMAO-induced atherosclerosis in ApoE ^−/− mice were attenuated after Lactobacillus plantarum ZDY04 supplementation [88]. Probiotics have been shown to attenuate endothelial dysfunction by increasing the bioavailability of nitric oxide (NO), reducing oxidative stress and inflammation, and recruitment of progenitor cells [89]. VSL#3 supplementation attenuated oxidative stress-mediated endothelial dysfunction in rat mesentery arteries following bile duct ligation [90]. Furthermore, kefir yogurt (which has a high concentration of Lactobacillus and Bifidobacterium) consumption improved endothelial function in spontaneous hypertensive rats by improving NO bioavailability and reducing oxidative stress [91]. Another study revealed that rats fed with GOS and HFD had significant decreases in serum TG, TC, LDL and VLDL levels along with increased levels of HDL [92]. Although a large body of evidence supports a beneficial role of pre- and probiotics use in cardiovascular and metabolic disease, there are some studies that show no effect or even the opposite, i.e., pro-atherogenic effects [74, 93]. More mechanism-based studies of pre- and probiotics with multiple concentrations and strains are needed to address these discrepancies in the literature.

4.2. Neurobehavioral disorders

In the past decades, the gut microbiota emerged as a key regulator of the gut-brain axis. Evidence from germ free mice showed that myelination, dendritic growth, neurogenesis, blood brain barrier (BBB), microglia, neurotransmitters, and synaptic plasticity are all affected in the absence of microbiota [94]. Alterations in behavior in animals and humans given specific strains of bacteria, and the long-lasting effects of antibiotics on the brain in early life indicate that the microbiota can exert a considerable influence over host cognitive, mood and behavioral processes [94]. Pathways of communication between the gut microbiota and the brain are bidirectional and include: i) the vagus nerve [95] - Vagotomy can decrease the benefits of L. rhamnosus [96] while enteroendocrine cells can release glutamate that activates the vagal pathways [97]; ii) Intestinal microbiota can generate bioactive molecules including SCFAs that regulate microglial inflammatory responses[98], and modulate neurotransmitter release and activity like dopamine, λ-aminobutyric acid (GABA) and norepinephrine, as well as serotonin (through tryptophan metabolism) [94]; iii) the immune system is considered the most important communication pathway since the intestines contains a condensed concentration of immune cells that produces a variety of inflammatory cytokines, which can be transmitted to the central nervous system (CNS) via the circulatory and nervous systems [99]. Pro-inflammatory cytokines can reach the CNS through the humoral route (via saturable transporters at the BBB), neural route (via afferent neurons), and cellular route (immune cells migrating to the brain vasculature and parenchyma).[100] CNS inflammation is initiated when the microglia become activated and release pro-inflammatory cytokines including IL-1β, which eventually induces the production of the stress hormone cortisol [101, 102]. It is well-documented that inflammation is the main underlying mechanism in many neurodegenerative and neuropsychiatric disorders such as Parkinson’s disease (PD), Alzheimer’s disease (AD), and major depressive disorder (MDD).[103] Thus, gut dysbiosis can be linked to neurobehavioral disorders [104]. Pre- and probiotics that influence the gut-brain -axis are called “psychobiotics” and a significant number of studies have recently shown anxiolytic, anti-depressant, emotional, and systemic effects, and also beneficial cognitive effects [105].

Some of the most recent RCTs evaluating the effects of pre- and probiotics in neurobehavioral disorders are discussed in Table 3. A systematic review of 24 RCTs examining the effects of pre- and probiotics on depression and anxiety showed that probiotics, but not supplementation with prebiotics exerted significant beneficial effects on depression and anxiety with a larger effect observed for clinical referral samples than community-based cohorts [106]. Another meta-analysis that evaluated the effects of probiotics on depressive symptoms in 19 RCTs with a total of 1,901 participants found that probiotics treatment led to significant improvements in depressive symptoms compared to placebo. The effects however were in participants with major depressive disorders (MDD), but not in those with other clinical conditions or in the general population.[107] On the other hand, other systematic reviews found no beneficial effects of probiotics in patients with dementia [108] or schizophrenia [109].

Table 3: Pre- and probiotic treatment effects on neurobehavioral parameters.

BDNF: brain-derived neurotrophic factor, CEBPD: CCAAT/enhancing-binding protein delta, IL: interleukin, IkBα: nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor-α, SCFA: short-chain fatty acid.

Pre-/Probiotic	Condition	Dose	Effects	Ref
Multi-strain probiotic (Bifidobacterium longum, Lactobacillus acidophilus, and Enterococcus faecalis)	Elderly patients scheduled for elective orthopedic or colorectal surgery	probiotic capsules (0.84 g) twice daily from admission until discharge (≥107 CFU/capsule each)	↓ Incidence of postoperative cognitive impairment ↓ IL-6 and cortisol levels 5-7 days after surgery. ↔ Postoperative intensity, sleep quality and GI function recovery	Wang et al.[191] 2020
Bifidobacterium bifidum BGN4 and Bifidobacterium longum BORI	Healthy elders	4 capsules/day (109 CFU/d) for 12 weeks	↑ Mental flexibility test and stress score ↑ Serum BDNF levels	Kim et al.[192] 2020
Multi-strain probiotic (contains Bifidobacterium and Lactobacillus strains) + glutamine	Students	2 capsules/day (109 CFU/d) for 28 days	↓ Perceived stress scale, depression anxiety stress scale, and state-trait anxiety inventory scores ↓ Serum cortisol levels	Venkataraman et al.[193] 2020
Bifidobacterium breve A1 (MCC1274)	healthy older adults with mild cognitive impairment	2 capsules/day (1010 CFU/d) for 16 weeks	↑ Repeatable Battery for the Assessment of Neuropsychological Status total score ↑ Domain scores of immediate memories, visuospatial/constructional, and delayed memory in both intention-to-treat analysis and per-protocol analysis.	Xiao et al.[194] 2020
Lactobacillus Plantarum 299v (LP299v)	patients with major depressive disorder	2 capsules/day (109 CFU/d) for 28 days for 8 weeks	↑ Attention and Perceptivity test and the California Verbal Learning Test total recall of trials 1-5 ↓ Kynurenine levels	Rudzki et al.[195] 2019
VSL#3 (multi-strain probiotic)	Ferrets	Syringe-fed 10 ml Kitty Milk Replacement with (4.5 × 109 CFU of freeze-dried VSL#3) for 6 days	↔ Microbiome composition ↑ Preference for novelty (object and conspecific)	Dugyala et al.[196] 2020
Inulin	Mouse model of Alzheimer’s disease)	Dietary (8% inulin) for 16 weeks	↑ Beneficial microbiota and decreased harmful microbiota ↑ metabolism in the cecum, periphery, and brain (increased the levels of SCFAs, tryptophan-derived metabolites, bile acids and glycolytic metabolites) ↓ Inflammatory gene expression in the hippocampus.	Hoffman et al.[197] 2019
Lactobacillus plantarum (Lp 286) and (Lp 81)	Swiss mice	oral gavage solution (109/0.1 ml CFU each) for 30 days	↔ Locomotor activity or learning and memory Lp 286: ↑ Forced swim test and plus-maze discriminative avoidance task	Barros-Santos et al.[198] 2020
multi-strain probiotic (contains Bifidobacterium and Lactobacillus strains)	Mouse model of Parkinson disease	Dietary (1010 CFU/mouse/day) for 16 weeks	↓ Motor impairments in gait pattern, balance function, and motor coordination. ↑ Tyrosine hydroxylase-positive cell in the substantia nigra	Hsieh et al.[199] 2020
Lactobacillus acidophilus and Bifidobacterium infantis	Specific pathogen free (SPF) pregnant C57/BL6J mice. Pups challenged with IL-6 injection (postnatal systematic inflammation model)	fed daily from E16 to weaning with either a combination of 109 L. acidophilus and 109 B. infantis	↓ Postnatal systemic inflammation associated with reduced astrocyte/microglia activation and downregulation of the transcriptional regulators CEBPD and IκBα) ↓ Blood-brain barrier permeability and tight junction protein expression in the offspring at pre-weaned age. ↓ Markers associated with leukocyte transendothelial migration, extracellular matrix injury and neuroinflammation. ↑ Neuronal and oligodendrocyte progenitor cell development in the offspring.	Lu et al.[200] 2020

A growing body of evidence from experimental studies has shown that psychobiotics can reduce anxiety and depression (Table 3). In a mouse model of early life-stress, administration of Lactobacillus plantarum PS128 decreased time spent immobile in the Forced Swim Test (FST) (a classic test of depression in rodents) and increased the time spent in the center of the arena in the Open Field Test (OFT), and more entries in the open arms of the Elevated Plus Maze (EPM) (both of these test evaluate anxiety levels). PS128 also increased levels of dopamine, serotonin in the prefrontal cortex and striatum, as well as decreased inflammation and corticosterone levels [110]. In a rat model of AD, multi-strain probiotic administration (Lactobacillus and Bifidobacterium) significantly improved spatial memory acquisition and retention measured by Morris Water Maze (MWM). Treatment also reduced Aβ plaques, MDA, IL-1β, TNF-α and other inflammatory markers in the hippocampus [111]. Prebiotics such as FOS have been shown to increase acetylcholine, serotonin, and adrenaline, as well as decreased damage to the CA1 region of the hippocampus in a rat model of AD [112]. FOS also improved cognition in a mouse model of AD by improving performance in cognitive tests and synaptic plasticity in the hippocampus possibly by regulating levels of glucagon-like peptide-1 (GLP-1) [113]. Increased expression of brain-derived neurotrophic factor (BDNF) and N-Methyl-d-aspartate receptor (NMDAR) subunits in the hippocampus and frontal cortex was observed in mice fed FOS and GOS. In healthy animals, dietary FOS and GOS improved parameters in OFT, FST and EPM [114]. Although our understanding of the gut-brain-axis is still in its early stages, much work is still required to prove that changes in microbiota are central to the pathophysiology of neural system-related disorders. Also, large-scale studies testing optimal doses, strains and timing of application with causative and mechanistic analysis are essential to the field of pre- and probiotics as therapeutic interventions.

From another perspective, many studies have pointed to a higher prevalence of OSA in many neurodegenerative disorders. In Alzheimer dementia, IH induces neuro-inflammation, releases of reactive oxygen species, and subsequently cellular apoptosis [115]. The inflammatory process leads to the accumulation of β amyloid protein in the central nervous system and hyperphosphorylation of tau protein [116, 117]. In Parkinson disease (PD), the exact correlation between the disease and OSA remains unclear. However, dysfunction in the upper airway dilator muscles has been shown in patients with PD which is probably part of the tremor, bradykinesia, and rigidity that is noticed in the disease in general [118]. This is a very similar picture to what is seen in patients with multiple systemic atrophy [119]. Interestingly, many studies confirmed the cognitive improvements in patients with neurogenerative diseases and OSA after using PAP therapy [120–124]. Moreover, a close association was noted between many mental diseases and OSA. A systematic review by Gupta et al [125]. concluded that OSA prevalence maybe increased in patients with MDD and post-traumatic stress disorders (PTSD) and treatment with PAP therapy improved many of psychiatric symptoms in these patients.

3. Gastrointestinal disorders

Inflammatory Bowel Disease (IBD) is characterized by chronic immune-mediated intestinal inflammatory condition driven by environmental, genetic and microbial factors.[126] IBD includes Ulcerative Colitis (UC) and Crohn’s Disease (CD) and is considered a major risk factor for malignancy development, such as colorectal cancer [127]. Recent metagenomic and metabolomic studies revealed changes in microbial composition and metabolites in patients with IBD [128]. Indeed, the depleted metabolites and related species were of anti-inflammatory nature while the enriched metabolites were pro-inflammatory [129]. For instance, SCFA-producing bacteria such as Faecalibacterium prausnitzii[128] and Roseburia intestinalis and hominis[130] were decreased in stool samples from patients with UC and CD. In addition to decreased SCFA levels, tryptophan metabolism (which promotes mucus production and down-regulates inflammation) is also reduced in IBD [131]. It is well documented that secondary bile acids play an important role in modulating host T regulatory function [132]. In patients with IBD, fecal primary bile acids and its conjugates (glycine and taurine) were increased, while secondary bile acids (lithocholate and deoxycholate) were decreased [126]. Bacteria associated with bile acid metabolism, such as Roseburia, Subdoligranulum, and Clostridium leptum were all reduced in IBD [126]. On the other hand, invasive virulent bacteria such as adherent invasive Escherichia coli (AIEC), Enterococcus faecium and Campylobacter concisus, and pro-inflammatory metabolites such as nictotinuric acid, taurine and acylcarnitines were all increased in IBD [129].

Many recent RCTs described the benefits of pre- and probiotics in the management of gastrointestinal disorders (Table 4). A recent meta-analysis included 5 RCTs examining the efficacy of prebiotics and probiotics in functional dyspepsia and found that treatment improved symptoms scores of functional dyspepsia [133]. Another meta-analysis identified 11 RCTs with 729 participants and assessed the effects of prebiotics in patients with IBS but found no differences for severity of abdominal pain, bloating and flatulence and quality of life score. Only flatulence improved by prebiotic doses <6 grams/day and by non-inulin type fructan prebiotics. However, there was a significant increase in Bifdobacteria [134]. One study included 18 RCTs with 1,491 patients with UC and reported that remission efficacy was improved only when probiotics containing Bifidobacterium were used [135]. Another meta-analysis included 10 studies with 1,049 patients evaluated the effects of probiotics on IBD. Results showed no differences in remission, relapse, and complication rates between Escherichia coli Nissle and mesalazine groups. Subgroup analysis suggested that VSL#3 presented a higher remission rate and lower relapse rate [136].

Table 4: Recent interventional studies using pre- and probiotics in gastrointestinal disorders.

COX-2: cyclooxygenase-2, CRP: C-reactive protein, hs-CRP: high-sensitivity C-reactive protein, IBD: inflammatory bowel disease, IBS: irritable bowel syndrome, IL: interleukin, PCNA: proliferating cell nuclear antigen, SCFA: short-chain fatty acid.

Pre-/Probiotic	Condition	Dose	Effects	Ref
Bifico® (Bacillus Coagulans GBI30 6086)	patients with IBD	2 tablets (2 x 1010 CFU/day) with 2 tablets of mesalazine	↓ Levels of fecal lactoferrin, 1-antitrypsin, β2-microglobulin, hs-CRP, and IL-6 ↓ Activity scores and recurrence rate	Fan et al.[201] 2019
Symprove™ (Lactobacillus and Enterococcus strains)	patients with ulcerative colitis and Crohn’s disease	Oral suspension 50ml/dose (1010 CFU/day) for 4 weeks	↔ IBD-quality of life scores ↓ Fecal calprotectin in patients with ulcerative colitis but not Crohn’s disease.	Bjarnason et al.[202] 2019
Multi-strain probiotic (contains Bifidobacterium and Lactobacillus strains)	patients with colorectal adenocarcinoma	2 capsules/day (109 CFU/d) for one month then 1 capsule/day for up to one year	↓ Reduction in postoperative complications in the localization of tumors on the rectum and the ascending colon	Bajramagic et al.[203] 2019
Xyloglucan (XG), pea protein and tannins (PPT) from grape seed extract, and xylo-oligosaccharides (XOS)	60 patients with diarrhea-predominant IBS	one capsule twice daily	↑ Stool consistency ↓ Abdominal pain and bloating ↑ Quality of life	Trifan et al.[204] 2019
Multi-strain probiotic (contains Bifidobacterium and Lactobacillus strains)	colorectal cancer patients at 4 weeks after surgery	orally twice daily (30 × 1010) for 6 months	↓ Plasma pro-inflammatory cytokine (TNF-α, IL-6, IL-10, IL-12, IL-17A, IL-17C and IL-22)	Zaharuddin et al.[205] 2019
Vivomix™ plus Bacillus subtilis	Mouse model of colitis	Oral gavage (50 × 109 probiotic CFU/kg) for 7 days	↓ Development of signs and symptoms of colitis ↓ Expression of pro-inflammatory cytokines (IL-6 and TNF-α) in colon ↑ Expression of IL-10 mRNA and the number of T regulatory cells in colon	Biagioli et al.[206] 2020
Lactobacillus rhamnosus 1.0320 + inulin	Mouse model of colitis	Intragastric (2 x 108 cfu/0.2ml) + 20mg inulin for 28 days	↓ Signs of colitis ↓ Disease Activity Index of colon tissue ↓ Myeloperoxidase activity ↑ Hemoglobin content ↓ Expression levels of inflammatory cytokines IL-1β, IL-6, TNF-α ↑ Expression of IL-10.	Liu et al.[207] 2020
Bacillus coagulans MTCC5856 plus green banana resistant starch (GBRS)	Mouse model of colitis	Dietary (2 × 109 CFU/day/mouse) and GBRS (400 mg/day/mouse) for 7 days followed by 7 days of colitis induction	↓ Disease Activity Index score ↑ Expressions of tight junction proteins ↓ Expression of IL-1β and CRP ↔ SCFA levels across the whole length of the colon.	Shinde et al.[208] 2020
Multi-strain probiotic (contains Bifidobacterium and Lactobacillus strains)	Mouse model of colorectal cancer	Oral gavage (5 x 1011 cfu/0.2ml) for 3 weeks	↓ Disease activity ↓ Incidence and progress of tumors to higher stages and grades	Parisa et al.[209] 2020
Lactobacillus acidophilus + Djulis	Rat model of colorectal cancer	Dietary (10% djulis + 5 × 106 CFU/g) for 10 weeks	↓ Numbers of total aberrant crypt foci, sialomucin-producing aberrant crypt foci, and mucin-depleted foci in the distal colon ↓ Expressions of PCNA and COX-2 ↓ Apoptosis-related proteins	Lee et al.[210] 2019

In murine models of colitis, several probiotics such as Bifidobacterium (infantis, bifidum PRL2010 and bifidum 231), Lactobacillus (plantarum 299, animalis BB12 and reuteri 5454), Faecalibacterium prausnitzii and VSL#3 reduced symptoms of disease, enhanced mucosal barrier and exhibited anti-inflammatory effects [137]. Furthermore, Lactobacillus fermentum V3 and plantarum YYC-3 ameliorated colitis-associated tumorigenesis in mice mainly through downregulation of the expression of IL-6, IL-17 f and IL-22, and suppression of the activation of nuclear factor kappa B (Nf-Kb) and Wnt transcriptional pathways [138, 139]. Moreover, in a mouse model of intestinal adenoma, administration of the SCFA-producing Clostridium butyricum exhibited decreased proliferation and increased apoptosis of intestinal tumor cells mainly through suppression of Wnt signaling. Additionally, C. butyricum modulated gut microbiota composition and increased cecal SCFA levels [140]. Symbiotic treatment of rats with colitis (Lactobacillus casei 01 and chitosan-Ca-alignate microparticles) reduced colonic damage and increased lactobacilli counts in feces. Treatment also reduced inflammation and lesions associated with a significant reduction in myeloperoxidase (MPO) activity [141]. Considering the pivotal role of the gut microbiota in IBD and the negative effects of conventional therapy, the supplementation of pre- and probiotics open a potential avenue in managing IBD. Also, standardizing therapeutic protocols could improve future studies and minimize potential bias.

5. Potential novel therapeutic interventions targeting the gut microbiome for the treatment of OSA

Continuous positive airway pressure (CPAP) is the mainstay treatment for OSA. However, the main challenge facing CPAP therapy is adherence. CPAP adherence is overall low and estimated to be around 35% over the last twenty years, with no significant improvements despite improved machines and patient-machine interfaces [142]. Mandibular advancement devices (MAD) are another option of treating OSA, however, they are not optimal for every OSA patient (for example severe OSA patients are unlikely to benefit) and can cause temporomandibular joint discomfort long-term [143]. The remaining options (including surgery and hypoglossal nerve stimulator) are invasive and may carry a higher risk for OSA patients (especially the elderly). We should not forget the role of conservative management (e.g. weight loss and exercise) in the armamentarium aimed at minimizing OSA severity and consequences. Peppard et al [144]. concluded that 10% weight loss reduced the AHI by 26% and 10% weight gain increases 6-fold the risk of developing OSA. Moreover, weight loss (dietary or through bariatric surgery) has been shown to impact the gut microbiome [145].

Therapies that target gut dysbiosis are potential novel therapies that can be adjuvant to OSA standard treatment and can potentially translate into benefits regarding the morbidities associated with OSA.

5.1. Pre- and probiotics

As stated above, experimental and clinical studies using pre- and probiotics treatments point to favorable outcomes in dysbiosis-related conditions. However, very few studies have explored the effects of pre- and probiotics in OSA-mediated dysbiosis. Ganesh et al. treated an rats with OSA (endotracheal obstruction device implantation) and fed HFD with either one billion CFU of Clostridium butyricum or 20% high amylose corn starch Hylon VII and reported that both treatments prevented OSA-induced hypertension, altered the gut microbiota composition, reduced OSA-induced damage to the gut wall, normalized cecal acetate levels and prevented neuroinflammation associated with OSA. They also reported that continuous infusion of exogenous acetate for 2 weeks prevented OSA-induced hypertension [146].

In another study, Liu et al. investigated the effect of Lactobacillus rhamnosus GG strain (LGG) on the development of OSA-induced hypertension in rats exposed to IH. As mentioned above, they found that IH combined with high salt content in the diet exacerbated the severity of hypertension by increasing levels of plasma TMAO, IFN-λ release from inflammatory cells, and reductions in TGF-β1, while also displaying depleted Lactobacillus content in the gut. They also reported increased expression of phosphorylated extracellular signal-regulated kinase1/2 (ERK1/2), phosphorylated protein kinase B (Akt) and phosphorylated mammalian target of rapamycin (mTOR) in the aorta. Treatment with LGG prevented increase in hypertension, modulated Th1/Th2 cytokine imbalance and blunted the phosphorylation of ERK1/2, Akt and mTOR.[42] Another study found that LGG treatment in obese mice exposed to IH for 15 weeks prevented cardiac remodeling and inflammation. The cardioprotective effect was associated with up-regulation of nuclear factor erythroid2-related factor (Nrf2) – mediated antioxidant pathways [147].

A very recent study by Liu et al. reported glucose intolerance, insulin resistance, increased adipose tissue inflammation and adipocyte size and reduced circulation levels of adiponectin in obese mice exposed to IH for 15 weeks, all parameters were restored by LGG treatment and was associated with increased hepatic fibroblast growth factor 21 (FGF21) mRNA expression and circulating FGF21 protein levels. The increased FGF21 was associated with upregulated hepatic peroxisome proliferator-activated receptor-α (PPAR-α) expression and fecal butyrate concentration. Furthermore, LGG treatment significantly reduced IH-induced hepatic fat accumulation and apoptosis [148]. Collectively, the data from these experimental studies show metabolic, neuronal and cardiovascular benefits which indicate to a potential adjuvant therapy in patients with OSA. Nevertheless, large RCTs need to be performed using multiple species of probiotics and types of prebiotics to determine their health benefits in OSA-induced dysbiosis.

5.2. Fecal matter transplantation or fecal microbiota transfer (FMT)

FMT simply implies the transfer of stool from healthy donor’s colon to a recipient’s colon [149]. This procedure is utilized in patients with difficult to treat or resistant Clostridium difficile with or without inflammatory bowel disease [150]. This is the only approved clinical indication for the use of FMT as per the European consensus conference on fecal microbiota transplantation in clinical practice [151]. However, FMT has also shown very promising results in the treatment of obesity and diabetes mellitus. In obesity, FMT trials in animal and human models showed an increase in Bacteroidetes and decrease in Firmicutes [152], increase in butyrate-producing bacteria (e.g., Roseburia intestinalis or Eubacterium hallii) [153], and increase in Verrucomicrobia, Akkermansia, and Alistipes [154]. In type II diabetes mellitus, FMT has been shown to correct insulin resistance and promote the repair and recovery of islet cells [155]. FMT plays an important therapeutic intervention in systemic hypertension in both animal and human models. Adnan et al. reported that FMT of cecal microbiota from spontaneous hypertensive and stroke prone rats to normotensive WKY rats resulted in an elevation in blood pressure in the latter [156]. In humans, Li et al. [157] examined the fecal microbiota in three groups in humans (Normal, pre-hypertensive and hypertensive) and noticed significant increases in the relative abundance of Prevotella and Klebsiella in both pre-hypertensive and hypertensive groups compared to the normotensive group. They also showed that hypertension is transferrable through FMT. Fecal microbiota from three groups were transferred to germ-free normotensive mice which led to increases in blood pressure in the mice who received fecal microbiota from the hypertensive and pre-hypertensive groups.

FMT has been utilized to test the transferability of OSA-induced hypertension in animal models. Durgan et al. [47] investigated the effect of high fat diet (HFD) alone, OSA alone, and both HFD and OSA on systemic hypertension in a rat model implanted with an endotracheal device that can be inflated/deflated mimicking an apnea/hypopnea at a rate of 60 times per hour for 8 hours every day. They concluded that HFD and OSA caused significant elevation in systolic blood pressure compared to HFD alone or OSA alone suggesting a synergistic effect of both on blood pressure. To further prove this hypothesis, they transferred the gut microbiota from donor rats fed (HFD) with no OSA and rats fed HFD with OSA into recipient rats fed normal chow diet with OSA and found that recipients who received gut microbiota from OSA-HFD donors developed significant elevation in systolic blood pressure compared to HFD alone donors. They also noticed that treating rats with oral antibiotics to deplete the gut microbiota prevented the elevation in systolic blood pressure. Such findings support the role of OSA- induced gut dysbiosis in inducing or exacerbating systemic hypertension. Different parameters of gut dysbiosis were tested in this trial. First, the F/B ratio which is the signature of gut dysbiosis was the highest in HFD rats alone. Combined HFD and OSA showed lower increases in F/B ratio but still higher than sham rats fed normal chow diet. Second, the Chao index (which reflects the total number of distinct genera) was decreased in rats fed HFD. Third, Shannon index (which reflects the relative abundance and richness of genera) was not statistically significant in rate fed HFD. It was noticed also that rats that are fed HFD and have OSA have significant reduction in the relative abundance of SCFA-producing bacteria, especially butyrate. Ruminococcaceae are prominent producers of butyrate and their abundance was reduced from 20% to 10% in this study. At the same time, there was an increase in the relative abundance of lactate-producing bacteria, namely (Lactococcus and Coprobacillus). Of note, lactate in gut is known to increase blood pressure [158].

Currently, the only clinical indication for FMT in humans is severe and resistant Clostridium difficile to antibiotic therapy. The procedure per se is effective and relatively safe. Cammarota et al.[159] conducted a systematic review which included one randomized controlled trial and showed resolution of diarrhea in 87% of total participants with infusion of the fecal suspension matter in the colon using colonoscopy being the most effective site of infusion. Up to date, FMT is not indicated for any other GI or non-GI diseases in humans.

5.3. Short chain fatty acids (SCFAs)

SCFAs include butyrate, propionate, and acetate. They are produced by the gut microbiota and have many beneficial effects locally (in the colon) and extracolonic. SCFAs, especially butyrate, are the major source of energy and nutrients for colonocytes [160]. Butyrate plays a role in the oxidative phosphorylation process that takes place in the mitochondria of the colonic epithelial cells [161]. Butyrate has anti-inflammatory features via the inhibition of IL-12, upregulation of IL-10 in monocytes [162], inhibition of many proinflammatory mediators (such as IL-1β, TNF-α), and suppression of NF-κB pathway which plays an important role in inflammation.[163] Similarly, acetate has anti-inflammatory features in vitro by enhancing the release of reactive oxygen species that have a bactericidal activity against pathogens [164]. Butyrate showed an anti-tumorigenic activity on variable human cancer cell lines (such has human hepatoma cells and human colon cancer cells).[165, 166] Locally, SCFAs play a vital role in maintaining the gut integrity. The mucus layer that separates the colonocytes from the lumen consists of secretory and epithelial membrane-bound mucin glycoproteins [167]. Butyrate and propionate can induce the secretion of some of these mucin glycoprotein to protect the mucus layer in some diseases (e.g., mucositis) [168]. In addition to the mucus layer, tight junction proteins (e.g. ZO-1 and occludin) are located between the epithelial cells laterally [169]. In vitro, butyrate was shown to enhance the assembly of ZO-1 and occludin proteins which increase trans-epithelial resistance (TER) and consequently maintain the gut integrity and prevent gut permeability [170]. Another SCFA, acetate, was shown in vitro to prevent the translocation of shiga toxin from the lumen into the blood stream [171] (Table 5).

Table 5. The use of SCFAs in some diseases (Human and animal models).

SCFA= short chain fatty acids; DBP= diastolic blood pressure; SBP= systolic blood pressure; DOCA= deoxycorticosterone acetate; HDAC= histone deacetylase; HFD= high fat diet; GLP-1= Glucagon-like peptide-1; GPR= G-protein-coupled Receptor.

SCFA	Condition	Effects	Reference
Cardiovascular disorders
Propionate	Hypertensive mice	↓ Blood pressure ↓ Vascular inflammation and Atherosclerosis ↓ Cardiac immune cell infiltration and remodeling ↓ Susceptibility to Ventricular Arrythmias	Bartolomaeus et al.[211] 2019
Formate	Patients with hypertension	↓ Systolic and diastolic blood pressure	Holmes et al.[212] 2008
Acetate	Hypertensive mice (DOCA)	↓ Blood pressure ↓ Cardiac hypertrophy ↓ Cardiac and renal fibrosis	Marques et al.[213] 2017
Metabolic disorders
Butyrate and propionate	Diabetic mice	Both: ↑ Pancreatic β-cells by incretins ↑ plasma insulin ↑ Oral glucose tolerance test Propionate: ↓ Fasting blood glucose	Lin et al.[214] 2012
Butyrate	Diabetic mice	↓ Weight gain ↓ Insulin resistance ↑ energy expenditure ↑ Mitochondrial function	Gao et al.[215] 2009
Butyrate	Diabetic rats	↑ β-cell proliferation ↓ β-cell apoptosis ↓ HbA1c and plasma glucose ↑ Plasma insulin	Khan et al.[216] 2014
Butyrate and propionate	Obese mice	Butyrate: ↓ Food intake Propionate: ↔ Food intake	Lin et al.[214] 2012
Butyrate	Obese mice	↓ Total body, liver and epididymal fat pad weights ↑ Adiponectin pathway ↑ Mitochondrial function ↑ Fatty acid β-oxidation ↓ Lipid deposition in the muscle	Hong et al.[217] 2016
Butyrate	Obese mice	↓ Weight gain ↓ Food intake ↑ release of GLP-1 ↑ glucose tolerance	Yadav et al.[218] 2013
Neoplastic disorders
Butyrate	Human liver cancer cell lines	↓ telomerase activity ↔ levels of reverse transcriptase component (as HDAC inhibitor)	Nakamura et al.[165] 2001
Butyrate	Human colon cancer cell line	↑ cancer cell apoptosis as a GPR109A ligand	Thangaraju et al.[219] 2009

Since OSA is associated with gut dysbiosis, theoretically, supplementation with SCFAs could protect against developing OSA-induced complications. To test this theory, Ganesh et al. [146] conducted an animal trial in which rats were randomized into three groups (HFD fed rats, HFD + Hylon “prebiotic”, and HFD + C. butyricum “probiotic”). Since its known that OSA alter the gut microbiota profile, the next step was to test if the metabolomic measures (i.e. the metabolites produced by altered gut microbiota) and they found that OSA decreased acetate concentration in the rat’s colon in HFD group. The addition of Hylon and C. butyricum increased acetate concentration independent of OSA and OSA did not have any impact on butyrate or propionate levels. After concluding that OSA reduced acetate concentration in the cecum, the next step was to examine the effect of supplementing acetate. They continuously replaced acetate in the colon using an indwelling catheter to see if that can prevent the development of HTN in OSA group. Cecal infusion of 20 μmol/(kg/min) sodium acetate or phosphate buffered saline (PBS) started and continued for two weeks in sham and OSA rats. After two weeks, OSA rats treated with PBS showed reduction in acetate in the cecum while OSA rats treated with acetate showed 2-fold increase in acetate concentration compared to sham. Acetate also prevented any increase in IL-1α and IL-6 in OSA group.

6. Conclusions

OSA is a chronic disease associated with gut dysbiosis. Recent findings reveal complex interactions between the microbiota and other biological systems, whereby perturbations in the microbiota structure and metabolites are linked to cardiometabolic, neurobehavioral and gastrointestinal diseases. Gut dysbiosis is characterized by compromising the intestinal barrier and induction of low systemic inflammation. Dysbiosis treatment with pre,pro-biotics, SCFAs and FMT have shown to mitigate dysbiosis-induced pathophysiology in many disorders. Current pre-clinical studies show beneficial effects of pre- and probiotics in animal models of OSA. However, randomized control trials in OSA patients and more mechanistic studies are needed to justify recommending those interventions as adjuvant therapies in OSA with the aim of preventing, abrogating or at least attenuating OSA-induced morbidities.

7. Expert opinion

The field of the gut microbiome and gut dysbiosis is rapidly evolving. Since the launch of the Human Microbiome Project by the National Institutes of Health in 2007 and with the current advances in metagenomics, metatranscriptomics, and metabolomics, our understanding of the human microbiome and its complicated interactions with the host has enormously progressed [172]. Recent studies indicate that perturbations in the gut microbiome can lead to a plethora of disorders across biological systems [173]. However, more mechanistic pre-clinical studies and large cohort studies are required to further elucidate these interactions. Questions have also been raised regarding the reproducibility of investigations on the gut microbiome research in experimental animals [174]. Environmental factors, rodent husbandry, and treatment protocols should be standardized and show the desired reproducibility. Although many of the microbes have been identified, many others (including viruses and fungi) are still undetermined [175]. Defining the host microbiome is not a straightforward process. For instance, High F:B ratio is usually associated with pathological states [176] but this is not always the case [177, 178]. In this context, the “pathobiome” is still not well defined.

OSA is associated with alterations in the gut microbiome profile and metabolites [179]. Linking OSA to gut dysbiosis might open new avenues for targeting chronic illnesses associated with OSA, however, the literature so far does not prove causal relationship. For example, Moreno et al. examined the impact of reversing IH in mice for six weeks and surprisingly this did not completely reverse gut dysbiosis [45]. Accordingly, we need large cohort studies to help establish a correlation between OSA and gut dysbiosis, and whether treating OSA can restore the gut microbiota, notwithstanding the potential confounders. In this regard, we need to tackle numerous issues such as sample collection, storage, processing, and potential contaminations [74]. Although the exact mechanisms behind OSA-induced dysbiosis are still not fully identified, it is probable that OSA-induced IH and sleep fragmentation can foster inflammation and oxidative stress, the latter playing a role in initiating gut dysbiosis, all the while dysbiosis can exacerbate inflammation and oxidative stress in a vicious cycle. Furthermore, gut dysbiosis and OSA share similar bidirectional relationships with chronic conditions such as obesity, type 2 diabetes mellitus, and systemic hypertension, which further complicates the OSA-dysbiosis relationships, and makes therapeutic interventions more difficult to delineate. Large interventional studies in OSA patients using CPAP and its effect on the microbiome with consistent follow-up, may provide a better insight on the efficacy of CPAP in targeting dysbiosis and may also suggest whether adjuvant interventions are worthwhile pursuing.

Current evidence from clinical and experimental studies show beneficial effects of pre- and probiotics treating gut dysbiosis in many conditions [180]. While no studies have been conducted in OSA patients, animal models of OSA treated with pre- and probiotics could modulate metabolism, blood pressure and immunity [146]. Given the useful effects of pre-and probiotics in mitigating dysbiosis-related inflammation in cardiometabolic, gastrointestinal and neurobehavioral conditions, it is plausible that pre- and probiotics treatment can target OSA-induced dysbiosis. Additionally, the anti-inflammatory effects of pre- and probiotics may be able to modify risk factors of OSA (such as obesity and hypertension) and in turn reduce the incidence or severity of OSA (Figure 3). However, there is minimal guidance from regulatory entities with conflicting information on probiotics in the literature sources used for clinical guidelines [181]. Furthermore, more studies in OSA patients and animal OSA models with multiple species and doses of pre- and probiotics are needed to better tailor treatments for OSA patients. SCFAs (especially butyrate and acetate) are very promising interventions targeting OSA-induced dysbiosis and OSA-induced co-morbidities especially that SCFAs have been shown to ameliorate OSA-induced inflammation and OSA-induced hypertension in animal models. Finally, FMT is another promising approach to ameliorate OSA-induced morbid conditions, yet it is too early to consider this option due to the lack of sufficient evidence. FMT has proven that some pathological features and diseases (e.g. hypertension and obesity) are transferrable by the gut microbiome in animal models. More animal studies are required to test the transferability of OSA symptoms (e.g. excessive daytime sleepiness) and OSA-induced co-morbidities by the gut microbiome.

Figure 3: Treatment of gut dysbiosis may ameliorate chronic diseases and reduce the development or progression of OSA.

IH: intermittent hypoxia, FMT: fecal matter transplantation, OSA: obstructive sleep apnea, SCFA: short chain fatty acid, SF: sleep fragmentation

Additional mechanistic and large cohort studies are needed to establish cause and effect relationships. Pre-and probiotics, beneficial bacterial metabolites and fecal matter transplantation from healthy subjects are novel therapeutic options that may benefit patients with OSA and OSA-induced co-morbidities. However, we are still in the early stages of establishing the recommendations of such use. Yet, we could envision a scenario soon where a well-defined microbiota profile associated with the production of toxic molecules would be appropriately modulated to lessen the disease risk of OSA, and that such approaches may open the way for precision-based interventions, one patient at a time.

Article Highlights.

• Obstructive sleep apnea (OSA) is a chronic condition associated with multiple chronic disorders affecting multiple systems

• The gut microbiome is integral to human health and imbalance in gut commensal bacteria is linked to many diseases

• OSA can alter the gut microbiome leading to dysbiosis

• Pre- and probiotics and short chain fatty acids (SCFA) can protect against dysbiosis by strengthening the gut barrier integrity and production of immunomodulatory compounds

• Given the useful effects of pre-and probiotics in mitigating dysbiosis-related inflammation in cardiometabolic, gastrointestinal and neurobehavioral conditions, it is plausible that pre- and probiotic treatment can target OSA-induced dysbiosis

List of abbreviations:

AD

Alzheimer’s disease

AHI

apnea-hypopnea index

Akt

protein kinase B

ApoE−/−

apolipoprotein-E knockout

BA

bile acid

BBB

blood brain barrier

BDNF

brain-derived neurotrophic factor

CD

Crohn’s disease

CKD

chronic kidney disease

CNS

central nervous system

CPAP

continuous positive airway pressure

CRP

c- reactive protein

EPM

elevated plus maze

ERK1/2

extracellular signal-regulated kinase1/2

F/B ratio

Firmicutes/Bacteroidetes ratio

FGF21

fibroblast growth factor 21

FIO2

fraction of oxygen inspired

FMT

fecal matter transfer/fecal microbiota transplantation

FOS

fructooligosaccharides

FST

forced swim test

GABA

λ-aminobutyric acid

GLP-1

glucagon-like peptide-1

GOS

galactooligosaccharides

GSH

glutathione

HDL

high-density lipoprotein

HFD

High fat diet

HTN

hypertension

IBD

inflammatory bowel disease

IH

intermittent hypoxia

LDL

low-density lipoprotein

LPS

lipopolysaccharides

MAD

Mandibular advancement devices

MDA

malondialdehyde

MDD

major depressive disorder

MPO

myeloperoxidase

mTOR

mammalian target of rapamycin

Nf-ƘB

nuclear factor kappa B

NO

nitric oxide

NMDAR

N-Methyl-d-aspartate receptor

Nrf2

nuclear factor erythroid2-related factor

OSA

obstructive sleep apnea

OFT

open field test

PaO2

oxygen partial pressure

PD

Parkinson’s disease

PPAR-α

proliferator-activated receptor-α

RCT

randomized control trail

ROS

reactive oxygen species

SCFA

short-chain fatty acids

TAC

total antioxidant capacity

TC

total cholesterol

TER

trans-epithelial resistance

TMA

trimethylamine

TMAO

trimethylamine N-oxide

WBCs

white blood cells

UC

ulcerative colitis