Liver disease are associated with a wide spectrum of neurological changes, of which the best known is hepatic encephalopathy (HE). Historically, hyperammonemia, was considered the main etiological factor; however, studies demonstrated a key role of neuroinflammation in the development of neurological complications in this setting. Neuroinflammation is characterized by activation of microglial cells and brain secretion of pro-inflammatory cytokines, such as tumor necrosis factor (TNF)-α, interleukin (IL)-1β, and IL-6, which alter neurotransmission, leading to cognitive and motor dysfunction. Changes in the gut microbiota resulting from liver disease play a crucial role in the pathogenesis of neuroinflammation. Dysbiosis and altered intestinal permeability, resulting in bacterial translocation and endotoxemia, are responsible for systemic inflammation, which can spread to brain tissue and trigger neuroinflammation. In addition, metabolites derived from the gut microbiota can act on the central nervous system and facilitate the development of neurological complications, exacerbating clinical manifestations.
1. Introduction
Hepatic encephalopathy (HE) is one of the most important complications related to acute liver failure (ALF) and liver cirrhosis. It is characterized by a wide spectrum of neurological symptoms, ranging from subtle cognitive impairment to coma
[1].
According to the type of underlying liver disorder, HE can be divided into type A, B, or C
[2]. Type A is related to the development of ALF in patients without a previous history of liver disease, typically in the setting of acute viral infections, drug-induced liver injury, and vascular disorders. The presence of HE in patients with ALF correlates with high mortality rates, being characterized by cerebral oedema and intracranial hypertension, which may induce brain herniation. Type B HE is due to the presence of portosystemic shunts in the absence of underlying liver disease, while type C HE develops in patients with liver cirrhosis and is mainly characterized by impaired neurological function.
Whatever the type, HE is a predictor of poor prognosis, with not only a great impact on patients’ survival and quality of life, but also a heavy burden for caregivers
[3].
Although HE pathogenic mechanisms are still not fully elucidated, ammonia has always been considered the main causative factor [4]. However, it has been shown that ammonia levels do not correlate with the severity of HE and that HE can also manifest in patients with normal ammonia levels, hinting at the presence of other contributing factors, such as systemic inflammation and oxidative stress [5][6]. Recently several studies have suggested a key role of neuroinflammation in this setting [7]. Indeed, systemic inflammation and hyperammonemia stimulate in concert neuroinflammation [8].
2. Pathophysiology of Neuroinflammation
Neuroinflammation refers to the inflammatory response that develops within the central nervous system following several insults, such as infections, traumatic injury, or exposure to toxic metabolites
[9]. Microglia and astrocytes, the main brain innate immune cells, drive this process by producing several pro-inflammatory cytokines, such as interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α, chemokines, including C-C motif chemokine ligand 2 (CCL2), CCL5, and C-X-C motif chemokine ligand 1 CXCL1, secondary messengers, such as nitric oxide (NO) and prostaglandins, and reactive oxygen species
[10]. Additionally, endothelial cells and peripheral immune cells act in promoting this inflammatory status
[11].
Under normal conditions, the central nervous system (CNS) is protected from the entrance of potentially pathological agents into the cerebral circulation thanks to the blood–brain barrier (BBB), a highly selective structure made of endothelial cells and astrocytes
[12]. The integrity of the BBB is guaranteed by tight junction (TJs) proteins such as occludin and claudin-5
[13]. Following injury and systemic inflammation, TJs undergo a dysregulation process that affects the integrity of the BBB, increasing the permeability of dangerous molecules that promote brain inflammation. Activated microglia and astrocytes in turn favor BBB dysfunction, exacerbating this process
[14].
Microglia is made of CNS resident innate immune cells derived from myeloid lineage. They constitute about 10% of the total CNS cells
[15]. Microglia play an active role in fundamental brain processes, such as neurogenesis, synaptic pruning and plasticity, and immune surveillance
[16]. In physiological conditions, microglia are quiescent, but they actively monitor the surrounding parenchymal environment with their branching processes
[17]. In response to stimulations such as pro-inflammatory cytokines or other pathological molecules, microglial cells become activated and produce inflammatory cytokines and chemokines to prevent CNS damage
[18]. However, when chronically activated, microglia plays a key role in the paradoxical propagation of neuroinflammation, leading to neurodegeneration
[19].
Astrocytes are the most abundant glial cells in the CNS, representing, with their processes, a critical component of the BBB together with endothelial cells
[20]. Astrocytes also provide metabolic support to neurons, regulate cerebral blood flow, and modulate synapses formation and synaptic transmission, through the uptake and release of neurotransmitters
[21][22]. Activated microglia releases IL-1α, TNFα, and complement 1q (C1q), and is responsible, together with peripheral inflammatory cytokines and signals, for astrocytes activation. In this way, a gliosis response occurs, which is characterized by the upregulation of glial fibrillary acid protein (GFAP) expression and gliotic scar formation
[23][24].
Neuroinflammation is normally part of a protective physiological process. However, its chronic and excessive activation triggers the development of brain damage with synaptic consequences, cell loss, and impaired neurogenesis
[15] that, altogether, lead to manifestations related to nervous system dysfunctions, such as anxiety, depression, memory loss and cognitive impairment
[25].
3. Gut–Liver Axis Contribution to Systemic Inflammation
The term gut microbiota refers to all the microorganisms, including bacteria, viruses, fungi, archaea, and protozoa that inhabit the human gut and live in a mutualistic and symbiotic relationship with the host
[26].
More than 100 trillions of microorganisms form the gut microbiota, the composition of which varies along the gastrointestinal tract and is influenced by genetic and environmental factors, such as early life events (i.e., mode of delivery, breastfeeding), diet, lifestyle, and exposure to drugs
[27].
A growing body of evidence highlighted the importance of a balanced gut microbiota in maintaining host’s health given its role in several important functions for the organism
[26][28][29][30]. Indeed, the gut microbiota is involved in the metabolism of undigested carbohydrates, producing short chain fatty acids (SCFAs) such as butyrate, propionate, and acetate, which not only are a source of energy for the organism and enterocytes
[30][31], but also guarantee the integrity of the intestinal barrier and maintain intestinal motility . Butyrate intervenes in the maintenance of the gut barrier integrity by regulating tight junction proteins, such as claudin-1 and zonula occludens-1
[32]. SCFAs, and especially butyrate, were shown to modulate the immune response, and consequently, liver inflammation
[33].
Gut microbiota, and in particular,
Lactobacillus,
Bifidobacterium, and
Enterococcus, convert primary bile acids derived from the liver into secondary bile acids, which exert antimicrobial effects and contribute to the homeostasis of the intestinal epithelial barrier and vascular barrier, through the interaction with the farnesoid X receptor (FXR)
[34][35].
In recent decades, increasing attention was paid to the close relationship between the gut microbiota and the liver. This strong bidirectional connection, known as the gut–liver–axis, is realized by the portal vein and the biliary tract; thus, gut-derived metabolites can reach the liver, which, in turn, releases bile acids as well as other mediators back into the intestine
[36]. The intestinal barrier, composed of structural elements such as mucus layer, epithelial cells, vascular barrier, immune cells, and soluble mediators, plays a critical role in this interaction, limiting the systemic spread of toxins and pathogenic molecules
[37]. Indeed, through the portal vein the liver receives about 70% of its blood supply from the intestine, it is constantly exposed only to small amounts of bacteria and bacterial products
[34]. These products, in normal conditions, are eliminated by resident immune cells like Kupffer cells, dendritic cells, natural killer (NK) cells and lymphocytes
[35] preventing their systemic spread, thus preserving a condition of immune tolerance
[38][39].
Based on these premises, dysbiosis and the alteration of the intestinal barrier, have been correlated with the development and progression of liver disease. Indeed, the loss of beneficial autologous taxa leads to a reduced production of SCFAs and conversion of primary into secondary bile acids that further exacerbate gut dysbiosis, alter the integrity of intestinal barrier and decrease gut motility, also favoring small intestinal bacterial overgrowth (SIBO)
[40]. These alterations increase the rate of bacterial translocation and promote endotoxemia, with a huge amount of pathogen-associated molecular patterns (PAMPs) that reach the mesenteric lymph nodes, with the consequent spread to the liver through the portal circulation
[41][42][43], promoting a chronic inflammation of the liver, worsening liver impairment in a vicious circle, and contributing to systemic inflammation.
4. Role of the Gut Microbiota in Hepatic Encephalopathy and Neuroinflammation
A strong interplay was demonstrated between the gut microbiota and the central nervous system. This bidirectional network realizes the gut–brain–axis
[25][44]. Different systems act together in this key channel, especially the enteric nervous system, the endocrine, and the immune system
[45][46]. Indeed, the gut microbiota influences the function and development of the CNS by modulating signals via the vagus nerve, through the production of hormones and neurotransmitters, and the stimulation of the neuroimmune pathway by cytokines secretion
[47]. On the other hand, the CNS uses these pathways to modulate intestinal secretory and immune functions, motility, and barrier permeability
[48].
In the setting of liver disease, HE is typically related to gut–liver–brain axis dysfunction. The pathogenesis of HE is still not fully clarified, although high brain ammonia levels were always considered a major etiological factor
[49]. Ammonia is a by-product of nitrogen metabolism, principally derived from the metabolic activity of urease-producing bacteria in the gut and the deamination of glutamine by the enzyme glutaminase present in the enterocytes of the small intestine and the colon
[4]. Other organs, such as muscles, brain, and kidney, participate to a lesser extent in ammonia metabolism
[40]. In normal conditions, ammonia is transported to the liver through the portal vein, where it enters in the urea cycle and is converted into urea, which is subsequently excreted through the kidneys
[50]. In case of liver dysfunction, ammonia metabolism is impaired, resulting in a significant increase in serum ammonia
[51].
However, it was shown that ammonia levels do not correlate exactly with the severity of HE
[52]. This indicates that other factors ranging from intestinal dysbiosis, systemic inflammation, and neuroinflammation intervene in the pathogenesis of HE
[53]. Many studies confirmed the role of gut microbiota dysregulation in HE; therefore, most of the therapies used in its treatment act on microbiota modulation [51,53]. The over-abundance of ammonia in HE can be in part explained by an overgrowth of urease-producing bacteria, as demonstrated by the presence of a greater population of urease-producing
Proteobacteria in patients with HE and poor cognition
[54].
Neuroinflammation was recently suggested to represent another crucial factor in the pathogenesis of neurological impairment in liver disease
[55]. Following liver dysfunction, hyperammonemia, circulating bile acids, and systemic inflammation are able to activate microglia, promoting neuroinflammation
[7]. Gut microbiota, being one of the main actors in the development of systemic inflammation and in ammonia metabolism, plays a critical role in the pathogenesis of neuroinflammation (
Figure 1)
[25]. This relation establishes a connection between hepatic inflammation and neuroinflammation. Indeed, gut dysbiosis, SIBO, and intestinal barrier dysfunction lead to increased bacterial translocation and release in circulation of bacterial products, such as LPS, peptidoglycan, flagellin, and bacterial DNA
[53][56]. These PAMPs interact with TLR-4 on the membrane of reticuloendothelial cells of the liver, such as Kupffer cells. This interaction in turn favors the activation of NF-kB and MyD88, triggering the release of inflammatory cytokines such as TNF-α, IL-6, and IL-1β by immune cells, leading to systemic inflammation
[57][58][59]. This inflammatory process is responsible for blood–brain barrier dysfunction and neuroinflammation.
Figure 1. Gut–liver–brain axis and neuroinflammation. Gut dysbiosis and intestinal barrier impairment occurring during chronic liver disease lead to increased bacteria and their products/fragments that reach the liver through the portal vein. PAMPs interaction with TLR-4 on liver reticuloendothelial cells activates NF-kB and MyD88, leading to the release of pro-inflammatory cytokines, which trigger systemic inflammation. Systemic inflammation and hyperammonemia derived from impaired liver function cause BBB dysfunction, microglia, and astrocytes activation, which in turn promote neuroinflammation. Abbreviations: AMPs: antimicrobial peptides; BBB: blood brain barrier; IL: interleukin; LPS: lipopolysaccharide; MYD88: myeloid differentiation primary response 88; NF-kB: nuclear factor NF-kappa-B; NH4+: ammonium; NO: nitric oxide; PAMPs: pathogen associated molecular patterns; pBA: primary bile acid; ROS: reactive oxygen species; sBA: secondary bile acids; SCFAs: Short-chain fatty acids; TGF-β: transforming growth factor beta; and TNF-α: tumor necrosis factor α. Created with BioRender.com.
LPS, a component of the Gram-negative bacteria cell wall, represents one of the major contributors to systemic inflammation. Intravenous administration of LPS transiently caused systemic inflammatory responses with an increase in IL-6 and TNF-α serum levels
[60]. LPS, together with inflammatory cytokines and other factors (e.g., glutamate), promote microglial activation and the consequent release of inflammatory cytokines, leading to neuronal damage
[55].
TNF-α induces microglia to release CCL2, leading to the recruitment of monocytes in the brain and being responsible for neurological decline
[61]. In an animal model of azoxymethane-induced ALF, the use of etanercept, a TNF-α neutralizing molecule, reduced both systemic and cerebral inflammation and prevented microglial activation
[62].
5. Neuroinflammation in Acute Liver Failure and Chronic Liver Disease
Neuroinflammation appears to be implicated in the pathogenesis of both ALF and chronic liver disease. ALF is a life-threatening condition characterized by the development of HE and coagulative disorder in patients without a previous history of liver disease
[63]. The mechanisms responsible for HE onset are not yet fully elucidated, although hyperammonemia and systemic inflammation, acting synergistically, appear to play an important role in its onset
[64].
Recently, a growing body of evidence linked neuroinflammation to the development of CNS-related complications such as HE, brain oedema, brain herniation, and intracranial hypertension occurring in patients with ALF
[65]. Both microglia and astrocytes are involved in this process, producing local pro-inflammatory cytokines under the influence of systemic inflammatory signals deriving from the failing liver. Moreover, hyperammonemia causes astrocytes swelling, leading to brain edema
[66].
Systemic inflammation during ALF also contributes to neuroinflammation. ALF induced by azoxymethane (AOM) in mice favors the release of hepatic transforming growth factor β 1 (TGFβ 1) into circulation, which binds TGFβ-receptor2 (TGFβR2) present on neurons, leading to the increase in CCL2 and decrease in CX3CL1 expression, which results in the activation of microglia. Indeed, neuroinflammatory responses were attenuated in mice receiving pharmacological inhibition of TGFβ1 or in TGFβR2 null mice
[67].
Rats with induced chronic liver failure after portocaval shunt had increased brain levels of IL-6 and cyclooxygenase (COX) and inducible nitric oxide synthase (iNOS) activity, which are markers of neuroinflammation. These rats also presented a decreased function of glutamate-(NO)-cyclic guanosine monophosphate (cGMP), leading to cognitive impairment with a lower ability to learn Y-maze task. Treatment with the anti-inflammatory drug ibuprofen decreased neuroinflammation and restored rats’ cognitive ability through the normalization of NO-cGMP function, confirming the contribution of neuroinflammation in cognitive alterations of HE
[68].
Evidence of neuroinflammation in patients with chronic liver disease and cognitive impairment is limited to few studies. Cagnin et al. demonstrated in five patients with cirrhosis and MHE who underwent positron emission tomography (PET) scans an increased binding of [11C]-PK11195 to peripheral benzodiazepine binding sites (PBBS) in the brain, confirming microglia activation. The highest increase in [11C](
R)-PK11195 binding was seen in the patient with the worst cognitive impairment
[69].
6. Intestinal Microbiota Modulation as Treatment Strategy and Emerging Therapies
6.1. Rifaximin
Rifaximin is an eubiotic compound currently approved for the treatment of overt HE
[2][70]. Several studies looked at how rifaximin can help the nervous system recover from neuroinflammation.
Mangas-Losada et al. administered rifaximin 1200 mg/day for six months to 22 cirrhotic patients with MHE. No significant changes in liver function, hemoglobin, or ammonia serum level were found, while immunological alterations showed a remarkable improvement in responder patients. In particular, pro-inflammatory CD14++CD16+ monocytes decreased in favor of anti-inflammatory CD14++CD16− monocytes; auto-reactive CD4+CD28− T-lymphocytes also decreased, while non-reactive CD4+CD28+ T-lymphocytes increased with the disappearance of CD69, a marker of early activation. Th22 CD4+ subsets and follicular Th diminished, as well as many pro-inflammatory cytokines, and levels of immunoglobulins normalized. Conversely, non-responders showed only a reduction in IL-6, CCL20, and T lymphocytes differentiation to Th22, and did not present increased expression of CD69 before treatment
[71].
Other evidence in mice showed that rifaximin reduces neuroinflammation and cognitive impairment through microbiota modulation and promotion of the gut barrier integrity
[72]. Furthermore, rifaximin favors the growth of gut bacteria associated with production of SCFAs
[73], which are able to cross the BBB and exert anti-inflammatory properties
[74].
6.2. Lactulose
Lactulose is a nonabsorbable disaccharide approved for the treatment, prevention, and secondary prophylaxis of overt HE
[75]. Some evidence showed effectiveness also in MHE and covert HE
[2].
Both systemic inflammation and hyperammonemia, which lead to lactate accumulation in the brain, are responsible for microglial activation and neuroinflammation and contribute to HE
[76][77]. Studies in both rat models and cirrhotic patients with MHE demonstrate that lactulose administration lowers serum endotoxins and pro-inflammatory cytokines such as TNFα, IL-2, IL-6, IL-13, and IL-18
[78][79][80]. Through its cathartic action and the acidification of the intestinal environment, lactulose also reduces ammonia levels in the blood. Indeed, gut bacteria metabolize lactulose producing SCFAs, such as lactic and acetic acid, which lower colonic pH. An acidic environment decreases the content of urease-producing bacteria and favors the production of non-absorbable ammonium (NH4+), which cannot pass the gut barrier
[81].
6.3. Non-Steroidal Anti-Inflammatory Drugs (NSAIDs)
A recent study using rat models of HE with hyperammonemia reported a beneficial effect of the NSAID ibuprofen. A significant improvement in spatial memory and anxiety was registered after treatment with ibuprofen; the combination of ibuprofen and the antioxidant 1,8-cineol also increased the superoxide dismutase activity and significantly reduced oxidative stress
[82]. Another pre-clinical study in rat models of HE reported a complete reversal of hypokinesia due to increased extracellular glutamate in substantia nigra pars reticulata (SNr) in rats with portacaval shunts (PCS) treated with ibuprofen 30 mg/kg. At the molecular level, this therapy normalized the amount of glutamate transporters GLT-1 and of excitatory amino acid carrier 1 (EAAC-1) and decreased by 53% extracellular glutamate in SNr of PCS rats
[83]. Despite these positive results, NSAID therapy is burdened by unacceptable toxicities, such as renal damage and gastropathy in cirrhotic patients
[84][85].
6.4. Fecal Microbiota Transplantation
FMT, through its ability to modulate gut microbiota, can potentially reverse all the consequences of gut dysbiosis, such as increased gut barrier permeability, bacterial translocation, and systemic inflammation. Several animal models suggested a beneficial effect of FMT on neuroinflammation.
In a rat model of HE induced by the administration of CCl4, FMT was able to improve cognitive functions and HE, improved gut barrier permeability and significantly decreased ammonia serum levels and the expression of TLR4 and TLR9, two important receptors involved in the inflammatory response. Overall, these effects led to a strong reduction in pro-inflammatory cytokines such as IL-1β, IL-6, and TNF-α, pointing out how FMT could be useful in modulating systemic inflammation and, consequently, neuroinflammation
[86].
6.5. Probiotics, Prebiotics and Postbiotics
Several studies report the gut microbiota modulatory properties of probiotics, thus they are supposed to help in counteracting the mechanisms of neuroinflammation and HE in patients with liver disease. Probiotics reduce the overgrowth of pathogenic bacteria, maintain the integrity of tight junction proteins strengthening the gut barrier, and decrease intestinal bacteria translocation, with consequent reduction in endotoxemia and systemic inflammation
[87][88][89][90][91][92].
In addition,
Lactobacillus has the ability to inhibit gut urease-producing bacteria and to acidify intestinal environment with the consequent reduction in serum ammonia levels
[93].
A randomized controlled trial involving 120 cirrhotic patients who recovered from an episode of HE proved the superiority of VSL#3, a group of eight probiotics, over placebo in improving Child Turcotte Pugh (CTP) and MELD scores and lowering the rate of HE recurrence and hospitalization
[90]. However, the effectiveness of probiotics compared to lactulose is uncertain
[94].
6.6. Challenges of Proposed Treatments
Rifaximin and lactulose are effective, low-cost drugs characterized by a good tolerability and safety profile
[95][96][97]; however, some adverse effects, such as bloating, may reduce quality of life and decrease the adherence to therapy
[96]. In addition, in cirrhotic patients, circulating levels of rifaximin, which has usually a negligible absorption, may rise due to increased intestinal permeability, with the risk of altering the safety profile of the drug
[98].
Several studies demonstrate the efficacy of FMT as a new treatment of HE. However, there is still no standardization about the route of administration, dosage, or the ideal bacterial consortium to be adopted for the transplant. This is made more difficult by the fact that each donor has a peculiar microbiome, which is complex to analyze
[99][100]. Up to date, FMT appears to be a safe treatment, although risks for the potential bacterial dissemination in the bloodstream were reported
[101].
In conclusion, neuroinflammation appears to be a promising and blooming area of study for the treatment and prevention of HE. The currently available therapeutic strategies appear to be partially effective in modulating neuroinflammation, so it is desirable to identify new effective weapons that are also easily applicable in clinical practice.