3.1. Pulmonary Involvement in Long COVID
The main impact of SARS-CoV-2 infection is on the respiratory system, the entrance of the virus into the epithelial lining of the lungs is facilitated by pneumonocytes that express the angiotensin-converting enzyme 2 (ACE2) receptors [
17]. Upon infection by SARS-CoV-2, lung epithelial cells, which serve as the primary origin of inflammatory cytokines, engage in interactions with immune cells that have been recruited to the site of infection. This interaction plays a significant part in the development of inflammatory lung damage and subsequent respiratory failure [
18]. The persistence and unregulated activation of cytokine storms can result in the impairment of the epithelial barrier. In the presence of such circumstances, the progression of lung fibrosis occurs due to the impairment of lung epithelial regeneration [
19].
The injury inflicted by SARS-CoV-2 on a range of lung epithelial cells is attributed to the ACE2 receptors or transmembrane serine protease 2 (TMPRSS2) expression. ACE2 has been observed to be expressed in various types of cells in the airway epithelium, such as basal cells, ciliated cells, mucous cells, club cells, and intermediate cells [
20]. According to a single-cell ribonucleic acid (RNA) sequencing analysis, it was observed that the average amount of ACE2 was comparatively higher in mucous cells as compared to other types of epithelial cells [
21]. Alveolar type 1 (AT1) and alveolar type 2 (AT2) cells are susceptible to SARS-CoV-2 infection, with AT2 cells being the initial target [
22]. Nonetheless, academic studies propose that a minor fraction of AT2 cells exhibit the expression of ACE2 while the expression of TMPRSS2 is minimal in the basal cells of the undifferentiated airway epithelium and more prominently expressed in the differentiated airway epithelium [
23,
24,
25]. This proposed mechanism is accountable for the susceptibility of the lungs to COVID-19.
Dyspnea is the most common pulmonary symptom, because of Long COVID. Dyspnea is defined as an unpleasant and intolerable pattern of breathing, related to an inability to ventilate enough to provide the required amount of air for one normal breath. According to Hentsch L. et all, in research, published in 2021, three possible mechanisms were proposed for dyspnea induced by COVID-19 [
26]. One of the mechanisms implies the interruption of afferent sensory signaling pathways by SARS-CoV-2 which may result in the inability of cortical structures responsible for processing the sensory aspects of dyspnea to receive afferent inputs from the brainstem. It is plausible that the virus may cause direct harm to the mechano- or irritant receptors located in the respiratory tract and/or chest wall, thereby impeding the transmission of afferent signals to the brainstem and higher brain structures. The second mechanism pertains to the possibility that SARS-CoV-2 may impede the capacity of cortical structures to identify or manage incoming sensory signals related to breathlessness originating from the brainstem. The occurrence of the conditions may arise either because of the direct impact of SARS-CoV-2 on nervous tissue or indirectly through the manifestation of inflammatory acute encephalopathy or cerebrovascular complications, including but not limited to ischemic or hemorrhagic stroke. The last mechanism indicates the cortical structures implicated in the perception of dyspnea may exhibit a facilitative influence on the perception of breathlessness, like that of pain, which may be disrupted by SARS-CoV-2. The induction of dyspnea-related panic attacks through experimental inhalation of 35% CO
2 has been demonstrated in patients with bilateral amygdala damage [
27]. The self-reported levels of panic and fear in this group were found to be remarkably higher than those acknowledged by the comparison group with intact neurological functioning. The findings of this study may indicate that the activation of extra-limbic brain structures by CO
2 occurs directly, and subsequently, these structures are controlled in a downward manner by the amygdala [
27].
Several mechanisms of pulmonary damage in COVID-19 have been identified, with viral and immune-mediated pathways being implicated. Pulmonary fibrosis may arise because of chronic inflammation or as an idiopathic, age-related fibroproliferative process that is influenced by genetic factors [
28]. Pulmonary fibrosis is a recognized consequence of acute respiratory distress syndrome (ARDS). Nevertheless, the clinical relevance of persistent radiological abnormalities after ARDS is limited and has decreased with the implementation of protective lung ventilation techniques [
29]. Research has indicated that a significant proportion of individuals diagnosed with COVID-19, specifically 40%, are prone to developing ARDS. Furthermore, it has been observed that 20% of ARDS cases are classified as severe [
30]. The manifestation of post-COVID-19 fibrosis will require further observation, however, initial examination of patients with COVID-19 upon hospital release indicates that over 33% of recuperated individuals exhibit fibrotic irregularities. The defining characteristic of ARDS is the presence of diffuse alveolar damage (DAD). This is marked by an initial phase of acute inflammatory exudation, which is characterized by the presence of hyaline membranes. This is then followed by an organizing phase and a fibrotic phase [
31]. Prior research has emphasized the significance of the duration of illness as a crucial factor in the development of pulmonary fibrosis after ARDS. The findings of this investigation indicate that a small proportion of patients (4%) who had a disease duration of less than one week, a notably larger proportion (24%) of patients with an illness lasting between 1-3 weeks, and a majority (61%) of patients who had a disease duration exceeding three weeks, experienced the development of fibrosis [
32]. The development and progression of pulmonary fibrosis may be instigated and facilitated by a cytokine storm resulting from an atypical immune response. The release of matrix metalloproteinases during the inflammatory phase of ARDS leads to epithelial and endothelial injury. The process of fibrosis involves the participation of vascular endothelial growth factor and cytokines, including interleukin (IL) 6 and tumor necrosis factor (TNF) a. The etiology behind the differential outcomes of individuals who either recuperate from an insult or develop progressive pulmonary fibrosis characterized by the accumulation of fibroblasts and myofibroblasts, along with excessive collagen deposition, remains unclear [
33]. While COVID-19-induced ARDS appears to be the primary indicator of pulmonary fibrosis, various studies have indicated that it differs from classical ARDS in terms of its high and low elastance types.
The computed tomography (CT) results of numerous cases of COVID-19 do not indicate classical ARDS. In addition, the presence of abnormal coagulopathy is a notable pathological characteristic of this ailment. The mechanism underlying pulmonary fibrosis in COVID-19 differs from that observed in other fibrotic lung diseases, such as idiopathic pulmonary fibrosis (IPF). Notably, pathological observations suggest that the site of injury in COVID-19-induced pulmonary fibrosis is primarily the alveolar epithelial cells, rather than the endothelial cells.
The act of coughing is a reflexive action that necessitates minimal conscious control. This reflex is initiated by the activation of peripheral sensory nerves that transmit signals to the vagus nerves. These nerves, in turn, provide sensory input to the brainstem at the solitary nucleus and the spinal trigeminal nucleus. The phenomenon of cough hypersensitivity has been established in the context of chronic cough, wherein the pathways responsible for coughing are believed to have undergone sensitization due to an increase in the magnitude of afferent signals transmitted to the brainstem. Coronaviruses, including SARS-CoV-2, gain access to host cells through specific receptors and proteases, namely ACE2, TMPRSS2, and furin [
34]. SARS-CoV-2 may have the ability to directly interact with sensory neurons, as evidenced by the prevalence of sensory dysfunction such as coughing, as well as olfactory and taste impairments, among individuals who have been infected with the virus [
35]. The expression of ACE2 or TMPRSS2 in human airway vagal sensory neurons and their susceptibility to SARS-CoV-2 infection remain unknown. The bronchopulmonary vagal sensory neurons in mice were subjected to single-cell sequencing, which revealed the absence of murine ACE2 expression [
36].
The potential involvement of supplementary viral entry factors in the interplay between SARS-CoV-2 and neurons cannot be disregarded. One such factor is neuropilin-1, which is present in vagal and other sensory neurons [
37]. The study by D. H. Brann et al, conducted a sequencing analysis on human olfactory mucosal cells, revealing the absence of ACE2 and TMPRSS2 in olfactory epithelial neurons [
38]. However, a significant expression of these genes was observed in support cells of the olfactory epithelium and stem cells [
38]. The veracity of the results was validated through the cellular histological localization of ACE2 in the specialized neuroepithelium of supporting cells surrounding neuronal dendritic projections. It is noteworthy that the neuroepithelium is comprised of odour-sensing cilia [
39]. Hence, it is plausible that the onset of anosmia resulting from SARS-CoV-2 infection could be attributed to the impact of the infected epithelium on neuronal function.
The ACE2 gene has been identified in a specific group of sensory neurons located in the thoracic ganglia of humans. These neurons are also known to provide innervation to the lungs. It is worth noting that a particular group of nociceptive neurons, which express calcitonin-related polypeptide alpha (CALCA) or purinergic receptor P2X 3 (P2RX3), have been found to exhibit expression [
40]. These neuronal subtypes are like those found in the vagal sensory ganglia, which play an important role in triggering coughing. The similarity in developmental lineage and molecular phenotype between certain vagal sensory neurons, particularly those implicated in cough, and dorsal root ganglion neurons suggests a potential correlation between ACE2 expression in human vagal sensory neurons. The etiology of persistent cough following SARS CoV-2 infection is presently unknown, despite the possibility that the involvement of dorsal root ganglion neurons that contain nociceptors could account for the joint and chest pain, headache, and dyspnea symptoms experienced Long COVID. The S1 spike protein of SARS CoV-2 can traverse the blood-brain barrier (BBB) in mice through absorptive transcytosis, indicating that a functional virus is not necessary for brain involvement [
41]. Additional research is required to explore the potential direct interactions between the virus and the nervous system in the development of cough and other sensory symptoms in individuals with SARS-CoV-2 infection. The exclusion of pathological or structural causes is crucial in the clinical management of chronic cough following COVID-19. This includes assessing fibrosis damage of lung parenchyma or damage to the airways resulting from SARS-CoV-2 infection or critical care management [
42]. The presence of lung parenchymal alterations is a frequent observation on computed tomography (CT) scans in adult individuals affected by COVID-19. Additionally, a proportion of 10-20% of patients may experience the development of lung fibrotic changes [
42]. The presence of lung fibrosis has been found to potentially heighten the sensitivity of the cough reflex in reaction to mechanical stimuli applied on the chest wall, as evidenced in individuals diagnosed with idiopathic pulmonary fibrosis [
43]. The presence of a persistent cough in individuals experiencing post-COVID symptoms may be attributed to neuroinflammation, resulting in a state of heightened laryngeal and cough hypersensitivity. This phenomenon serves as the underlying cause of chronic refractory or unexplained cough [
44,
45]. Neuromodulators such as gabapentin and pregabalin have demonstrated efficacy in managing chronic refractory cough. The aforementioned strategy could be deemed as a viable option for addressing the Long COVID, as these pharmaceutical agents may have utility in mitigating additional symptoms that coincide with coughing, such as discomfort, albeit with the possibility of exacerbating any cognitive impairment. The pulmonary manifestations of Long COVID are summarized in
Figure 2.
3.4. Cardiovascular Involvement in Long COVID
The evaluation of prolonged cardiovascular issues is ongoing and understanding the actual impact of post-COVID-19 cardiovascular complications remains uncertain. COVID-19 appears to worsen preexisting cardiac pathology, but also causes new-onset cardiac diseases such as arrhythmias or hypertension,
Figure 3.
While the causes behind enduring cardiac injury post-recovery are not yet fully grasped, a potential scenario points to a persistent inflammatory reaction reflected by the presence of circulating biomarkers for 3 to 8 months after acute infection: C-reactive protein, IL 6 and 8, ferritin, IFN beta and gamma1, procalcitonin (correlated with microvessel disease), chemokine ligand (CXCL) 9 (small cytokines that induce chemotaxis, promote differentiation and multiplication of leukocytes and cause tissue extravasation), CXCL10 (an ‘inflammatory’ chemokine), T-cell immunoglobulin mucin-3 (TIM-3) (involved in immune response), plasma ACE2 activity, pentraxin 3 (PTX3) (activates complement and facilitates pathogen recognition by macrophages). Some of these biomarkers were also found elevated in asymptomatic post-COVID patients during the lasting up to 8 months [
127]. This chronic inflammatory status may be caused by persistent viral reservoirs such as intestines or other immune-privileged sites following the acute infection [
106].
Immune exhaustion after prolonged antigen stimulation or immunosuppressive treatment may also facilitate the development of cardiac injury. Within the cardiac structure, SARS-CoV-2 primarily resided within interstitial cells and infiltrating macrophages within the myocardium. Viral entry depends upon the interaction between viral spike glycoprotein, the ACE2 receptor, a transmembrane aminopeptidase in the host cell membrane, and the host cell protease system such as TMPRSS2. ACE2 plays a crucial role in the neurohumoral regulation of the cardiovascular system and is primarily present in the vascular endothelium, cardiomyocytes, pericytes, cardiac fibroblasts, and epicardial adipose tissue. ACE2 naturally transforms angiotensin 1 and 2 into beneficial peptides like angiotensin 1-7 and angiotensin 1-9, which hold cardioprotective properties. SARS-CoV-2 causes downregulation of ACE2, thus preventing the synthesis of cardioprotective peptides and predisposes to cardiovascular damage.
Another proposed mechanism of myocardial damage is mediated by reactive oxygen species, which can induce the release of internal histones, damage-associated molecular patterns (DAMPs), and oxidized lipid-protein complexes. During the post-acute phase, these compounds trigger an inflammatory response that results in substantial myocardial tissue injury, leading to chronic myocardial scarring. This scarring may cause issues like reduced ventricular compliance, impaired blood flow within the heart muscle, decreased cardiac muscle contraction, and potential irregular heartbeats. Cytokine-induced damage can also manifest as blood clot formation, reduced oxygen delivery, destabilization of coronary plaques, the transition of chronic heart conditions into unstable states, increased metabolic requirements, diminished cardiac capacity, and inflammation in the heart valves [
127].
A growing number of reports on cardiovascular autonomic dysfunction (CVAD) in PASC patients, namely patients who experience symptoms suggestive of postural orthostatic tachycardia syndrome (POTS), with exaggerated rise in heart rate and intolerance to standing, commonly affecting females and individuals aged between 15 and 45 years old.
Infections commonly trigger dysautonomia, and several potential ways in which SARS-CoV-2 might contribute to this condition have been suggested based on initial evidence: hypovolemia, brainstem involvement, and autoimmunity [
129]. Hypovolemia is a recognized characteristic in patients with PASC, leading to a hyperadrenergic response. This reaction subsequently leads to cerebral hypoperfusion and disruption of central autonomic networks. Brainstem dysfunction encompasses various mechanisms, including direct invasion by the virus, neuroinflammation, brainstem compression, and vascular activation. Autoimmunity has an important role in the pathophysiology of post-viral POTS as supported by recent evidence [
130]. A high prevalence of specific autoantibodies has been found in the sera of PASC patients presenting dysautonomia. They include G-protein coupled receptor (GPCR) antibodies which can activate adrenergic receptors and elicit a negative allosteric effect on muscarinic GPCRs. Autoantibodies can also activate cholinergic receptors and cause peripheral vasodilation. Patients experiencing POTS linked to PASC have exhibited the presence of serum anti-nuclear, anti-thyroid, anti-cardiac protein, anti-phospholipid, and Sjogren’s antibodies [
131]. Alarmingly a possible shift in age distribution of Long Covid can be expected due to an increased number of young infected unvaccinated individuals in whom morbidity and mortality are difficult to predict according to the European Society of Cardiology [
132].
3.5. Renal Involvement in Long COVID
Impairment of the kidneys in individuals who contracted COVID-19 have been widespread worldwide and manifests as hematuria, proteinuria, acute kidney injury (AKI), end-stage kidney disease (ESKD) and major adverse kidney events (MAKE) [
133,
7]. Even in the absence of coexisting elevation in serum creatinine concentration or a decline in estimated glomerular filtration rate (eGFR), the presence of low molecular weight proteinuria or hematuria indicates a subclinical AKI. The mechanisms behind these injuries are multiple and include direct injury to the renal cells, indirect injury through other organ destruction etc.
The ACE2 receptor is expressed in the epithelium of the kidneys and plays a major role in the entrance of the SARS-CoV-2. Previous studies have shown that ACE2 is the functional receptor for SARS-CoV in vivo and in vitro. Recent cryo-electron microscopy has shown that the spike protein of SARS-CoV-2 binds directly to ACE2 with an even greater affinity than SARS-CoV [
138].
Several studies have revealed that SARS-CoV-2 relies on the widely expressed transmembrane glycoprotein, CD147, as a crucial element in infiltrating target cells [
137,
8]. CD147 is expressed in high amounts on the cell surface of the proximal tubular epithelial cells and is believed to have contributed to the pathogenesis of several renal diseases through its involvement in the immune-inflammatory response and dysregulated cell cycle [
137]. It is established that the CD147 partners, cyclophilins, have a crucial role in the replication process of the SARS-CoV-2 and inhibiting these by cyclosporins has demonstrated effective suppression of intracellular viral propagation [
136]. Su H. et all, in their autopsy results of patients with COVID-19, demonstrated by electron microscopic examination agglomerates of coronavirus-like particles having distinctive spikes within the tubular epithelium and podocytes [
136]. They further described notable tubular impairment, especially within the early segment of the proximal tubule marked by an absence of microvilli, acute tubular necrosis, a notable decrease in megalin expression within the microvilli and the existence of intraluminal debris. All denoting the presence of AKI [
136]. Further findings of the virus in urine only strengthen this discovery [
139]. Diao B. et al, further cemented this discovery when they confirmed the findings of the SARS-CoV-2 in renal cells, further demonstrating that SARS-CoV-2 can replicate in vivo [
140].
A study conducted by Chand S. et al., on 300 survivors of critical COVID-19, conducted in 2022 found that out of the patients who survived 120 days after covid resolution, 74.4% had AKI resolution while 25.6% did not have a resolution. Out of the 25.6% with AKI resolution 60% developed chronic kidney disease (CKD) and 40% required renal replacement therapy (RRT) [
135].
According to Bowe B. et all in their cohort study conducted for 1 year on 1,726,683 US Veterans, and 89,216 patients who were 1-month survivors of SARS-CoV-2 infection, patients who successfully recovered from COVID-19 displayed kidney-related symptoms during the post-acute stage, such as eGFR decline, AKI, ESKD or MAKE. These symptoms increased in the post-acute period with the severity of the acute infection, and the risk for renal problems was higher in hospitalized patients, who developed AKI [
134].
3.6. Dermatological Involvement in Long COVID
Dermatological manifestations and complications of COVID-19 are becoming increasingly acknowledged in the literature. There are many types of skin manifestations described in conjunction with COVID–19 such as maculopapular rash, urticaria, petechiae, purpura, vesicles, chilblains, livedo racemose and ischemia of the distal segments of the limbs [
141,
142,
143]. These dermatological findings are significant as they can help the clinician to make a COVID-19 diagnosis. There are various hypothesized causes of rash in COVID-19 patients. Diffuse microvascular vasculitis, brought on by complement system activation, is the first. In one study carried out by Magro C. et al, in 2020, interstitial and perivascular neutrophilia with pronounced leukocytoclasia, considerable complement protein deposition in the dermal capillaries, and other findings point to vasculitis phenomena [
141]. Other studies, performed by Sanchez A. et al, in 2020, suggest that the rash occurs as a direct effect of the virus. The justification is the presence of lymphocytosis (without eosinophils), papillary dermal oedema, epidermal spongiosis and lymphohistiocytic infiltrates [
142]. Five clinical patterns were identified in a recent study conducted in Spain, that included 375 cases: maculopapular eruptions (47%), urticarial lesions (19%), various vesicular eruptions (9%) and livedo or necrosis (6%) [
144]. Although the face is often spared, and the lesions are mostly restricted to the trunk and extremities (hands and feet).
The average lesions last a few days, although others have been reported to persist as short as 20 minutes or as long as four weeks. Nevertheless, in some individuals, lesions emerged 2 to 5 days before the beginning of COVID-19 symptomatology. In most cases, the mean latency that is, the amount of time it took for skin lesions to arise following the onset of the first typical COVID-19 symptoms was between 1 and 14 days [
145]. There is still a lack of knowledge on the pathophysiological mechanisms of skin lesions in COVID-19 patients. The skin manifestations presented in COVID-19 may be divided into two categories regarding their pathological mechanisms [
146]. The first is an immune response to the viral nucleotides of COVID-19, which clinically manifests in a similar way to viral exanthems. The second group is skin eruptions that result from the systemic effects of COVID-19, this is especially the case in vasculitis and thrombotic vasculopathy.
The possible actions of SARS-CoV-2 on the skin may be mediated by non-structural viral proteins (NSPs) that block the innate immune system (NSP3, NSP16), inhibit the effect of IFN (NSP5), or synthesize cytokines (NP3). The cytokine storm can stimulate dermal cells, leading to the appearance of various maculopapular or vesicular rashes. Secondary activation of complement by the surface viral antigen and microangiopathy may lead to the appearance of purpuric manifestations [
145,
147]. TMPRSS2 activation is crucial for the virus to attach to ACE2 through its spike protein. Androgen receptor elements near the TMPRSS2 gene on chromosome 21 suggest COVID-19 could affect men more due to androgen levels. The higher COVID-19 rates in Spanish men, possibly linked to androgenetic alopecia prevalence, hint at androgens playing a role in worsening the severity of the disease [
148]. Another possible effect of SARS-CoV-2 is a direct impact through ACE2 in the epidermis, causing certain skin conditions (acantholysis and dyskeratosis) [
149]. ACE2 appears in the skin’s basal epidermal layer, dermal blood vessel ECs, and eccrine tissue [
145,
150]. The reduced microcirculatory function across arterial beds in COVID-19 patients and its effects may be linked to COVID-19-related endotheliitis involving ACE2 [
145,
6].
3.7. Neuropsychiatric Mechanism of Long COVID
Up to one-third of individuals with long COVID, also known as PASC, may have ongoing neurological problems that manifest as anosmia, hypogeusia, lethargy, headaches, “brain fog,” dysautonomia, cognitive impairment, and peripheral neuropathy [
151].
The neurological manifestations observed in cases of acute COVID-19 have been associated with various interconnected pathogenetic mechanisms. These mechanisms include viral invasion of the nervous system accompanied by abnormal immune responses, dysfunction of the blood-brain barrier due to epitheliopathy, coagulation disorders leading to neuronal injury caused by reduced oxygen supply, imbalances in metabolic processes, cascades of oxidative stress, and cellular apoptosis [
152]. These factors are posited as potential causes for the enduring symptoms observed in individuals affected by COVID-19. SARS-CoV-2 can invade the stem and support cells located in the olfactory epithelium, which is situated outside the central nervous system. This invasion can lead to persistent alterations in the sense of smell. Anomalies in innate and adaptive immunity that may arise from SARS-CoV-2 infection include monocyte enlargement, T-cell exhaustion, and prolonged cytokine release [
151]. These abnormalities may lead to neuroinflammatory reactions and microglia activation, abnormalities in the white matter, and microvascular alterations. Additionally, viral protease activity and complement activation can cause endotheliopathy, which can lead to hypoxic neuronal damage and BBB failure, and microvascular clot formation, which can obstruct capillaries [
151].
3.7.1. SARS-CoV-2 Neurotropism
It is now believed that SARS-CoV-2’s direct neurotropic actions have a limited impact on the etiology of Long COVID, like the part that invasion of the nervous system played in acute COVID-19 [
153]. A detailed study by Meinhardt J. et all proved the SARS-CoV-2 presence in the olfactory cells [
153]. The serine protease TMPRSS2 is known to act on the cellular receptor ACE2 and to bind to the SARS-CoV, especially SARS-CoV-2, allowing the virus to enter human host cells. It has been demonstrated that non-neuronal cells in the human olfactory mucosa physiologically express ACE2 [
153]. It was determined that the olfactory mucosa had the highest concentration of S protein [
153]. Using immunohistochemistry, unique immunoreactivity for the S protein was discovered in cell types that differ morphologically and are suggestive of neuronal/neural origin [
153]. This immunoreactivity included a distinctive granular, partially perinuclear pattern. The RNA of SARS-CoV-2 was identified in cells of the olfactory epithelium and nasal mucus [
153].
Coronaviruses and other neurotropic viruses employ sensory and motor neural pathways to penetrate the central nervous system (CNS). The olfactory nerve is a type of neural route. The way olfactory nerves are structured differently in the nasal cavity and forebrain, as well as the olfactory bulb, mediates this. Inflammation and a demyelinating reaction may result from the virus getting into the brain and cerebrospinal fluid (CSF) in this way. In less than 7 days, once the infection is established, the viruses can infect the entire brain and CSF [
155]. Hematogenous spread from severely infected airways and lungs is one way that SARS-CoV-2 may enter the CNS. This sort of spread would be made easier by systemic inflammation that raises BBB permeability. The brain’s circumventricular organs are fenestrated, highly permeable structures. This often enables circulating but non-BBB-crossing mediators to have an impact on brain activity directly. This permeability, however, can also allow for pathogen neuroinvasion during an acute infection. This can happen directly or by an indirect process where host immune cells actively carry intracellular infections into the central nervous system (CNS) [
91].
SARS-CoV-2 showed choroid plexus basal (vascular side) epithelial tropism in an organoid model made from human pluripotent stem cells, despite more abundant ACE2 on the apical (CSF) side. Additionally, SARS-CoV-2 led to epithelial destruction and barrier leakage, which may have facilitated neuroinflammation by increasing the entry of immune cells, including those infected in a Trojan horse form, and circulating cytokines into the CNS [
91].
Invading the stem cells, pericytes, columnar epithelial cells, and Bowman’s gland cells in the olfactory epithelium using the ACE2 receptor cause persistent filia thinning and volume loss in the olfactory bulb. Although there is limited proof of direct neuroinvasion in these regions, there is a correlation between the geographical distribution of ACE2 receptors and areas of hypometabolism in the brain. Instead, it is thought that these areas suffer from increased levels of oxidative stress, cytotoxic T lymphocyte infiltration, neurodegeneration, and demyelination because of neuroinvasion. These mechanisms probably continue because of persistent viral shedding, particularly in the GI tract where ACE2 co-regulates dopa-decarboxylase (DDC) and the dopamine metabolic pathway is active [
151].
3.7.2. Brain-Gut Axis - Dysbiosis
The intestinal microbiome is a dynamic ecosystem composed of diverse communities of microorganisms including bacteria, archaea, and fungi, it begins its colonization early in life, undergoing dynamic changes in the first five years and achieving relative stability in adulthood. However, it remains responsive to various factors such as environmental changes, antibiotic use, diet, infections, or diseases. Changes in its composition are associated with multiple health conditions, although causality remains unclear in many cases. The association between microbiome changes and various health conditions like inflammatory bowel disease, obesity, and mental health disorders suggests a potential role of the microbiome in the pathogenesis of these conditions. However, establishing direct causation is challenging due to the complexity of microbiome-host interactions and numerous influencing factors affecting its composition and functionality.
During the COVID-19 pandemic, research has highlighted the impact of the infection on the intestinal microbiome. Cheng X et al., in a systematic analysis of 16 studies, observed a reduction in intestinal bacterial diversity in patients with acute COVID-19 and those who recovered from the illness [
156]. Moreover, a decrease in bacteria producing anti-inflammatory effects mediated by butyrate production (Megasphaera, Dialister, Ruminococcus, Faecalibacterium, Roseburia, Lachnospira, and Prevotella) was noted, alongside an increase in microorganism’s species with pro-inflammatory properties (Streptococcus, Enterococcus, Corynebacterium, Blautia, and Dorea), leading to increased synthesis of proinflammatory cytokines [
156,
157]. These disruptions in the gut microbiome persisted into the recovery phase after the resolution of the initial infection, lasting for months, suggesting their potential contribution to PASC. Additionally, increases in fungi (Candida albicans) and Inoviruses [Pseudomonas phages (Pf1)], as well as a decrease in beneficial bacteria (Faecalibacterium prausnitzii, Eubacterium rectale, Bifidobacterium adolescentis), have been identified in patients with severe forms of COVID-19, leading to an increased risk of intestinal infections, heightened intestinal permeability, reduced intestinal immune response, and proinflammatory effects [
158,
159]. Furthermore, a significant increase in proinflammatory pathobionts (Prevotella species, Veillonella species, genus Actinomyces, Streptococcus anginosus, Gemella sanguinis) was observed in the oral microbiome of patients experiencing PASC [
160]. The changes identified in the oral microbiome of these patients resembled those observed in individuals suffering from chronic fatigue syndrome, suggesting a potential link between the two conditions [
160].
In COVID-19, several actions can lead to the onset of dysbiosis, including hypoxia, inflammation, antibiotic use, or mediated by decreased tryptophan absorption,
Figure 4.
The absorption of dietary tryptophan, an essential amino acid absorbed in the small intestine, is mediated by ACE2, the enzyme involved in SARS-CoV-2 cell entry. SARS-CoV-2 leads to a decrease in tryptophan absorption through this pathway [
157]. The majority (90%) of tryptophan binds to albumin and crosses the blood-brain barrier, entering the cells of the central nervous system (neurons, astrocytes, and microglia). Some tryptophan remains free in the blood, and the unabsorbed portion serves as a metabolic substrate for intestinal bacteria. Tryptophan is subsequently metabolized via three pathways: the kynurenine pathway, the serotonin pathway, and the Indole pathway [
161]. Tryptophan is the sole precursor of serotonin (from which melatonin is later synthesized). Its synthesis primarily occurs in the intestine (enterochromaffin cells) and at neuronal levels (serotonergic neurons in the enteric nervous system and neurons in the central nervous system). A low level of serotonin can be responsible for brain fog, fatigue, depression, sleep disturbances, headaches, immunosuppression, and constipation. Constipation can, in turn, lead to alterations in the intestinal microbiome. The kynurenine pathway, primarily occurring hepato-renally, is the main metabolic route for tryptophan. It is stimulated by proinflammatory cytokines and corticosteroids, resulting in proinflammatory neurocatabolites (kynurenic acid and quinolinic acid). Almulla A.F. et all in a meta-analysis of 14 studies have shown that the kynurenine degradation pathway of tryptophan is highly stimulated in severe COVID-19 [
162]. The resulting products are pro-inflammatory, leading to the activation of the aryl hydrocarbon receptor (AhRs), and the direct association with SARS-CoV-2 action on this receptor is responsible for the onset of systemic aryl hydrocarbon receptor activation syndrome (SAAS), responsible for fibrosis, chronic inflammation, impaired cognitive function, increased oxidative stress, thrombosis, and disorders in the functionality of various organs [
162]. The third metabolic pathway is the indole pathway, producing indole derivatives, namely indole-3-propionic acid and indole-3-aldehyde, within the intestinal microbiome. The obtained metabolites have an immune-modulating effect, preventing the harmful effects of free radicals and oxidative stress. Additionally, they have a neuroprotective effect and prevent intestinal dysbiosis [
161].
In cases of severe COVID-19, it has been observed that SARS-CoV-2 can continue to be present in the upper respiratory and gastrointestinal epithelium for around 1 month. This prolonged viral shedding is believed to contribute to an imbalance in the brain-gut axis, as the virus tends to deplete symbionts and disrupt the gut microbiome in the gastrointestinal tract. The disturbance of the microbiome is further perpetuated by the continued shedding of the virus. An analysis of human intestinal organoids infected with SARS-CoV-2 has shown that the ACE2 gene is co-regulated with DDC and genes involved in the dopaminergic pathway metabolism and the absorption of amino acid precursors for neurotransmitter synthesis. This finding provides additional evidence of an altered brain-gut axis [
151].
3.7.3. Long COVID Neuroinflammation
The CNS has seen neuropathological alterations because of COVID-19. The production of pro-inflammatory mediators because of these modifications includes the induction of uncalled-for inflammatory reactions. Inflammation was induced in COVID-19 instances, according to recent clinical findings, and this induction was linked to an elevated level of cytokines, such as IL-1, IL-6, IL-10, and TNF. Previous research has shown that tight junction (TJ) proteins are degraded by cytokines, which change BBB integrity. Emerging data suggested that BBB permeability is increased by TJ degradation, specifically in claudin-5 and zonula occludens (ZO)-1. When the BBB’s integrity is altered, more viruses and cytokines have a chance to cross it and enter the CNS. This causes cerebral immune cells like microglia and astrocytes to become activated, which leads to cytokines-induced neuroinflammation [
163].
It has been established that astrocytes and microglia have a very important role in neuroinflammation. Microglial cells have been demonstrated to activate in the CNS because of systemic infection, and they are more vulnerable to pathogens than astrocytes. Molecular signals like IL-1 and TNF stimulate astrocytes in response to the activation of microglia. In response to microglial activation, activated astrocytes can release a variety of inflammatory substances, such as TNF, reactive oxygen species (ROS), and nitric oxide (NO). This two-way interaction between astrocytes and microglia intensifies the ‘chain reaction pattern’ neuroinflammation [
163].
TMPRSS2 was also highly expressed by astrocytes in the cerebral cortex of those patients. In general, ACE2 activates the AT1 receptor, which promotes neuroinflammation and oxidative stress. Additionally, it has been demonstrated that the high expression of inflammatory mediators and nuclear factor kappa-light-chain-enhancer (NF-kB) is linked to the elevated level of cathepsin L (CTSL) [
163]. It is interesting to note that TMPRSS2 aids virus entrance into host cells, according to many studies. All the relevant studies on SARS-Cov-2 neuroinflammation suggest that the virus activated the CNS cells involved in the inflammation [
163].
Additionally, COVID-19 encourages mast cell activation, neuroinflammation, and a high intracranial level of proinflammatory cytokines. Alcohol use and drug use disorders are two factors that can exacerbate the neuroinflammatory response caused by SARS-CoV-2, which differs between patients. To give one example, SARS-CoV-2 infection results in TJ loss, mast cell activation, and the production of inflammatory mediators, all of which have the potential to result in neuroinflammation, oedema, and bleeding, particularly in individuals with concurrent neurodegenerative disorders [
163]. The impairment of the endothelial barrier and the subsequent increase in blood-brain barrier (BBB) permeability by the SARS-CoV-2 spike protein has been experimentally validated through the utilization of a three-dimensional tissue model of the BBB. The spike protein of SARS-CoV-2 has the potential to activate brain ECs, leading to an inflammatory reaction that could worsen the breakdown of the blood-brain barrier [
163]. The in vitro studies demonstrated that the recombinant SARS-CoV-2 spike protein can decrease the expression of TJ proteins, such as ZO-1, ZO-2, Claudin-5, and junctional adhesion molecule (JAM-2), in human brain microvascular ECs [
163]. Changes in junctional protein integrity could affect the BBB as a whole. It’s interesting to note that the level of cytokines, including TNF, IL-6, and IL-10, increased when the production of TJ proteins was downregulated. The anti-inflammatory enzyme indoleamine-2, 3-dioxygenase 1 (IDO1) is expressed by a variety of immune cells, including macrophages, monocytes, and microglia. Infection with SARS-CoV-2 leads to abnormal IDO1-mediated inflammation. Therefore, SARS-CoV-2-induced neurological problems may be caused by IDO1 [
163].