Pathogenic mechanisms of post-acute sequelae of SARS-CoV-2 infection (PASC)

PASC co, which encompasses Long COVID or post-COVID syndrome, is a condition marked by the continuation of COVID-19 symptoms—or the emergence of new ones—4 or more weeks after infection with SARS-CoV-2 (http://www.CDC.gov↗), or those persistently symptomatic for >30 days from onset of infection (http://www.NIH.gov↗). Alternatively, the World Health Organization describes PASC (or Post COVID-19 condition) as occurring in individuals with a history of probable or confirmed SARS-CoV-2 infection, usually 3 months from the onset of COVID-19 with symptoms that last for at least 2 months and cannot be explained by an alternative diagnosis (http://www.WHO.int↗). The current PASC definition in the NIH RECOVER (REsearching COVID to Enhance Recovery) Initiative program is: 'ongoing, relapsing, or new symptoms, or other health effects occurring after the acute phase of SARS-CoV-2 infection (i.e., present four or more weeks after the acute infection)'. The RECOVER Initiative is an NIH-funded project whose goal is to rapidly improve our understanding of and ability to predict, treat, and prevent PASC (post-acute sequelae of SARS-CoV-2), including Long COVID.

The symptoms of PASC whether in hospitalized or non-hospitalized patients are numerous and diverse and include shortness of breath, chest pain, cognitive impairment, and fever, among others (Michelen et al., 2021). Reports also reveal that there are wide and country-specific variations in the prevalence of PASC ranging from 3.3% in the UK, 13.9% in the USA, to 39% in Denmark and the Faroe Islands (Mantovani et al., 2022; Perlis et al., 2022; Petersen et al., 2022; Subramanian et al., 2022). The pathophysiologic mechanisms underlying PASC remain to be determined. However, the clinical features and epidemiology may provide insights into possible mechanisms. Although PASC may be a consequence of long-term organ damage attributable to acute-phase infection, it is postulated that specific mechanistic pathways exacerbate the initial illness adversely affecting many organs and resulting in the development of later symptoms. As such, this multifactorial condition displays over 200 symptoms and affects multiple tissues, organs, and biological systems (i.e., ears, eyes, head, nose, throat, kidney, brain, cardiopulmonary, endocrine, gastrointestinal (GI), musculoskeletal, neuropsychiatric, and systemic) to varying degrees. So far, immune dysregulation, autoimmunity, dysautonomia, endothelial dysfunction (ED), occult viral persistence, reactivation of pre-existing chronic viral infections as well as coagulopathies are some of the main underlying pathophysiological mechanisms described (Ahamed and Laurence, 2022; Al-Ramadan et al., 2021; Amenta et al., 2020; Camargo-Martínez et al., 2021; Carod-Artal, 2021; Choutka et al., 2022; Cortés-Telles et al., 2021; Doykov et al., 2020; Fernández-de-Las-Peñas et al., 2021; Gottschalk et al., 2023; Lee et al., 2021; Mantovani et al., 2022; McFarland et al., 2021; Oronsky et al., 2023; Silva Andrade et al., 2021; Song et al., 2021; Toniolo et al., 2021; Touyz et al., 2021; Townsend et al., 2021; Wirth and Scheibenbogen, 2021; Wong et al., 2021). Although PASC has been described as 'a mysterious mix of symptoms with no clear pattern', it bears similarities to other post-viral syndromes, and to myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), certain phenotypic clusters are emerging in the literature.

This review will delineate the current knowledge about the underlying mechanisms and/or biological pathways responsible for tissue damage including pulmonary, cardiac, GI, neuronal, vascular, etc. associated with COVID-19 and the subsequent generation of symptoms. It will highlight the underlying causes of PASC and suggest ways to understand the clinical symptomatology for improved diagnostic and therapeutic procedures for the condition. We will also postulate on the pathophysiology and pathobiology of PASC by providing working hypotheses that could aid in further exploration of this multifaceted syndrome. This review will attempt to decipher the possible underlying tissue injury mechanisms for long-term residual illness post-SARS-CoV-2 infection.

Epidemiological studies in the UK and USA have found similar symptoms, but differing rates and durations of Long COVID (Groff et al., 2021). A UK study discovered that SARS-CoV-2 infection was linked to 62 symptoms lasting beyond 12 weeks, including anosmia, hair loss, sneezing, ejaculation difficulties, and reduced libido. The risk of Long COVID was higher among women, ethnic minorities, those with lower socio-economic status, smokers, obese individuals, and those with underlying health conditions, and increased with decreasing age (Subramanian et al., 2022). In contrast, Long COVID affects 7.3% of 3042 US survey respondents, equivalent to 18.8 million adults. Impacts on daily activities were severe for 25.3% of those with Long COVID and 28.9% had SARS-CoV-2 infection over a year ago. Women, those with comorbidities, and those unvaccinated or not boosted were at higher risk (Robertson et al., 2022). Studies from the USA and Europe have found that the prevalence of Long COVID is higher among women and those with multiple chronic conditions (Chen et al., 2022; Fernández-de-Las-Peñas et al., 2022; Robertson et al., 2022; Thompson et al., 2022). However, there is variation in risk factor data due to differences in study design, population studied, and definition of Long COVID. Often, studies lack a COVID-free control group, focus on hospitalized or healthcare-seeking populations, and have different definitions of Long COVID. These studies emphasize the high burden of Long COVID and the need for further research on its prevalence and risk factors.

The clinical manifestations of PASC vary widely. Common PASC symptoms include fatigue, dyspnea, cough, 'brain fog', headaches, sleep disturbance, impaired smell and taste, palpitations, chest pain, and orthostatic dizziness (Davis et al., 2021; Hernandez-Romieu et al., 2022; Ozonoff et al., 2022; Sneller et al., 2022; Sudre et al., 2021). Using machine-learning algorithms, Estiri et al. uncovered 33 phenotypes of PASC and it is likely that multiple different pathophysiologic mechanisms account for this heterogeneity (Estiri et al., 2021). Indeed, detailed multi-omic profiling of COVID-19 patients revealed unique associations between immunological signatures and specific PASC endotypes (Su et al., 2022). Future mechanistic studies will need to recognize the heterogeneity of the disorder. PASC is more common in individuals who suffered a severe vs mild acute COVID-19 illness (Hernandez-Romieu et al., 2022; LaVergne et al., 2021; Xie et al., 2021). In such patients, the illness shares features of post-intensive care syndrome (PICS) (Rawal et al., 2017) and findings observed in survivors of the SARS and MERS coronavirus epidemics (Greenhalgh et al., 2020; Tansey et al., 2007). Pathophysiologic mechanisms in this group may include the effects of SARS-CoV-2 infection, reactivation of other viruses such as the Epstein–Barr virus (EBV), autoimmunity, incomplete recovery from acute organ injury (e.g., lung, heart, kidney, brain, thrombosis), effects of treatments (e.g., mechanical ventilation, steroids), and exacerbation of pre-existing conditions (e.g., pre-diabetes, CKD, CHF). However, individuals with mild COVID-19 (Augustin et al., 2021; Estiri et al., 2021), and even individuals with no serologic evidence of prior SARS-CoV-2 infection (Krishna et al., 2022; Sneller et al., 2022), can experience PASC symptoms, suggesting that some features of PASC may represent responses to the stress of the pandemic, for example, social isolation, anxiety, loss of friends or family, work stress, and financial stress, in addition to the viral infection itself. The prevalence of PASC appears to be lower in the more recent COVID 'waves', especially with the Omicron variant, and lower in people who were vaccinated prior to contracting COVID-19 (Azzolini et al., 2022; Zisis et al., 2022), suggesting that SARS-CoV-2 variants, availability of newer treatments and immune responses may impact the pathogenesis of PASC. For example, recent research suggests that the use of nirmatrelvir/ritonavir during acute SARS-CoV-2 infection reduces the risk of PASC regardless of vaccination status or history of prior infection (Xie et al., 2022). Long COVID may also manifest as residual or new organ dysfunction which could be the result of acute organ injury during the COVID-19 infection, for example, ARDS, acute kidney injury, or myocarditis, or the result of chronic inflammation, ED, or autoimmunity triggered by the infection. The high incidence of new-onset type I diabetes after COVID-19 infection, for example, could be due to direct viral injury of pancreatic beta cells, induction of autoimmunity to beta cells or glucose-regulating pathways or decreases in peripheral insulin sensitivity associated with chronic inflammation that is exhibited in PASC patients (Barrett et al., 2022; Rubino et al., 2020). There are commonalities of clinical manifestations and symptoms between COVID-19 and Long COVID-19 that require detailed mechanistic understandings (Davis et al., 2021; Mehandru and Merad, 2022).

Many PASC features resemble (ME/CFS), which is also commonly preceded by viral infections, prompting some to ask whether PASC and ME/CFS are similar diseases (Bornstein et al., 2022; Tate et al., 2022). Fatigue, exertion intolerance, and post-exertional malaise are among the most frequent symptoms cited for PASC. These same symptoms manifesting in PASC patients had been designated as 'core criteria' of ME/CFS by the National Academy of Medicine in a 2015 report (https://batemanhornecenter.org/wp-content/uploads/filebase/providers/Recognizing-PVS-and-Tools-ECHO-5_3_2022-V2.pdf↗). It is estimated that about half of people with Long COVID will meet the criteria for ME/CFS, whether they are given that specific diagnosis or not. It is also important to note the multiple types of tissues and organs adversely affected by ME/CFS including cardiovascular problems, central nervous system (CNS), immune system disturbances, ion transport dysfunction, cell energy metabolism, GI dysfunction, cognitive impairment, myalgia, arthralgia, orthostatic intolerance, and chronic inflammation (Rasa et al., 2018), are also commonly reported in PASC patients. The trigger for COVID-19 is the SARS-CoV-2 infection. ME/CFS is a complex disease frequently triggered by an infection with EBV (the virus that causes mononucleosis) or parvovirus B19 even though various other viral and nonviral triggers have also been reported (Chu et al., 2019; Fluge et al., 2021; Jason et al., 2022; Sotzny et al., 2018). The evidence is clear that in COVID-19 and potentially in PASC patients, the initial SARS-CoV-2 infection resurrects latent viruses such as EBV (Fluge et al., 2021), which could trigger ME/CFS. Although EBV has been associated with post-viral fatigue in ME/CFS, the mechanisms leading to post-viral fatigue syndrome and Long COVID (PASC) are unresolved. The similarities raise two questions: Do these two multi-symptomatic diseases share common molecular and pathophysiological pathways in their disruption of homeostasis post-infection? Although the triggers may be different, because of their overlapping symptoms, could there be a common pathobiological explanation? Indeed, the lessons learned from three decades of studying ME/CFS should help with identification of a variety of conceptually distinct medical comorbid conditions found in PASC.

Postural orthostatic tachycardia syndrome (POTS), which can be seen after other viral infections has also been observed in PASC and may point toward autoimmune dysfunction of the autonomic system, including autoantibodies to acetylcholine and adrenergic receptors (Blitshteyn and Whitelaw, 2021; Fluge et al., 2021). However, a recent German study reports that PASC is not associated with autoantibodies (Schultheiss et al., 2022). Nevertheless, SARS-CoV-2 may be responsible for PASC symptoms by activating the host's immune response and leading to long-term autoantibody production. Autoantibodies were isolated in acute COVID-19 patients by several research teams including Wang et al. who used REAP to screen a group of 194 SARS-CoV-2-positive COVID-19 patients for autoantibodies against 2770 extracellular proteins. Results showed that COVID-19 patients had a significant increase in autoantibody reactivity compared to uninfected controls (Jiang et al., 2023; Wang et al., 2021). It is currently not well understood what role autoantibodies may play in the pathophysiology of Long COVID or PASC. Evidence linking autoantibodies to Long COVID as an etiological factor is limited and further research is needed to fully understand their role in the development of PASC. Similarities between PASC symptoms and those of Mast Cell Activation Syndrome (MCAS) have also been noted (Akin, 2017; Blitshteyn and Whitelaw, 2021; Schieffer and Schieffer, 2022).

Longitudinal studies need to be conducted to determine the similarities and differences between PASC and ME/CFS and other post-infectious syndromes by frequently collecting biological samples (as it is being implemented by the RECOVER study—https://recovercovid.org↗), and up-to-date information on the presence of the infectious agent and the severity of various symptoms. Such studies should also include continual laboratory studies of the immune system, metabolism, gene structure and function, and transcriptome, as well as tests of cognition, sleep, and the functioning of the nervous system, heart, and cardiovascular system.

Although the COVID-19 pandemic disproportionately affected patients who had comorbid diabetes mellitus, it is now clear from numerous studies across the globe that COVID-19 causes dysregulation of glucose homeostasis leading to new-onset hyperglycemia and type I diabetes (Genç et al., 2021) in patients with no previous risk factors for diabetes mellitus (Bode et al., 2020; Sathish et al., 2021c; Sathish and Chandrika Anton, 2021b; Tittel et al., 2020; Unsworth et al., 2020; Watson, 2022). New-onset COVID-induced type 1 or 2 diabetes has also been documented in children (Tittel et al., 2020; Trieu et al., 2021; Unsworth et al., 2020). This novel form of COVID-induced diabetes with anomalous glycemic parameters and heightened rates of DKA (diabetic ketoacidosis), occurs in severely ill patients with higher mortality rates and poorer outcomes compared to COVID-19 patients with pre-existing diabetes (Chandrashekhar Joshi and Pozzilli, 2022). The underlying pathobiological mechanism for this induced diabetes in COVID-19 patients has not been clearly delineated although it is suspected that SARS-CoV-2 could damage the beta cells of the pancreas and precipitate insulin resistance (Aluganti Narasimhulu and Singla, 2022). There are some plausible mechanisms that underlie this new diabetes onset. SARS-CoV-2 infection can damage beta cells or increase glucose levels via oxidative stress and inflammation; the virus can also bind the angiotensin-converting enzyme receptor 2 (ACE2), on acinar cells and cause tissue detriment and pulmonary fibrosis via increasing concentration of the fibrosis-promoting Angiotensin II (Morganstein et al., 2021) and hyperglycemia can inhibit lymphocyte proliferation to repair the tissue damage. The initial binding of the virus to ACE2 requires the presence of transmembrane serine protease 2 (TMPRSS2), a protein that is regulated by androgens. The ACE2 mRNA is also expressed in several endocrine glands, including the pancreas, ovaries, testes, and thyroid gland (Lazartigues et al., 2020). At postmortem, both follicular and parafollicular cells of the thyroid gland were extensively damaged in patients who died of COVID (Wei et al., 2007). These systems, which are physiologically regulated by ACE2 upon viral infection, induce acute cardiopulmonary failure, and coagulopathy. Furthermore, the overexpression of the ACE2 receptor can deregulate these systems and cause cardiovascular instability (renin–angiotensin system), acute inflammatory pulmonary edema (kinin–kallikrein system), and thromboembolism (coagulation system) (Sidarta-Oliveira et al., 2020). In fact, ACE2 gene therapy in gene knockout mice promoted better survival and function of the beta cells, as well as improved glucose homeostasis (Bindom et al., 2010). Additionally, a downregulation of the transcription factor REST (RE1-silencing transcription factor) observed in COVID-19 patients is linked to the altered gene expression of glucose and lipid metabolism involving apelin, myostatin myeloperoxidase (peptides and proteins important for the insulin signaling pathway) (He et al., 2021). COVID-19 patients also harbored upregulated short-chain fatty acids, propionic acid, and isobutyric acid, which are possibly linked to insulin resistance (He et al., 2021).

Stress is another factor that can be induced by a viral infection, which causes elevated glucose levels in subjects with diabetes via the secretion of glucocorticoids and catecholamines contributing to further impairment of the immune system (Wang et al., 2020). In addition, the incidence of DKA observed in COVID-induced diabetic patients was unusually high in the absence of autoantibodies suggesting that acute viral-induced pancreatic damage might have occurred (Li et al., 2020b; Vellanki and Umpierrez GE - Center ofDiabetes and Metabolism, Emory University School of Medicine, 2021). There are also a few reports of cases that have described autoantibody-negative insulin-dependent diabetes (Hollstein et al., 2020; Kuchay et al., 2020). Mechanistically, the consequence of stress caused by SARS-CoV-2 infection may be explained by the stimulation of a signaling pathway designated as the integrated stress response (ISR) that promotes the activation of a family of serine/threonine kinases that is, double-stranded RNA-dependent protein kinase (PKR) and PKR-like ER kinase (PERK), and the subsequent induction of serine phosphorylation of insulin receptor substrates (IRS) which leads to downregulation of the insulin signaling pathway. Both the viral components (RNA or proteins) of SARS-CoV-2 and 'cytokine storm', which activates a family of serine/threonine kinases linked to the ISR and is represented by pro-inflammatory cytokines such as IL-6 and tumor necrosis factor (TNF-α, can cause insulin resistance Santos et al., 2021). Moreover, insulin resistance can be caused by the downregulation of insulin receptors in skeletal muscle via virally induced interferon gamma (IFN-γ) (Sestan et al., 2018). This could be an additional contributor to the insulin resistance syndrome that has been described in COVID-induced diabetes patients.

There are also several other postulated mechanisms including molecular mimicry, that may explain the rise in autoimmune diabetes following COVID-19 infection. SARS-CoV-2 also downregulates ACE2, which degrades Angiotensin II. These mechanisms of unrestricted actions of Angiotensin II lead to deleterious effects including inflammatory activity by increasing the infiltration of macrophages and monocytes, reducing blood flow to pancreatic islets, and promoting the instability of the beta cells leading to glycemic dysregulation (Carlsson et al., 1998; Chee et al., 2020; Montezano et al., 2014). There is emerging evidence showing that the new-onset diabetes is also a feature of Long COVID (Sathish et al., 2021a) even though the precise mechanism might be multifaceted. It is essential to collect biospecimens to design mechanistic studies that might shed light on the underlying causes precipitating the various types of symptoms experienced by Long COVID sufferers and specifically to follow-up COVID-related diabetes patients by implementing longitudinal and prospective cohort studies to distinguish between COVID-induced diabetes from conventional diabetes. There is a digital registry that is specifically designed to establish the extent and characteristics of new-onset, COVID-19-related diabetes, and to investigate its pathogenesis and outcomes (http://covidiab.e-dendrite.com↗).

The upper respiratory system is the primary site of the SARS-CoV-2 viral infection. In the scenario of an excessive inflammatory response, the consequence can be irreversible lung fibrosis and the compromise of respiratory effector function (Tanni et al., 2021). Fibrotic remodeling with characteristic findings consists of fibroblast proliferation, micro-honeycombing, and airspace obliteration, which were uncovered in pulmonary postmortem findings in a series of COVID-19 cases (Grosse et al., 2020). This suggested a link to respiratory distress syndrome (ARDS) as a long-term complication of COVID-19. In addition to ARDS, fibrotic remodeling resulting in respiratory symptoms can originate from pathologic conditions including dyspnea, and even from hospitalization, intensive care unit stays, use of high-flow oxygen support, and need for mechanical ventilation (Grillo et al., 2021). Additional mechanisms include epithelial and endothelial to mesenchymal transition and the 'cytokine storm'. Prolonged exposure to supplemental oxygen is known to lead to increased oxidative stress and could maintain the inflammatory status favorable for activation of post-COVID-19 pulmonary fibrosis (George et al., 2020). A fraction of PASC patients present with fatigue, exercise limitation, persistent chest pain, and breathlessness. These symptoms may be caused in part by compromised cardiorespiratory function, lung vascular damage in microvessels, and hemodynamic complications such as resolved thrombus (Kosuge et al., 2012; Yaméogo et al., 2011). Fatigue and breathlessness may also be caused by chronic overstimulation of the ergoreflex, activated by exercise to centrally couple ventilation and cardiovascular function to exercise intensity (Sze et al., 2022). Exaggerated ergoreflex leads to excessive response relative to work performed, and to the sensation of breathlessness and fatigue. The hypercatabolic state, cytokine storm, and stimulation of the renin–angiotensin–aldosterone system by SARS-CoV-2 use of ACE2 to facilitate cell entry might cause a reduction in skeletal muscle mass and function, excessive ergoreflex, and sensations of fatigue and breathlessness (Medina-Enríquez et al., 2020). Further study of the mechanistic involvement and interactions between respiratory systems and extrapulmonary systems with reciprocal crosstalk and influence is needed.

Pulmonary dysfunctions are among the most destructive events associated with impaired immune responses caused by SARS-CoV-2 infection. The resulting cytokine storm activates defense events that stimulate biochemical pathways that eventually lead to the production of tissue injury markers and the collapse of lung tissue (Piazza et al., 2020; Sakr et al., 2020; Sidarta-Oliveira et al., 2020). These hematological changes form pulmonary thrombi and capillary microthrombi, which may represent thromboembolism instead of thrombosis in bone marrow emboli and septic pulmonary thromboembolism (Ackermann et al., 2020; D'Errico et al., 2020; Grosse et al., 2020; Khan et al., 2020). Transcriptomic analyses of coexpression data from human lung cells have shown that there are three physiological systems directly involved in the pathogenesis of COVID-19: (1) the kinin–kallikrein system; (2) the renin–system angiotensin; and (3) the coagulation system coexpressed with the ACE2 receptor in alveolar cells (Sidarta-Oliveira et al., 2020). These systems, which are physiologically regulated by ACE2 upon viral infection of the lungs, induce acute cardiovascular failure, coagulopathy, acute inflammatory pulmonary edema (kinin–kallikrein system), and thromboembolism (coagulation system) (Sidarta-Oliveira et al., 2020). Hyperinflammation induced by immunopathogenic, and biochemical mechanisms of SARS-CoV-2 suggests that cytokines and chemokines contribute to alveolar endothelial damage that promotes apoptosis of alveolar cells (pneumocyte type I) and degeneration of pulmonary pneumocyte type II alveolar cells (Hussman, 2020). These cell types are operationally connected via narrow junctions that control the transfer of ions, minerals, and fluid through the epithelium (Ackermann et al., 2020). SARS-CoV-2 infects ACE2-expressing pneumocyte type II cells, which act as epithelial immune cells and produce TNF, IL-6, IL-1β, and MCP-1.

Figure 1 depicts the initial entry of SARS-CoV-2 into the host cell by binding to ACE2 receptor in the respiratory tract and causing disruption of homeostasis by persisting in several organ systems and leading to the eventual development of PASC in a subset of patients. The depiction is for immunological, neurological, and pulmonary dysregulations, which may likely contribute to the hyperactivation of monocyte-derived macrophages in both acute and post-acute phases of the disease. Type I IFN-γ production is delayed leading to increased sensing of microbial threats. This in turn enhances the release of monocyte chemokines by alveolar epithelial cells (pneumocyte types 1 and 2), which direct blood monocytes into the lungs. There are several signaling molecules that induce monocytes to differentiate into pro-inflammatory macrophages via activation of Janus kinase (JAK)–signal transducer and activator of transcription (STAT) pathways. Activated T cells and natural killer cells enhance the recruitment and stimulation of macrophages (derived from monocytes) through the production of granulocyte–macrophage colony-stimulating factor (GM-CSF), IFN-γ, and TNF (see Figure 1).

Alveolar macrophages lead to the activation and release of cytokine and chemokine and aggregation of neutrophils and monocytes that produce toxic mediators resulting in the death of the alveolar cells, fibrin deposition, hyaline membrane formation, the influx of red blood cells, alveolar edema, and dyspnea (Ackermann et al., 2020; Al-Khawaga and Abdelalim, 2020).

Whole-transcriptome sequencing of postmortem lung tissue from COVID-19 patients reveals two distinct molecular signatures: Lung damage caused by a massive metabolic reprogramming due to the upregulation of the unfolded protein response, steroid biosynthesis, and complement activation; and secondarily, 'cytokine release syndrome' (CRS) represented by the upregulation of cytokines such as IL-1 and CCL19, but absence of complement activation (Budhraja et al., 2022). Although most patients cleared the viral infection, they succumbed to acute dysbiosis overrepresented by Staphylococcus cohnii in 'classical' patients and Pasteurella multocida in CRS patients (Budhraja et al., 2022).

The association between insulin resistance and lung dysfunction in humans has also been documented in the literature. Another recent study confirmed other previous studies that revealed that the insulin/insulin-like growth factor 1 (IGF) signaling pathway, which plays an important role in energy metabolism, is impaired in organs and tissues of severe COVID-19 patients (Shin et al., 2022). AKT/mTOR/MAPK pathway and ligand/receptor interactions that initiate cellular signaling through IRS/PI3K lie downstream of insulin/IGF signaling pathway but are also additional molecular pathways that are associated with energy uptake and utilization. An impaired IGF pathway engenders insulin resistance or abnormal insulin/IGF response, which also leads to an enhanced metabolic syndrome manifestation such as diabetes, hyperglycemia, hyperlipidemia, and obesity. A COVID-19-impaired insulin/IGF signaling pathway is also associated with the downregulation of various metabolic pathways including the citrate cycle, lipid metabolism, beta-oxidation, amino acid metabolism, carbohydrate metabolism, and respiratory electron transport chain affecting the adipose tissue, liver, lungs, and pancreas resulting in the development of multiorgan dysfunction (Shin et al., 2022).

Chest pain, dyspnea, and heart palpitations are some of the key symptoms that draw attention to the cardiovascular system (Gluckman et al., 2022). However, cardiovascular disease (CVD) also includes congenital heart disease, coronary artery disease, hypertension (HTN), peripheral artery disease, stroke, and heart failure. Regardless of the severity of COVID-19, many patients have a complicated course of COVID-19 with regard to CVD. This is especially true for viral pneumonia (remembering that COVID-19 is primarily a respiratory infectious disease) that may result in severe systemic inflammation, acute myocardial infarction (MI), acute heart failure, cardiac arrest, venous thromboembolism (VTE), tachyarrhythmias, and stroke (Baldi et al., 2020; Basso et al., 2020; Gao et al., 2020; Lala et al., 2020; Parohan et al., 2020; Shi et al., 2020; Wallentin et al., 2020; Zhou et al., 2020). In a single-centered observational study, up to 329 days following diagnosis with COVID-19, approximately 75% reported persistent cardiac symptoms (Gluckman et al., 2022). In the absence of validated biomarkers for PASC, imaging evidence using cardiac magnetic resonance (CMR) was obtained for myocardial inflammation, pericardial enhancement, and diffuse myocardial edema. Similarly, various abnormalities on CMR as well positron emission tomography have been reported in COVID-19 patients, even in the absence of cardiac symptoms (Hanneman et al., 2022; Puntmann et al., 2020). There are indicators of CVD such as cardiomyocyte injury, quantified by cardiac troponin (cTn) concentrations, heart failure, quantified by N-terminal pro B-type natriuretic peptide (NT-proBNP), and hemodynamic cardiac stress, quantified by natriuretic peptide concentrations that were detected in COVID-19 patients (Flores et al., 2019; Lala et al., 2020; Mueller et al., 2021; Parohan et al., 2020; Shi et al., 2020; Zhou et al., 2020). The potential mechanisms underlying elevations in acute cTn and therefore myocardial injury from non-ischemic causes in COVID-19 patients include direct effect on myocardial cells via ACE2 receptor, cytokine storm, and hypoxia-induced apoptosis as well as myocarditis, Takotsubo syndrome, and pulmonary embolism. The rise in cTn from ischemic causes in COVID-19 patients include type I MI, and type II MI comprising shock, hypoxia, and tachycardia (Mueller et al., 2021).

SARS-CoV-2 can infect the brain, causing neuroinflammation (Pan et al., 2020). Moreover, inflammation elsewhere in the body can activate the innate immune system (Figure 1) in the brain via both humoral and retrograde neural signals, largely involving the vagus nerve (Poon et al., 2015; VanElzakker, 2013). Different study model systems such as autopsies, animals and organoids used to study the infective potential of the SARS-CoV-2 virus show that SARS-CoV-2 can reach and infect neurons and cells of the CNS and produce neuroinflammation (Matschke et al., 2020; Song et al., 2020). Neuropsychological symptoms are frequent in PASC survivors of COVID-19 who often experience lingering neurological symptoms analogous to cancer-therapy-related cognitive impairment, which involves white matter microglial reactivity and consequent neural dysregulation. SARS-CoV-2 infection has been shown to affect white-matter-selective microglial reactivity in mice and humans by impairing hippocampal neurogenesis, decreasing oligodendrocytes, and promoting myelin loss with a concomitant increase in CSF cytokines/chemokines including CCL11 (Fernández-Castañeda et al., 2022a). Accordingly, cognitive impairment manifested even with mild COVID symptoms parallel the neuropathophysiology that is observed in people with cancer therapy (Fernández-Castañeda et al., 2022b). The question remains: How does the virus manage to get to the brain? One potential pathway to reach the CNS is via hematogenous dissemination from severely infected upper and lower respiratory system by way of systemic inflammation that raises the permeability of blood–brain barrier (BBB) and permit circulating non-BBB-crossing molecules to adversely affect brain function (Proal and VanElzakker, 2021). More importantly for PASC, during acute infection this porousness can also allow for viral neuroinvasion that may persist past the acute infection phase. Although the mechanism is not well known, this neuroinvasion can occur directly or indirectly through host immune cells infected with the virus and are actively transported into the CNS (Dando et al., 2014; Lauer et al., 2018). A large MRI registry study that examined changes in the brain before and a mean of 141 days following SAR-CoV-2 identified reductions global brain size and gray matter thickness in orbitofrontal cortex/parahippocampal gyrus. Importantly, the study also identified increased markers of tissue damage in regions associated with the primary olfactory cortex following infection. These data suggest that COVID-19 has profound impact on the brain, especially the olfactory cortex, which may explain changes of taste and smell and may indicate CNS infection from direct extension from the olfactory system (Douaud et al., 2022).

The extrapulmonary disease that SARS-CoV-2 inflicts upon the body also extends to the GI tract where there is evidence that the virus may persist where the ACE2 receptor is highly expressed (Settanni et al., 2021). Given that the severity of COVID-19 is typically engendered by a set of comorbidities such as HTN, diabetes, obesity, and/or advanced age, the GI symptoms, such as diarrhea, vomiting, or abdominal pain during the early phases of the disease can be debilitating (Lamers et al., 2020). Symptoms can last for years as is the case for post-infectious irritable bowel syndrome (PI-IBS), which is punctuated by the persistence of abdominal discomfort, bloating and diarrhea that continue despite clearance of the inciting pathogen, usually bacterial origin (Dunlop et al., 2003; Kim et al., 2006; Settanni et al., 2021). Again, the culprit in all these acute and post-acute COVID-19 symptoms including the impairment of bowel physiology may be a dysregulation of ACE2-mediated functions due to a competitive mechanism of the virus on ACE2 receptor or from a downregulation of its anti-inflammatory activity. To date, there is a dearth of data on the GI sequelae of SARS-CoV-2 infection. This may be due to the uncertainty over COVID-19's path of action on the gut physiology.

While much of our current understanding of PASC comes from adults, one of the first PASC phenotypes occurred in children. Shortly after the emergence of SARS-CoV-2, rare cases of multisystem inflammatory syndrome in children (MIS-C) were documented (Riphagen et al., 2020). It is now understood that MIS-C occurs weeks after SARS-CoV-2 infection and is characterized by persistent fevers, massive inflammation, and multiorgan involvement including long-term neurologic complications. The most striking clinical manifestation is profound myocardial dysfunction resulting in shock; however, extracardiac organ system dysfunction is also common, particularly GI (Cheung et al., 2020; Feldstein et al., 2020; Riphagen et al., 2020). The mechanisms which underlie MIS-C are not fully understood but emerging evidence points toward immune system dysfunction. Acute MIS-C is defined by massive elevations in inflammatory cytokines, particularly those related to myeloid recruitment and mucosal immunity (Consiglio et al., 2020; Gruber et al., 2020). The production of autoreactive antibodies is also characteristic of acute MIS-C with antibodies specific for endothelial proteins detected in many patients (Consiglio et al., 2020; Gruber et al., 2020; Ramaswamy et al., 2021). Aberrant activation of T cells is a consistent finding as well (Vella et al., 2021), with a potentially pathognomonic role for biased expansions of T cell clones (Porritt et al., 2021). Activated T lymphocyte cell expansions resemble those seen following bacterial superantigen stimulation with additional evidence suggesting that motifs from the SARS-CoV-2 spike protein may be capable of promoting autoimmunity (Noval Rivas et al., 2022). These immune derangements along with elevated biomarkers suggesting mucosal barrier breakdown have been linked to detectable antigenic persistence in circulation (Yonker et al., 2021), suggesting an additional role for viral persistence in the pathogenesis of MIS-C. Treatment for MIS-C has included steroids, intravenous immunoglobulin, and monoclonal antibodies targeting IL-6 and IL-1Ra (Feldstein et al., 2020). Fortunately, following treatment for MIS-C, short to mid-range reported health outcomes have been excellent (Farooqi et al., 2021; Matsubara et al., 2022). Pediatric PASC outside of MIS-C has been more difficult to quantify, with estimates ranging from 4% to 25% of previously infected children experiencing PASC symptoms (Izquierdo-Pujol et al., 2022). Risk factors for development of pediatric PASC include severity of acute infection, younger age (<5 years), and presence of chronic health conditions (Rao et al., 2022). In case reports or small series, endothelial damage and encephalomyelitis have been identified in children with PASC (Buonsenso et al., 2021; Lindan et al., 2021) but underlying mechanisms remain undefined. Based on our current knowledge of MIS-C and the evidence that pediatric immune responses to SARS-CoV-2 are distinct from adults (Pierce et al., 2020; Weisberg et al., 2021; Yoshida et al., 2022), understanding immune responses in pediatric PASC must be prioritized. To better characterize pediatric PASC subtypes and define the mechanisms giving rise to persistent symptomatology, awareness, and identification followed by study enrollment of relatively rare cases is critical to maximize our ability to understand and treat this emerging pediatric health threat.

What are PASC factors? When can the factors be assayed in the course of disease progression? Are these factors interrelated or independent? What are the immune and genomic signatures of PASC? What is the eventual public disease burden of PASC? What does PASC mean for patients and what can be done about it? The unknowns about PASC are too numerous to list. We still do not know if there is a correlation between PASC and the duration/frequency/severity of the initial SARS-CoV-2 infection. Also, the subsequent iterations of the virus (i.e., omicron variants) with factors like age, gender, demographics, and comorbidities are not clear. The longitudinal collection of biospecimens and deep phenotyping of PASC subjects in the RECOVER study will provide a rich set of resources to answer key prognostic, epidemiologic and pathophysiologic questions about PASC and will serve as a platform for testing interventions to prevent or treat PASC.

We urgently need comprehensive epidemiologic studies via surveys/questionnaires and by interrogating the electronic health record databases and through longitudinal study enrollment that allow for a more complete understanding of the prevalence ad incidence of PASC within various populations. Furthermore, comprehensive examination of blood and tissues using synergistic methods such as proteomics, metabolomics, transcriptomics, viral nucleic acid and protein testing, and virus-specific and auto-reactive antibody characterization will be needed. Additionally, comprehensive immunophenotyping of PASC can be performed to identify cell markers and to provide further pathophysiologic understanding of the interplay between immune responses, inflammation, and PASC phenotypes. It must also be noted that the mechanisms underlying Long COVID/PASC may not be uniform across all individuals. Among the more prominent sociodemographic and clinical risk factors, the severity of acute COVID-19 infection, female sex, advanced age, pre-existing diabetes mellitus, more than five early symptoms, early dyspnea, and prior psychiatric disorders have been identified (Carfì et al., 2020; Subramanian et al., 2022; Yong, 2021). Additional variables that need to be considered include the variant of SARS-CoV-2 infection and those who develop PASC and the role of vaccination and antiviral therapy in acute infection (Antonelli et al., 2022; Ledford, 2022). Double-vaccinated participants with COVID-19 caused by the Omicron BA.1, BA.4, and BA.5 variants were less likely to develop PASC, as compared to those infected by the Delta variant. It is also likely that most of the symptoms and pathologies associated with COVID-19 persist in Long COVID perhaps because of the spike protein, which is also the target of the vaccination. The currently known and suspected mechanisms and molecular features for COVID and potentially for Long COVID/PASC pathologies are shown in Table 1.

There are various hypotheses underlying Long COVID/PASC symptoms. The first points to the immune system going haywire after the initial virus infection, triggering a long-lasting inflammatory response that inflicts havoc on multiple organ systems. The second hypothesis is that this long-lasting inflammation may be due to the continued presence of virus or viral particles in organs such as the gut but may also be due to the reactivation of latent pathogens such as EBV, the most common cause of mononucleosis. The third hypothesis states that COVID-19 elicits a radical response analogous to an autoimmune disease that causes antibodies to target and destroy the body's own tissues. The fourth leading hypothesis stipulates the formation of microclots, tiny blood clots that cause coagulopathy and lead to thrombosis. Therefore, the conclusion from this review on tissue damage is that there may be more than a single mechanism in motion for Long COVID (PASC) patients who experience a constellation of symptoms. Certain mechanisms may dominate over others in individual patients. There may be potential biomarkers or signaling pathways for different tissues or organs affected as well as therapeutic applications mostly in the form of inhibitors of the presumptive biomarkers of various tissues presented in Table 1, that could be employed for PASC sufferers until the underlying individual or collective mechanisms giving rise to their symptoms are clearly delineated. Furthermore, there is an urgent need to design experimental protocols that can distinctively identify the underlying mechanisms or pathways that are responsible for the clinical manifestations exhibited by Long COVID or PASC patients.

Authorship has been determined according to ICMJE recommendations.