Hyaluronan in the pathogenesis of acute and post-acute COVID-19 infection

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) recently emerged as the cause of a global pandemic. Infection with SARS-CoV-2 can result in COVID-19 with both acute and chronic disease manifestations that continue to impact many patients long after the resolution of viral replication. There is therefore great interest in understanding the host factors that contribute to COVID-19 pathogenesis. In this review, we address the role of hyaluronan (HA), an extracellular matrix polymer with roles in inflammation and cellular metabolism, in COVID-19 and critically evaluate the hypothesis that HA promotes COVID-19 pathogenesis. We first provide a brief overview of COVID-19 infection. Then we briefly summarize the known roles of HA in airway inflammation and immunity. We then address what is known about HA and the pathogenesis of COVID-19 acute respiratory distress syndrome (COVID-19 ARDS). Next, we examine potential roles for HA in post-acute SARS-CoV-2 infection (PASC), also known as “long COVID” as well as in COVID-associated fibrosis. Finally, we discuss the potential therapeutics that target HA as a means to treat COVID-19, including the repurposed drug hymecromone (4-methylumbelliferone). We conclude that HA is a promising potential therapeutic target for the treatment of COVID-19.

Introduction

In late 2019, the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a new and highly pathogenic strain of coronavirus, was identified in China [1]. It soon became clear that infection with SARS-CoV-2 can lead to coronavirus disease (COVID-19), including potentially severe respiratory tract disease [2,3]. By February of 2020, Chinese healthcare workers encountered an average of 3000 new cases per day, and on March 11, 2020, the World Health Organization (WHO) officially categorized the worldwide outbreak of COVID-19 a pandemic [1,3]. The virus has caused a worldwide health crisis causing nearly 604 million confirmed cases and 6.5 million deaths as of September 2022 [1,[4], [5], [6], [7]].

Three years after it emerged and despite the development of effective vaccines, the COVID-19 pandemic continues to spread. Moreover, many individuals suffer persistent, debilitating post-acute sequelae of SARS-CoV-2 infection (PASC), also known as “long COVID” [8], [9], [10]. Therefore, there is great interest in identifying additional inflammatory factors and pathways that contribute to COVID-19, in the hope that these might lead to novel therapeutic targets and biomarkers.

The purpose of this article is to review the existing data on one such factor, the extracellular matrix polymer hyaluronan (HA), in COVID-19 disease and to critically evaluate the hypothesis that HA drives COVID-19 pathogenesis. We first provide a short overview of HA in inflammation and cellular metabolism, particularly in the context of pulmonary disease. We then briefly review the literature on COVID-19 acute respiratory syndrome (ARDS) and PASC. Next, we address what is known about HA in acute-COVID and PASC. Finally, we discuss potential therapeutics for COVID-19 that target HA. This review provides a foundation for understanding the role of HA in the pathogenesis and treatment of COVID-19.

HA is elevated in airway inflammation, lung infection, and chronic fibrotic diseases of the lung

HA is a glycosaminoglycan with important roles in inflammation and cellular metabolism [11]. HA is a polysaccharide comprised of a repeating disaccharide that plays important roles in healthy tissues. Unlike other glycosaminoglycans, it is not linked to a protein core and instead exists as elongated, highly charged strands [12]. HA is extremely hygroscopic and is able to absorb water over 1000 times its own molecular weight [13,14]. In healthy tissues its high molecular weight and flexible polymer chain structure allow it to form viscoelastic barriers. In healthy tissues, HA provides structural support to skin, joints, and other tissues [12]. Within the lung, HA is a crucial component of bronchial basement membranes, bronchiolar epithelium, alveolar tissues, and the endothelial glycocalyx. In addition, HA is found at the surface of alveolar macrophages and Type II alveolar epithelial cells [15].

At these sites, HA is typically bound to a diverse group of HA binding proteins, called hyaladherins. These HA–hyaladherin macromolecular complexes interact with a variety of cell-surface proteins, growth factors, chemokines, and proteases to modulate the adhesive properties and activation state of inflammatory cells [16], [17], [18]. Examples of hyaladherins include inter-α-trypsin-inhibitor (IαI) and TNF-stimulated gene-6 (TSG-6) [19], [20], [21], [22], [23].

TSG-6 catalyzes the covalent transfer of heavy chains (HCs)(also known as HA-associated protein SHAP) from IαI to HA [24]. This results in the formation of the HC—HA complex, a form of HA that promotes the adhesion of leukocytes to HA-rich matrices and thereby influences the inflammatory response [21,25,26]. Elements of IαI and versican, including the heavy chains (HC) of IαI, can interact with HA, forming a new HC—HA complex called HA cables. These HC—HA complexes promote the binding of leukocytes to the ECM and monocytes to the cell membrane, contributing to tissue destruction [21,25] can also be potently anti-inflammatory (for example, in amniotic membrane where it polarizes macrophages to M2 phenotype [27]. There is evidence that TSG-6 drives pathology in a number of lung conditions (e.g., ozone-induced airway hyperresponsiveness), whereas it can be protective in others (e.g., LPS-induced lung injury), as recently reviewed [28]. In these and other ways, hyaladherins modulate interactions between ECM and the innate immune system.

At sites of tissue injury and infection, both HA production and catabolism are greatly upregulated leading to the accumulation of HA fragments. In this context, HA promotes leukocyte migration and activation within inflamed tissues [29,30]. The capacity of HA to absorb water drives edema and swelling in ways that increase leukocyte extravasation and tissue stiffness [31].

The biological function of HA is influenced by its size. High-molecular weight HA (HMW-HA); ∼2  ×  10⁵ - 7  ×  10⁶ Da, predominates in most tissues under healthy conditions and dampens inflammation [32], [33], [34], [35], [36]. In contrast, fragmented low-molecular weight HA (LMW-HA); <120 kDa, has been reported to be pro-inflammatory and predominates at sites of active inflammation [37]. Recent work suggests LMW-HA octa-saccharides prove effective monovalent competitors HMW-HA receptor interactions, potentially abrogating the anti-inflammatory effects of HMW-HA [38]. Moreover, LMW-HA is reported to function as an endogenous danger-associated molecular pattern molecule (DAMPs) that triggers inflammatory responses through TLR4 [39], [40], [41]. In this way, LMW-HA fragments may promote inflammatory cell recruitment and release of inflammatory cytokines [42,43]. Alternatively, HA may also distinctly activate a complex of TLR4/MD2/CD44 receptors [44].

However, this body of literature is complicated by publications which suggest that bacterial contaminants as well as DNA may have played a role in some of these pro-inflammatory effects [15,45,46]. Direct data demonstrating binding between HA and TLR4 are also absent. There is clearly a need to learn more about the role of HA in the host's response to pathogens. Several excellent reviews summarize the current state of knowledge regarding the role of HA size in inflammatory responses and a detailed investigation of the receptors involved in this biology [42,[47], [48], [49]].

The amount and size of HA polymers are regulated both at the level of HA synthesis and HA catabolism. HA is synthesized by three membrane-bound synthase enzymes (HAS1, HAS2, HAS3), found on the inner surface of the plasma membrane, which extrude HA as a polymer of repeating disaccharide units of N-acetylglucosamine and glucuronic acid [50]. The HA synthases generate HA polymers of distinct lengths [51] in response to inflammatory cytokines (i.e. TNF and interferons [52]), including in viral respiratory infections [53]. HA is degraded in part by the activity of hyaluronidases [30,54,55] which catalyze the cleavage of HMW-HA strands into smaller, LMW-HA fragments generally of <120 kDa [30,56]. Mechanical force, injury, infection, and oxidative stress also contribute to HA catabolism and result in LMW-HA [57,58]. The biology of HA synthesis and catabolism has been addressed in excellent reviews elsewhere [42].

There is an abundance of data implicating HA in inflammatory diseases of the lung. Research has found heightened HA levels in cases of chronic obstructive pulmonary disease (COPD) [59,60], asthma, and pulmonary hypertension (pH) [61]. HA is also associated with fibrotic diseases in the lung, including IPF [62], [63], [64], a disease characterized by progressive lung scarring, chronic respiratory failure, and severe hypoxemia [65]. While the causes of IPF remain unknown, HA has been shown to be important for regulating disease progression and HA-based treatments have shown efficacy in treating chronic inflammation and collagen deposition in IPF [66], [67], [68], [69]. Enhanced expression of HA through overexpression of HAS2 has been demonstrated to exacerbate fibrotic injury of the lung [70]. Similarly, in models of pulmonary hypertension, increased HA has been detected [71], [72], [73], [74], and its role in promoting vascular remodeling has been identified using gain and loss of function approaches targeting HAS2 [72]. Herein, HAS2 deletion protected mice from developing pH yet overexpression of HAS in smooth muscle cells led to worsening pH.

In line with these studies, inhibition of HA synthesis ameliorates disease in many of these settings [71,[75], [76], [77]]. While HA is increased in many diseases featuring airway injury or inflammation, the degree of accumulation and localization within the tissue parenchyma [15] can vary greatly. However, a more nuanced view of HA in these contexts is needed as in some settings HA can be anti-fibrotic while in others it can be pro-fibrotic [62,78].

HA is also implicated in ARDS [79,80]. Ventilator-induced lung injury promotes the synthesis of LMW-HA via HAS3 upregulation in fibroblasts, leading to increased inflammatory cell infiltration, vascular leak, and both interstitial and alveolar edema [81]. In combination with alveolar collapse, this contributes to the ventilation-perfusion mismatch, hypoxia, and respiratory failure seen in ARDS. HA levels in bronchoalveolar lavage fluid are associated with organ failure in ARDS [79] and there is an inverse correlation between the concentration of HA (BAL and serum) and the pulmonary oxygenation index in patients with ARDS [82,83]. Using experimental models of ARDS, investigators have demonstrated that inhibition of HA reduced markers of cell injury [84,85].

The role of HA in lung inflammation is a large and complex subject and the sections above only provide a short overview of this biology. For further information, we refer the reader to several excellent reviews [15,30,51,59].

COVID-19 infection is associated with both acute disease as well as long-term sequalae post-infection

Following transmission via respiratory droplets, SARS-CoV2 binds to multiple cell types, including epithelial cells, alveolar epithelial cells, vascular endothelial cells, fibroblasts and macrophages via the angiotensin-converting enzyme 2 (ACE2). The molecular mechanisms involved in SARS-CoV2 infection and the pathophysiology of COVID-19 infection have been addressed in depth elsewhere [86, 91, 87, 88] However, in brief, upon infection, host cells undergo necrosis, pyroptosis and necroptosis as a result of direct viral toxicity. Cellular debris and viral components then trigger innate immune pathogen recognition receptors. They similarly activate pathways downstream of the ACE2 receptors which SARS-CoV-2 uses for host entry [89], [90], [91]. Induction of cytokine “storm” and endothelial dysfunction, topics addressed later in this review, drive the inflammatory cascade that underlies acute COVID-19 [92,93]. This pathophysiology can impact nearly every organ system, including the lungs, heart [94], neurologic tissues [95], and gastrointestinal tract [96].

There is substantial heterogeneity in the patterns of disease associated with COVID-19 [97,98]. Mild cases of acute COVID-19 often display symptoms typical of upper respiratory tract infections, including fever, fatigue, and sore throat [2]. Older populations and patients with risk factors (e.g. diabetes, immune suppression, etc.) are prone to develop more severe respiratory symptoms [3,99]. Mortality is linked to risk factors including age, diabetes, underlying immune suppression, and host ethnic background and genetics [100], [101], [102], [103].

Severe cases of COVID-19 can manifest as ARDS [104,105], a respiratory syndrome characterized by widespread pulmonary inflammation, airway edema, and poor oxygenation [50] which often necessitates prolonged, challenging mechanical ventilation. Many computed tomographic images of patients with severe COVID-19 disease revealed “ground glass opacities”, often associated with areas of airway edema, in the patient's lungs. Consistent with ARDS, dense hyaline material was found in the lungs of autopsy cases [106,107]. Thick, respiratory secretions are also a prominent feature of severe COVID-19 respiratory disease [105,108].

In addition to acute COVID-19 respiratory disease, many individuals suffer persistent, debilitating symptoms of PASC [8], [9], [10]. PASC cases vary in severity and duration and can present symptoms ranging from respiratory difficulty and fatigue to neurological issues [109]. While acute COVID-19 lasts for 1–4 weeks, PASC symptoms can persist for months [8,98,110,111]. Approximately a third of patients developing PASC experienced asymptomatic acute COVID, suggesting that PASC is possible regardless of the severity of the acute COVID infection [112]. One study on Italian Healthcare Workers found obesity significantly predisposes participants to PASC [113]. Other metabolic and autoimmune diseases have been implicated [113,114]. In light of the ongoing COVID-19 pandemic and the uncertain pathogenesis of PASC, many countries are implementing programs dedicated to understanding PASC and similar post-infection syndromes [99].

One of the most severe, long-term manifestations of COVID-19 is lung fibrosis [115,116]. This can be highly debilitating, requiring long-term oxygen therapy [117]. There are indications that up to a third of patients who recover from COVID-19 ARDS develop fibrotic abnormalities [118]. The severity of ARDS and the duration of illness seem to be important predictors of pulmonary fibrosis in COVID-19 [119,120]. For a subset of these patients, lung transplantation is the only treatment option [121]. In these patients, extensive fibrotic lung injury presents on average 79 days from the time of the first COVID-19 symptoms or first positive test for SARS-CoV-2 [122,123]. COVID-19 pulmonary fibrosis appears to be distinct from idiopathic pulmonary fibrosis (IPF) and other fibrotic lung diseases that can take decades to develop [124], [125], [126], [127]. However, only a handful of studies have probed into the mechanisms that lead to fulminant lung fibrosis in patients with COVID-19 [122,128,129].

In addition to lung fibrosis, SARS-CoV-2 has been demonstrated to induce profound effects in the vasculature including intussusceptive angiogenesis, endothelial cell injury, vascular remodeling and changes in vascular tone. These changes have been summarized in a pair of articles [130,131]. The effects of SARS-CoV2 in the pulmonary circulation are significant since not only were they impacted by thrombotic process in the early stages of SARS-CoV2 infection, but pulmonary vascular sequelae following COVID-19 is increasingly recognized as a potential public health problem [132].

For further information, we refer the reader to several excellent reviews summarizing the clinical features and epidemiology of PASC [133], [134], [135].

HA is abundant in acute COVID-19 ARDS

HA is increased in the lungs of individuals with COVID-19. This was first demonstrated by Hellman et al., who showed HA staining in cadaveric histologic lung tissue sections from deceased COVID-19 patients [14]. Similar findings were reported in pre-prints or speculated upon earlier in the pandemic [13,136]. In these and other studies, significant HA staining was found in intra-alveolar spaces, particularly in areas of necrosis and inflammation, together with diffuse hemorrhage and hyaline membranes [137,138]. Consistent with this, genes involved in HA metabolism were over-represented in bronchoalveolar cells infected by SARS-CoV-2 [139]. HA synthases expression (HAS1–3) is likewise significantly upregulated in lung tissue of COVID-19 patients [140].

HA is abundant in the respiratory secretions of patients with severe COVID-19 compared to healthy patients [14,136]. A recent study observed greatly increased HA content (predominantly LMW-HA) in sputum from recently intubated patients along with increased DNA from dead or dying cells [137]. Treatment with hyaluronidase and DNase demonstrated that these two polymers contribute to the thick, tenacious respiratory secretions seen in COVID-19 ARDS [137]. The increase in the amount of HA may contribute to the respiratory pathophysiology of COVID-19 ARDS, including fluid accumulation, airway plugging, and reduction of oxygen exchange in the lung, leading to respiratory failure [141]. This LMW-HA may drive further inflammatory responses. Alternatively, it may be that the accumulation of HA fragments is not "an accumulation of pro-inflammatory HA fragments" but rather removing the anti-inflammatory HMW-HA via a competitive inhibitor mechanism [38]. More research is needed to tease these relationships apart.

Along with HA, hyaladherins levels are also altered in the lung. Increased versican and diminished TSG-6 were observed in lung sections from deceased patients with COVID-19 ARDS compared to other ARDS cases [137]. However, as discussed above, the impact of TSG-6 on HA in the lung is complex and the impact of TSG-6 levels on COVID-19 pathophysiology is unclear.

Serum HA is likewise associated with the pathogenesis of severe COVID-19. This is supported by Queisser et al., who reported that circulating HA fragments and serum hyaluronidase were strongly associated with organ failure and increased inflammatory cytokine levels in patients with COVID-19 [55]. Furthermore, serum HA has been identified as a predictor of COVID-19 disease severity [138]. Plasma HA levels during early infection and admission to healthcare facilities may similarly distinguish severely ill patients with COVID-19 [14,142]. Gene expression studies of cells in BAL fluid from COVID-19 patients further implicated HA in a “bradykinin-storm” signature associated with severe disease [140]. It has further been proposed based on transcriptomic and metabolomic studies that glutamine deficiency and overproduced HA is the central metabolic characteristic of severe, acute COVID-19 [143]. Together, these reports support a pathological role for HA in humans with COVID-19 disease.

Several factors may contribute to HA production during acute COVID-19. As noted above, SARS-CoV-2 infection triggers a “storm” of pro-inflammatory cytokines [144], leading to pronounced airway inflammation and systemic lymphocytopenia [145], [146], [147], [148], [149]. This includes injury to the endothelial glycocalyx, a pericellular structure which contains hyaluronan [150], [151], [152]. At an early stage, the infected cells also release inflammatory cytokines and chemokines such as IL-1β, IL-8, IL-18 [104], IL6, Type I interferon, and TNF-α [153]. Increased levels of the inflammatory cytokine IL-6 correspond to a greater need for mechanical ventilation [14]. Type 1 interferon production has likewise been associated with acute respiratory disease [154,155]. Along with driving HA production, these factors further promote immune activation and neutrophil degranulation, leading to the release of reactive oxygen species, proteases, and other factors that perpetuate the inflammatory cascade [156].

Several of these cytokines, including IL-6, IFNy, and TNFα, are known to drive HA production in other contexts [52]. These include in-vitro experiments observing the effect of pro-inflammatory cytokines on HA synthesis in human umbilical cord endothelial cells [157] and orbital fibroblasts [158].

In a mouse model of COVID-19 infection, HA production in the lung could be triggered by IL-13 administration and reduced by blockade of IL-13 in conjunction with less HAS1 expression, suggesting that this cytokine could contribute to HA levels in COVID-19. Consistent with this, IL-13 was increased in serum samples from a cohort of 178 COVID-19 patients [159] and in a second study of 138 COVID-19 patients [160]. However, in another study of 82 patients with severe COVID-19, IL-13 was decreased [159]. IL-13 was also not increased in a small study of human respiratory secretions from 22 patients [137]. Similarly, IL-13 was not found to be elevated in cadaveric lung tissues from a small cohort of 6 COVID-19 patients compared to controls with influenza [161].

Along with these conflicting data on IL-13 levels, the data on IL-13 contributions to COVID-19 pathophysiology are also unclear. Blockade of IL-4/IL-13 signaling with dupilumab led to significantly reduced ventilation and death in COVID-19 patients already on this drug. [162]. However, IL-13 was found to be protective against SARS-CoV-2 infection in vitro [163,164]. Further study and data from additional cohorts are needed to resolve the role of IL-13 and the value of targeting this cytokine therapeutically in COVID-19.

The hypoxic conditions in the COVID-19 lung can also trigger the hypoxic adenosinergic response [130] . This results in elevated expression of the adenosine A2B receptor [165], which has been shown to mediate HAS2 expression and subsequently lead to production of HA from macrophages and smooth muscle cells [74,166].

The cells that produce HA in acute COVID-19 infection are likewise unclear. HA is produced by many cell types in the lung, as reviewed elsewhere [59,167], including many of the same epithelial cells, endothelial cells, fibroblasts and macrophages that can be infected by SARS-CoV-2. However, we find that cell types that can be infected with SARS-CoV-2, including respiratory tract epithelial cells in air-liquid-interface (ALI) cultures and lung organoids, produce minimal amounts of HA upon viral culture (unpublished observations). This suggests that bystander cells, and not primary infected cells are likely to produce HA in response to COVID-19 infection. Ravindra N, et al. observed increased interferon-stimulated gene (ISG) induction in bystander and some infected cells in response to interferons released by specific infected cells [168]. While both bystander and infected cells experience heightened ISG activity, differentiated gene and cytokine expression found by the study between the two cell states support a unique role for bystander cells.

Together, based on this report we propose that SARS-CoV-2 infection triggers HA production by bystander cells as part of an anti-viral inflammatory response (Fig.1 ) and suggest that inhibition of HA could be a therapeutic target in severe COVID-19 ARDS. Additional work is needed to implement this idea and understand how HA shapes the inflammatory microenvironment. dummy citation Fig.2

Fig. 1.

A model for HA production in the COID-19 infected lung. A. HA production by airway cells in response to viral triggers. Inflammatory cytokines and viral PAMPs, including dsRNA, trigger HA production by respiratory epithelial cells. This HA production is mediated by HA synthases, most notably HAS2. The accumulation of HA fragments may perpetuate the inflammatory casecade associated with infection. B. While dying cells infected with SARS-CoV-2 make minimal HA, infection drives robust HA production by respiratory epithelial cells. This, together with DNA released from dead cells, results in thick, tenacious airway secretions in COVID-19 acute lung infections.

Fig. 2.

A model for how HA might contribute to PASC pathophysiology. Thick, HA-rich hyaline material fills lung alveolae in response to acute SARS-CoV-2 infection. In severe acute COVID-19 infection, this can lead to thick respiratoty secretions, edema, and impaired gas exchange typical of ARDS. This HA is catabolized over time into LMW-HA, which drives further inflammation. Most individuals clear this HA via hyaladherin-mediated cellular uptake. However, when hyaladherin levels or function are impaired, HA persists and promotes fibrosis and poor gas exchange (respiratory PACS).

HA in “long” COVID (PASC)

PASC is both common and devastating [169,170]. In one UK-based study of 1.1 million patients with PASC, the duration of symptoms was over 12 weeks for 65%, and over 20% experienced “persistent COVID-19″ disabling symptoms [171]. Chronic shortness of breath or dyspnea is a particularly common and debilitating symptom of PASC and is thought to reflect post-infectious sequalae in the lung [172]. A recent meta-analysis found that up to 80% of patients display one or more post-acute COVID-19 symptoms, with fatigue and shortness of breath (dyspnea) being among the most common [173]. Some groups have reported that over 18–25% of PASC patients have structural lung abnormalities and dyspnea [169,170,174].

The pathophysiology underlying these phenotypes is unclear. Moreover, it is unknown whether the same disease mechanisms are involved across the entire spectrum of PASC. This uncertainty, together with the increasing relevance of PASC in healthcare, highlight the need to develop biomarkers and therapeutic strategies to identify, prevent, and treat lung disease in PASC.

While several independent reports have documented increased serum and sputum HA levels in patients with acute COVID-19, there are very little data on HA levels in PASC. A recent report indicated that perivascular HA levels were increased in lung tissue samples of patients with COVID-19 and non-resolvable COVID-19 (NR-COVID-19) [175]. Interestingly, this was not associated with increased in HAS1–2 expression levels but rather reduced levels of CD44, potentially implicating impaired HA clearance in these effects. A study of post-COVID-19 liver fibrosis likewise reported that serum HA was elevated in individuals with this condition and that HA levels were associated with liver stiffness and elevated liver enzymes [176]. However, while reports of post-COVID-19 lung fibrosis are beginning to appear in the literature [177], much remains unknown about this devastating condition.

Despite the lack of published data, a role for HA in PASC would be consistent with the abundance of HA in acute COVID-19 and the well-established role of HA in other chronic inflammatory and fibrotic diseases [69,78,[178], [179], [180], [181], [182], [183], [184], [185], [186]], a role for HA in long-COVID is worthy of further study. We propose that robust HA production in the lungs and peripheral tissues contributes to the pathogenesis of severe acute COVID-19 and a subsequent inability to clear these HA fragments contributes to post-COVID-19 fibrotic syndromes and to the chronic inflammatory and metabolic derangements associated with long COVID-19.

Taken together, these data support that HA is abundant in COVID-19 PASC, and inhibition of HA could be a therapeutic target with the potential to significantly improve clinical outcomes for patients with severe COVID-19 PASC.

HA is a novel therapeutic target in COVID-19

Existing therapies to treat COVID-19 infection which target viral replication (e.g. Remdesivir) or systemic inflammation (e.g. Dexamethasone) are not indicated as treatments for post-COVID-19 fibrosis. Similarly, anti-fibrotic agents such as Nintedanib and Pirfenidone [115,187,188] have shown modest effects in COVID-19 induced fibrosis [115,[189], [190], [191]]. This highlights the need to develop strategies to identify, prevent, and treat lung fibrosis in COVID-19 patients. The data presented in this review suggest that it may be beneficial to target HA in COVID-19, as was originally suggested by Shi et al., early in the pandemic. Potential options include targeting the production of HA during infection, post-production HA degradation via hyaluronidases, and using HMW-HA as an anti-inflammatory therapy.

Dupilumab, an IL-13 antagonist and an already approved therapy, is another potential therapeutic option to block HA production [162]. In a recent clinical trial of 19 patients who received dupilumab and 21 controls, there were no differences in the proportion of patients on mechanical ventilation at either 28 or 60 days. There was a difference of 2 deaths in the dupilumab group (2/19; 10.5%) versus 5 deaths in the control group (5/21; 23.8%) at 60 days but this difference was not significant (Unadjusted HR: 0.40 (0.08–2.05)) until the data were adjusted for sex and mechanical ventilation as a time-varying predictor (Adjusted HR: 0.05 (0.004–.72)) [162]. Larger, better controlled studies are needed to explore dupilumab as a potential therapy in COVID-19.

One molecule known to inhibit HA synthesis is 4-methylumbelliferone (4-MU), also known as hymecromone [192], [193], [194], [195], [196]. 4-MU is a competitive substrate for UDP-glucuronyl transferases (UGTs), depleting one of the HA precursors, UDP-glucuronic acid [197]. 4-MU also indirectly reduces the expression of mRNA transcripts involved in HA synthesis [192,198]. The main metabolite of 4-MU, 4-methylumbelliferyl glucuronide (4-MUG), has also been shown to be bioactive [199]. In the lung, 4-MU reduces HA and ameliorates disease in mouse models of lung infection [85,200,201], lung metastases [202], pulmonary hypertension [71], and pulmonary fibrosis [203]. Work from our lab and others has shown that the immunosuppressive effect of 4-MU is mediated in large part through inhibition of antigen presentation and the induction of anti-inflammatory Foxp3+ Tregs [204], [205], [206].

Tantalizingly, 4-MU is already an approved drug in Europe, currently used to treat biliary spasm [207], [208], [209], [210]. It is given orally and is safe and well-tolerated [192,197,[208], [209], [210], [211], [212], [213], [214], [215], [216], [217]]. A Stage 1 clinical trial of 4-MU in healthy human volunteers was recently completed and demonstrated inhibition of sputum HA levels at drug dosages currently approved in Europe. This suggests that it may be possible to repurpose 4-MU to inhibit HA production and prevent fibrosis in COVID-19 and other conditions. [75].

In pursuit of this goal, a single-center, randomized, placebo-controlled, double-blind clinical trial is currently exploring this at Shanghai Zhongshan Hospital (NCT05386420). This follows on the heels of an earlier trial that was performed in China that yielded tantalizing results that were unfortunately difficult to interpret due to issues with study design [218].

Dupilumab, hyaluronidase and 4-MU have all been clinically approved for other indications and could be repurposed to treat COVID-19. HAdases such as PH20 have been used to enhance the absorption and dispersion of injected drugs, for subcutaneous fluid administration for hypodermoclysis, and for subcutaneous urography to improve absorption of radiopaque agents. PH20 could be repurposed to treat COVID-19. Thinning of the fluid to improve lung clearance is a common goal across a range of diseases with respiratory inflammation [219], [220], [221], [222]. Indeed, intranasal administration of exogenous HAdase was reported to sufficiently reduce the level of lung HA content, thereby restoring the lung function in the influenza mouse model [53]. In vitro treatment of acute COVID-19 respiratory secretions with HAdase decreased the flow resistance of thick samples [137].

A number of studies have shown encouraging results in treating inflammatory lung diseases with aerosolized HMW-HA [63,179,223], including in patients with bronchial asthma [224], cystic fibrosis [225,226], in children with recurrent upper respiratory infections [227], and in patients with COPD [228,229]. A feasibility and outcomes trial for the use of HMW-HA in patients with severe COVID-19 to prevent late-stage COVID-19-associated cytokine storm is currently underway (NCT04830020). Specifically, this trial aims to determine whether inhaled HMW-HA protects against progression of COVID-19-induced respiratory failure and promotes recovery from COVID-19 lung disease in hospitalized patients. Further investigation is warranted to identify whether HMW-HA treatment will improve COVID-19 severe outcomes, optimal administration methods, and dosing regimens that effectively ameliorate disease.

Conclusions and outstanding questions

The data reviewed here implicate HA in the pathogenesis of severe, acute COVID-19. In particular, HA appears to be linked to COVID-19 respiratory secretions and to ARDS. The available studies further suggest that serum HA could serve as a biomarker for severe acute COVID-19 disease. These associations require further validation in larger, independent cohorts.

It is not known whether HA levels are sustained or increased in PASC. This is an exciting avenue of investigation that requires additional research. If an association with HA is demonstrated in COVID-19, it may be possible to use sputum and/or serum levels to identify patients at risk of developing lung fibrosis.

Numerous other questions remain. It is unclear how HA accumulates and persists in COVID-19 ARDS. A more nuanced understanding of the cells and stimuli that drive HA production in COVID-19 will be essential to define mechanisms of HA accumulation.

The receptors and pathways involved in HA contributions to COVID-19 pathophysiology are unknown [230,231]. There are limited data on the role of CD44 in COVID-19. Donlan et al. [138], demonstrated that blocking of CD44 in mice infected with SARS-CoV-2 reduced clinical scores (an observational metric which includes data on eye rubs, weight loss, activity, and posture) but no survival benefit. Further complicating interpretation of these data, anti-CD44 antibodies are known to trigger neutropenia, platelet depletion, and impaired leukocyte migration [232] and could impact this model and SARS-CoV-2 susceptibility in non-specific ways. Studies with CD44-/- mice have not been performed. Further investigation is needed on this topic.

More generally, we need to learn more about how HA shapes selective pro- and anti-inflammatory microenvironments in the lungs to develop effective treatments that protect against the development of a pathological HA matrix while maintaining an anti-inflammatory or pro-resolving matrix.

More studies that determine how HA promotes pro-fibrotic responses in COVID-19 and the pathways implicated are needed. Nonetheless, HA is a promising potential target and biomarker in this terrible disease.

Data availability

No data was used for the research described in the article.