Immunopathology of galectin-3: an increasingly promising target in COVID-19

The pandemic brought on by the outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) has become a global health crisis, with over 22 million confirmed cases and 777,000 fatalities due to coronavirus disease 2019 (COVID-19) reported worldwide. The major cause of fatality in infected patients, now referred to as the “Cytokine Storm Syndrome” (CSS), is a direct result of aberrant immune activation following SARS-CoV2 infection and results in excess release of inflammatory cytokines, such as interleukin (IL)-1, tumor necrosis factor α (TNF-α), and IL-6, by macrophages, monocytes, and dendritic cells. Single cell analysis has also shown significantly elevated levels of galectin 3 (Gal-3) in macrophages, monocytes, and dendritic cells in patients with severe COVID-19 as compared to mild disease. Inhibition of Gal-3 reduces the release of IL-1, IL-6, and TNF-α from macrophages in vitro, and as such may hold promise in reducing the incidence of CSS. In addition, Gal-3 inhibition shows promise in reducing transforming growth factor ß (TGF-ß) mediated pulmonary fibrosis, likely to be a major consequence in survivors of severe COVID-19. Finally, a key domain in the spike protein of SARS-CoV2 has been shown to bind N-acetylneuraminic acid (Neu5Ac), a process that may be essential to cell entry by the virus. This Neu5Ac-binding domain shares striking morphological, sequence, and functional similarities with human Gal-3. Here we provide an updated review of the literature linking Gal-3 to COVID-19 pathogenesis. Dually targeting galectins and the Neu5Ac-binding domain of SARS-CoV2 shows tentative promise in several stages of the disease: preventing viral entry, modulating the host immune response, and reducing the post-infectious incidence of pulmonary fibrosis.


Introduction
Galectin 3 (Gal-3) is a carbohydrate-binding protein that exhibits pleiotropic effects throughout the body, including the modulation of apoptosis, cell migration and adhesion, angiogenesis, tumorigenesis, and post-injury remodeling ( The continued lack of an effective standard of care for treating patients with COVID-19 has brought on an urgent need to identify effective therapies. In a prior review article, we had discussed promising indications for Gal-3 targeted therapy in the treatment of COVID-19, with the goal of inspiring further research on the topic (Caniglia et al., 2020). In recent months, however, a substantial amount of new evidence has emerged that further links Gal-3 to severe COVID-19 infection. As such, the authors see a need to achieve two aims in this review: highlighting novel discoveries to expand upon previously discussed treatment indications, and to detail a further potential role for anti-galectin therapy in reducing post-infectious pulmonary fibrosis. This article may be particularly useful for immunologists studying COVID-19, as well as any researchers with a structural or functional focus on galectins.

SARS-CoV2: host cell attachment and entry
A critical step prior to viral infection is the entry of the virus into host cells, a process that in severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) is mediated by the S1 subunit of the spike protein (Blaas, 2016; Zhai et al., 2020). Within coronaviridae, it is commonplace to refer to the S1 protein as consisting of two distinct regions: the C-terminal domain (CTD) and N-terminal domain (NTD) (Li, 2016). In most cases, the CTD binds peptide receptors and the NTD binds sugar receptors (Li, 2016). The main entry mechanism of SARS-CoV2 has been shown to be via the CTD binding to angiotensin converting enzyme receptor 2 (ACE2) receptors (Wang et al., 2020b). Until recently, the role of the NTD has been largely overlooked. A study from Baker et al. has shown evidence that SARS-CoV2 also binds N-acetylneuraminic acid (Neu5Ac), with this interaction being mediated by the NTD of the S1 subunit (Baker et al., 2020). This is the first in vitro evidence of this occurring, although several prior bioinformatics and modeling studies have hypothesized that a Neu5Ac binding site exists, with one suggesting its affinity for Neu5Ac (0.88) is only slightly lower than that of influenza hemagglutinin (0.94) (Alban et al., 2020; Behloul et al., 2020; Fantini et al., 2020; Kim, 2020; Milanetti et al., 2020; Robson, 2020). Binding of sialic acids by the NTD is the main entry mechanism in several other coronaviruses known to infect humans, most notably members of the bovine coronavirus family (Li, 2015). Additionally, the closely related middle eastern respiratory syndrome coronavirus (MERS-CoV) has been shown to exhibit a dual attachment model similar to SARS-CoV2, where the CTD binds a peptide receptor and the NTD binds sialic acids (Li et al., 2017). Depletion of sialic acids with neuraminidase inhibitors prevented MERS-CoV infection of Calu-3 human airway cells, indicating that NTD-targeted therapies may be effective in preventing cell entry by coronaviruses possessing this function (Li et al., 2017). Additionally, a neutralizing antibody against the SARS-CoV2 S1-NTD has been shown to completely inhibit cell entry by the virus (Chi et al., 2020). This indicates the NTD region is essential for viral entry and a promising therapeutic target (Chi et al., 2020). The dual mechanism by which SARS-CoV2 may enter host cells is seen in Figure 1.
The binding of Neu5Ac may also explain the greater infectivity of SARS-CoV2 as compared to SARS-CoV (Alban et al., 2020). While the CTD of SARS-CoV2 has been shown to exhibit higher affinity for ACE2 receptors than that of SARS-CoV, this is likely insufficient to fully explain the marked disparity in transmissibility (Tai et al., 2020). The NTD of

Amendments from Version 1
The updated article provides additional sources where necessary to further highlight the roles of galectin-3 in the innate immune system. Additionally, recent evidence is included that validates the S1-NTD of SARS-CoV2 as a promising therapeutic target. We hope this updated text to be a more detailed review with enhanced readability compared to the prior copy. . The structures of Gal-3 and the S1-NTD of betacoronaviridae are so similar, in fact, that it is hypothesized that coronaviruses incorporated a host galectin gene into their genome (and then the NTD) at some point in their evolution (Caniglia et al., 2020;Li, 2015). Structural analysis comparing the SARS-CoV2 NTD to Gal-3 resulted in a Z-score of 6 (p < 0.00001), indicating a high degree of similarity between the structures (Behloul et al., 2020). In fact, human Gal-3 was shown to be equally similar to SARS-CoV2 NTD as the NTD of NL63-CoV and infectious bronchitis coronavirus, accounting for both sequence and structure (Behloul et al., 2020). Given the high degree of structural and promising sequence similarity (12%) of the NTD with Gal-3, it may be possible that existing Gal-3 inhibitors possess dual-binding capabilities (Behloul et al., 2020). Such a mechanism shows promise in reducing viral entry to host cells (Milanetti et al., 2020).

Gal-3 in severe infection: promoting immunologic sequelae of COVID-19
The major cause of death in patients infected with SARS-CoV and MERS-CoV infection was found to be the "Cytokine Storm Syndrome" (CSS), and this is likely to be the case in COVID-19 as well (Channappanavar & Perlman, 2017; Zhang et al., 2020). CSS develops due to hyper-activation of macrophages, monocytes, and dendritic cells, which are stimulated to release a variety of inflammatory mediators including IL-1, IL-6, and TNF-α (Zhang et al., 2020). This in turn leads to systemic organ dysfunction that may result in death (England et al., 2020). Notably, a study of nearly 4,000 patients has found the levels of IL-1, IL-6, and TNF-α to be significantly elevated in the sera of patients suffering from severe COVID-19 as compared to those with mild disease (Wang et al., 2020a). Similar findings were reported in a cohort of over 1,5000 patients, where serum IL-6 and TNF-α were found to be independent predictors of disease severity and mortality in COVID-19 (Del Valle et al., 2020). This data speaks to the urgency of identifying therapeutics to reduce the incidence of CSS ( There is a plethora of evidence that makes Gal-3 a promising target to achieve this aim. First, the most concerning sequelae of CSS is evolution to acute respiratory distress syndrome (ARDS), a condition which often leads to respiratory failure despite proactive measures such as mechanical ventilation and intubation (  In SARS-CoV infection, particularly in patients who suffered from ARDS, marked pulmonary fibrosis was found in a cohort of patients following prolonged infection (Ye et al., 2007). Though long term outcomes remain to be seen, lung tissue in the acute phase of COVID-19 shows similar changes (Xu et al., 2020a). Following a 24 hour incubation of SARS-CoV2, human airway cells showed upregulation of ACE2, vascular endothelial growth factor (VEGF), connective tissue growth factor (CTGF), fibronectin (FN), and transforming growth factor ß (TGF-ß), a molecular signature highly similar to that of patients with diagnosed pulmonary fibrosis (Xu et al., 2020a). It is believed that a large number of COVID-19 patients will go on to develop pulmonary fibrosis, and that these changes are mediated by a number of cytokines including TGF-ß, IL-1, IL-6, and TNF-α (Delpino & Quarleri, 2020).
The role of Gal-3 as a mediator of lung fibrosis has long been studied since the discovery that its levels are elevated in alveolar macrophages following lung injury (Kasper & Hughes, 1996;Nishi et al., 2007). Higher levels of Gal-3 have now been extensively associated with the development of restrictive lung diseases (Ho et al., 2016). Following cellular stress, the secretion of Gal-3 by macrophages upregulates TGF-ß receptors on fibroblasts and myofibroblasts (Henderson et al.,  2008). This in turn activates these cells, initiating the formation of granulation tissue (via collagen deposition) that is eventually remodeled to a fibrous scar (Henderson et al., 2008;  Mackinnon et al., 2012). This Gal-3 mediated pathway is widespread throughout the body and fundamental to the development of fibrotic change in the liver, kidneys, and heart as well (Hara et al., 2020). Gal-3 mediated fibrosis often has deleterious effects; for example, pathologic scar formation is the likely explanation for serum Gal-3's utility as an independent predictor of mortality and heart failure post-myocardial infarction (Asleh et al., 2019). The mechanism by which Gal-3 may contribute to post-infectious pulmonary fibrosis in COVID-19 patients can be seen in The drug TD139 has recently begun phase II trials for the treatment of COVID-19, the first clinical trial of a galectin inhibitor in COVID-19 to date (University of Edinburgh DEFINE trial, 2020).

Conclusions and future directions
In summary, Gal-3 is a lectin that exhibits a pleiotropic role in mediating the acute and chronic consequences of infection and inflammation. Multiple studies have shown Gal-3 to be highly upregulated in patients suffering from severe COVID-19 (De Biasi et  A key domain in the spike protein exhibits a high degree of morphological and sequence similarity to human Gal-3 (Behloul et al., 2020). This NTD has been shown to bind Neu5Ac  (Xu et al., 2020a). Gal-3 secreted by macrophages during injury promotes the upregulation of TGF-ß receptors, leading to fibroblast activation and collagen deposition (Delpino & Quarleri, 2020). Gal-3 inhibition has been shown to reduce adenovirus-induced lung fibrosis, and an inhibitor is currently in Phase IIb clinical trials for IPF treatment (Mackinnon et al., 2012; Saito et al., 2019). The indications for targeting Gal-3 in the treatment of COVID-19 are widespread. Processes directly mediated or affected by Gal-3 have been shown to be deleterious in several stages of the disease process. As such, Gal-3 represents a highly promising target for COVID-19 treatment that should urgently be investigated.

Literature search methodology
Eligibility criteria This review consists of original studies that provided information about SARS-CoV2, Gal-3, or Gal-3 inhibitors. Compiled results from both in vivo, in vitro, and clinical studies were used for analysis. Studies with only an abstract or no full-text available were excluded from the review.

Search methodology
To retrieve primary literature, electronic searches were performed on PubMed and Google Scholar. A list of search terms can be seen in Table 1.

Risk of bias
To minimize the risk of error, all authors involved assessed the cited studies for quality. To discuss important claims in the article, including that SARS-CoV2 binds sialic acids with the S1-NTD, that Gal-3 is upregulated in human immune cells, and Gal-3 inhibitors' ability to reduce fibrosis, multiple sources were included. Additionally, the use of open-ended searches ensured that an accurate profile of results was obtained on the topics discussed.

Data availability
No data are associated with this article. The quality of the figures in the article is good and clearly explains the concept discussed.

○
The authors are requested to address the following specific comments mentioned below: 1) The authors have mentioned Gal-3 inhibitors could target Neu5Ac-binding domain, thereby reducing viral entry to host cells. Since ACE2 receptors serve as the main entry mechanism ofSARS-CoV2, To what extent Gal-3 inhibitors alone can provide mitigatory effects?. Also, could ACE2 inhibitors be used as adjuvants along with Gal-3 inhibitors?. What are the possibilities?. These need to be explained.
2) The authors are advised to cite the article Garcia-Revilla, J. et al., 2020in their manuscript. The article discusses similar concepts mentioned in the current review by the authors. Therefore, citing it could provide more support to the authors claim.
a very effective approach in reducing the spread, cytokine storm, and postinfection pulmonary fibrosis.
The quality of the figures in the article is good and clearly explains the concept discussed.

○
The authors are requested to address the following specific comments mentioned below: 1) The authors have mentioned Gal-3 inhibitors could target Neu5Ac-binding domain, thereby reducing viral entry to host cells. Since ACE2 receptors serve as the main entry mechanism ofSARS-CoV2, To what extent Gal-3 inhibitors alone can provide mitigatory effects?. Also, could ACE2 inhibitors be used as adjuvants along with Gal-3 inhibitors?. What are the possibilities?. These need to be explained The reviewer brings up a great point. To address this question, we have added a citation that shows the detection of a neutralizing antibody against the NTD of SARS-CoV2 S1 protein. This antibody is effective at completely neutralizing SARS-CoV2 cell entry even in the absence of other antibodies that bind the receptor binding domain or ACE2 receptors. With this in mind, it follows that a drug such as a galectin inhibitor targeting the NTD may be effective as a standalone therapy. The reviewer brings up a good point regarding combination therapy, in that combination of NTD / CTD targeted therapies may be a better long term treatment strategy given the mutagenicity of the virus' spike protein. Thank you for providing this citation. We are certainly glad to see other laboratories participating in this field of research. However, given that this article is also a review article and not primary research, we do not see a role for it to be cited in our current review article, which aims to summarize primary research findings regarding galectin-3 and COVID-19.

○ ○
Overall the article is well-written and recommended for publication once the minor corrections have been addressed.  The authors note on p. 5 "A study found that tests for SARS-CoV2 RNA in the serum of infected individuals did not become negative until a median of 24 days post-symptom onset, with some individuals remaining positive even greater than a month from the beginning of symptoms (Gombar et al., 2020). This indicates that for some, COVID-19 infection may run a particularly long course. Findings such as this have led to the question of whether or not anti fibrotic therapy would be beneficial for such patients (George et al., 2020)." The persistence of SARS-CoV-2 RNA should not be equated with replication competent virus; there is significant literature to suggest residual nucleic acid that does not represent infectious virus. This should not be equated with long term infection or increased risk of fibrotic disease and these patients should not be treated with anti-fibrotic therapy for that reason. Severe lung injury from ARDS would be much more plausible an explanation for fibrotic lung disease than persistent SARS-CoV2 RNA.
○ patients suffering from severe COVID-19 as compared to those with mild disease (Wang et al., 2020a)." The authors should additionally cite Del Valle et al. Nature Medicine 2020 that noted similar findings in 1500 patients. 2 Thank you for providing the additional citation. We will certainly add these findings to the manuscript. ○ Figure 1 does not add richly to this work and perhaps could be a panel combined with Figure 2.
Thank you for this comment. We believe Figure 1 to be essential as it details the likely role of the galectin-like S1-NTD in COVID-19 infection. To further highlight the importance of this figure and the NTD, we have added a recent publication (Chi et. al, 2020) which shows a neutralizing antibody against the NTD inhibits viral entry. It then follows that if a drug such as a galectin inhibitor is able to also bind this region of the NTD, it may also inhibit cell entry by SARS-CoV2. The reviewer makes an excellent point here. We have added multiple sources to this section that we believe better characterize the chronic inflammatory signature some COVID-19 patients report months after the initial infection. We have also included an additional citation of an excellent commentary on the concern for development of post-infectious IPF in COVID-19 patients.