Non-anticoagulant heparin derivatives for COVID-19 treatment

Abstract

The ongoing pandemic of COVID-19, caused by the infection of SARS-CoV-2, has generated significant harm to the world economy and taken numerous lives. This syndrome is characterized by an acute inflammatory response, mainly in the lungs and kidneys. Accumulated evidence suggests that exogenous heparin might contribute to the alleviation of COVID-19 severity through anticoagulant and various non-anticoagulant mechanisms, including heparanase inhibition, chemokine and cytokine neutralization, leukocyte trafficking interference, viral cellular-entry obstruction, and extracellular cytotoxic histone neutralization. However, the side effects of heparin and potential drawbacks of administering heparin therapy need to be considered. Here, the current heparin therapy drawbacks were covered in great detail: structure-activity relationship (SAR) mystery, potential contamination, and anticoagulant activity. Considering these unfavorable effects, specific non-anticoagulant heparin derivatives with antiviral activity could be promising candidates to treat COVID-19. Furthermore, a structurally diverse library of non-anticoagulant heparin derivatives, constructed by chemical modification and enzymatic depolymerization, would contribute to a deeper understanding of SAR mystery. In short, targeting non-anticoagulant mechanisms may produce better therapeutic effects, overcoming the side effects in patients suffering from COVID-19 and other inflammatory disorders.

1. Introduction

COVID-19 outbreak, caused by SARS-CoV-2, has been affecting the public health and the global economy since 2019 [1], [2], [3]. Due to the high mortality and spread rate, COVID-19 has become an unprecedented humanitarian crisis in modern history with >600 million infections and 6.47 million deaths reported at the time of writing. Investigators are trying to unleash effective treatment strategies against COVID-19, halting its spread and boosting public health. Vaccinations are undermined by variants and do not give immediate and 100 % protection [4], [5], [6], [7], [8], [9]. Undoubtedly, developing a novel therapeutic is a complicated and time-consuming process. So, repurposing the accustomed and approved therapies with demonstrated safety profiles is an exciting strategy.

Heparin has been used for over 100 years as a well-tolerated anticoagulant drug, and biological clinical research began in 1935, it was effective and then widely used. Heparan sulfate (HS) is a co-receptor used by many viruses to invade cells, which allows for a localized increase in viral particle concentration to boost the infection rate [10]. A SARS-CoV strain, obtained from a severely infected patient, was 50 % reduced by 100 μg/mL exogenous heparin treatment in vitro Vero cells test [11]. Also, there is a growing body of literature that explains how heparin works against SARS-CoV-2. However, almost all heparin used in clinical practice is from animal sources at present. Although the natural extraction process is strictly controlled to avoid some side effects, there are still certain problems in clinical application, e.g., potential contamination, and anticoagulant activity. Furthermore, because of the heterogeneity, heparin is quantitated by anticoagulant potency, and structure-activity relationship still is not clear. Hence, these adverse effects should be considered when heparin is a candidate drug for anti-COVID-19. Not surprisingly, non-anticoagulant heparin derivatives, providing direct antiviral activity without anticoagulant side effects, have been attracting considerable interest. Furthermore, chemical or chemoenzymatic approach offers a feasible method to construct a structurally diverse library of non-anticoagulant heparin derivatives, fostering the discovery of functions and structure-activity relationship, as well as the development and application of heparin-based new drugs for COVID-19 treatment.

This review aims to critically interpret the literature concerning the demerits of heparin therapy for COVID-19, including structure-activity relationship (SAR) mystery, potential contamination, and anticoagulant activity. Besides, non-anticoagulant heparin derives were discussed, which may avoid the problems of heparin therapy with equivalent therapeutic effects for COVID-19. Moreover, the structurally diverse library of non-anticoagulant heparin derivatives was described, and constructed by chemical modification and enzymatic depolymerization, which may contribute to a deeper understanding of SAR mystery, accelerating the development of precise structures for specific treatments. To date, there is no review about non-anticoagulant heparin derivatives for COVID-19 treatment, which are promising candidates for the said purpose.

2. SARS-CoV-2 and heparin

2.1. Blocking the invasion of SARS-CoV-2 with heparin

It would be more beneficial in preventing hospitalization and long-term sequelae of infection by inhibiting viral replication in the initial stages of infection as opposed to treating the symptoms brought on by immune activation and inflammation. For this reason, the cell entry of SARS-CoV-2 has emerged as an appealing and repurposing target for COVID-19, and the development of direct-acting antiviral compounds has been a research priority. The research showed that binding of SARS-CoV-2 spike protein (S glycoprotein) onto cell surface HSPG and angiotensin-converting enzyme 2 (ACE2) is generally the first step required for the initiation of infection, which involves a cascade of interaction [12]. Accordingly, exogenous heparin, or HS, as a competitive inhibitor, competes with the HS chain of HSPG on the cell surface to bind with viral S protein for virus entry inhibition (Fig. 1 ) [13], [14].

Fig. 1.

Exogenous heparin/HS inhibits COVID-19 infection by competitively blocking the viral entry into the host cell.

2.2. Preclinical evidence for heparin as a SARS-CoV-2 antiviral drug

In preclinical studies of viral infections, heparin possesses a direct antiviral activity upon SARS-CoV-2. The early reports indicated that heparin could bind to the proteolytic cleavage site of the S1 and S2 protein based on combined surface plasmon resonance (SPR) and computational study [15], [16]. Zhang et al. found that heparin, HS, and non-anticoagulant low molecular weight heparin at micromolar doses could effectively compete with spike glycoprotein of COVID-19 to block the viral invasion pathway [17]. Mycroft-West et al. demonstrated that heparin (200 μg/mL) inhibited SARS-CoV-2 invasion of Vero cells up to 80 % [18]. Linhardt et al. illustrated that unfractionated heparin (UFH), enoxaparin, 6-O-desulfated UFH, and 6-O-desulfated enoxaparin efficiently neutralized pLV-S particles which pseudotyped SARS-CoV-2 SGP [19]. Boons et al. identified a common octasaccharide sequence (K_D = 38 nM) as the most potent in preventing the S protein-heparin interaction [20]. Overall, based on accumulated evidence, heparin and non-anticoagulant heparin derivates may be promising drug candidates for COVID-19 treatment.

2.3. Structure and function of heparin

Heparin or HS is a linear sulfated hetero-polysaccharides, consisting of alternating α1-4-linked D-glucosamines (GlcN) and 1-4-linked uronic acids (UA), →GlcN→UA→. The UA residue is either α-L-iduronic acid (IdoA) or β-D-glucuronic acid (GlcA), and these residues could be sulfated at 2-O-position [21]. In addition, one or more modifications on the GlcN residue, include N-sulfation, N-acetylation, 6-O-sulfation, and 3-O-sulfation (Fig. 2 ) [22]. Heparin, a unique form of HS, has a higher level of sulfation and IdoA content than HS due to extensive sulfate modification during its biosynthesis, and displays prominent anticoagulant activity making it a commonly used anticoagulant drug [23], [24]. Furthermore, heparin, a unique class of pharmaceuticals, has effective antidotes available to make its use safer. Heparin or HS, exhibiting a wide range of biological functions, e.g. inflammatory responses, blood coagulation, tumor metastasis, lipid metabolism and so on, is an attractive synthetic target [25]. The clinical implementation of heparin or HS as an anti-Cov-19 drug demands management of various potential problems, including the anticoagulant effect.

Fig. 2.

The major repeating disaccharide unit in heparin (a) and HS (b).

3. The current drawbacks of heparin therapy

3.1. SAR mystery

The chain length and structure of heparin extracted through porcine intestinal mucosa, which is permitted in most countries, are highly heterogeneous. [25]. The structure of the obtained heparin is influenced by the method of extraction and purification, and the type of animal species and organs. Accordingly, due to the uncontrolled degradation site and degree, the compositional structures of different batches of products are not uniform, resulting in the variable activities and inconsistent structure-activity relationship [26] (Table 1 ). Accordingly, there are extensive descriptions about discrepancies in the antiviral potencies of the heparin and heparin analogues in the literature. Tandon et al. reported IC₅₀ values of 5.99, 1.08 × 10³, 1.77, and 5.86 × 10³ μg/L for UFH, enoxaparin, 6-O-desulfated UFH, and 6-O-desulfated enoxaparin, respectively [19]; Tree et al. reported antiviral IC₅₀ values of 41, 7.8 × 10³ μg/mL for UFH and enoxaparin, respectively [27]; Gasbarri et al., found that UFH and enoxaparin could not be effective up to 1 × 10⁶ μg/L [28]. In terms of the absolute values of the dissociation constants, the values reported by the Linhardt, Boons, and Tan research teams vary significantly. They suggested that the discrepancy between the K_D values might be due to the method of analysis and/or the experimental materials used in the studies [29]. In a word, it is now well established from a variety of studies that different descriptions exist concerning the antiviral properties of heparin and heparin analogues. As a universal antiviral drug, homogeneous heparin would be desirable.

Table 1.

Antiviral activity of heparin mimicking compounds against SARS-CoV-2.

Compound	IC50a (μg/L)	KDb (M)	Ref.
UFH	5.99		[19]
Enoxaparin	1.08 × 103
6-O-desulfated UFH	1.77
6-O-desulfated enoxaparin	5.86 × 103
UFH	4.1 × 104		[27]
Enoxaparin	7.8 × 106
UFH	>1 × 106		[28]
Enoxaparin	>1 × 106
Heparin		7.3 × 10−11	[19]
Heparin		6.4 × 10−8	[20]
Heparin		1.6 × 10−5	[29]
Fondaparinux		2.8 × 10−5

^a50 % inhibitory concentration.

^bEquilibrium dissociation constant (K_D = Koff / Kon) of SARS-CoV-2-S-trimer.

3.2. Potential contamination

Heparin, isolated from dog liver, was discovered by Jay McLean in 1916 [30], then was approved by the FDA in 1939 and used in Europe and the United States. Until the 1990s, bovine-derived heparin gradually disappeared from the markets due to the outbreak of BSE in Europe [31]. In 2013, FDA issued the “Industrial Guidelines for Quality Control of Crude Heparin Products”, which stipulated that crude heparin is derived from pigs and cannot be mixed with heparin derived from ruminants [32]. However, potential viruses or impurity contaminations still exist for obtaining heparin derived from animals, leading to concerns about the clinical application of heparin [33]. In addition, artificial pollution cannot be ignored. Unfortunately, serious adverse events occurred between 2007 and 2008, as a result of heparin supplies being contaminated with oversulfated polysaccharides [34], [35], [36]. More stringent testing requirements, such as the proton nuclear magnetic resonance spectroscopy for the identification of compounds and the implementation of a new potency assay as additional tests for impurities, were introduced by the U.S. Food and Drug Administration and other national authorities to address this problem [37], [38]. Although the issues have been partially addressed by the stringent testing, some contamination cannot be detected by current detection methods because some low-level contamination was below the sensitivity of the assay [39]. Notably, the variability of natural heparin isolated from animals may be affected by environmental factors, animal species, and organ tissue, leading to commercial heparin being complex and heterogeneous mixtures. Accordingly, finding alternatives to heparin derived from animals is clearly needed.

3.3. Non-animal source heparin with homogeneous structure

Synthetic heparin can eliminate the adverse effects as noted above, caused by inherently heterogeneous heparin derived from natural sources, and the positive effects of UFH could be further enhanced in HS compounds with optimized oligosaccharide structure, leading to an increase in anti-inflammatory and antiviral properties over animal-derived heparin. Chemical method is still the major way of preparing homogeneous oligosaccharides. Target-oriented, modular, combinatorial, one-pot, and solid-phase syntheses, as well as other chemical synthetic techniques, have all been developed to generate HS/heparin oligosaccharides that range from di- to octasaccharides of various sequences and sulfation patterns. These long and elaborate chemical synthesis processes need for extremely specialized techniques, adding cost and decreasing overall yields [40], [41]. Furthermore, chemoenzymatic syntheses of various HS/heparin derivatives have been realized by the characterization of HS biosynthetic enzymes. It is an alternative technique to catalyze uridine diphosphate N-acetylglucosamine (UDP-GlcNAc) and uridine diphosphate glucuronic acid (UDP-GlcA) using the glycosyltransferase PmHS2 (Pasteurella multocida heparosan synthase 2) to synthesize sugar chains, followed by sugar chain modification (N-deacetylation, N-sulfation, O-sulfation, etc.) to construct a well-defined heparan sulfate [42], [43], [44]. Chemoenzymatic synthesis, which integrates the flexibility of chemical synthesis with the high efficiency of enzymatic catalysis, is still a step-by-step synthesis, accompanied by relatively complicated processes of enzyme separation and purification. Thus, the synthesis is usually not amenable to large-scale commercial heparin preparation. More, heparosan, an E. coli K5 strain fermentation product that is composed of a [→4) β-D-glucuronic acid (GlcA) (1→4) N-acetyl-α-D-glucosamine (GlcNAc) (1→]_n repeating disaccharide unit, serves as a starting material, after which the backbone is modified by enzymatic or chemical reactions [43], [45], [46], [47]. The modification involves a series of cascade reactions, mimicking in vivo heparin biosynthesis, including the conversion of N-acetyl heparosan into an N-acetyl, N-sulfo heparosan, C5-epimerization, 2-O-sulfation, 6-O-sulfation, and 3-O-sulfation (Fig. 3 ). The approach is reasonably simple to scale up and can avoid duplicate stages, but complete control over polymer size (chain length), composition, and sequence must be given up.

Fig. 3.

(a) Chemical synthesis of structurally defined oligosaccharides, where R and R' are protecting groups, and LG is leaving group; (b) chemoenzymatic synthesis assembly, and (c) semisynthetic production of heparin and heparan sulfate polysaccharides starting from heparosan.

3.4. Side-effect: excessive anticoagulant activity

The sulfated nature confers heparin with high negative charge density, enabling it to engage strongly and selectively with an immense number of proteins [48]. It interacts most classic with the serine protease inhibitor antithrombin-III (AT3), which presents anticoagulant activity. A pentasaccharide sequence motif determines the interaction of heparin-AT binding, →GlcNS6S→GlcA→GlcNS3S6S→IdoA2S→GlcNS6S→ [49]. In addition, heparin's global anticoagulant activity is dependent on the formation of the ternary heparin-AT-FIIa complex, which is facilitated by a repeating trisulfated disaccharide sequence, →IdoA2S→GlcNS6S→, corresponding to heparin's thrombin or factor IIa, FIIa binding site (Table 2 ) [43]. Interestingly, heparin, providing a continuous anticoagulant effect with a recycling effect, can effortlessly separate from the heparin/AT-III complex and bind to extra AT-III. However, the properties, which may make heparin have a pivotal role in the management of COVID-19, are often hindered by the presence of anticoagulant oligosaccharides, which leads to a significant risk of bleeding [50].

Table 2.

Anticoagulation structure in heparin, and reported non-anticoagulant heparin derivatives.

Anticoagulation structure in heparin	Pentasaccharidea
	Repeating trisulfated disaccharideb
Compound	Structure	Anticoagulant activityc	Function	Ref.
Non-anticoagulant version of heparin (NACH)		Anti-Xa potency: 6 U/mg	Antibody immune responses up-regulation	[56]
2-O,3-O-desulfated heparin (ODSH)		Heparin  Anti-Xa: 165–190 U/mg  Anti-IIa: 165–190 U/mg ODSH  Anti-Xa: 1.9 U/mg  Anti-IIa: 1.2 U/mg	Anti-inflammatory	[57], [58]
N-desulfated/acetylated heparin (NAH)		Heparin  Anti-Xa: 191.8 ± 5.7  Anti-IIa: 183.0 ± 8.9 NAH  Anti-Xa: 20.3 ± 1.8  Anti-IIa: 9.4 ± 0.8	Anti-apoptotic activities	[59]
6-O desulfated heparin (6-OdeSH)		Heparin  Anti-Xa: 191.8 ± 5.7  Anti-IIa: 183.0 ± 8.9 6-OdeSH  Anti-Xa: 20.3 ± 0.9  Anti-IIa: 33.1 ± 1.8	Anti-apoptotic activities	[59]
2,6-de-O-sulfated heparin (2,6-OdeSH)		Heparin  Anti-Xa (IC50): 0.93 μg/mL  Anti-IIa (IC50): 1.03 μg/mL 2,6-OdeSH  Anti-Xa (IC50): NI  Anti-IIa (IC50): NI	Anti-metastasis/cancer	[60]
Ultralow molecular weight glycol-split heparin (Gs-hepULMWH)	Mn: 2.828 kDa	UFH  Anti-Xa: 163 U/mg  Anti-IIa: 163 U/mg Gs-hepULMWH  Anti-Xa: 0 U/mg  Anti-IIa: 0 U/mg	Hearanase inhibition	[61]
Non-anticoagulant heparin-carrying polystyrene (NAC-HCPS)		Coagulation time  APTT-Control: 30 s  Heparin: 800 s  NAC-HCPS: 160 s	Angiogenesis and tumor invasion inhibition	[62]
Roneparstat	Mw: 15– 25 kDa	–	Anti-cancer and heparanase inhibition	[63]
Non-anticoagulant oxidized ultra-LMWH (NA-LMWH)		Control  Anti-Xa: 0.02 U/mL  Anti-IIa: 0.02 U/mL Enoxaparin (10 mg/kg, SC)  Anti-Xa: 1.2 U/mL  Anti-IIa: 0.6 U/mL NA-LMWH (10 mg/kg, SC)  Anti-Xa: 0.02 U/mL  Anti-IIa: 0.02 U/mL	Anti-metastatic	[64]

NI, no inhibition of coagulation up to 100 μg/mL compounds.

^aThe antithrombin III-binding pentasaccharide.

^bThe thrombin or factor IIa, FIIa binding repeating trisulfated disaccharide.

^cThe anticoagulation tests of relevant studies were different, and we have tried our best to keep the data comparable.

4. Non-anticoagulant heparin derivatives

Given that heparin is a treatment for COVID-19, there is merit in conducting clinical studies that assess therapeutic doses of heparin in patients with COVID-19 in terms of the risk of bleeding. Heparin is thought to exert many of its non-anticoagulant actions through binding and modulating proteins, including cytokines, growth factors, adhesion, cytotoxic peptides, and tissue-degrading enzymes, responding to the treatment of clinical events associated with COVID-19, including neutralization of inflammatory chemokines, and cytokines; neutralization of extracellular cytotoxic histones and by interfering with leukocyte trafficking [51]. Since heparin plays versatile biological roles in the life process, the related mechanism is unclear for COVID-19 patients. Therefore, its therapeutic potential must be explored by clinical studies that focus on identifying and obtaining non-anticoagulant heparin derivatives and further elucidating the structure and mechanisms of these non-anticoagulant heparin derivatives.

4.1. The potential antiviral mechanisms of non-anticoagulant heparin derivatives

A considerable literature suggests the potential non-anticoagulant mechanisms underlying the treatment of COVID-19 patients with non-anticoagulant heparin derivatives (Fig. 4 ) [52], [53], [54], [55], which includes: 1. Heparanase inhibition (HPSE). The loss of endothelial barrier function in the case of COVID-19, which leads to pulmonary oedema and proteinuria, can be partially attributed to increased HPSE activity-degrading the endothelial glycocalyx. Notably, non-anticoagulant heparin derivatives, proved to be potent HPSE inhibitors, may benefit COVID-19 patients by preventing glycocalyx dysfunction. 2. Chemokine and cytokine neutralization. COVID-19 is related to the generation of elevated levels of pro-inflammatory cytokines. Non-anticoagulant heparin derivatives are able to bind to the majority of chemokines and cytokines to neutralize. 3. Leukocyte trafficking interference. Non-anticoagulant heparin derivatives could attenuate the process of leukocyte adhesion and migration, having a pivotal role in the inflammatory response of COVID-19. 4. Viral cellular-entry obstruction. As outlined, heparin and heparin derivatives could inhibit the binding of SARS-CoV-2 to cells. The relevant mechanism is not directly related to the anticoagulant mechanism, indicating that non-anticoagulant heparin derivatives might have such a function. 5. Extracellular cytotoxic histone neutralization. It has been shown that negative-charged heparin neutralizes the cytotoxic effect of positive-charged histones, potentially reducing organ damage in COVID-19 patients. Hence, non-anticoagulant heparin derivatives may serve the same purpose, circumventing the drawbacks of the excessive anticoagulant activity of heparin.

Fig. 4.

The potential mechanisms of non-anticoagulant heparin derivatives for the treatment of COVID-19 patients.

4.2. Reported non-anticoagulant heparin derivatives

Non-anticoagulant heparin derivatives have attracted the attention of researchers, and recent investigations demonstrated their functions without any significant impact on coagulation (Table 2) [56], [57], [58], [59], [60], [61], [62], [63], [64]. Various non-anticoagulating heparin derivatives, including non-anticoagulant versions of heparin (NACH), 2-O,3-O-desulfated heparin (ODSH), N-desulfated/acetylated heparin (NAH), 6-O-desulfated heparin (6-OdeSH), 2,6-de-O-sulfated heparin (2,6-OdeSH), ultralow molecular weight glycol-split heparin (Gs-hepULMWH), non-anticoagulant heparin-carrying polystyrene (NAC-HCPS), Roneparstat, and non-anticoagulant oxidized ultra-LMWH (NA-LMWH), have been reported which differ from heparin regarding functional domains. For instance, 3-O-sulfation, although it is rare, is essential to form a specific pentasaccharide domain for heparin, being responsible for binding to anti-thrombin with high affinity that is crucial for anti-coagulant activity [65]. The report showed that removing N-sulfoglucosamine (GlcNS3S) C-3 sulfate from pentasaccharide resulted in a 10⁵-fold reduction in binding affinity, and almost maintained its affinity for COVID-19 [19]. Although heparin is known to benefit in COVID-19, the balance between its benefits and risks should be taken into account. Non-anticoagulant heparin derivatives can also be a choice to explore in COVID-19 treatment, circumventing the risk of the bleeding complications.

5. Outlook and conclusion

COVID-19 is not only an enormous public health burden, but has significantly impacted civil societies and the global economy. There is a mountain of preclinical evidence demonstrating the advantages of heparin therapy for SARS-CoV-2 infection. In this review, the inherent harms of heparin and the potential challenges in the implementation of heparin were described. 1) SAR mystery: preclinical trials are starting to show some results of discrepancies in the antiviral potencies of heparin and heparin analogues due to the heterogeneous structures. 2) Potential contamination: heparin, sourced from animals, has a chance to be contaminated with bioactive entities, viruses, or prions, presenting safety issues. 3) Side-effect (excessive anticoagulant activity): bleeding complications due to the anticoagulant effect of heparin have been limiting their clinical implementation in treating viral infections.

Given the co-existence of clinical benefits and safety concerns, efforts should be focused on maximizing therapeutic effects while minimizing the side effects of heparin. Well-defined, heparin-derived compounds could be beneficial in terms of safe production routes to replace animal-sourced heparin, providing a novel class of compounds for vital therapeutic applications. Non-anticoagulant heparin derivatives can be synthesized for direct antiviral activity by controlling the length, sulfation degree, and sulfation position of heparin. Such efforts could contribute to the development of anti-viral drugs that could be effective against SARS-CoV-2 and other unforeseeable viruses.

CRediT authorship contribution statement

All authors wrote the manuscript, read and approved the manuscript.

Declaration of competing interest

We declare that we have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.