Abbreviations:

AIDS: Acquired Immune Deficiency Syndrome; AH: Actinohivin; BanLec: Banana lectin; BzB: Benzoboroxole; CBA: Carbohydrate Binding Agents; Con A: Concanavalin A; CVL: Chaetopterus variopedatus lectin; CV-N: Cyanovirin-N; DC-SIGN: Dendritic Cell-Specific Intercellular adhesion molecule-3-Grabbing Non-integrin; FIPV: Feline Infectious Peritonitis Virus; Fuc: Fucose; Gal: Galactose; GlcNAc: N-acetylglucosamine; GRFT: Griffithsin; HexNAc: N-acetylhexoseamine (N-acetylglucoseamine, N-acetylgalactoseamine); HIV: Human Immunodeficiency Virus; HIV gp120: HIV envelope glycoprotein 120; HIV gp160: HIV glycoproteins precursor; HIV gp41: HIV envelope glycoprotein 41; Man: Mannose; MHC: Major Histocompatibility Complex; MHL: Myrianthus holstii lectin; MRC 1 & 2: Mannose Receptor C-Type 1 & 2; MVL: Microcystis viridis lectin; MVN: Microvirin; Neu5Ac: N-Acetylneuraminic acid (sialic acid); NPL: Narcissus pseudonarcissus lectin; OAA: Oscillatoria agardhii agglutinin; PBA: Phenylboronic acid; SARS: Severe Acute Respiratory Syndrome; SIV: Simian immunodeficiency virus; SVL: Serpula vermicularis lectin; SVN: scytovirin

1. Introduction

Since its discovery in 1983 as the virus responsible for the Acquired Immune Deficiency Syndrome (AIDS), the Human Immunodeficiency Virus (HIV) has remained a scientific challenge; as a complete eradication strategy has yet to be successful ^[1]. Although early-generation microbicides, such as surfactant-containing spermicides, acid-buffering gels, and polyanionic gels have failed to demonstrate efficacy, advances in microbicide development using HIV-specific antiretroviral agents (ARV) have shown significant promise ^[2]. Part of the success demonstrated by ARV comes from their ability to target and block key stages in HIV replication including viral entry, reverse transcription, integration, and maturation. Despite recent progress in anti-HIV microbicide development and advances in access to antiretrovirals, an average of 1.9 million (1.9-2.2 million) new HIV infections still occur globally every year ^[3]. More alarmingly, the rate of new HIV-1 infections is believed to be outpacing the rate of new individuals receiving antiretroviral therapy by an average of 2.5:1 ^[4]. Furthermore, the widely promoted “ABC” approach (abstinence, being faithful, condom usage) aimed at fostering HIV prevention and reducing the rate of infection spread, has shown some limitations, especially in third world countries, which remain the most affected by the pandemic infection ^[5]. In fact, Cohen stated that the practice of abstinence “is only theoretical, since one can only be certain of one's own behavior, not the behavior of one's partner” ^[6]. Although HIV infection declines in Uganda and Thailand were attributed to reduction in partner number, its long-term application remains a challenge in rural areas, where polygamy is often deeply rooted in traditional and religious beliefs ^[7,8]. Condoms, which are known to be effective when used consistently and correctly, still face a strong rejection from certain users who often report reduced physical pleasure, embarrassment of purchasing condoms, and a general perception that condom use represents a sign of infidelity and/or HIV/STD-seropositive status ^[9]. Therefore, there is a critical need for the development of alternative, long-lasting, self-applied and effective microbicides. Such topically (vaginally or rectally) applicable microbicides are projected to ultimately protect women, since more than half of all new HIV infections worldwide occur in females ^[10].

Among the different classes of anti-HIV microbicides currently in use, agents targeting and preventing viral entry into target cells have shown remarkable promises, partly favored by fewer barriers that could potentially hinder their mechanism of action. HIV entry inhibitors are subdivided into three main groups: Attachment, co-receptor binding and fusion inhibitors ^[11]. Attachment inhibitors such as zintevir, BMS-378806, and BMS-488043 block a non-specific adsorption step between HIV virions and target cells' membrane, which is due to an interaction between the positively charged regions of the envelope glycoprotein gp120 and oppositely charged proteoglycans on cell surface. Co-receptor binding inhibitors are generally CCR5 antagonists such as aplaviroc, vicriviroc, and maravirok that bind to CCR5 and prevent further gp120-CCR5 interaction, which is critical for viral entry into host cells. Fusion inhibitors such as tifuvirtide, enfuvirtide, and sifuvirtide prevent the formation of the fusion pore by mimicking either heptad repeat 1 or 2 (HR1 or HR2) sequences in gp41. These sequences block the formation of a six-helix bundle structure necessary for HIV entry into host cells ^[11].

Lectins, which are carbohydrate binding proteins, have long being considered for their diagnostic and therapeutic potentials, as well as their pathogenic implication in many human diseases and conditions including various cancers ^[12], type 2 diabetes ^[13,14], cardiovascular disease ^[15], weight management ^[16,17], and HIV/AIDS ^[18]. The ability of lectins to recognize and bind several mannose oligosaccharides was long considered a viable example of anti-HIV therapeutic strategy. Primarily, anti-HIV lectins act as viral entry inhibitors by binding to oligosaccharide epitopes on HIV surface glycoproteins, which either hinder a proper interaction between HIV and receptors on target cell membranes or affect post-binding conformational alterations of key viral envelope and transmembrane glycoproteins (HIV gp120/HIV gp41). In this manuscript, we report the current trend in anti-HIV lectins research and emerging lectins formulations aiming at improving the delivery of these sugar-binding proteins.

2. HIV surface glycoproteins and glycans

HIV surface glycoproteins (HIV gp120 and HIV gp41) mediate host cell entry through interactions with CD4 receptor and CCR5/CXCR4 co-receptors on target cells. These glycoproteins are first expressed as HIV gp160 precursor before the proteolytic cleavage in the trans-Golgi by cellular furin or furin-like proteases that lead to the formation of envelope glycoprotein HIV gp120 and transmembrane glycoprotein HIV gp41. In mature HIV viruses, HIV gp120 and HIV gp41 remain linked by noncovalent interactions ^[19]. Most anti-HIV lectins target and bind specific glycan structures on HIV envelope glycoproteins. Understanding the glycosylation pattern of these glycoproteins is useful not only for anti-HIV vaccine design, but also for the selection of appropriate lectins for potential anti-HIV therapy. Glycans found on HIV surface glycoproteins may also help in understanding anti-HIV lectins overall mechanism of action. Moreover, the extent and variation in glycosylation pattern between HIV strands, as well as changes in the glycosylation pattern during HIV maturation may help explain resistance to certain anti-HIV vaccines and lectins, as well as the lack of broad activity usually observed with anti-HIV lectins ^[20,21].

2.1. HIV gp120 glycans and their function

HIV gp120 is the external envelope glycoprotein of HIV. It is a homotrimer with each subunit having a nominal molecular weight of 120 kDa. HIV gp120 plays an essential role in HIV entry into host cells. In fact, HIV entry into host cells is initiated by the binding of gp120 to CD4 receptor. This binding triggers a conformational change in HIV gp120, which, in turn, enhances its affinity to chemokine receptors CXCR4 or CCR5. This secondary binding induces another conformational change in the transmembrane glycoprotein HIV gp41 resulting in an intimate contact and fusion of both viral and host cell membranes. The membrane fusion process leads to the internalization of HIV viral capsid containing the viral mRNA and key viral proteins into host cells' cytoplasm. Ultimately, new HIV virus particles are produced which then propagate the infection ^[22,23]. Literature expounds on HIV gp120 biosynthesis, trafficking, and incorporation. Rather than the underlying biological mechanism involved in these processes, we will focus on the nature of glycosylation and its role in the membrane fusion.

Ratner et al. ^[24], Allan et al. ^[25] and Montagnier et al. ^[26] published some of the first studies that explored HIV gp120 glycosylation. Although these pioneering studies did not investigate HIV gp120 glycan structures in detail, they did report HIV gp120 glycosylation to account for approximately 27 to 50% of the overall glycoprotein molecular weight. Some of these early investigations also demonstrated the presence of 31 ^[25] or 32 ^[27] potential N-glycosylation sites on HIV gp120. Subsequent studies by Mizuochi et al. ^[28,29] further investigated HIV gp120 glycan structures. Mizuochi's findings showed in part that HIV gp120 is unique in its diversity of oligosaccharide structures. These studies also reported HIV gp120 glycans to be predominantly oligomannoses that are mostly comprised of five to nine mannose residues, and accounting for approximately 33% of the overall glycoprotein's carbohydrate structures ^[30]. Furthermore, this study projected the number of potential N-glycosylation sites on HIV gp120 to be 20. Besides the high-mannose type glycans, Mizuochi et al. also identified complex type glycan chains (34% of carbohydrate structures) mainly composed of four categories: mono-, bi-, tri- andtetra-antennary, with or without N-acetyllactosamine repeats, and with or without a core fucose residue ^{[28,30,31,32]}. A previous study by Geyer et al. ^[31] reported similar findings and showed that predominant oligomannose glycans in HIV gp120 are composed of seven to nine mannose residues, depending on whether the glycoprotein is excreted or expressed intracellularly. Geyer et al. also showed that HIV gp120 complex-type oligosaccharide are fucosylated with partial sialylated bi- and triantennary structures. Recent advances in glycoscience, genomic, proteomics, and mass spectrometry have led to more detailed and in-depth characterization studies of HIV gp120 glycosylation. In fact, recent mass spectrometry studies have confirmed HIV gp120 high mannose proportion for various viral clades and expression systems ^{[32,33,34,35,36]} and it is widely accepted that the number of HIV N-glycosylation sites range from 20-35 ^[19]. Following a matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) analysis, Bonomelli et al. showed that HIV gp120 oligosaccharides, derived from virus [clade A (92RW009), clade B (JRCSF), clade C (93IN905)] isolated from infected peripheral blood mononuclear cells (PBMCs), are almost entirely oligomannoses and varies from 62-79% for virion-associated versus 30% for recombinant monomeric HIV gp120 ^[33,37]. These studies also identified an “intrinsic” mannose patch in HIV gp120 composed essentially of Man_5-9GlcNAc₂ and conserved across primary isolates and geographically divergent HIV clades. Many other studies have confirmed the presence of the mannose patch on HIV gp120 and its relevance in the development of a successful anti-HIV vaccine ^{[38,39,40,41]}. HIV gp120 main glycan structures are summarized in Figure 1.

HIV gp120's heavy glycosylation is believed to play four key roles: Host immune evasion, pathogenesis, proper glycoproteins folding, and host cell surface recognition ^[42]. In fact, several studies have compared HIV gp120 extensive glycosylation to a protecting shield that prevents antibody access to underlying amino acid sequences and therefore limits their efficacy ^{[40,43,44,45,46]}. More specifically, Sanders et al. determined that the carbohydrate at asparagine 386 on HIV-1 gp120 enhances HIV immune evasion ^[47]. Furthermore, the general role of HIV gp120 glycosylation in HIV pathogenesis has widely been reported ^[48]. Besides its major implication in HIV gp120 proper folding and lysosomal degradation, Francois et al. ^[49] and Mathys et al. ^[50] showed that cleavage of glycan at asparagine 260 of HIV-1 gp120 results in loss of viral infectivity. Similarly, Huang et al. demonstrated that deletion of HIV gp120 glycans from asparagine proximal to the CD4-binding region (156, 262 and 410) impairs HIV viral infectivity ^[51]. The essential role of glycosylation in proper HIV gp120 folding was also elucidated by numerous reports ^[50,52]. For example, Li et al. showed that N-linked glycosylation is highly essential for a proper conformation of HIV gp120 capable of binding to CD4 receptor ^[53]. In a separate study Li et al. determined that deletion of the glycan at asparagine 448 can profoundly influence CD4 T-cell recognition of HIV-1 gp120 ^[54].

Figure 1. High mannose and complex glycan structures found in HIV gp120. Structures are adapted from the following references ^{[28,30,31,55,56]}.

DownLoad: Full-Size ImgPowerPoint

2.2. HIV gp41 glycans and their function

HIV gp41 is composed of 345 amino acids that are organized into three major domains: extracellular (ectodomain), transmembrane, and C-terminal cytoplasmic tail ^[19,57,58]. Unlike HIV gp120, the transmembrane glycoprotein contains fewer N-glycosylation sites. Nonetheless, there is an inconsistency pertaining to the number of N-glycosylation sites in HIV gp41, as various communications often report different numbers. This may be due to differences in expression systems, cell lines, and HIV strands. According to the current literature, the number of potential N-glycosylation sites in HIV gp41 vary between three to eight. In fact, Perrin et al. reported poor glycosylation of HIV gp41 ectodomain with only four or five potential glycosylation sites ^[59]. Following the screening of ten HIV-1 amino acid sequences, Johnson et al. determined that HIV gp41 typically contains three or four N-glycosylation sites, localized within a short stretch (20 to 30 amino acid residues) of the C-terminal half of the ectodomain ^[60]. The same number of HIV gp41 potential N-glycosylation sites was reported by Fenouillet et al. ^[61,62], Lee et al. ^[63], Ma et al. ^[64], and Wang et al. ^[65]. Furthermore, citing the work of Montefiori et al. ^[48] and Checkley et al. reported HIV gp41 N-glycosylation sites to vary between three to five ^[19]. The work of Balzarini et al. reported some of the highest number of HIV gp41 potential N-glycosylation sites which was thought to be seven with only four seemingly glycosylated ^[66]. Further studies by Mathys and Balzarini have reported N-glycosylation sites in HIV gp41 between four-eight with all four to five N-glycans on the ectodomain composed of complex-type glycans ^[67,68].

In contrast to HIV gp120, the transmembrane glycoproteins' glycans are known to be primarily composed of more complex carbohydrate types. In fact, following the analysis of HIV gp41 expressed from two different producer cells (Chinese hamster ovary cells and human embryonic kidney cells [293T]), Pritchard et al. determined that in combination with the presence of less oligomannose glycans (19-34%) compared to HIV gp120 (60-65%), HIV gp41 contains large populations of complex-type glycans on its ectodomain ^[56]. Regardless of the expression system, HIV gp41 oligomannose population was also found to be composed of Man_5-9GlcNAc₂. Like HIV gp120, complex glycans in HIV gp41 are composed of sialilated and asialilated bi-, tri-, and tetra-antennary structures usually containing N-lactosamine repeats with or without core fucose residue ^[34,55,56]. HIV gp41 main glycan structures are summarized in Figure 2.

Figure 2. High mannose and complex glycan structures in HIV gp41. Structures are adapted from the following references ^[34,55,56].

DownLoad: Full-Size ImgPowerPoint

HIV gp41 plays an equally critical role in HIV entry into target cells by mediating the membrane fusion process required for the internalization of the HIV viral capsid ^[23]. Glycans in HIV gp41 are reported to play key roles in viral entry, immune evasion, and infectivity. In fact, Fenouillet et al., reported a loss of HIV ability to enter target cells after complete removal of the glycan cluster from asparagines at positions 621, 630 and 642 ^[69] in HIV gp41. A follow up study by Perrin et al., determined the critical role of HIV gp41 glycosylation for an effective membrane fusion process. This study also reported that out of the four or five glycosylation sites in HIV gp41, only two sites are sufficient for efficient membrane fusion with a single site at asparagine 621 being the most critical of all positions ^[59]. Yuste et al. have also suggested that the function of HIV gp41 glycosylation from both HIV and SIV may be shielding underlying epitopes sequences, thereby allowing the virus to escape neutralizing antibodies ^[70]. Furthermore, Wang et al. reported that the glycan at asparagine 637 in HIV gp41 is composed of Man₉GlcNAc₂ and plays a critical role in immune evasion through the facilitation of the membrane fusion process ^[65]. A later study by Mathys and Balzarini lead to a rather different conclusion regarding the importance of this glycan ^[68]. By following the generation of several HIV-1 mutants lacking HIV gp41 N-glycans and assessing their influence on viral infectivity, this study determined that besides the glycan at asparagine 616 that when deleted, leads to a complete loss of HIV-1 infectivity, deletion of glycans on asparagines 611 and 637, displayed marginal effect on overall viral infectivity. In addition, this study concluded that glycans on asparagines 625 and 674 do not play any significant role in HIV infectivity, since their deletion did not influence viral infectivity.

3. Anti-HIV lectins

3.1. Natural lectins

Owing to HIV-gp120 high mannose content, various mannose binding natural lectins have been investigated as HIV entry inhibitors. Primarily, these lectins specifically bind mannose oligosaccharides on HIV-gp120, thus hindering a proper interaction between the envelope glycoprotein and its host cell receptor (CD4). This may ultimately prevent the membrane fusion step and the production of new HIV virions. Actinohivin (AH), a prokaryotic lectin derived from the gram-positive bacteria actinomycete Longispora albida (K97-0003T) successfully prevented HIV-1 entry into CD4⁺ T lymphocytes (IC₅₀ ≈ 2-110 nM) ^[71]. It was determined that AH binds α(1-2)-mannose oligosaccharides present in HIV gp120 and HIV gp41. Furthermore, AH did not induce any mitogenic activity or cytokine/chemokine production in PBMC cultures, suggesting that this lectin could be a safe and effective microbicide candidate ^[72]. Recently, Zhang et al. reported that Man₁ and Man₂ residues, found on HIV gp120 high-mannose type glycans structures, occupy two of the three carbohydrate binding sites of AH, while Man₃ residues interact with conserved hydrophobic amino acid residues (Tyr and Leu) of AH ^[73]. Cyanovirin-N (CV-N) is a cyanobacterial lectin with broad-spectrum antiviral activity. The potential use of CV-N as anti-HIV microbicide has widely been reported (IC₅₀ ≈ 3.9-31 nM) ^{[74,75,76,77,78]}. CV-N inhibits HIV replication in part by binding to HIV-gp120 high mannose glycans, thus preventing the envelope glycoprotein from binding to its cell surface receptor (CD4), thereby blocking the glycoprotein-mediated membrane fusion process required for HIV-1 entry ^[79]. Hu et al. have determined that CV-N binding interaction is mediated through three to five high-mannose residues from 289 to 448 in the C2-C4 region of HIV gp120 and deglycosylation of these residues resulted in a resistance to CV-N ^[80,81]. It was also shown that CV-N inhibits HIV replication by interacting with the chemokine receptors CXCR4 and CCR5 ^[82]. Recently, a CV-N oligomer (CV-N₂) was designed and demonstrated an increased HIV-1 neutralization activity by up to 18-fold compared to the wild-type CV-N (IC₅₀ ≈ 0.1-41 nM) ^[81]. Oscillatoria agardhii agglutinin (OAA) is a newly discovered cyanobacterial lectin that was shown to prevent HIV transmission, replication, and syncytium formation between HIV-1-infected and uninfected T cells (IC₅₀ ≈ 24-30 nM) ^[83]. OAA is known for having two sugar binding sites that recognize Manα(1-2)Man, Manα(1-6)Man and the branched core unit of Man₉ (3α,6α-mannopentaose) ^{[83,84,85,86]}. Similar to OAA, scytovirin (SVN) is a cyanobacterium lectin isolated from Scytonema varium ^[87]. SVN was also shown to possess potent anti-HIV activity through its binding interaction with HIV gp160, HIV gp120, and HIV gp41 and binds Manα(1-2)Manα(1-6)Manα(1-6)Man tetrasaccharide in high mannose type oligosaccharides (IC₅₀ ≈ 24.1 nM) ^[88,89,90]. In addition, MVL, a lectin isolated from the cyanobacterium Microcystis viridis also showed strong anti-HIV activity at nanomolar concentrations by binding to Manα(1-6)Manβ(1-4)GlcNAcβ(1-4)GlcNAc oligosaccharides on the surface of HIV gp120 (IC₅₀ ≈ 30 nM) ^[91,92]. Another cyanobacterial lectin, microvirin (MVN), isolated from Microcystis aeruginosa has shown anti-HIV activity comparable to CV-N with a much better cytotoxicity profile (IC₅₀ ≈ 2-12 nM) ^[93]. It was further shown that MVN binds Manα(1-2)Man residues on HIV gp120 ^[93,94]. Plant lectins such as Narcissus pseudonarcissus lectin (NPL) (EC₅₀ ≈ 0.17-2.76 µg/mL) ^[95,96] and Myrianthus holstii lectin (MHL) (EC₅₀ ≈ 150 nM) ^[97] have also shown potential HIV inhibition in vitro. Concanavalin A (Con A), one of the most studied plant lectins, is a mannose binding lectin extracted from jack bean. Con A binds sugars, glycoproteins, and glycolipids, containing internal and nonreducing terminal α-D-mannosyl and α-D-glucosyl groups (K_D ≈ 0.05 µM to 1.5 µM) ^[98,99]. Several studies have demonstrated the ability of Con A to bind HIV gp120 and inhibit the fusion process during HIV infection (EC₅₀ ≈ 98 nM) ^{[100,101,102,103]}. Furthermore, BanLec is a lectin isolated from the banana fruit (Musa acuminata), which has shown potent anti-HIV activity (IC₅₀ ≈ 2.5-694 nM) ^[104]. Like Con A, BanLec inhibits HIV by binding to high mannose carbohydrate structures found in HIV gp120, thus blocking the virus entry into host cells. In fact, in a comparative study, BanLec showed similar inhibitory activity like T-20 and maraviroc, two FDA approved HIV entry inhibitor microbicides ^[4].

Griffithsin (GRFT), a lectin isolated from the red algae Griffithsia inhibited cell-to-cell fusion between HIV infected and uninfected cells (IC₅₀ ≈ 4 nM) ^[105]. GRFT also inhibited HIV-1 transmission by binding to glucose, mannose, and N-acetylglucosamine residues in HIV glycoproteins (HIV gp120, HIV gp41 and HIV gp160) ^[106]. Emau et al. ^[107] have also established that GRFT strongly blocked CXCR4- and CCR5-tropic viruses at concentrations less than 1 nM, with low cytotoxicity, rapid onset of antiviral activity and long-term stability in cervical/vaginal lavage. GRFT tandemers, recently reported by Moulaei et al., have shown anti-HIV activities five to ten-fold higher than native GRFT (IC₅₀ ≈ 0.02-0.274 nM) ^[108]. Table 1 summarizes natural anti-HIV lectins and major properties discussed above.

Table 1. Example of natural anti-HIV lectins.

Lectin	Glycan preference		Target	IC50/EC50	Origin	References
AH	Manα(1-2)Man, Manα(1-3)Man, Manα(1-6)Man, GlcNAc		gp120 and gp41	IC50 ≈ 2-110 nM	Actinomycete Longispora albida	[71,72]
CV-N	Manα(1-2)Man in Man8 or Man9		gp120, CXCR4 and CCR5	IC50 ≈ 0.1-41 nM	Nostoc ellipsosporum	[80,81]
OAA	Manα(1-2)Man, Manα(1-6)Man, 3α,6α-mannopentaose		gp120	IC50 ≈ 24-30 nM	Oscillatoria agardhii	[83,84,85,86]
SVN	Man		gp120	IC50 ≈ 24.1 nM	Scytonema varium	[88,89]
MVL	Manα(1-6)Manβ(1-4)GlcNAcβ(1-4)GlcNAc		gp120	IC50 ≈ 30 nM	Microcystis viridis	[91]
MVN	Manα(1-2)Man		gp120	IC50 ≈ 2-12 nM	Microcystis aeruginosa	[93,94]
NPL	Manα(1-3)Man; Manα(1-2)Man		gp120	EC50 ≈ 0.17-2.76 µg/mL	Narcissus Pseudonarcissus	[95,96]
MHL	GlcNAc		gp120	EC50 ≈ 150 nM	Myrianthus Holstii	[97]
Con A	α-D-Man and α-D-Glc		gp120	EC50 ≈ 98 nM	Jack bean	[98]
BanLec	Man		gp120	IC50 ≈ 2.5-694 nM	Musa acuminata	[4]
GRFT	Glc, Man and GlcNac		gp120, gp41, gp160, CXCR4 and CCR5	IC50 ≈ 4 nM	Griffithsia	[106]
CVL	β-Gal		gp41, gp120	EC50 ≈ 73 nM	Chaetopterus variopedatus	[110]
Mermaid	Manα(1-3)Manα(1-6)Man		gp120	IC50 ≈ 3.1 µg/mL	Laxus oneistus	[111,112]
SVL	GlcNAc		gp41, gp120	IC50 ≈ 0.15-0.23 µg/mL	Serpula vermicularis	[113]

 | Show Table

DownLoad: CSV

Chaetopterus variopedatus lectin (CVL) is a β-galactose-specific lectin extracted from the marine worm Annelida. CVL was shown to inhibit both HIV attachment to host cells and the fusion process between HIV and target cells (EC₅₀ ≈ 73 nM) ^[109]; suggesting that CVL might be exerting its action through interaction with complex glycan type found in HIV gp120 and HIV gp41 ^[110]. In addition, Mermaid, a calcium (Ca²⁺) dependent lectin isolated from the marine nematode (Laxus oneistus) was shown to have structural similarities and similar glycan specificity with the Dendritic Cell-Specific Intercellular adhesion molecule-3-Grabbing Non-integrin (DC-SIGN). Mermaid, which binds mannose oligosaccharides on HIV gp120 prevented HIV-1 binding to DC-SIGN on dendritic cells, which ultimately blocked HIV transmission (IC₅₀ ≈ 3.1 µg/mL) ^[111,112]. Another marine lectin Serpula vermicularis lectin (SVL) isolated from the sea worm Serpula vermicularis was also shown to bind GlcNAc and inhibited the production of viral p24 antigen and cytopathic effect induced by HIV-1 (EC₅₀ ≈ 0.15-0.23 µg/mL) ^[113,114].

3.2. Synthetic lectins

Carbohydrate binding agents (CBAs) can bind to carbohydrate residues on the surface of viral envelopes, as for HIV gp120. This binding could lead to an inhibition of viral entry. Moreover, mutations of the envelope glycoproteins can improve drug pressure and lead to viral immune response neutralization. Because manufacturing natural plant-based lectins can be expensive, synthetic lectins have been considered as potential inhibitory alternative ^[115]. Synthetic lectins are cheaper to mass-produce as compared to their plant-based counterparts. As a response to the high cost and potential mitogenecity of natural lectins, Mahalingan et al. developed a benzoboroxole (BzB) polymeric synthetic lectin. Like natural plant-based lectins, BzB targets and binds carbohydrates on HIV viral envelope. At pH 7, BzB demonstrated increased binding efficiency to reducing sugars, such as fructose and weak binding affinity to non-reducing sugars, such as galactopyranose, a terminal sugar found on the surface of HIV gp120 complex glycans. This study further showed that BzB polymers of high molecular weight increased antiviral activity, proving that polyvalent interactions between inhibitor and glycosylated sites on HIV viral envelope improved with increased molecular weight. For example, increasing the mole percent of BzB functionalization showed an increase in EC₅₀ [EC₅₀ = 15 uM (25 mol%); EC₅₀ = 15 nM (75 mol%)]. Moreover, substituting the polymer backbone with 10% sulfonic acid, resulted in an increased synergistic anti-HIV activity, as well as a 50-fold increase in aqueous solubility of the polymer. Furthermore, BzB-sulfonic acid showed an improved selectivity to HIV gp120, and the presence of fructose from seminal fluid did not decrease its antiviral activity ^[116]. Similarly, synthetic lectins containing phenylboronic acid (PBAs) could potentially exhibit carbohydrate recognition like that of CBA. For instance, Trippier et al. synthesized mannose selective PBA-based synthetic lectins that were tested for binding affinity against HIV gp120 ^[117]. Because the mono-PBA synthetic lectins tested did not bind HIV gp120 and were not good antiviral candidates, bis-PBA synthetic lectins were further investigated ^[118]. Although the bis-PBA did not demonstrate pronounced antiviral activity, these compounds were however found to be relatively noncytotoxic. It was also suggested that the lack of HIV gp120 binding could be due to the lack of multivalency and the small size of the PBA compounds.

4. Current anti-HIV lectins delivery strategies

4.1. Challenges in anti-HIV lectins drug development

The clinical translation of anti-HIV natural lectins faces numerous challenges including stability, solubility, resistance, toxicity, and manufacturing. These factors have seriously limited the progress in the field and often overshadow the potential benefits that anti-HIV lectins may have. For example, BanLec has been shown to partially dissociate into its monomeric forms in acidic conditions (pH 2) while maintaining a dimeric structure at neutral pH. The monomeric form of BanLec also offered more resistance towards chemical denaturation than the native dimeric form ^[119]. In addition, AH was shown to display low solubility in neutral buffer solutions with an enhanced solubility in acidic conditions ^[120]. This lack of solubility in neutral pH conditions could dramatically limit AH use as a topical microbicide for the prevention of HIV sexual transmission. It is established that vaginal pH increases from acidic (pH ≈ 4.5) to neutral (pH ≈ 7.5) during intercourse ^[121]. Furthermore, lectin resistant HIV strands have been reported ^[20]. The mutation of certain glycan structures in HIV gp120 was shown to be responsible for CV-N and Con A resistant HIV strands ^[100]. Although the development of HIV resistance to lectin may ultimately undermine the potency of these proteins, this is however viewed as an indirect route for exposing underlying amino acid sequences that could potentially be targeted by antibodies ^[80]. Anti-HIV lectins have also been associated with strong toxicity. In particular, Con A was shown to be mitogenic toward T-cells and induced cytotoxicity at high concentrations ^[122,123]. Similarly, CV-N and BanLec induced pronounced mitogenic activities on PBMCs and T-cells respectively ^[124,125]. Nonetheless, by replacing histidine 84 with a threonine in BanLec, Swanson et al. have demonstrated the possibility of bioengineering anti-HIV lectins to suppress their mitogenicity while maintaining their antiviral activity ^[104]. The high cost of natural anti-HIV lectins mass production and purification presents another particularly difficult challenge ^[126]. Although recombinant technology was proposed to overcome this limitation, improving fermentation yield, controlling mutation, and addressing potential immune system insults need to be studied ^[127]. Besides their ability to address some of these limitations and inhibit HIV transmission with relatively good safety profiles, synthetic lectins usually lack carbohydrate specificity and often require extensive optimizations to improve their binding affinity for HIV surface glycoproteins ^{[117,118,126]}.

4.2. Anti-HIV lectin formulations

A potential barrier to the development of antibody-based vaccines against HIV is the oligosaccharide layer that provide a protective covering to the underlying antigens on the viral envelope surface ^[128]. Carbohydrate-lectin complexes are a promising therapeutic strategy because various proteins interact with oligosaccharides on the surface of many human cells. Glycoproteins and glycolipids can also interact with lectins and enhance mucosal absorption of drugs and vaccines ^[129]. Taking advantage of the so-called “lectin direct targeting”, potential efficacious HIV vaccine nanoformulations have targeted endogenous lectins for antigen delivery to immune cells ^[130]. Dendritic cell lectins are often targeted in this strategy. Those anti-HIV vaccines strategy activate various receptors on antigen presenting cells or C-type lectins to illicit immune responses.

The mannose receptor, a C-type lectin found on the vaginal epithelium, is known to bind HIV gp120 ^[131,132]. Binding of the mannose receptor to HIV gp120 allows the virus to cross the vaginal epithelium ^[133]. Humans have two types of mannose receptors, type 1 (MRC1) and type 2 (MRC2) and both can stimulate active and adaptive immunity. Because mannose receptors are highly expressed on dendritic cells as well as macrophages, these receptors are important for antigen recognition. Mannose receptors on dendritic cells take up antigen which stimulates robust T-cell activation via both major histocompatibility complexes (MHC) I and II molecular uptake mechanisms. This T-cell activation plays a critical role in the successful anti-HIV vaccine development ^[131]. When HIV-1 DNA was encapsulated in mannan coated-cationic liposomes targeting MRC, the nanoformulation successfully activated immunological responses, such as cytotoxic T cells, IgA, and other hypersensitivity responses ^[131]. These cationic nanoparticles showed 50% higher uptake than non-coated mannan nanoparticles in the macrophage cell line J774E ^[133]. Similarly, Espuelas et al. showed that a liposome nanoformulation containing mono-, di-, and tetra antennary mannosyl lipid derivatives could potentially achieve identical mannose receptor targeting on dendritic cells for a potential mannose-targeted vaccination strategy ^[134]. Furthermore, this study proved that liposome formulations containing higher mannose density result in more efficient interactions with mannose receptors. DC-SIGN is a Ca²⁺ binding adhesion lectin present on the surface of immature dendritic cells that play an important role in modulating host response to infection and inflammatory stimuli ^[135]. Because of its implications for antigen targeting and stimulation of T-cell responses, DC-SIGN has been considered a potential receptor for HIV vaccine targeting. In fact, DC-SIGN recognizes various high mannose oligosaccharides on HIV gp120 ^[136]. In vitro studies using DC-SIGN-targeted PLGA nanoparticles have shown that these nanoformulations deliver antigens to human dendritic cells ^[137]. DC-SIGN also increased antigen presentation, which translated into an improved activation of CD4 + and CD8 + T-cells.

The development of additional HIV nanovaccine immunogens utilized envelope glycoprotein mimetics. Ingale et al. investigated liposomes-grafted high-density enveloped HIV glycoprotein trimers that were recognized by anti-HIV-1 antibodies and activated B-cells ^[138]. These liposome constructs may lead to a promising HIV neutralization vaccine. Moreover, He et al. designed nanoparticles containing native like trimeric structures of V1V2 and gp120. These nanoformulations presented a variety of gp140 trimers that displayed 20 spikes like that of other virus like particles. This study showed high B cell stimulation, which may lead to further investigation in the development of a multivalent HIV vaccine ^[139].

Other lectin-based anti-HIV strategies have focused on “lectin indirect targeting” instead. In this case, lectins (natural or synthetic) are included in formulations to target HIV envelope glycoproteins. This “virion capture” approach may lead to a successful HIV prevention by hindering a proper interaction between HIV virus and its targets. Virion and HIV gp120 antigen capture could potentially lay the foundation for a mucosal anti-HIV vaccine. Akashi et al. proposed a Con A immobilized polystyrene nanospheres capable of capturing HIV virions through binding interactions with HIV gp120 high mannose glycans ^[140]. Hayakawa et al. further investigated a similar strategy using nanoparticles prepared via co-polymerization of polystyrene and poly methacrylate^[141]. Recently, Coulibaly et al. developed a mannose specific lectin-based HIV-1 gp120 responsive microbicide formulation capable of the control release of the nucleotide reverse transcriptase inhibitor Tenofovir (TFV) ^[142]. In this study, Con A's ability to bind glycogen (a glucose-based polysaccharide) was used to engineer a self-assembled layer-by-layer drug delivery system. Drug release was achieved through a controlled and reverse disassembly of Con A/glycogen layers in seminal and vaginal fluid simulants at HIV-1 gp120 concentrations ≥ 25 µg/mL. Con A/glycogen layers disassembly was also shown to be primarily due to the lectin's higher binding affinity for mannose glycans in HIV-1 gp120. Moreover, the amount of TFV released was shown to potentially inhibit HIV sexual transmission. This system also appeared to be safer on vaginal (VK2), murine macrophage (RAW 264.7) and Lactobacillus crispatus cell lines. Although this system could be a safe and effective template for HIV vaginal microbicide drug delivery, future studies still need to prove its anti-HIV activity and in vivo safety. Current anti-HIV lectins' formulations, discussed above, have been summarized in Table 2.

Table 2. Summary of anti-HIV lectin formulations.

Formulation	Lectin	Target	References
Mannan coated-cationic liposomes	Mannose receptors, C-type lectins	MRC (Dendritic cells)	[131,133]
Mannosylated liposome	Mannose receptors, C-type lectins	MRC (Dendritic cells)	[134]
PLGA nanoparticles	DC-SIGN	Dendritic cells	[137]
High density enveloped HIV glycoprotein liposomes	N/A	BRC (B cells receptor)	[138]
Con A immobilized polystyrene nanospheres	Con A	HIV gp120	[140]
Con A immobilized polystyrene/methacrylate nanospheres	Con A	HIV gp120	[141]
Layer-by-layer engineered Con A/glycogen microparticles	Con A	HIV gp120, methylα-D-mannopyranoside	[142]

 | Show Table

DownLoad: CSV

5. Conclusion

In general, the field of lectinology has greatly contributed in the structural elucidation, the mechanistic understanding and the advancement of lectin based alternative antiviral therapy for various enveloped viruses including HIV, Zika, Ebola, Marburg, Herpes, Hepatitis-C, influenza, Severe Acute Respiratory Syndrome (SARS), Feline Infectious Peritonitis Virus (FIPV), and many more ^{[143,144,145,146,147,148,149,150]}. Despite test tube promises shown by lectins (natural and synthetic) against these pathogens, lectin-based antiviral clinical translation still faces great challenges including, resistance, cytotoxicity, immunogenicity, antigen specificity, and limited stability ^[151]. Nonetheless, current research on selected lectin candidates, such as BanLec and Griffithsin, may potentially lead to the first clinically available lectin-based antiviral therapy in the future ^[4,152]. Although anti-HIV lectins research is projected to grow, future investigations in the field would likely have to address novel delivery strategies to significantly improve CBA clinical translation.

Acknowledgements

This work is supported by Grant Number R01AI087304 from the National Institute of Allergy and Infectious Diseases. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Allergy and Infectious Diseases or the National Institutes of Health.

Conflict of interest

All authors declare no conflicts of interest in this manuscript.