Lectins (Molecular Biology)

Lectins (from Latin, legere, to select or choose) are proteins that bind mono- and oligosaccharides specifically and reversibly, but are devoid of catalytic activity and, in contrast to antibodies, are not products of an immune response. They are typically di- or polyvalent with respect to carbohydrate binding, and therefore they agglutinate cells and precipitate polysaccharides or glycoproteins; these reactions are inhibited by the sugars for which the lectins are specific. The erythrocyte agglutinating, or hemagglutinating, activity of lectins is their major attribute and is used routinely for their detection and characterization; another way is by primary sequence similarities. These proteins are found in most organisms, from viruses and bacteria to plants and animals. Tissues or cells may contain one or more lectins of the same or different specificity. Lectins are readily obtainable in purified form, mostly by affinity chromatography on the immobilized ligands and also by recombinant DNA techniques. Hundreds of lectins have been isolated and characterized (for examples, see Table 1), some 30 to the level of their three-dimensional structure in complex with ligands. Generally, they are oligomeric proteins of different types with diverse combining sites. Nonetheless, many of them belong to distinct protein families that share sequence homologies and similar tertiary and quaternary structures (see Protein Evolution). Microbial toxins that are carbohydrate-specific are also classified as lectins.


Table 1. Some Well-Characterized Lectins2

                                             Specificity

Lectin

Source

Monosaccharide Oligosaccharide Structure

Conanavalin A

Plants (legumes)

Man

Man a 3(Mana6) Simple Man

Calnexin

Animals

Man

Mosaic

Dolichos

Plants

GalNAc

Simple

biflorus

(legumes)

Galectins

Animals

Gal

Galb4GlcNAc

Simple

Hepatic

Animals

Gal/GalNAc

Mosaic

binding

protein

Influenza virus

Influenza virus

NeuAc(a2-3 / 6) Gal

Mosaic

hemagglutinin

Man-6-P

Animals

Man-6-P

Mosaic

receptor

Peanut

Plants

Gal

GalbGalNAc

Simple

agglutinin

(legumes)

P-type

E. coli

Gala4Gal

Macromolecular

fimbriae

assembly

Selectins

Animals

sLex

Mosaic

Sialoadhesins

Animals

NeuAc(a2-3 / 6) Gal

Mosaic

Type 1

E. coli

Man

Macromolecular

fimbriae

assembly

Wheat germ

Plants

GlcNAc, NeuAc

Simple

agglutinin

(cereals)

Lectins are invaluable tools for the structural and functional investigation of complex carbohydrates, especially glycoproteins, for the study of their biosynthesis and for the examination of changes that occur on cell surfaces during physiological and pathological processes, from cell differentiation to cancer. At present, they are the focus of intense attention because of the demonstration that they act as recognition determinants in diverse biological processes.

1. Carbohydrate specificity

Based on their specificity, lectins are classified into five groups, according to the monosaccharide for which they have the highest affinity: mannose, galactoses-acetylgalactosamine, N-acetylglucosamine, fucose, and N-acetylneuraminic acid (sugars are of the D-configuration except for fucose, which is L). Relevant for the functions of lectins is the fact that, of the numerous monosaccharides found in nature, these six alone are typical constituents of eukaryotic cell surfaces. Many lectins specific for mannose also react with glucose (the 2-epimer of mannose), but none of these reacts with galactose (the 4-epimer of glucose), nor do those specific for galactose bind mannose. Similarly, members of the N-acetylglucosamine specificity group do not combine with N-acetylgalactosamine (or vice versa). However, most lectins that bind galactose interact also with N -acetylgalactosamine, in some cases preferentially. For this reason, they are classified in one specificity group, Gal/GalNAc, even though a few (eg, peanut agglutinin) do not bind N-acetylgalactosamine at all. Occasionally, lectins combine with apparently unrelated monosaccharides that, however, share common topological features. For instance, wheat germ agglutinin (WGA) binds both N-acetylglucosamine and N-acetylneuraminic acid, and animal mannose-specific binding proteins (MBPs) bind fucose too. Lectins of the same specificity group may combine preferentially with either the a- or b-glycosides of the corresponding monosaccharide, whereas others lack anomeric specificity. They may also differ markedly in their affinity for other derivatives, especially oligosaccharides. Certain lectins interact with oligosaccharides only (Table 1). The association constants for the binding of monosaccharides and oligosaccharides to lectins are typically in the rangetmp14F-7_thumbrespectively; multivalent oligosaccharides bind more strongly.

2. Molecular Properties

2.1. Plant Lectins

The largest and most thoroughly studied family of lectins is that of the legumes, with over 100 members, some 40 of which have been sequenced. These lectins, as well as others from plants, often occur as isolectins that either are coded by closely-related genes or are the product of post-translational modification such as glycosylation or proteolysis. Concanavalin A (Con A) from Jack bean, the prototype plant lectin, was first isolated in 1919 by James Sumner (of urease fame) and shown by him, in 1936, to be specific for mannose and glucose. Members of this family consist of two or four identical, or nearly identical, subunits of 25 to 30 kDa that may carry up to two N-linked oligosaccharides. In some cases, the subunits are fragmented. Each subunit has a single carbohydrate combining site with the same specificity. One exception is phytohemagglutinin, which occurs as a family of five tetrameric isoforms in all possible combinations of E and L subunits (from E4 to L4) that differ in their specificity and biological properties. Legume lectins also contain one tightly 2 + 2 + bound Ca ion and one transition metal ion (usually Mn ) per subunit, which are required for  carbohydrate binding. Their sequences are about 40% homologous; invariant residues include several of those that participate in hydrogen bonding (an aspartic acid and an asparagine residue) and in hydrophobic interactions (an aromatic amino acid or leucine) with the ligand, and almost all those that coordinate the metal ions. Two animal lectins (MR60/ERGIC-53 specific for mannose, and VIP36 specific for N-acetylgalactosamine) are homologous to those of the legumes (1).

Concanavalin A occupies a special position, because it exhibits "circular homology" with the other legume lectins. This homology is obtained by aligning residue 123 of concanavalin A with the amino-terminal residue of the other lectins, proceeding to the carboxyl end of concanavalin A and continuing along its amino-terminal region. It is the result of an unusual post-translational processing of the lectin (2) (Fig. 1).

Figure 1. Post-translational modifications during the biosynthesis of concanavalin A; a summary of processing events converting glycosylated pro-concanavalin A to mature lectin. The amino- and carboxy-termini are indicated by N and C, respectively, and the numbers in parentheses are residue numbers in mature concanavalin A. During processing in the plant, the inactive glycosylated prolectin is deglycosylated (arrow a), resulting in appearance of lectin activity. An endopeptidase then cleaves (arrows b, c, d, and e) a carboxy-terminal nonapeptide and the glycosylated spacer (shown as solid black areas). Residues 118 (arrow d) and 119 are ligated enzymatically. Splicing thus results in a transposition of the linear arrangement of the protein sections designated B and A.

Post-translational modifications during the biosynthesis of concanavalin A; a summary of processing events converting glycosylated pro-concanavalin A to mature lectin. The amino- and carboxy-termini are indicated by N and C, respectively, and the numbers in parentheses are residue numbers in mature concanavalin A. During processing in the plant, the inactive glycosylated prolectin is deglycosylated (arrow a), resulting in appearance of lectin activity. An endopeptidase then cleaves (arrows b, c, d, and e) a carboxy-terminal nonapeptide and the glycosylated spacer (shown as solid black areas). Residues 118 (arrow d) and 119 are ligated enzymatically. Splicing thus results in a transposition of the linear arrangement of the protein sections designated B and A.

High-resolution X-ray crystallography analysis of the structures of legume lectins revealed that the subunits are dome-shaped, made up largely of two antiparallel b-sheets that form a jellyroll or lectin fold (3). The subunits are nearly superimposable, irrespective of the specificity of the lectins, and associate into different types of dimers or tetramers (Fig. 2). The carbohydrate-binding site is located at the top of the subunit dome, in close proximity to the metal ions, which help to position the amino acid residues that form contacts with the carbohydrate. The invariant binding site residues (aspartic acid, asparagine, and an aromatic amino acid) occupy identical spatial positions in all these lectins, and discrimination between glucose/mannose and galactose is achieved by the different orientation of the respective monosaccharides in the binding sites (4). Thus, concanavalin A binds glucose and mannose such that the Od1 and Od2 of aspartic acid are hydrogen-bonded with the 6-OH and 4-OH, respectively, and the Nd2 of asparagine is bonded with the 4-OH of the sugar. On the other hand, in lectins (such as peanut agglutinin, ECorL and SBA) that bind galactose, Od1 and Od2 of aspartic acid form hydrogen bonds with the 4-OH and 3-OH, respectively, and the Nd2 of asparagine forms a hydrogen bond with the 3-OH.

Figure 2. Three-dimensional model of concanavalin A. (a) Subunit dimer obtained by the antiparallel side-by-side alignment of b-sheets, leading to the formation of a contiguous 12-stranded b-sheet that extends across the dimer interface. (b) Subunit tetramer formed by the association of the central parts of both dimers; the dimer contacts are mainly through loop interactions. Prepared using program MS I/BIOSYM ING, San Diego, CA. PDB entry 5CNA.

Three-dimensional model of concanavalin A. (a) Subunit dimer obtained by the antiparallel side-by-side alignment of b-sheets, leading to the formation of a contiguous 12-stranded b-sheet that extends across the dimer interface. (b) Subunit tetramer formed by the association of the central parts of both dimers; the dimer contacts are mainly through loop interactions. Prepared using program MS I/BIOSYM ING, San Diego, CA. PDB entry 5CNA.

Cereal lectins, a prominent example of which is WGA, consist too of two identical subunits, but they differ markedly from those of the legumes. For instance, they are exceptionally rich in cysteine residues, which are rare in legume lectins, are devoid of metals, and possess multiple binding sites (5). Although WGA is not glycosylated, its precursor is a glycoprotein. X-ray crystallography of the sugar complexes of WGA show that the combining site of this lectin contains several tyrosine residues that interact hydrophobically with the bound GlcNAc or NeuAc and also form hydrogen bonds with the ligand.

2.2. Animal Lectins

The galectins are a family of soluble animal lectins that bind exclusively b-galactosides, such as lactose and N-acetyllactosamine. They are found inside the cytoplasm and nucleus of cells, and occasionally also on the cell surface and outside the cell, and their expression is developmentally regulated. They are synthesized without a leader sequence and possess a relatively simple structure, occurring as monomers and homodimers of subunits with molecular weights of about 14 kDa, as well as larger polypeptides (30 to 35 kDa). Each galectin contains one or two copies of a homologous domain, known as the S-carbohydrate recognition domain (S-CRD). The tertiary structure of the galectins also exhibits the jellyroll topology found in the legume lectins, despite the absence of significant sequence homology and a different location of the combining site.

Numerous animal lectins are of the C-type (Ca -dependent). They are mosaic (or multi-domain) molecules, characterized by an extracellular carbohydrate recognition domain (C-CRD) consisting of 115 to 130 amino acid residues, of which 14 are invariant and 18 are highly conserved (6). To the CRD is attached a variable number of domains of different kinds, which form the bulk of the molecule, and also a membrane-spanning domain. Lectins of this class are grouped into three families: selectins, collectins and endocytic lectins. Only three selectins are known: E-selectin, P- selectin, and L-selectin. The first two are specific for the sialylated Lewisx blood group determinant, NeuAca(2-3)Galb(1-4)[Fuca(1-3)]GlcNAc), abbreviated sLex, and its positional isomer, sialylated Lewisa blood group determinant, NeuAca(2-3)Galb(1-3)[Fuca(1-4)]GlcNAc, or sLea, with both L-fucose and sialic acid (or another negatively-charged group such as sulfate) required for binding. L-Selectin binds sialylated, fucosylated, and sulfated oligosaccharides on diverse mucin-like glycoproteins. Recognition of the carbohydrates is possible only when they are present on particular glycoproteins, such as cell surface mucins, pointing to the importance of the carrier molecule and carbohydrate presentation in the interaction with the lectins.

The collectins are soluble proteins; they include the mannose-binding proteins MBP-A and C, the structural unit of which is a trimer of 32-kDa subunits based on a triple helix formed by the collagen-like portion of the molecule. MBP-A circulates in serum of, for example, rodents and humans as a hexamer of the trimeric units. As shown by X-ray crystallography, MBP-A and MBP-C bind mannose via Ca that serves as the nucleus of the combining site and interacts with the 3-OH and 4-OH of the ligand. Four of the five additional bonds that coordinate the metal ion are provided by the side chains of two glutamic acid and two asparagine residues that also are hydrogen bonded to the same (3 and 4) mannose hydroxyls. The four amino acids just mentioned are conserved in all C-type lectins specific for mannose, two of them in the sequence Glu-Pro-Asn (positions 185 to 187 in MBP-A). When Glu185 and Asn187 in MBP-A were replaced by glutamine and aspartic acid, respectively, as found in galactose specific C-type lectins, galactose became the preferred ligand (7).

A prominent representative of the endocytic lectins is the rabbit hepatic asialoglycoprotein receptor [or hepatic binding protein (HBP)], the first mammalian lectin to be described. It is found on hepatocytes of different mammals and is specific for galactose and N-acetylgalactosamine, whereas its avian homologue is specific for N-acetylglucosamine. Other endocytic lectins are a fucose- and galactose-specific receptor found on Kupffer cells and the mannose-specific lectin of macrophages and hepatic endothelial cells. Except for the latter, these lectins are type II transmembrane proteins, consisting of a short amino-terminal cytoplasmic domain, a hydrophobic, membrane-spanning domain, a neck region, and a carboxy-terminal CRD. The mannose-specific macrophage surface lectin differs from the other endocytic lectins primarily in that it is a type I transmembrane protein and that its extracellular part contains a domain, closest to the membrane, with eight CRDs.

The sialoadhesins (I-type lectins) are a family of sialic acid-specific type I membrane glycoproteins with variable numbers of extracellular immunoglobulin (Ig)-like domains and are thus members of the immunoglobulin superfamily (8). They include the sheep erythrocyte receptor for macrophages (referred to simply as sialoadhesin), the lymphocyte surface antigen CD22 found only on B cells, CD33 present on early myeloid cells, and MAG, a glycoprotein associated with myelin. CD22 recognizes specifically NeuAc(a2-6)Gal(b1-4)GlcNAc. In contrast, all other known I-type lectins bind structures containing N-acetylneuraminic acid that is a2-3-linked. The P-type CRD has been found only in two closely related lectins, the mannose-6-phosphate (Man-6-P) receptors.

2.3. Microbial Lectins

Several viruses (eg, influenza virus and polyoma virus) contain lectins specific for N-acetylneuraminic acid. Many bacterial species express surface lectins, usually in the form of fimbriae (or pili). These filamentous, heteropolymeric organelles, a few nanometers in diameter and 100 to 200 nm in length, consist of helically arranged subunits (pilins) of several different types. Only one of the subunits, usually a minor component of the fimbriae, possesses a carbohydrate-binding site— for example, for mannose (in type 1 fimbriae) or galabiose, Gala(1-4)Gal (in P fimbriae).

3. Functions

Participation of lectins in cell recognition (Table 2) was first demonstrated in the 1940s for the influenza virus hemagglutinin and in the 1970s for bacterial surface lectins. These lectins mediate the binding of the pathogens to host cells, a step essential for the initiation of infection. Inhibitors of bacterial lectins protect animals against experimental infection by the lectin-carrying organisms, providing a basis for the development of antiadhesion therapy of microbial infections (9). Some bacterial surface lectins allow the specific binding of the bacteria to human polymorphonuclear cells and human and mouse macrophages in the absence of opsonins, which may lead to activation of the phagocytes and ingestion and killing of the bacteria (lectinophagocytosis ).

Table 2. Functions of Lectins

Lectin

Role in

Microorganisms

Influenza virus

Infection

Amoeba

Infection

Bacteria

Infection

Animals

Calnexin

Glycoprotein synthesis

Galectins

Embryogenesis, metastasis (?)

C-type lectins

Mannose-binding

Host antimicrobial defense

protein

L-selectin

Lymphocyte homing

E-selectin

Leukocyte trafficking to sites of inflammation

P-selectin

Leukocyte trafficking to sites of inflammation

I-type lectins

Cell-cell interactions in the immune and neural

systems

Man-6-P receptors

Targeting of lysosomal enzymes

Natural killer cell lectins

Cytolysis of target cells

The galectins are postulated to function in cell adhesion. When present on the surface of metastatic murine and human cancer cells, they may be responsible for the adhesion of the cells to target organs, a step necessary for metastasis. Exposing highly metastatic cells to compounds containing lactose before injecting them into mice reduced the metastatic spread almost by half. Therefore, antiadhesive drugs may turn out to be antimetastatic.

The endocytic lectins have been assumed to facilitate clearance from the circulation of glycoproteins with complex oligosaccharide units (eg, ceruloplasmin and a j-acid glycoprotein) from which the terminal sialic acid has been removed, exposing the subterminal galactose. It is uncertain, however, whether this represents a physiological mechanism for regulating the turnover of serum glycoproteins (and cells), because disruption of the receptor does not result in decreased levels of desialylated forms of predominant circulating glycoproteins (10).

The mannose-6-phosphate receptors are responsible for targeting the appropriate enzymes to the lysosome subcellular compartment. The recently discovered intracellular lectins calnexin, MR60/ERGIC-53, and VIP-36 participate in the biosynthesis of glycoproteins, as well as in their intracellular sorting. The mannose-specific macrophage lectin has been implicated in innate antimicrobial defense (11). It binds infectious organisms that expose mannose-containing glycans on their surface, leading to their killing by lectinophagocytosis. The MBPs of mammalian serum and liver combine with oligomannosides of yeast and fungi, causing activation of complement in an antibody- and C1q-independent manner and subsequent lysis of the pathogens. They also enhance phagocytosis of the invading organisms by acting as opsinins . A mutation of a single amino acid residue in the collagen-like domain of the lectin is associated with recurrent, severe bacterial infections in infants.

The selectins provide the best paradigm for the role of sugar-lectin interactions in biological recognition (12). They mediate the initial adhesion of circulating leukocytes to endothelial cells of blood vessels, a prerequisite for the exit of the former cells from the circulation and their migration into tissues. L-Selectin is found on all leukocytes and is involved in the recirculation of lymphocytes, directing them specifically to peripheral lymph nodes. The two other selectins are expressed on endothelial cells, and only when these cells are activated by inflammatory mediators (eg, histamine, interleukin-2, and tumor necrosis factor ) released from tissue cells in response to, for example, infection or ischemia. Individuals unable to synthesize the selectin ligands sLex and sLea suffer from recurrent bacterial infections. Prevention of adverse inflammatory reactions by inhibition of leukocyte-endothelium interactions, another application of antiadhesion therapy, has become a major aim of many pharmacological industries. As shown in animal models, oligosaccharides recognized by the selectins protect against experimentally induced lung injury and tissue damage caused by myocardial ischemia and reperfusion. In addition to their involvement in inflammation, selectins may play a role in the spread of cancer cells from the primary tumor throughout the body.

3.1. Plant Lectins

The role of plant lectins is still not well understood. Plant lectins may function in the establishment of symbiosis between nitrogen-fixing bacteria, mainly rhizobia, and leguminous plants and in defense of plants against phytopathogenic fungi and predatory animals.

4. Applications

Native lectins are used predominantly for applications that are based on precipitation and agglutination reactions or for mitogenic stimulation. Lectins derivatized with fluorescent dyes, gold particles, or enzymes are employed in histo- and cytochemistry. Immobilized lectins, for instance on Sepharose, are indispensable for the isolation by affinity chromatography of glycoproteins, glycopeptides, and oligosaccharides. Mouse and human cortical (immature) thymocytes can readily be separated from the medullar (mature) ones with the aid of peanut agglutinin, making it possible to examine in vitro their developmental and functional relationships. SBA is used clinically for the removal, from bone marrow of haploidentical donors, of the mature T cells responsible for the lethal graft versus host reaction, affording a fraction enriched in stem cells. This fraction is employed routinely for transplantation into children born with severe combined immune deficiency ("bubble children") with close to 70% success, and experimentally for transplantation into leukemic patients. Another clinical application of lectins is in blood typing—for example, to identify blood type O cells and to differentiate between M and N cells.

Lectins are potent mitogens; they are polyclonal activators stimulating lymphocytes irrespective of their antigenic specificity. Phytohemagglutinin in particular serves clinically to assess the immunocompetence of patients—for example, under chemotherapy or in those with AIDS. It is also employed for the preparation of chromosome maps for karyotyping, sex determination, and detection of chromosome defects, because the chromosomes are easily visualized in the stimulated cells. Highly toxic lectins (such as ricin) and moderately toxic ones (such as concanavalin A, phytohemagglutinin, and WGA) serve for the selection of lectin-resistant cell mutants that are widely employed in studies of the genetics, biosynthesis, and functions of complex carbohydrates.

Next post:

Previous post: