Patent Application
20150253336
Glycosylation is one of the most abundant and biologically significant post-translational modifications to occur in cells. It is a highly complex non-template driven process that results in the addition of oligosaccharide moieties to a variety of biomolecules. Cell surfaces, both prokaryotic and eukaryotic, are covered in a dense layer of complex oligosaccharide structures that are attached to proteins and lipids. This layer is called the “glycocalyx” and the nature of the glycans displayed can be organism and cell type specific. As the interface between a cell and its environment it is not surprising that interactions between these glycans with carbohydrate binding proteins, called lectins, mediate a vast array of biological processes, play a central role in the orchestration of the immune system and mediate interactions between cells and various infectious agents such as prions, viruses and microorganisms. The glycans present on proteins also have a very significant impact on their physiochemical properties and biological activity. Changes in the glycans presented on glycoproteins or cell surfaces can result in, or be indicative of, changes in the physiological status of a cell or signify the development of a disease state (1-6) including many types of cancer (7-10) and autoimmune disorders such as rheumatoid arthritis (11-14). Many biopharmaceutical molecules are also glycosylated proteins and the glycans attached to these products can have a significant impact on their safety and efficacy. Given the biological significance of glycosylation, there is a requirement for methods that enable efficient isolation of glycosylated biomolecules and informative glycoanalysis of biomolecules and cell surfaces.
Lectins are carbohydrate proteins that are capable of recognizing and binding reversibly to specific carbohydrate structures. They display exquisite specificity for their cognate glycans and their ability to discriminate between different glycan structures has been exploited for many years for glycoanalytical applications. Their ability to bind to glycans in situ on proteins and cell surfaces, without the need for prior release and derivatization, makes them particularly attractive when compared to alternative MS and HPLC based approaches as these treatments can often result in the loss of significant biological data. When immobilized to solid support matrices, lectins can be used to effect the separation and purification of glycosylated molecules. Lectin affinity chromatography is often used as a preliminary step to isolate or separate oligosaccharides, glycopeptides, glycoproteins and glycoprotein glycoforms to facilitate their identification and characterisation.
The most commonly used lectins are plant lectins and these have traditionally exhibited significant problems, particularly with respect to product quality and performance. Many plant lectins are purified from source material, due to incompatibility with recombinant production methods, and this results in batch to batch variations and variability from one supplier to another (5, 15, 16). Production methods usually generate relatively low yields and final products are expensive which has meant that lectins have been restricted to analytical scale applications where only small quantities are required (16).
Prokaryotic lectins offer new opportunities for the development of superior glycoselective bioaffinity tools but, to date, they have been relatively underexploited. They usually exhibit greater affinities for their glycan targets and less structural complexity than plant lectins (17). They are also more amenable to recombinant production, particularly in Escherichia coli, which simplifies production but also opens up opportunities for the development of novel enhanced recombinant prokaryotic lectins (RPL's) with diversified and optimized binding properties (18-20).
We will demonstrate herein how the carbohydrate binding properties of the α-galactophilic PA-IL protein, from the opportunistic pathogen Pseudomonas aeruginosa (21-23), were significantly altered through random mutagenesis of specific amino acid residues in the proteins carbohydrate binding site. We will generate a series of novel RPL's that exhibit specificity and high affinity for glycoprotein targets displaying lactosamine (LacNAc) and demonstrate that binding was dependent on terminal β1,4-linked galactose. Lactosamine is commonly displayed as part of glycan structures found on cell surfaces and as part of the antenna of N-linked glycans displayed on glycoproteins including serum IgG's where it is important for the ability of these molecules to elicit CDC (complement dependent cytotoxicity) and ADCC (antibody dependent cellular cytotoxicity) effector functions (11, 12, 14,24). RPL's with specificity for LacNAc therefore represent potentially valuable tools for glycoselective applications throughout the life sciences.
These novel RPL's carried multiple simultaneous substitutions in the carbohydrate binding site of the PA-IL (Pseudomonas aeruginosa lectin 1 or Pseudomonas aeruginosa lectin I) protein. As a result, it was difficult to fully determine the specific contribution of individual substitutions to the observed carbohydrate binding properties of the mutant PA-IL proteins. In this work, we also undertook a progressive site directed mutagenesis approach to assess the significance of specific amino acid residues in dictating binding specificity and affinity and, through in silico modelling, we explored the potential structural basis for the observed carbohydrate binding properties. In doing so, we identified optimal amino acid substitutions that promote specific carbohydrate binding activities and produced an array of novel RPL's with diverse carbohydrate binding activities that will be of use for a broad spectrum of glycoselective applications.
In a first embodiment, there is provided a peptide analogue of PA-IL of SEQ ID NO: 1, wherein the peptide analogue has altered carbohydrate binding specificity, and wherein the peptide analogue comprises an amino acid substitution at one, two or three of positions 50, 52 and 53, wherein the amino acid substitution at position 50 is selected from the group consisting of Ala, Val, Leu, Ile, Met, Phe, Trp, Pro, Ser, Thr, Cys, Tyr, Gly, Asn, Asp, Gln, Glu, Lys, and Arg; optionally from the group consisting of Ala, Val, Leu, Phe, Pro, Ser, Thr, Gly, Asn, Asp, Gln, Glu, Lys, and Arg; and further optionally from the group consisting of Ala, Val, Leu, Ser, Thr, Gly, Asn, Gln, Glu, Lys and Arg; wherein the amino acid substitution at position 52 is selected from the group consisting of Ser, Thr, Cys, Asn, Gln, Asp, Glu, Lys, Arg and His; optionally from the group consisting of Thr, Cys, Asn, Arg and His; and further optionally from the group consisting of Asn, Thr, Arg and His; and wherein the amino acid substitution at position 53 is selected from the group consisting of Ala, Val, Leu, Ile, Met, Phe, Trp, Pro, Gly, Ser, Thr, Cys, Tyr, Asn, Asp, Glu, Lys, Arg and His; optionally from the group consisting of Ala, Val, Leu, Gly, Ser, Tyr, Asn, Asp, Glu, Lys, Arg and His; and further optionally from the group consisting of Ala, Val, Leu, Gly, Ser, Asn, Asp, Glu, Lys, Arg and His.
Optionally, the peptide analogue has improved binding to a carbohydrate having a terminal galactose, optionally a terminal β1,4-linked galactose, over PA-IL of SEQ ID NO: 1 and wherein the peptide analogue comprises an amino acid substitution at position 50 selected from Ala, Val, Leu, Ile, Met, Pro, Ser, Thr, Cys, Asn, Gln, Glu, Lys and Arg; optionally from the group consisting of Ala, Val, Leu, Pro, Ser, Thr, Asn, Gln, Glu, and Lys; and further optionally from the group consisting of Asn, Gln, Glu, and Val. Further optionally, the peptide analogue comprises Asn at position 50 and an amino acid substitution at position 53 selected from the group consisting of Ala, Val, Leu, Ile, Met, Gly, Ser, Thr, Asn, Asp, Glu, Lys, Arg and His; optionally from the group consisting of Ala, Val, Leu, Gly, Ser, Asn, Asp, Glu, Lys, Arg and His; and further optionally from the group consisting of Ala, Val, Gly, Ser, Lys, Arg and His. Alternatively, the peptide analogue comprises Asn at position 50 and Gly at position 53; the peptide analogue optionally comprising Gln, Asp, Glu or Asn at position 52; the peptide analogue further optionally comprising Asn at position 52. Further alternatively, the peptide analogue comprises Val at position 50 and an amino acid substitution at position 53 selected from the group consisting of Ala, Val, Leu, Ile, Met, Gly, Ser, Thr, Asn, Asp, Glu, Lys, Arg and His; optionally from the group consisting of Ala, Val, Leu, Gly, Ser, Asn, Asp, Glu, Lys, Arg and His; and further optionally from the group consisting of Ala, Val, Gly, Ser, Lys, Arg and His.
Optionally, the peptide analogue has altered carbohydrate binding specificity for a carbohydrate having a terminal α-linked galactose over PA-IL of SEQ ID NO: 1 and wherein the peptide analogue comprises an amino acid substitution at position 50 selected from Ala, Val, Leu, Ile, Met, Ser, Thr, Cys, Asn, Gln, Glu, Lys and Arg; optionally from the group consisting of Ala, Val, Leu, Ser, Thr, Asn, Gln, Glu, Lys and Arg; and further optionally from the group consisting of Val, Leu, Asn, Gln and Lys. Further optionally, the peptide analogue comprises Asn at position 50 and an amino acid substitution at position 53 selected from the group consisting of Ala, Val, Leu, Ile, Met, Gly, Ser, Thr, Cys, Asn, Asp, Glu, Arg, Lys and His; optionally from the group consisting of Ala, Val, Leu, Gly, Ser, Asn, Asp, Glu, Arg, Lys and His; and further optionally from the group consisting of Ala, Ser, Gly, Arg, Lys and His.
Optionally, the peptide analogue has enhanced carbohydrate binding specificity for a carbohydrate having a terminal α-linked galactose over PA-IL of SEQ ID NO: 1 and wherein the amino acid substitution at position 53 is selected from the group consisting of Asn, Asp, Glu, Arg and His; and optionally from the group consisting of Glu and Arg.
Optionally, the peptide analogue has altered carbohydrate binding specificity for a carbohydrate having a terminal α-linked galactose over PA-IL of SEQ ID NO: 1 and wherein the peptide analogue comprises Val at position 50 and an amino acid substitution at position 53 selected from the group consisting of Ala, Val, Leu, Ile, Met, Phe, Trp, Pro, Gly, Ser, Thr, Cys, Tyr, Asn, Asp, Glu, Lys, Arg and His; wherein optionally the peptide analogue comprises Val at position 50 and an amino acid substitution at position 53 selected from the group consisting of Ala, Val, Leu, Gly, Ser, Tyr, Asn, Asp, Glu, Lys, Arg and His and further optionally, the peptide analogue comprises Val at position 50 and an amino acid substitution at position 53 selected from the group consisting of Arg and Lys.
Optionally, the peptide analogue has altered carbohydrate binding specificity for a carbohydrate having a terminal α-linked galactose over PA-IL of SEQ ID NO: 1 and wherein the peptide analogue comprises Gln at position 50 and an amino acid substitution at position 53 selected from the group consisting of Ala, Val, Leu, Ile, Met, Phe, Trp, Pro, Gly, Ser, Thr, Cys, Tyr, Asn, Asp, Glu, Lys, Arg and His; wherein optionally the peptide analogue comprises Gln at position 50 and an amino acid substitution at position 53 selected from the group consisting of Ala, Val, Leu, Gly, Ser, Tyr, Asn, Asp, Glu, Lys, Arg and His and further optionally, the peptide analogue comprises Gln at position 50 and an amino acid substitution at position 53 selected from the group consisting of Arg and Lys.
Optionally, the peptide analogue has enhanced carbohydrate binding specificity for a carbohydrate having a terminal β- or a α-linked galactose over PA-IL of SEQ ID NO: 1 and wherein the peptide analogue comprises an amino acid substitution at position 50 selected from Asn, Leu and Gln.
Optionally, the peptide analogue comprises an amino acid substitution at position 50 selected from Ala, Val, Leu, Phe, Pro, Gly, Ser, Thr, Asn, Gln, Asp, Glu, Lys, and Arg.
Optionally, the peptide analogue comprises Asn at position 50 and an amino acid substitution at position 53 is selected from the group consisting of Ala, Val, Leu, Gly, Ser, Tyr, Asn, Asp, Glu, Lys, Arg and His.
Optionally, the peptide analogue comprises an amino acid substitution at position 53 selected from Glu, Lys and Arg.
Optionally, the peptide analogue comprises Val at position 50 and an amino acid substitution at position 53 selected from the group consisting of Ala, Val, Leu, Gly, Ser, Tyr, Asn, Asp, Glu, Lys, Arg and His.
Optionally, the peptide analogue comprises an amino acid substitution at position 52 selected from the group consisting His, Asn, Cys, Thr and Arg and, further optionally, an amino acid substitution at position 50 selected from the group consisting of Leu, Thr, Val, Asn, Gly and Pro and an amino acid substitution at position 53 selected from the group consisting of Arg, Glu, Ser, Gly, Leu and Asn.
Optionally, the carbohydrate, for which the peptide analogue of the present invention has altered carbohydrate binding specificity, is a carbohydrate or is selected from the group consisting of glycoprotein, glycoconjugate and cell surface. Further optionally, the glycoprotein, glycoconjugate and cell surface comprises an oligosaccharide or a polysaccharide linked to a protein or other conjugate.
In a second embodiment, there is provided a method for detecting changes in the glycosylation of a carbohydrate that is optionally selected from the group consisting of glycoprotein, glycoconjugate and cell surface, the method comprising qualitatively or quantitatively assessing terminal galactosylation using a peptide analogue of any aspect of the first embodiment. These methods find utility for detecting changes in the presence of, or exposure of, terminal α- and β-linked galactose, whether for purification or analytical work, or for diagnostic purposes.
In a third embodiment, there is provided a method of separating and isolating/purifying biomolecules/cells comprising a glycoprotein or glycoconjugate, the method comprising contacting the peptide analogue of any aspect of the first embodiment with a solution or suspension containing biomolecules/cells; separating any biomolecules/cells not bound by the peptide analogue and, optionally, subsequently recovering biomolecules/cells bound by the peptide analogue by disassociating them from the peptide analogue.
The abbreviations used are: PA-IL, Pseudomonas aeruginosa lectin 1; rPA-IL, recombinant PA-IL; ELLA, enzyme linked lectin assay; iGb3, isoglobotriaosylceramide (Gal-α1,3-Gal-β1,4-Glc); PBS, Phosphate Buffered Saline; TBS, Tris Buffered Saline; TBST, Tris Buffered Saline with Tween 20; IPTG, Isopropyl-β-D-thiogalactopyranoside.
The glycoconjugate used as a representative of glycoproteins displaying glycans with terminal β-linked galactose was BSA-LacNAc (Gal-β1,4-GlcNAc-BSA). Lectins representative of those showing binding to a terminal β-linked galactose include ECL (Erythrina cristagalli Lectin) and RCA (Ricinus communis Agglutinin).
The glycoconjugate used as a representative of glycoproteins displaying glycans with terminal α-linked galactose was BSA-αGal (Gal-α1,3-Gal-BSA). The lectin used as a representative of those showing binding to a terminal α-linked galactose was GSLI (Griffonia simplicifolia isolectin B4).
Plasmid Construction—pQE30PA-IL & pQE60PA-IL—
All strains and plasmids used or constructed as part of this study are listed and described in Table A1 set out below. The lecA gene encoding the PA-IL protein was amplified from Pseudomonas aeruginosa PAO1 (This strain can be obtained from a wide variety of sources including many cell culture banks) genomic DNA by PCR to facilitate cloning into the pQE series of E. coli expression vectors from Qiagen. PCR reactions were carried out using high fidelity Phusion Taq and PCR conditions recommended by the manufacturer (New England BioLabs). The lecA gene was amplified using the PA-IL-F1 and PA-IL-R1 primers (Table A2 below) to generate a product that could be cloned as a BamHI-HindIII fragment into the pQE30 expression vector. The resulting plasmid, pQE30PA-IL (
Escherichia coli
Pseudomonas aeruginosa
Protein Expression and Purification—
For protein expression plasmids were transformed into the protease deficient E. coli strain KRX (30). Expression clones were cultured at 30° C. in Terrific Broth (TB) broth and protein expression induced by addition of IPTG to a final concentration of 50 μM. Cells were harvested by centrifugation and cell pellets resuspended in lysis buffer [10 mM NaH2PO4, 300 mM NaCl, 40 mM imidazole, pH 8.0). Cell disruption was achieved by high pressure using a Constant Systems™ cell disrupter and cell debris was removed by centrifugation. Clarified cell lysates were applied to 10 mL IMAC™ columns (IMAC Hypercel from Pal) and a high stringency wash buffer with 100 mM imidazole was used to remove non-specifically bound contaminating proteins. The desired 6HIS tagged proteins were ultimately eluted using 250 mM imidazole and eluted proteins were aliquoted and stored at −80° C. in the elution buffer. Typical yields were around 200 mg per 250 mL starting culture. Purified proteins were analysed by SDS-PAGE to assess purity (
Gel Permeation Chromatography (GPC)—
The estimated molecular weights of the rPA-IL proteins were determined by GPC, which was performed on a Superdex™ 75 10/300 GL column (GE Healthcare) using an AKTA Purifier 100 FPLC system. The molecular weight of commercially obtained untagged PA-IL (Sigma Aldrich) was also experimentally determined and used for comparison with 6HIS tagged rPA-IL proteins to enable determination of their quaternary structure (
Hemagglutination Assays—
The hemagglutination assay is widely used to study lectin activity and is dependent on the multi-valency typically displayed by lectins. The assay was essentially performed according to the method described by Garber et al (31). The assay was performed using Papain treated Rat red blood cells (RBC's), obtained from the Bioresource unit at DCU such that the final concentration of cells in reaction wells was 3.5% w/v. Lectins to be tested were prepared in TBS (20 mM Tris, 150 mM NaCl, 1 mM CaCl2, 1 mM MnCl2, 1 mM MgCl2, pH 7.6). Hemagglutination was observed, after 1 hour incubation at 25° C., as a thin film of cells coating the bottom of wells in U-bottomed 96 well plates compared to a concentrated spot of sedimented cells observed in negative controls to which no lectin was added (
General Enzyme Linked Lectin Assay (ELLA) Method—
The Gal-α1,3-Gal-BSA (BSA-αGal) and Gal-β1,4-GlcNAc-BSA (BSA-LacNAc) glycoconjugates used were from Dextra Laboratories. Biotinylated plant lectins GSLI (Griffonia simplicifolia isolectin B4), ECL (Erythrina cristagalli Lectin), RCA (Ricinus communis Agglutinin), SNA (Sambucus nigra Agglutinin), MALII (Maackia amurensis Lectin) and LCA (Lens culinaris Agglutinin) were from Vector Laboratories. The glycoproteins fetuin, asialofetuin and invertase were from Sigma Aldrich while asialotransferrin, agalactotransferrin and agalactofetuin were generated by treatment using glycosidases, neuraminidase (Clostridium perfingens) and β1,4-galactosidase (Bacteroides fragilis), in accordance with manufacturer's guidelines (New England Biolabs). ELLA's were essentially performed according to the method described by Thompson et al (2011) (33). More specifically, glycoproteins were prepared in PBS and typically immobilized at a concentration of 5 μg mL−1. For qualitative ELLA's, lectins were assayed at a concentration of 10 μg mL−1 in TBST (20 mM Tris, 150 mM NaCl, 0.05% Tween-20, 1 mM CaCl2, 1 mM MnCl2, 1 mM MgCl2, pH 7.6). For lectin dose response experiments, each lectin was evaluated at a range of concentrations prepared by serial 1:2 dilution of an initial lectin solution of 10 μg mL−1 to a final concentration of 156 ng mL−1. Binding of 6HIS tagged rPA-IL proteins was detected after 1 hour incubation at 25° C. using a HRP conjugated anti-HIS antibody diluted 1:10,000 in TBST (Sigma Aldrich). Biotinylated plant lectins were detected using a HRP conjugated anti-biotin antibody diluted 1:10,000 in TBST (Sigma Aldrich).
Lectin Affinity Constant Determination by ELLA—
Affinity constants were determined according to the method described by Kirkeby et al 2002 (27). ELLA's were performed using a constant concentration of 2 μg mL−1 for the rPA-ILNm proteins and 4 μg mL−1 of ECL to ensure all lectins were evaluated at equimolar concentrations. Each lectin was evaluated against BSA-LacNAc glycoconjugate immobilized at a range of concentrations from 10 μg mL−1 to 19.5 ng mL−1 (prepared by serial 1:2 dilution of a 10 μg mL−1 stock). The resulting glycoconjugate dose response curve obtained enabled the calculation of Bmax and the affinity constant KD for each lectin against BSA-LacNAc. Bmax is defined as being the maximum plateau value of absorption and represents the maximum number of lectin binding sites expressed in the units of the Y-axis (AU). KD is defined as being the glycoconjugate concentration required to fill half of the available lectin binding sites at equilibrium. KD is therefore the glycoconjugate concentration that generates a signal equivalent to half Bmax and the unit for KD is nanograms of glycoconjugate. These values are specific for the defined experimental conditions used.
Site Directed Random Mutagenesis of the rPA-ILN Protein—
PCR based site directed mutagenesis of the lecA gene was achieved through whole vector amplification in which the pPC30PA-IL vector was used as the parental template DNA. Whole vector amplification was achieved using the primers PA-ILmutF and PA-ILmutR (Table A2 above). These primers were 5′ phosphorylated and designed to anneal within the lecA sequence with their 5′ ends exactly next to each other. The reverse primers were designed to overlap the region to be mutagenized enabling the introduction of mutations through manipulation of reverse primer sequences (
Fabrication of Lectin Affinity Resins—
lectin affinity resins were prepared by immobilization onto cyanogen bromide (CNBr) activated Sepharose 4B prepared according to the manufactures guidelines (GE Healthcare). Lectin to be immobilized was buffer exchanged into coupling buffer (20 mM NaH2PO4, 500 mM NaCl, pH 8.5) and the lectin solution mixed with the CNBr Sepharose at a concentration of approximately 30 mg mL−1 of resin. This mixture was then left mixing by inversion overnight at 4° C. Unbound protein was then decanted and un-reacted CNBr groups on the resin were capped with 1 M ethanolamine in coupling buffer and pH 8.5. This was left mixing by inversion at room temperature for 4 hours. Finally non-specifically bound protein was removed by 4 successive washes with coupling buffer and actetate buffer (100 mM NaOAc, 500 mM NaCl, pH 4.0). Resins were ultimately washed with TBS and for long term storage sodium azide was added to a final concentration of 2 mM.
Evaluation of Lectin Affinity Resins—
lectin affinity resins were evaluated by performing small scale pull down assays in 1.5 mL eppendorf tubes or by packing into 1 mL FPLC cartridges (FliQ Column Housings, Generon) to enable easy connection to FPLC systems. For pull down assays, 50 μL of lectin affinity resin was mixed with 100 μL it of a test protein mixture and mixed by inversion for 1 hour. Unbound protein was then removed using a pipette and the resin washed with three 1 mL aliquots of TBST. Bound protein was eluted by addition of 100 μL of TBST with 0.5 M galactose followed by incubation for 1 hour. Lectin affinity FPLC columns were connected to an AKTA Purifier 100 FPLC system. Columns were equilibrated using TBS and typically run at a flow rate of 0.5 mL per minute. Samples were prepared in TBS and 2 mL sample volumes were injected onto the 1 mL lectin affinity columns. Bound glycoproteins were eluted using 0.5M galactose prepared in TBS.
Production of Affinity Tagged Recombinant PA-IL (rPA-IL)—
Commercially available untagged PA-IL (PA-ILU) is typically purified by exploiting its natural affinity for Sepharose 4B (23,34) but alteration of the protein's carbohydrate binding specificity could prevent purification in this manner. One of the first steps required for this study was, therefore, the incorporation of an affinity tag that would enable simple and rapid purification of recombinant PA-IL (rPA-IL) proteins independently of glycan binding specificity. The lecA gene, encoding the wild type PA-IL protein, was therefore cloned into two E. coli plasmid expression vectors to enable expression of rPA-IL with either an N-terminal (rPA-ILN) or C-terminal (rPA-ILC) polyhistidine (6HIS) tag and thereby enabling purification by IMAC. Both proteins were expressed to relatively high levels in the soluble cytoplasmic fraction of E. coli KRX, from which they were subsequently purified by IMAC. Assessment of the proteins by SDS-PAGE verified that both exhibited very high levels of purity (
Structural and Functional Assessment of Poly-Histidine Tagged rPA-IL Proteins—
Incorporation of a poly-histidine tag into any protein can have an impact on the structure, activity and other physiochemical properties (solubility, stability) of the protein. We therefore assessed the impact of the incorporated 6HIS tags on the quaternary structure and functionality of the rPA-IL proteins. Gel permeation chromatography (GPC) was used to experimentally determine the molecular weight of commercially obtained untagged PA-IL (PA-ILU), which is known to be a tetramer of four identical subunits under physiological conditions (
Selection of a Target rPA-IL Molecule for Mutagenesis Studies—
The steric accessibility of the 6HIS tags within the quaternary structures of the rPA-IL proteins was assessed by ELLA. Both of the rPA-IL proteins were tested for their ability to bind an immobilized BSA-αGal glycoconjugate with subsequent detection of the bound lectins using an anti-HIS antibody (33). Biotinylated GSL-I, a plant lectin with a binding specificity for terminal α-galactose (27,35), was included in the assays as a positive control. Binding of both rPA-ILN and GSLI could be detected in ELLA's with high sensitivity but binding of the rPA-ILC could not be detected using an anti-HIS antibody, even at relatively high lectin concentrations of 10 μg mL−1 (not shown in
Random Mutagenesis of the rPA-ILN Protein—
Residues from three separate parts of the PA-IL monomer are involved in coordinating calcium and binding of the iGb3 trisaccharide Gal-α1,3-Gal-β1,3-Glc (23) (
The pQE30PA-IL plasmid was mutagenized using a PCR based method that resulted in the introduction of random simultaneous substitutions at positions corresponding to residues His50, Asp52 and Gln53 in the wild type native PA-IL protein. An array of mutant clones, containing 154 individual mutants, was prepared and cell lysates from these clones were analysed by ELLA. A number of different glycoproteins were used as immobilized targets in ELLA screens to identify clones expressing mutated rPA-ILN proteins (rPA-ILNm) exhibiting altered binding specificities compared to the parental rPA-ILN protein. Glycoproteins tested included fetuin (3 N-linked and 3 O-linked glycan structures highly sialylated with terminal α2,3 and α2,6-Neu5Ac) (36), asialofetuin (terminal LacNAc and terminal β1,3-Gal) and invertase (high mannose structures) (37). PA-IL is known not to bind strongly to these glycoprotein targets (28) and so any rPA-ILNm proteins identified as generating altered responses to these targets were considered to have altered carbohydrate binding properties. Of the 154 clones screened, none of the rPA-ILNm proteins displayed binding to invertase or significant binding to fetuin but a number of rPA-ILNm proteins were observed to exhibit altered binding to asialofetuin (data not shown). Nine of the rPA-ILNm proteins that exhibited the highest signals against these glycoproteins in the ELLA screen were selected and recovered from the array for further analysis. The selected rPA-ILNm proteins were named according to the well from which they were recovered in the original mutant array (i.e. rPA-ILNmE6 was recovered from row E and column 6).
The initial ELLA mutant library screen had been performed using soluble fractions of cell lysates and therefore differences observed in ELLA responses may have been impacted by differences in the expression levels of specific rPA-ILNm proteins or differences in the overall concentrations of cell lysates. To validate the results of the initial ELLA screen, the selected rPA-ILNm proteins were first purified by IMAC and then re-evaluated in qualitative ELLA's against an expanded set of glycoproteins and specifically generated glycoprotein glycoforms (
Identification of Amino Acid Substitutions in Selected rPA-ILNm Proteins—
Plasmid DNA from each of the rPA-ILNm expressing clones was isolated and sequenced to determine the nature of the amino substitutions present in each protein (
Lectin Dose Response Curves Against Defined Glycoconjugate Targets—
The specificity of the selected rPA-ILNm proteins was further assessed by generating lectin dose response curves against two specific glycoconjugate targets; BSA-LacNAc and BSA-αGal (
The parental rPA-ILN protein only showed very weak binding to the BSA-LacNAc glycoconjugate, even at relatively high lectin concentrations of 10 μg mL−1 (not shown in
Lectin Dose Response Curves on Natural Glycoproteins Glycoforms—
Lectin dose response curves were also generated against asialotransferrin and asialofetuin (
Determination of Affinity Constants for rPA-ILNm Proteins—
A relative affinity constant, KD, was calculated for selected rPA-ILNm proteins against the BSA-LacNAc glycoconjugate according the method described by Kirkeby et al 2002. KD is defined as being the concentration of the glycoconjugate required to fill half of the available lectin binding sites at equilibrium. If a lectin has a high affinity for the glycoconjugate, then the KD will be low as it will take a lower concentration of glycoconjugate to bind half of the lectin molecules. The calculated KD for rPA-ILNmE6, rPA-ILNmF3 and ECL for BSA-LacNAc were 4 ng, 6.3 ng and 21 ng, respectively (
Application of Immobilized rPA-ILNmE6 for Glycoprotein and Glycoform Isolation—
to evaluate the ability of rPA-ILNmE6 to be used for selective glycoprotein and glycoform isolation and purification. The lectin was immobilized onto CNBr activated Sepharose 4B. The lectin readily immobilized at high densities and lectin immobilization densities of approximately 20 mg mL−1 of resin were reproducibly achieved. To evaluate the ability of this lectin affinity resin to isolate glycoproteins displaying terminal LacNAc, we first performed simple lectin pull down assays in eppendorf tubes using a test protein mixture prepared by mixing asialotransferrin, glucose oxidase (displays high mannose), cytochrome C and carbonic anhydrase (both non-glycosylated). Fractions of unbound and bound protein were ultimately evaluated by SDS-PAGE (
In the present study, we have demonstrated how the carbohydrate binding specificity of the α-galactophilic PA-IL protein could be significantly altered through random mutagenesis of specific amino acid residues in its binding site. We identified a number of novel RPL's exhibiting specificity and high affinity for glycoproteins displaying LacNAc and an affinity for this glycan epitope significantly greater than that of commercially available plant lectin ECL (
The Role of Specific Amino Acid Substitutions in Dictating Carbohydrate Binding Properties—
Of the rPA-ILNm proteins identified, the rPA-ILNmE6 protein appeared to exhibit a very high relative affinity for BSA-LacNAc. While this protein also showed some capacity to bind to the BSA-αGal glycoconjugate, it only did so relatively weakly when compared to its response to the BSA-LacNAc glycoconjugate (
The responses of another group of rPA-ILNm proteins indicated that a H50V substitution could also support high affinity binding to LacNAc. This substitution was present in rPA-ILNmF3, rPA-ILNmB10 and rPA-ILNmG3 and all of these proteins were observed to bind strongly to the BSA-LacNAc glycoconjugate. Analysis against asialofetuin and asialotransferrin again indicated however that the rPA-ILNmB10 and rPA-ILNmF3 proteins displayed a higher relative affinity for LacNAc than the rPA-ILNmG3, although not as high as that displayed by the rPA-ILNmE6 protein. The rPA-ILNmF3 and rPA-ILNmB10 proteins also displayed strong binding to the BSA-αGal, albeit not as strong as the parental rPA-ILN protein, and this was not observed for the rPA-ILNmG3 protein. As these three proteins only differ from each other by possessing different substitutions at positions 52 and 53, this again demonstrates that amino acids at these positions play a role in further defining the specificity and affinity of the rPA-ILNm proteins. The rPA-ILNmB10 and rPA-ILNmF3 proteins actually only differ from each other at one position, carrying a Q53E and a Q53R substitution respectively. The Q53R substitution was also present in rPA-ILNmB4 which also binds well to the BSA-αGal. These data indicated that substitution of Gln53 with a basic (Arg) or acidic (Glu) residue could be linked with a higher affinity for α-linked galactose.
Another interesting observation was that, while the lectin dose response curves for rPA-ILNmF3 and rPA-ILNmB10 proteins on asialotransferrin did not increase as rapidly as that of rPA-ILNmE6, indicative of a lower relative affinity, they ultimately reached a higher absorbance plateau indicating a greater final density of these proteins bound to the surface. However, we had observed that both of these proteins had a tendency to form aggregates when high protein concentration stock solutions, stored at −80° C., were being defrosted. These aggregates generally went back into solution when samples were fully thawed but all samples were centrifuged to ensure removal of any residual protein aggregates prior to use. These proteins occasionally also generated high signals in negative control wells. This was also observed for the rPA-ILNmE12 protein, albeit more consistently, leading to it being excluded from further analysis. The one common feature of all three of these proteins was the occurrence of a D52C substitution and it is possible that this residue could mediate protein aggregation at high lectin concentrations through disulfide bond formation. As a result, the higher saturation signals obtained for these lectins on asialotransferrin may be the result of binding of protein aggregates formed at high lectin concentrations.
Novel Glycoanalytical Tools for Applications in the Life Sciences—
Lectins have found widespread applications within the field of glycobiology and have been implemented in a diverse range of formats to characterise the glycosylation status, and to detect changes in glycosylation, of biomolecules. Changes in the glycosylation of proteins or cell surfaces can be concurrent with, and indicative of, a change in the physiological status of a cell or the development of a disease state and can therefore be used as a means of diagnosis (1-5). LacNAc is an important glycan epitope commonly displayed on cell surfaces and as part of the antenna of complex N-linked glycan structures of glycoproteins. For example, serum IgG's, unlike many other serum glycoproteins, are not heavily sialylated and the N-linked glycans present in the Fc region of the glycoproteins usually bear biantennary glycans terminating in LacNAc (11,24). A reduction in terminal β1,4 galactosylation of these N-linked glycans has been diagnostically linked with a number of autoimmune disorders including rheumatoid arthritis while increased galactosylation indicates remission of the disease (11-14,40). The RPL's we have developed were clearly demonstrated to be capable of sensitively detecting glycoproteins displaying terminal LacNAc and of being capable of differentiating between different glycoprotein glycoforms (
Many biopharmaceutical products are glycosylated molecules and variations in glycosylation of bio-therapeutics can have a very significant impact on a products physiochemical properties, efficacy, and immunogenicity (11,44-47). Sialylation of some bio-therapeutics, such as Erythropoietin (EPO), can have a significant impact on their physiochemical properties, blood retention and overall efficacy (44,46,48,49). Monitoring of sialylation of these products is often an important determinant in the production of these products and methods using lectins, such as ECL, to monitor for changes in sialylation have been reported in the literature (48,50). Monoclonal antibodies (MAb's) represent a very significant and rapidly growing class of biotherapeutics (11,24). The N-linked glycans in the Fc region of MAb's are usually terminated in galactose and these glycans are essential for the ability of MAb's to elicit ADCC (Antibody Dependent Cellular Cytotoxicity) and CDC (Complement Dependent Cytotoxicity) effector functions vital for their efficacy (11,12,24,40,51). As with the analysis of other glycoproteins, LAC can enable more efficient analysis and characterisation of glycosylated biotherapeutics. With their specificity for LacNAc, the RPL's reported here could be particularly useful in the analysis of MAb's to determine the extent of terminal galactosylation which is often a major source of heterogeneity in these products (12). In addition to the many potential analytical scale applications, the ability to readily scale the production of our novel RPL's, could also enable them to ultimately overcome the many barriers that have limited the application of other eukaryotic lectins and enable them to be applied at a production scale, in a way analogous to Protein A, for the selective purification of optimal biotherapeutic glycoforms to produce safer more efficacious drugs.
Site Directed Mutagenesis—
PCR based site directed mutagenesis of the lecA gene, which encodes the PA-IL protein, was achieved as described in Example A. The pQE30PA-IL vector is an Escherichia coli expression vector which expresses the rPA-ILN protein (recombinant PA-IL protein with an N-terminally positioned hexa-histidine (6HIS) affinity purification tag) and this was used as a template for whole vector amplification (See Example A—Table A1). Whole vector amplification was achieved using 5′ phosphorylated primers designed to anneal within the lecA sequence with their 5′ ends exactly next to each other. The reverse primers were designed to overlap the region to be mutagenized. This enabled the introduction of mutations through manipulation of reverse primer sequences while the sequence of the forward primer, PA-ILmutF, was kept constant (Table B2). Successful PCR reactions were purified and subjected to digestion with the restriction enzyme Dpnl to selectively digest the parental template vector DNA. Digestions were ultimately run on agarose gels and PCR products corresponding to the expected size of linear vector were gel extracted. The final purified PCR products, with blunt phosphorylated ends, were re-circularised by simple self-ligation and the ligated DNA transformed into E. coli strain KRX. Typically three transformants were picked into overnight 10 mL Terrific Broth (TB) cultures supplemented with 50 μM IPTG and expression of mutant rPA-ILN proteins (rPA-ILNm) confirmed by SDS-PAGE analysis of total cellular protein. Plasmid DNA was isolated from clones expressing proteins of the expected size and successful introduction of the desired mutations confirmed by DNA sequencing (MWG-Eurofins). All of the plasmids used in this study are described in Table B1 and all of the primers used are described in Table B2.
Protein Expression and Purification—
For protein expression, plasmids were transformed into the protease deficient E. coli strain KRX (30). Expression clones were cultured in Terrific Broth (TB). Cultures were grown at 37° C. with shaking at 200 rpm until an optical density of 0.6 at 600 nm was reached and then induced by addition of IPTG to a final concentration of 50 Cultures were then placed at 30° C. with shaking at 200 rpm for overnight incubation. Cells were harvested by centrifugation and cell pellets resuspended in lysis buffer (10 mM NaH2PO4, 300 mM NaCl, 40 mM imidazole, pH 8.0). Cell disruption was achieved by high pressure using a Constant Systems cell disrupter and cell debris was removed by centrifugation. Clarified cell lysates were applied to 10 mL IMAC columns (IMAC Hypercel from Pal) and a high stringency wash buffer with 100 mM imidazole was used to remove non-specifically bound contaminating proteins. The desired 6HIS tagged proteins were ultimately eluted using 250 mM immidizole and eluted proteins were aliquoted and stored at −80° C. in the elution buffer. Typical yields were around 200 mg per 250 mL starting culture. Purified proteins were analysed by SDS-PAGE to assess purity and routinely buffer exchanged and concentrated using Vivaspin centrifugal membrane devices (Sartorius-Stedim), with a molecular weight cut off of 10 kDa, according to the manufacturer's guidelines.
General Enzyme Linked Lectin Assay (ELLA) Method—
The Gal-α1,3-Gal-BSA (BSA-αGal) and Gal-β1,4-GlcNAc-BSA (BSA-LacNAc) glycoconjugates used were from Dextra Laboratories and presented on average 20 glycan moieties per BSA molecule. Biotinylated plant lectins GSL-I (Griffonia simplicifolia isolectin B4), ECL (Erythrina cristagalli Lectin), RCA (Ricinus communis Agglutinin), SNA (Sambucus nigra Agglutinin) and MALII (Maackia amurensis Lectin) were from Vector Laboratories. Human transferrin was from Sigma Aldrich and asialotransferrin (AsT) was generated by treatment using neuraminidase (Clostridium perfingens) in accordance with manufacturer's guidelines (New England Biolabs). ELLA's were essentially performed according to the method described by Thompson et al (2011) (33). More specifically, glycoproteins were prepared in PBS and typically immobilized at a concentration of 5 μg mL−1. For qualitative ELLA's lectins were assayed at a concentration of 2 μg mL−1 in TBST (20 mM Tris, 150 mM NaCl, 0.05% Tween-20, 1 mM CaCl2, 1 mM MnCl2, 1 mM MgCl2, pH 7.6). For lectin dose response experiments, each lectin was evaluated at a range of concentrations prepared by serial 1:2 dilution of an initial lectin solution of 2 μg mL−1 to a final concentration of 31 ng mL−1. Binding of 6HIS tagged rPA-IL proteins was detected after 1 hour incubation at 25° C. using a HRP conjugated anti-HIS antibody diluted 1:10,000 in TBST (Sigma Aldrich). Biotinylated plant lectins were detected using a HRP conjugated anti-biotin antibody diluted 1:10,000 in TBST (Sigma Aldrich).
Protein Structural Modelling and Image Rendering—
In silico analysis of the PA-IL protein and its carbohydrate binding site was carried out using the PDB file 2VXJ (23). Structural models were generated using Deep View (Swiss Model) (25) and models generated were ultimately rendered using the CCP4MG software (26).
In Example A, we constructed an expression vector, pQE30PA-IL, enabling the expression of rPA-IL with an N-terminally positioned poly-histidine tag (rPA-ILN). This poly-histidine (6HIS) tag enabled rapid and simple purification of the rPA-ILN protein by IMAC and therefore independently of its carbohydrate binding specificity. Positioning of the poly-histidine tag at the N-terminus of the rPA-IL protein was found to disrupt the quaternary structure of the protein resulting in the formation of homodimers rather than the native tetrameric configuration adopted by the wild type untagged PA-IL protein. Despite this, the rPA-ILN protein, and derived rPA-ILNm proteins, were demonstrated in Example A to be active and binding could be detected with high sensitivity in ELLA's against defined glycoprotein targets. The pQE30PA-IL vector was therefore selected as the target DNA molecule for mutagenesis studies undertaken in this work. As all of the rPA-ILNm generated would exhibit an equivalent quaternary structure to the parental rPA-ILN protein, comparative analysis of carbohydrate binding specificity and affinity could be performed by ELLA to assess the impact of specific amino acid substitutions.
The His50 Residue is Critical for Dictating the Binding Specificity of rPA-ILN—
Work conducted in Example A indicated that the His50 residue in the binding pocket of the PA-IL protein played a critical role in determining the carbohydrate binding specificity of the protein. To independently examine the role of this amino acid residue, it was substituted with 13 alternative amino acids selected to be representative of the full spectrum of potential amino acid prosperities (Table B1 above). All of the resulting rPA-ILNm proteins were successfully expressed in E. coli and purified by IMAC with the exception of one carrying a H50D mutation, which was found to be insoluble. An rPA-ILNm protein carrying a H50P substitution was observed to express weakly in E. coli and, when purified protein was assessed by SDS-PAGE, it generated multiple high molecular weight bands indicating that it potentially formed aggregates (data not shown). The carbohydrate binding properties of each of the remaining 12 successfully purified His50 substituted rPA-ILNm proteins was qualitatively assessed by performing ELLA's against two specific BSA glycoconjugate targets; Gal-α1,3-Gal-BSA (BSA-αGal) and Gal-β1,4-GlcNAc-BSA (BSA-LacNAc) (
Specific His50 Substitutions Result in High Affinity Binding to Terminal β1,4-Linked Galactose
Initial screens identified a number of His50 substitutions that generated rPA-ILNm proteins exhibiting strong binding to BSA-LacNAc (
Of all of the His50 substitutions made, the H50N protein exhibited the highest relative affinity for BSA-LacNAc. This was only slightly weaker than that observed for the random mutant rPA-ILNmE6 protein which carries the same H50N substitution but also carries two additional D52N and Q53G amino acid substitutions (
His50 Substitutions Result in Reduced Binding Affinities for α-Linked Galactose—
All of the His50 substitutions negatively impacted on the ability of rPA-ILNm proteins to bind to the BSA-αGal glycoconjugate. Proteins carrying either a H50Q or a H50K substitution exhibited the strongest binding to the BSA-αGal glycoconjugate (
The H50N protein generated significant binding signals against BSA-αGal (
The Q53 and D52 Residues Play a Role in Modulating Carbohydrate Binding Specificity and Affinity—
Analysis of the rPA-ILNm proteins with single His50 substitutions, and comparison with the sugar binding properties of closely related randomly mutated rPA-ILN proteins, clearly indicated that the Gln53 and D52 residues played a role in further modulating the carbohydrate binding properties of the rPA-ILNm proteins. We first examined the impact of Gln53 substitutions when made in conjunction with a H50N substitution. The plasmid encoding the H50N protein was mutagenized using primers specifically designed to introduce an additional specific amino acid substitution in place of the Gln53 residue (Table B2 above). This generated twelve new expression vectors, each expressing an rPA-ILNm protein with the H50N substitution in combination with one of 12 different amino acid substitutions in place of the Gln53 residue (Table B1). One of the resulting proteins, H50N:Q53Y, was found to be insoluble and was not characterised further. The remaining 11 mutants were evaluated as before by performing ELLA analysis against BSA-LacNAc, BSA-αGal and AsT (
Q53R Substitutions Promote Binding to α-Linked Galactose—
Comparison of the binding specificities of the H50T and H50V proteins with that of the rPA-ILNmB4 and rPA-ILNmF3 proteins respectively, suggested that a Q53R substitution might promote binding to terminal α-linked galactose. We therefore introduced a Q53R substitution into the H50V protein and the resulting H50V:Q53R protein was observed to bind more strongly to the BSA-αGal glycoconjugate (
The above results suggested that a Q53R substitution could promote binding to glycans with terminal α-linked galactose. However, neither the H50Q:Q53R protein nor the H50V:Q53R proteins bound to the BSA-αGal glycoconjugate as strongly as the rPA-ILN protein due to substitution of the His50 residue (
In Example B, we set out to independently assess the roles of the His50, Asp52 and Gln53 residues in the carbohydrate binding site of the rPA-ILN protein in dictating and modulating its carbohydrate binding properties. This was achieved through extensive site directed mutagenesis to introduce specific amino acid substitutions in place of these residues and subsequent evaluation of the carbohydrate binding specificity and affinity of each of the resulting proteins. In doing so, we also aimed to identify specific amino acid substitutions that promoted specifically enhanced carbohydrate binding activities.
The Role of His50 in Defining the α-Galactophilic Selectivity of the PA-IL Protein—
The PA-IL protein has been shown to be α-galactophilic with a preference for glycans displaying α1,4-linked terminal galactose (23). X-ray crystal structures of the protein have been obtained with bound D-galactose and α-galactophilic ligands (23,29). In all of the structures obtained to date, the terminal galactose is bound in the same orientation and this is likely due to the large number of interactions between it, the coordinated calcium and specific amino acid side chains in the binding pocket (
The Impact of His50 Substitutions on the Carbohydrate Binding Specificity and Affinity of rPA-ILN—
Our earlier work herein had indicated that substitution of the His50 residue was particularly critical in generating lectins capable of binding with high affinity to glycans displaying LacNAc and terminal β1,4-linked galactose. In Example B, we assessed the role of this residue in dictating carbohydrate binding properties by introducing 13 independent specific amino acid substitutions in its place. Initial qualitative screens of these rPA-ILNm proteins verified that substitution of this residue alone could significantly alter the carbohydrate binding specificity and affinity of the protein. Our results also demonstrated that observed changes in carbohydrate binding activities were not simply due to the alleviation of steric restraints imposed by the His50 residue in the carbohydrate binding site as they were dependent on the His50 substitutions made. Some amino acid substitutions simply had a deleterious impact on the overall carbohydrate binding activity of proteins. However, a number of specific amino acid substitutions generated proteins capable of binding with high affinity to glycans with terminal β1,4-linked galactose. Among these were proteins carrying H50N and H50V substitutions which had also been observed in rPA-ILNm proteins we generated through random mutagenesis in our earlier study herein. Also of particular interest was the H50Q protein, which exhibited a dual specificity binding to both BSA-αGal and BSA-LacNAc glycoconjugates. Through the generation of in silico structural models of these proteins, we explored the potential structural basis for the observed carbohydrate binding specificities of these proteins.
The Carbohydrate Binding Properties of the H50N Protein—
The H50N protein exhibited the highest relative affinity for the BSA-LacNAc glycoconjugate of all of the His50 substitutions made (
The Carbohydrate Binding Properties of the H50V Protein—
The H50V protein was also observed to bind strongly to BSA-LacNAc albeit not as strongly as H50N (
The Carbohydrate Binding Properties of the H50Q Protein—
The H50Q protein displayed strong binding to the BSA-LacNAc but it had a significantly lower relative affinity for this glycoconjugate than either the H50N or H50V proteins (
The H50Q protein exhibited stronger binding to BSA-αGal in ELLA's than any of the other proteins with single His50 substitutions (
The Role of Gln53 and Asp52 in Modulating Carbohydrate Binding Activities—
Characterisation of the H50N, H50V and H50T proteins, and comparison with the carbohydrate binding activities of rPA-ILNmE6 (and rPA-ILNmC5), rPA-ILNmF3 and rPA-ILNmB4 respectively, clearly indicated that that additional Asp52 and Gln53 substitutions play a role in further modulating binding carbohydrate binding specificities and affinities. The Asp52 residue does not participate in forming productive interactions with bound iGb3 in the wild type PA-IL binding site (
Comparison of the binding properties of the H50V and H50T proteins with those of the rPA-ILNmF3 and rPA-ILNmB4 respectively implied that a Q53R substitution could promote binding to glycans with terminal α-linked galactose. To explore this, we introduced a Q53R substitution into the H50V protein to generate a H50V:Q53R double mutant that therefore only differed from the rPA-ILNmF3 protein by a single D52C substitution. ELLA analysis demonstrated that the resulting H50V:Q53R double mutant did bind to BSA-αGal and that its affinity for this glycoconjugate was comparable to that of the rPA-ILF3 protein (
Final Conclusions—
This work successfully demonstrated the critical role that the His50 residue plays in dictating the specificity of the PA-IL protein. We clearly demonstrated that substitution of this residue alone was sufficient to significantly alter the carbohydrate binding properties of the protein. The observation that only specific amino acid substitutions promoted high affinity binding to glycans with LacNAc, and terminal β1,4-linked galactose, demonstrated that this was not simply due to alleviation of steric restraints that might be imposed by the His50 residue in the carbohydrate binding site of the protein. Through the use of structural models generated in silico, we were able to explore the potential structural basis for the carbohydrate binding specificities and affinities displayed by a number of rPA-ILNm proteins. We also demonstrated that both Gln53 and Asp53 substitutions played significant roles in further modulating the binding specificities and affinities of proteins. Predictive structural models could not explain the differences in the carbohydrate binding properties of the rPA-ILNmE6 protein compared to those of the H50N and H50N:Q53G proteins. These may be due to conformational changes in structure of the carbohydrate site induced by substitution of the Asn52 and Gln53 residues that could not be predicted and so verification of this will require future solving of the structure of these proteins. However, it is also clear from the results obtained that the final carbohydrate binding properties of rPA-ILNm proteins is the result of the combined effects of substitutions at His50, Asn52 and Gln53.
Many of the novel lectins generated in this study will be of use for glycoanalytical applications. While proteins like rPA-ILNmE6 would be of use for the detection of terminal β1,4 linked galactose, and LacNAc, others like the H50E protein could provide further biologically relevant information about a sample as binding is potentially dependant on the density and spatial distribution of glycans. The H50Q, with its dual specificity for terminal α-linked or β-linked galactose, could be used for general detection of terminal galactose while the Q53R and Q53E proteins, which display enhanced affinity for terminal α-linked galactose could be used to detect the presence of this potentially immunogenic sugar moiety. Inclusion of these novel recombinant prokaryotic lectins (RPL's) into any of the currently evolving glycoanalytical platforms, such as lectin microarrays, would significantly expand the utility of these platforms. If immobilized to solid support matrices, these RPL's may also facilitate enhanced glycoselective separations and the purification of glycoproteins and biotherapeutic molecules.
The invention is not limited to the embodiments described herein but can be amended or modified without departing from the scope of the present invention.
| Number | Date | Country | Kind |
|---|---|---|---|
| 12188702.0 | Oct 2012 | EP | regional |
This application is a continuation of U.S. application Ser. No. 13/933,195, filed Jul. 2, 2013, which claims priority of European patent application no. 12188702.0, filed Oct. 16, 2012, the disclosures of which are incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 13933195 | Jul 2013 | US |
| Child | 14718355 | US |
