**3. Grafted glycopolymers**

The various saccharide monomers have been reported, which were shown in Figure 2. There are various saccahride vinyl compounds. Styrene [18, 19], methacrylate [20], acrylate [21], acrylamide [22] and methacryl amide [23] with saccharide were reported. Living radical pol‐ ymerizations were reported with them. Norbornene saccharide derivatives were also report‐ ed, which provides the fine-tuned polymers via ring opening metastasis polymerization (ROMP) [24]. Reactive functional monomers were also utilized for glycopolymer synthesis. Monomers with acetylene [25] and active ester [26] were reported, where glycopolymers

An Integrated View of the Molecular Recognition and Toxinology - From Analytical Procedures to Biomedical

Saccharide recognition proteins are called lectin, which basically have multiple domain structures [27]. The multivalent saccharides gain in enthalpy due to multiple binding to sug‐ ar recognition sites, and gain in entropy due to the various binding modes. The glycopoly‐ mers are large sugar cluster to gain the Gibbs free energy in both enthalpy and entropy, and lectins have multiple and valuable structure, which is advantage for binding. The distance of sugar binding sites is different with each lectin, which is easily tuned by copolymeriza‐ tion. The density, distance, and the size of multivalent compounds can be easily adjusted by

The glycopolymers are water soluble polymers, which can be utilized as artificial polymeric ligands or polymer drugs. Choi et al reported polyacrylic acid with sialic acid, and the poly‐ mer efficiently inhibited the sialidase of Influenza virus [6] Kobayashi et al reported the var‐ ious glycopolymers. The lactose substituted polystyrene (poly(*N*-vinylbenzyl-O-β-Dgalactopyranosyl-(1-4)-D-gluconamide (PVLA)) strongly interacted with lectin, and it was applied for hepatocyte culture [18]. Polystyrene with sialyl lactose showed the strong bind‐ ing to influenza virus A [28]. Gestwicki et al prepared various glycopolymers via metastasis reaction, and reported the glycopolymers to bind lectin and *E. coli* [29]. Nishimura reported the glycopolymers interacting with glycosyl transferases to synthesize oligosaccharides, and

We defined glycopolymers as polymers with pendant saccharides. As we described above, the glycopolymers showed the strong multivalent effect based on the multivalency, with lec‐ tins, cells, viruses and bacteria. Another interesting property of glycopolymer is amphiphi‐ licity. Glycopolymers via addition polymerization have hydrophobic backbones with C-C bond, and are amphiphilic due to the hydrophilic side chain. The side chain of glycopoly‐ mer is bulky structure, and so the glycopolymer easily form the self-assembling structure. The structure of PVLA in aqueous solution was analyzed by small angle X-ray scattering [31], and it was found that PVLA formed rod-like structure, where the rod had the structure with long axis 10 nm and shot axis 5 nm.The rod-structure was similar to some polysacchar‐

It has been known that amylose is a host of hydrophobic compound as it is known starch-amylose complex. PVLA also became a host compound of hydrophobic substan‐ ces. We investigated the supramolecular polymer complex of PVLA. PVLA formed su‐

were obtained by polymerization and successive sugar addition.

Applications

458

copolymerization, which can be applied to variable lectins.

**2.2. Amphiphilic property of glycopolymer**

ides of amylose and sizofiran.

they developed the oligosaccharide synthesizer with glycopolymers [30].

Glycopolymer-coated substrates were facilely prepared by self-assembly of hydrophobic in‐ teraction. However, it is difficult to control the density of glycopolymer by self-assembling process. In addition, the physical adsorbed polymers were fragile in a specific solvent condi‐ tion. The coatings with spin-coat and Langmuir-Blodgett (LB) technique also provide the well-defined coating, but they are also fragile.

On the other hand, the surface-attached polymers are robust and practical to various pur‐ poses. In order to attach the polymer to the substrate, the covalent bond formation between polymers and substrates was necessary. The polymers with functional groups on the side chain and the polymer terminal were subjected to covalent bond formation with substrate. Those method is called "grafting-to" process. The grafting of the polymer was also reported via surface-initiated polymerization, which is called "grafting-from" method. The polymeri‐ zation was possible to start from the substrate by surface activation with γ-beam, VUV and plasma irradiation, and by the radical initiator immobilization. The properties of the sub‐ strates depend on the polymer density, thickness and flatness. The grafting polymers are categorized as "pancake", "mushroom", and "brush". Generally, the grafting to method provides the non-dense grafting substrate like "pancake" or "mushroom" and the grafting from method enables "polymer brush" structure [37].

**Figure 4.** Prepration of a glycopolymer modified nanoparticle via RAFT living radical polymerization. (a) Properties and

Molecular Recognition of Glycopolymer Interface

http://dx.doi.org/10.5772/54156

461

A glycopolymer modified gold nanoparticle stained pink color with peak top at 520 nm. ConA (α-Man recognition protein) was added to the α-Man-moidified nanoparticle solution, and the glycopolymer-gold nanoparticle showed the lectin recognition property. The color of the particle solution changed to blue, and the spectra showed the red-shift (Figure 4). The color changed occured based on the aggregation of nanoparticle by α-Man-lectin binding. The nanoparticle showed the affinity to a sugar recognition protein, and bacterium. The sug‐ ar recognition *E.coli* (ORN 178) was also added to the solution with glycopolymer-modified gold nanoparticle. The nanoparticle was adsorbed onto the periphery of *E.coli*, which was observed by TEM observation. On the other hand, *E.coli* without sugar recognition property (ORN 258) didn't show the change. The color change slowly occurred in 8 hours, while the color change with protein occurred quickly for 1-3 min. The color change occurred specifi‐

Sugar modified gold nanoparticles were reported by other groups. Otsuka et al reported lac‐ tose substituted gold nanoparticles with PEG linker [42]. The gold nanoparticle also showed red-shift by addition of lactose-recognition lectin. Narain et al reported nanoparticle of gly‐

Advantage of the glycopolymer-modified materials is the specific recognition and bioi‐ nert property. The detailed protein affinity was investigated with surface plasmon reso‐ nance of glyco-polymer-modified gold substrate. The glycopolymer-modified gold substrate had affinity constants of 10<sup>7</sup> (M-1) order, which was much stronger than the monovalent sugar of 103 (M-1) order. At the same time, the glycopolymer-modified sub‐ strate showed the highly specificity to proteins. The amount of specific protein bounds (α-Man-ConA) was more than 15 times larger than that of non-specific binding (BSA, fi‐ brinogen, and lysozyme) [44]. Interestingly, the glycopolymer-interface showed much better protein specificity than the artificial glycoplipid monolayers of self-assembled monolayer (SAM) and LB membrane. The hydrophilicity of the glycopolymer-modified

(b) a synthetic scheme of particle with a color change image of nanoparticle.

cally with the corresponding lectin and glycopolymer.

copolymer having biocompatibility [43].

gold substrate contributed the bioinert property.

### **3.1. Glycopolymer-grafted nanoparticle via RAFT polymerization**

In order to prepare the polymer-grafted materials, the living radical polymerizations are ac‐ tively utilized by many groups. Living radical polymerization provides the uniform poly‐ mer, and polymer terminals can be modified. Atom-transfer-radical-polymerization (ATRP) enabled the dense-polymer brush. Since living radical polymer has active terminal end, the polymer terminal is possible to be modified. Specially, the polymers via RAFT process have the active terminal end with dithio- or trithioester. The polymer terminal with reversible-ad‐ dition-fragmentation chain-transfer polymerization (RAFT) is converted to thiol by reduc‐ tion or hydrolysis. Thiol is highly reactive and relates to various reaction like thiol-ene reaction, thiol-maleimide coupling, and Au-S interaction [38].

The glycopolymer conjugates have been synthesized via RAFT polymerization. Mancini et al reported a protein with glycopolymer via RAFT polymerization and disulfide bond for‐ mation [39]. Narain et al reported the prepration of particle by RAFT polymerization and subsequent Au-S bond formation [40].

We prepared the glycopolymer with *p*-amidophenyl glycosides (α-Man, β-Gal and β-GlcNAc) and acrylamide via RAFT process with (thiobenzoyl)thioglycolic acid [41]. The pol‐ ydispersities were below 1.5 in spite of random copolymer. The obtained glycopolymers had dithioester terminal, which was reduced thiol by addition of NaBH4. The thiol-terminated glycopolymers were mixed with gold nanoparticle solution. The gold nanoparticle (40 nm) was successfully modified by glycopolymer, which was confirmed by TEM observation and zeta-potential measurement. The glycopolymer modified gold nanoparticle was water solu‐ ble and stably dispersed for more than half a year.

process. In addition, the physical adsorbed polymers were fragile in a specific solvent condi‐ tion. The coatings with spin-coat and Langmuir-Blodgett (LB) technique also provide the

An Integrated View of the Molecular Recognition and Toxinology - From Analytical Procedures to Biomedical

On the other hand, the surface-attached polymers are robust and practical to various pur‐ poses. In order to attach the polymer to the substrate, the covalent bond formation between polymers and substrates was necessary. The polymers with functional groups on the side chain and the polymer terminal were subjected to covalent bond formation with substrate. Those method is called "grafting-to" process. The grafting of the polymer was also reported via surface-initiated polymerization, which is called "grafting-from" method. The polymeri‐ zation was possible to start from the substrate by surface activation with γ-beam, VUV and plasma irradiation, and by the radical initiator immobilization. The properties of the sub‐ strates depend on the polymer density, thickness and flatness. The grafting polymers are categorized as "pancake", "mushroom", and "brush". Generally, the grafting to method provides the non-dense grafting substrate like "pancake" or "mushroom" and the grafting

In order to prepare the polymer-grafted materials, the living radical polymerizations are ac‐ tively utilized by many groups. Living radical polymerization provides the uniform poly‐ mer, and polymer terminals can be modified. Atom-transfer-radical-polymerization (ATRP) enabled the dense-polymer brush. Since living radical polymer has active terminal end, the polymer terminal is possible to be modified. Specially, the polymers via RAFT process have the active terminal end with dithio- or trithioester. The polymer terminal with reversible-ad‐ dition-fragmentation chain-transfer polymerization (RAFT) is converted to thiol by reduc‐ tion or hydrolysis. Thiol is highly reactive and relates to various reaction like thiol-ene

The glycopolymer conjugates have been synthesized via RAFT polymerization. Mancini et al reported a protein with glycopolymer via RAFT polymerization and disulfide bond for‐ mation [39]. Narain et al reported the prepration of particle by RAFT polymerization and

We prepared the glycopolymer with *p*-amidophenyl glycosides (α-Man, β-Gal and β-GlcNAc) and acrylamide via RAFT process with (thiobenzoyl)thioglycolic acid [41]. The pol‐ ydispersities were below 1.5 in spite of random copolymer. The obtained glycopolymers had dithioester terminal, which was reduced thiol by addition of NaBH4. The thiol-terminated glycopolymers were mixed with gold nanoparticle solution. The gold nanoparticle (40 nm) was successfully modified by glycopolymer, which was confirmed by TEM observation and zeta-potential measurement. The glycopolymer modified gold nanoparticle was water solu‐

well-defined coating, but they are also fragile.

Applications

460

from method enables "polymer brush" structure [37].

**3.1. Glycopolymer-grafted nanoparticle via RAFT polymerization**

reaction, thiol-maleimide coupling, and Au-S interaction [38].

subsequent Au-S bond formation [40].

ble and stably dispersed for more than half a year.

**Figure 4.** Prepration of a glycopolymer modified nanoparticle via RAFT living radical polymerization. (a) Properties and (b) a synthetic scheme of particle with a color change image of nanoparticle.

A glycopolymer modified gold nanoparticle stained pink color with peak top at 520 nm. ConA (α-Man recognition protein) was added to the α-Man-moidified nanoparticle solution, and the glycopolymer-gold nanoparticle showed the lectin recognition property. The color of the particle solution changed to blue, and the spectra showed the red-shift (Figure 4). The color changed occured based on the aggregation of nanoparticle by α-Man-lectin binding. The nanoparticle showed the affinity to a sugar recognition protein, and bacterium. The sug‐ ar recognition *E.coli* (ORN 178) was also added to the solution with glycopolymer-modified gold nanoparticle. The nanoparticle was adsorbed onto the periphery of *E.coli*, which was observed by TEM observation. On the other hand, *E.coli* without sugar recognition property (ORN 258) didn't show the change. The color change slowly occurred in 8 hours, while the color change with protein occurred quickly for 1-3 min. The color change occurred specifi‐ cally with the corresponding lectin and glycopolymer.

Sugar modified gold nanoparticles were reported by other groups. Otsuka et al reported lac‐ tose substituted gold nanoparticles with PEG linker [42]. The gold nanoparticle also showed red-shift by addition of lactose-recognition lectin. Narain et al reported nanoparticle of gly‐ copolymer having biocompatibility [43].

Advantage of the glycopolymer-modified materials is the specific recognition and bioi‐ nert property. The detailed protein affinity was investigated with surface plasmon reso‐ nance of glyco-polymer-modified gold substrate. The glycopolymer-modified gold substrate had affinity constants of 10<sup>7</sup> (M-1) order, which was much stronger than the monovalent sugar of 103 (M-1) order. At the same time, the glycopolymer-modified sub‐ strate showed the highly specificity to proteins. The amount of specific protein bounds (α-Man-ConA) was more than 15 times larger than that of non-specific binding (BSA, fi‐ brinogen, and lysozyme) [44]. Interestingly, the glycopolymer-interface showed much better protein specificity than the artificial glycoplipid monolayers of self-assembled monolayer (SAM) and LB membrane. The hydrophilicity of the glycopolymer-modified gold substrate contributed the bioinert property.

### **3.2. Biosensing with glycopolymer-modified nanoparticles**

We investigated the biosensing of the glycopolymer-modified gold nanoparticles.

First, the gold nanoparticles have been applied for biotechnology as a marker. We applied the glycopolymer-modified gold nanoparticle for lateral flow assay (immune-chromatogra‐ phy), where we tested the properties of particle with target analyte of lectin (ConA) [45]. Anti-ConA antigen was immobilized on the nitrocellulose strip, and the detection of target ConA was investigated with glycopolymer-modified gold nanoparticle.

Target protein of ConA was detected by the pink color of gold nanoparticle. We tested the glycopolymer with varying sugar ratio of 0, 6, 12 and 50 %. In terms of red-shift, the glyco‐ polymer with higher sugar ratio (50 %) exhibited more red-shift. However, the nanoparticle with higher sugar ratio (50 %) was not appropriate for lateral flow assay. The gold nanopar‐ ticle with higher sugar ratio aggregated at the bottom line with addition of ConA. The gly‐ copolymer with modest sugar content (6 %) exhibited the best indicator of ConA. The glycopolymer with modest sugar content was more flexible than that with higher sugar con‐ tent, which improves the sensitivity in lateral flow assay. What is interesting about lateral flow assay is the biosensing with naked eye, using a simple device. The detection of ConA was possible from 1 nM level with naked eye.

**Figure 5.** (a) Schematic Illustration of glycopolymer brush for protein and pathogen removal. (b) The amount of pro‐

We synthesized the glycopolymer with α-Man and trimethoxysilane units, and the glyco‐ polymer was immobilized onto the porous siliceous materials via Si-O-Si bond [48] (Fig‐ ure 5). The radius of the porous materials was 2 μm, which was much larger than the size of proteins and viruses. The porous membrane was connected to flow channel, and the protein solutions (ConA and BSA) was injected to the flow. ConA was selectively ad‐ sorbed onto the porous membrane, but BSA passed through the membrane due to α-Man-ConA interaction. On the other hand, the porous membrane adsorbed proteins by

Li et al reported the filter preparation with sialyl-lactose modified chitosan [49]. The modi‐ fied chitosan took up the influenza virus. The solution containing influenza virus A was passed through the filter, and the amount of virus was reduced about 1/200. The chitosan filter without sialyl –lactose didn't remove influenza virus. The influenza virus showed the affinity to sialyl-lactose via hemagglutinin. Muschin et al also reported the virus removal by

Bio-separation with nanomaterials was investigated. Nagatsuka et al reported the protein separation with glycopolymer-modified magnetite. The glycopolymer with lactose modified magnetite was prepared by biotin-streptavidin reaction. The toxic protein of ricin solution was mixed with lactose-substituted nanoparticle [50]. The ricin was separated with maginet‐

The affinity between saccharide and protein was strongly affected by multivalency. There‐ fore, the precise multivalent compound is useful to fabricate the efficient ligand and to clari‐

ic. El-Boubbou et al separated sugar recognition *E.coli* with a similar manner [51].

. The modification of glycopolymer provides the affinity to specific pro‐

, and that of BSA

Molecular Recognition of Glycopolymer Interface

http://dx.doi.org/10.5772/54156

463

non-specific interaction. The amount of ConA bound was 34 nmol/m2

tein and the bioinert properties to other proteins.

**4. Glyco-interface with precise structure**

tein adsorbed on the glycopolymer brush.

sulfated curdlan modified filter.

was 4.2nmol/m2

Electrochemical biosensing was also conducted with glycopolymer-modified gold nanopar‐ ticle [46]. The gold nanoparticle was assembled on anti-ConA antigen immobilized elec‐ trode. The amount of protein bound was estimated by the electrohemical signal of gold nanoparticle, where the gold nanoparticle was electrochemically reduced in differential pulse voltammetry. The amount of ConA bounds were more sensitively monitored than that with lateral flow assay. The detection limit was around 0.1 nM.

These experiments were conducted using the model target of ConA. Since the proteinsaccharide interactions are involved in various infection diseases, the detection of serious disease like influena and cancer will be realizable with the corresponding saccharide modified particle.

#### **3.3. Protein separation with glycopolymer materials**

The glycopolymer-modified interface showed specific affinity biomolecules, which can be applied not only for biosensing but also for protein purification devices. We modified the porous filter membrane with glycopolymer grafting, and prepared protein purification de‐ vice. Basically, the purification and removal of specific biomacromolecules are mainly con‐ ducted by the size-exclusion process. For example, bacteria are able to be removed by filtration, which are called "sterile filtration". The size of bacteria was μm order, and so the porous materials with μm order pore are applied for sterilization. However, the size of pro‐ teins and viruses are nm level, which is difficult to apply the size-exclusion way. In addi‐ tion, the filtration speed is strongly dependent on the radius of porous materials, and the flux speed of nano-level porous materials were too slow to use it practically. Therefore, it is almost impossible to attain the protein purification by nm porous membrane, and the affini‐ ty purification is appropriate to the purification and removal of protein and viruses [47].

**3.2. Biosensing with glycopolymer-modified nanoparticles**

was possible from 1 nM level with naked eye.

modified particle.

Applications

462

with lateral flow assay. The detection limit was around 0.1 nM.

**3.3. Protein separation with glycopolymer materials**

We investigated the biosensing of the glycopolymer-modified gold nanoparticles.

An Integrated View of the Molecular Recognition and Toxinology - From Analytical Procedures to Biomedical

ConA was investigated with glycopolymer-modified gold nanoparticle.

First, the gold nanoparticles have been applied for biotechnology as a marker. We applied the glycopolymer-modified gold nanoparticle for lateral flow assay (immune-chromatogra‐ phy), where we tested the properties of particle with target analyte of lectin (ConA) [45]. Anti-ConA antigen was immobilized on the nitrocellulose strip, and the detection of target

Target protein of ConA was detected by the pink color of gold nanoparticle. We tested the glycopolymer with varying sugar ratio of 0, 6, 12 and 50 %. In terms of red-shift, the glyco‐ polymer with higher sugar ratio (50 %) exhibited more red-shift. However, the nanoparticle with higher sugar ratio (50 %) was not appropriate for lateral flow assay. The gold nanopar‐ ticle with higher sugar ratio aggregated at the bottom line with addition of ConA. The gly‐ copolymer with modest sugar content (6 %) exhibited the best indicator of ConA. The glycopolymer with modest sugar content was more flexible than that with higher sugar con‐ tent, which improves the sensitivity in lateral flow assay. What is interesting about lateral flow assay is the biosensing with naked eye, using a simple device. The detection of ConA

Electrochemical biosensing was also conducted with glycopolymer-modified gold nanopar‐ ticle [46]. The gold nanoparticle was assembled on anti-ConA antigen immobilized elec‐ trode. The amount of protein bound was estimated by the electrohemical signal of gold nanoparticle, where the gold nanoparticle was electrochemically reduced in differential pulse voltammetry. The amount of ConA bounds were more sensitively monitored than that

These experiments were conducted using the model target of ConA. Since the proteinsaccharide interactions are involved in various infection diseases, the detection of serious disease like influena and cancer will be realizable with the corresponding saccharide

The glycopolymer-modified interface showed specific affinity biomolecules, which can be applied not only for biosensing but also for protein purification devices. We modified the porous filter membrane with glycopolymer grafting, and prepared protein purification de‐ vice. Basically, the purification and removal of specific biomacromolecules are mainly con‐ ducted by the size-exclusion process. For example, bacteria are able to be removed by filtration, which are called "sterile filtration". The size of bacteria was μm order, and so the porous materials with μm order pore are applied for sterilization. However, the size of pro‐ teins and viruses are nm level, which is difficult to apply the size-exclusion way. In addi‐ tion, the filtration speed is strongly dependent on the radius of porous materials, and the flux speed of nano-level porous materials were too slow to use it practically. Therefore, it is almost impossible to attain the protein purification by nm porous membrane, and the affini‐ ty purification is appropriate to the purification and removal of protein and viruses [47].

**Figure 5.** (a) Schematic Illustration of glycopolymer brush for protein and pathogen removal. (b) The amount of pro‐ tein adsorbed on the glycopolymer brush.

We synthesized the glycopolymer with α-Man and trimethoxysilane units, and the glyco‐ polymer was immobilized onto the porous siliceous materials via Si-O-Si bond [48] (Fig‐ ure 5). The radius of the porous materials was 2 μm, which was much larger than the size of proteins and viruses. The porous membrane was connected to flow channel, and the protein solutions (ConA and BSA) was injected to the flow. ConA was selectively ad‐ sorbed onto the porous membrane, but BSA passed through the membrane due to α-Man-ConA interaction. On the other hand, the porous membrane adsorbed proteins by non-specific interaction. The amount of ConA bound was 34 nmol/m2 , and that of BSA was 4.2nmol/m2 . The modification of glycopolymer provides the affinity to specific pro‐ tein and the bioinert properties to other proteins.

Li et al reported the filter preparation with sialyl-lactose modified chitosan [49]. The modi‐ fied chitosan took up the influenza virus. The solution containing influenza virus A was passed through the filter, and the amount of virus was reduced about 1/200. The chitosan filter without sialyl –lactose didn't remove influenza virus. The influenza virus showed the affinity to sialyl-lactose via hemagglutinin. Muschin et al also reported the virus removal by sulfated curdlan modified filter.

Bio-separation with nanomaterials was investigated. Nagatsuka et al reported the protein separation with glycopolymer-modified magnetite. The glycopolymer with lactose modified magnetite was prepared by biotin-streptavidin reaction. The toxic protein of ricin solution was mixed with lactose-substituted nanoparticle [50]. The ricin was separated with maginet‐ ic. El-Boubbou et al separated sugar recognition *E.coli* with a similar manner [51].
