Bioresponsive hydrogels for sensing applications

Abstract

This Highlight presents some of the recent efforts in the design of bioresponsive hydrogels, and their application to biosensing. These efforts extend philosophically from early work on glucose responsive gels, with current studies being focused both on gel responsivity and transduction of that response such that true sensor applications can be realized. The future outlook for the field is also discussed.

---

Grant R. Hendrickson

Grant Hendrickson received his B.S. and M.S. in chemistry from Furman University working under Prof. Lon B. Knight studying matrix-isolated radical species using ESR. He started his doctoral work at Georgia Institute of Technology in 2006 under the supervision of Prof. L. Andrew Lyon. His research is directed towards designing and synthesizing responsive microgel microlenses for biosensing applications.

L. Andrew Lyon

Dr Lyon began his career at Georgia Tech in January of 1999 following his postdoctoral work with Prof. Michael J. Natan at Penn State and Ph.D. work at Northwestern University under the direction of Prof. Joseph T. Hupp. He was promoted to Associate Professor in 2003 and Professor in 2007. Dr Lyon's current research involves the design of stimuli-sensitive hydrogel nanoparticles for biomedical applications.

---

1 Introduction – bioresponsive hydrogels

Responsive materials comprise an attractive field of study due to their possible applications in tissue replacement,^1–6 biological coating technologies,^7–9 drug delivery,^10–26 and biosensing.^21–24 Hydrogels in particular are interesting for these types of applications due to the flexibility of their networks in aqueous media, the degree to which their mechanical and chemical properties can be tuned, and therefore their intrinsic compatibility with biological systems. Importantly, the tunability of the dynamics of these constructs is critical for the responsiveness, and therefore the bioresponsivity, of such materials. Some inherent responsivity can be engineered into a gel based simply on the choice of monomers, co-monomers, and cross-linkers used in synthesis. The most common responsive hydrogels either incorporate a thermo-responsive monomer that shows a phase change at a given temperature, or a pH-responsive monomer that exhibits more or less intermolecular electrostatic repulsion due to the pH and salt content of the surrounding media. These two stimuli are advantageous for some applications, but for higher order responsiveness to biological stimuli, it is desirable to design responsive materials sensitive to antigen–antibody binding or enzyme–substrate interactions.^{21–24,27–30} These bioresponsive materials help bridge the gap between the robustness of synthetic polymers and the functionality and specificity of biological interactions.

In this Highlight, hydrogels used specifically for sensing applications will be discussed. Examples based on bulk phase transitions such as sol–gel transitions^31–33 will not be discussed in great detail; the primary focus will be on swelling or deswelling responses in chemically cross-linked networks. These responsive hydrogels have been shown to give a number of different responses to analyte recognition. These include a simple expansion or contraction of the polymer network, a change in fluorescence response of fluorophore in the gel, a change in the diffracted wavelength in a colloidal assembly, or a change in the optical properties of the gel such as in the case of microlenses. Some select examples of each response will be presented.

2 Physical expansion/contraction response

A physical change in the size of the gel due to a change in network density is the basis for all of the responsive hydrogel systems discussed below. This sensing modality is particularly useful given the preponderance of different physical observables that are modulated as a result of hydrogel swelling. These include, but are not limited to, size, porosity, density, refractive index, and modulus. Furthermore, the extreme porosity of hydrogels permits rapid analyte diffusion into the network, thereby taking advantage of the entire three-dimensional structure. As a result of these advantages, a wide range of sensor transduction methods could, in principle, be applied to hydrogel-based bioresponsive materials. Note however, that many of the examples described below have not been specifically engineered into a sensing system or device. For many bioresponsive hydrogels, this remains a relatively untapped area of research, with far more effort having been expended on the development of new materials than on their real-world application in sensors.

An early bioresponsive hydrogel was developed by Miyata et al.³⁴ that utilized a poly(2-glucosyloxyethyl methacrylate) (poly(GEMA)) hydrogel, which contains glucose moieties. The multivalent lectin concancavalin A (Con A) was then introduced to the polymer, which is accompanied by Con A binding to two to four glucose moieties thereby further cross-linking the poly(GEMA) hydrogel. This induced, non-covalent cross-linking was interrupted when free glucose was added into the solution, causing displacement of the glucose moieties and expansion of the hydrogel network. A compression apparatus was used to detect the swelling of the gel. This construct was slightly modified in another study by covalently attaching the Con A to the hydrogel network, thereby limiting the diffusion of Con A from the gel.³⁵ The covalently attached Con A could then re-cross-link the network after rinsing the gel with buffer, freeing the glucose, and binding the glucose moieties attached to the polymer network.

In a similar vein, Miyata et al.^36,37 have published two different examples of antigen–antibody responsive gels. In one example an acrylamide-based hydrogel was synthesized using a co-monomer displaying a covalently attached “antigen” (rabbit IgG). An anti-rabbit IgG antibody was added to the material, thereby forming cross-links by binding two polymer-bound antigen equivalents. Consequently, when free rabbit IgG was present in an analyte solution, it displaced the covalently attached antigen from the bound antibody, disrupting the cross-links, and causing swelling of the gel.³⁶ In the other example, both the antibody and antigen were covalently linked to co-monomers in an interpenetrating network (IPN).³⁷ The IPN was naturally cross-linked by the antigen–antibody interactions and upon addition of free antigen these cross-links were again disrupted and caused swelling of the gel. The advantage of a covalently attached sensing element is that it allows for reversibility of the cross-linking and therefore the possibility of regeneration of the sensor after washing out the free analyte. The fundamental concept of using antibody–antigen interactions to sense either polyvalent protein binding, or the ligands that bind to them is potentially useful for reversible/reuseable sensors and is employed by our group in the microlens construct discussed below.

Another method for introducing binding specificity into hydrogels is by molecular imprinting, which has been utilized in both bulk hydrogels and microgels (hydrogel microparticles). In a hydrogel example for recognizing the tumor specific marker glycoprotein, α-fetoprotein (AFP), the hydrogel was synthesized using both a Con A functionalized monomer and an anti-AFP functionalized monomer.³⁸ Free AFP was introduced to the pre-polymer solution, which bound to both the Con A and anti-AFP moieties during synthesis, creating a non-covalent cross-link. After synthesis, the AFP was washed from the system, and the hydrogel was left with specific binding sites that contained both Con A and anti-AFP recognition units, and were, in theory, organized appropriately for AFP binding. Therefore, when AFP was present the non-covalent cross-link was recreated and the gel collapsed. The efficacy of the imprinting process was illustrated via appropriate controls. In a microgel example, the imprinted molecules were a therapeutic drug, theophylline, and the steroid 17β-estradiol.³⁹ These molecules were introduced into a pre-polymer solution of methacrylic acid (MMA) and the cross-linker trimethylolpropane trimethacrylate (TRIM). After the photo-initiated synthesis and several rounds of washing to remove the imprinted molecules, affinity and competitive binding studies were performed showing good specificity and selectivity for the target molecules. Incorporation of molecular imprinting into hydrogel synthesis is an interesting way of designing a highly selective hydrogel network. However, the approach is not likely to be generally applicable to a wide variety of analytes, due to the difficulty of incorporating multiple affinity sites into the network for all analytes, the difficulty of removing the template under certain conditions, and also the likelihood that the synthesis conditions can be incompatible with the optimal analyte binding conditions.

Biospecificity can also be imparted to hydrogels by incorporating enzymes into the system.⁴⁰ Here we will focus on two specific gel transitions: sol-to-gel or gel-to-sol transitions and hydrogel swelling or deswelling. One example of a gel-to-sol transition was demonstrated in an acrylamide-based hydrogel synthesized with a tetrapeptide cross-linker.⁴¹ In this study, two different peptide cross-linkers were used in different hydrogel syntheses: a tyrosine-lysine linkage cleavable by α-chymotrypsin and a serine-lysine linkage that could not be cleaved by the enzyme. The hydrogels were polymerized in a circular disk in a microfluidic channel, and in the presence of α-chymotrypsin a gel-to-sol transition of the hydrogel containing the enzyme-cleavable peptide was observed by optical microscopy. Fig. 1 shows the hydrogel containing the enzyme-degradable cross-linker and the hydrogel containing the non-degradable cross-linker before and after exposure to the enzyme.

Fig. 1 Optical micrograph of an enzyme-responsive gel with the gels containing the cleavable peptide sequence (left) and the non-cleavable sequence (right) 0 (a), 5 (b), and 20 min (c) after enzyme addition. Scale bar = 500 µm.⁴¹

An example of a swelling response to enzymatic activity involved the cleavage of a peptide-linker that contained both anionic and cationic residues.¹⁷ In this case a poly(ethylene glycol)–poly(acrylamide) (PEGA) bead was synthesized with a pendent zwitterionic peptide sequence. Two different hydrogels were made with different peptide sequences, where those sequences contained the following critical components: a positively charged arginine directly attached to the hydrogel, a dialanine or diglycine linkage, a pendent aspartic acid with a 9H-fluoren-9-ylmethoxycarbonyl (FMOC) protected amine and an acid group contributing the anionic part of the polyelectrolyte. The diglycine or dialanine linkages were cleaved with various enzymes, the most effective for both being thermolysin. As these linkages were cleaved, the negative part of the polyelectrolyte was lost into solution, and the positive fragment was left attached to the gel increasing the electrostatic repulsion in the hydrogel and causing swelling. This change could conceivably be used for enzyme-responsive drug delivery, or in the screening of enzyme inhibitors, as many proteases are involved in specific disease states, including many cancers.

Physical changes in hydrogels have also been coupled to protein conformational changes that occur during a ligand-binding event.^42,43 In one example a PEG hydrogel was synthesized incorporating calmodulin (CaM) as a hydrogel cross-linker. CaM has an extended conformation in the presence of calcium ions, and upon ligand binding the protein collapses into a more compact structure. Due to this dramatic change in conformation, the CaM-cross-linked hydrogel collapsed significantly upon binding of the ligand, trifluoroperazine (TFP). This response is shown in Fig. 2. Using a protein that has a large conformational change due to ligand binding as a cross-linker is a clever way of introducing responsivity, although it may not be as widely applicable to many proteins that are either hard to incorporate into the gel or involve time-consuming synthesis and purification techniques.

Fig. 2 Crystal Structure, scheme, and optical micrograph of the CaM-containing gel before (left) and after (right) ligand binding.⁴²

3 Fluorescence response

In another sensing modality, hydrogels have been integrated with fluorescent tags to provide an easily detectable readout of a change in the hydrogel network density.⁴⁴ In one case, a glucose sensor was fabricated from an acrylamide-based hydrogel containing covalently attached rhodamine dye molecules and an amine moiety, which introduced pH sensitivity.⁴⁵ The enzymes glucose oxidase and catalase were physically entrapped in the gel by adding them into the pre-polymer solution before polymerization. The gel was cut into disks that were then affixed to the end of the optical fiber, providing a means of exciting fluorescence, and collecting the resultant emission. When the optical fiber + gel assembly was placed into a solution of glucose, the glucose was oxidized by β-D-glucose oxidase to D-gluconic acid and hydrogen peroxide. The catalase then converted the hydrogen peroxide to water and oxygen to prevent oxygen depletion within the gel. Glucose oxidation decreased the local pH, protonating the amine groups in the gel, thereby increasing the electrostatic repulsion within the network, which is accompanied by swelling of the gel. The swelling of the gel decreased the local concentration of fluorophore, and consequently the fluorescent response was strongly dependent on the solution glucose concentration. In this approach, however, the response was not linear in the physiological glucose concentration range and may not be easily transitioned to real in vitro diagnostics applications.

Another glucose-sensing construct was designed such that fluorescence was increased due to the presence of glucose.⁴⁶ In this case, a PEG-based hydrogel was synthesized with a covalently attached, fluorescently-labeled Con A and physically entrapped fluorescien isothiocyanate (FITC) labeled dextran. The fluorophore attached to Con A was tetramethylrhodamine isothiocyanate (TRITC), which undergoes fluorescence resonance energy transfer (FRET) with fluorescein at short distances due to their strong spectral overlap. Therefore when the dextran binds Con A, the two chromophores undergo FRET and a decreased amount of fluorescien fluorescence is observed. However, introduction of glucose displaces the dextran from the Con A and increases the fluorescein fluorescent signal, as shown in Fig. 3.

Fig. 3 Fluorescence spectra showing the FITC-dextran signal (left) and FRET signal (right) with 0 (●), 200 (■), 400 (▲), and 1000 (▼) mg dL⁻¹ concentrations of glucose.⁴⁶

4 Diffraction response

Hydrogels have also been coupled to the diffraction from photonic crystals by either encapsulation of a photonic crystal inside a hydrogel network, or by the direct assembly of photonic crystals using microgels as building blocks. These systems are potentially advantageous sensing constructs because the Bragg diffraction they display, which is due to the periodicity of the crystal, is tunable/responsive to gel swelling. Therefore if a hydrogel–photonic crystal hybrid is designed in which the interparticle spacing changes due to the presence of an analyte, a label-free colorimetric sensor is obtained.

One key example of photonic crystal sensors are embodied by polymerized crystalline colloidal arrays (PCCA) studied by the Asher group at the University of Pittsburgh. PCCAs are colloidal crystals embedded in a responsive hydrogel network, as shown in Fig. 4. A number of different sensing applications have been attacked using PCCAs including glucose sensing.^43,47–50 For this application acrylamide-based hydrogels were modified to contain boronic acid units that bind to glucose; therefore, when glucose is present, multiple acid groups complex the glucose molecule thereby cross-linking the gel. The deswelling caused by this induced cross-linking causes the interparticle spacing of the colloidal crystalline array to decrease, thereby blue-shifting the Bragg diffraction.^47,48,50 The blue shift response is seen in when the ionic strength of the medium is high enough to shield Coulombic repulsion between the negatively charged glucose–boronic acid complexes. Conversely, when the ionic strength is lower, electrostatic repulsion in the gel causes a swelling response, increasing the interparticle spacing and inducing a red shift in the diffraction spectrum (Fig. 4).⁴⁹ Obviously, the different responses due to different ionic strengths may be problematic if precise ionic strength control is not possible. However, this approach at least illustrates a valuable concept in how to couple the diffractive optical response to a biologically relevant sensing event.

Fig. 4 Scheme (left) and corresponding diffraction spectra (right) of a PCCA in the absence (top) and presence (bottom) of analyte.⁴⁹

PCCAs have also been employed for nerve agent sensing.^51,52 In this methodology, an enzyme is covalently attached to the hydrogel network allowing binding of the analyte to the enzyme. Upon analyte binding or enzyme conversion, the charge in the network is either increased or decreased and swelling or deswelling occurs. The analytes in two specific cases were organophosphorus (OP) compounds.^51,52 In one case the enzyme acetylcholinesterase was bound to the hydrogel network and upon OP binding an anionic complex was formed, causing the hydrogel to swell, increasing particle spacing, and shifting the Bragg diffraction peak.⁵¹ In a similar case, organophosphorus hydrolase was attached to the hydrogel along with a phenol moiety.⁵² When the analyte was introduced at a pH of 9.7, which is above the pK_a of the pendant phenol groups, those moieties became deprotonated and charged. When the OP was introduced and hydrolyzed by the enzyme, protons were released, thereby lowering the local pH below the phenol pK_a and protonating the phenols. As a result, the internal network charge, electrostatic repulsion, interparticle spacing, and diffraction wavelength were all decreased. The PCCA method has been widely applied to other sensing applications, including metal ion,⁵³ ammonia,⁵⁴ and creatinine sensing.⁵⁵

A different diffraction construct in which the colloidal crystal is directly formed from microgels has been studied as a glucose sensor.⁵⁶ In this approach, 3-acrylamidophenylboronic acid (APBA) was incorporated into pNIPAm microgels. The APBA is in equilibrium between neutral species and the hydrolyzed negatively charged species. Upon addition of glucose, the glucose binds to the charged form, yielding two water molecules and shifting the equilibrium towards the hydrolyzed products. This increases the number of locally charged species and the internal electrostatic repulsion, thereby causing a swelling of the microgels. Since these microgels are assembled into a colloidal crystal, the expansion of the particles yields an increase in the interparticle distance and an increase in the Bragg diffraction wavelength and therefore a color change, shown in Fig. 5. This is an important example because it is one of the only colloidal crystal sensing examples in which the building blocks are composed of intrinsically glucose-sensitive microgels. Unfortunately, this current system operates only above physiological pH, as the APBA needs to be ionized in order to bind glucose. Presumably this deficiency could be corrected with a different polymer design.

Fig. 5 Glucose-sensing microgel colloidal crystals with and without glucose.⁵⁶

5 Direct optical responses

This section focuses on hydrogel-based materials that change their optical properties due to a change in network swelling. Primary amongst these examples are microlenses that change their optical properties due to a change in refractive index or radius of curvature. The swelling response is therefore conveniently read out by observing the focusing power of the microlens by simply projecting an image through the microlens. Applying a stimulus that changes the focal length of the lens is observed as a focusing or defocusing of the projected image. There have only been a few examples of such optically responsive hydrogel materials. Most of the work done on such materials has been developed by our group using a microgel construct and will be discussed in detail. However, a few bulk hydrogel examples have also been reported.

The primary example of a “bulk” hydrogel approach to responsive microlenses involved microlenses synthesized by polymerizing a hydrogel precursor solution onto a glass substrate in the form of a “microdome”.⁵⁷ The microdome was an acrylamide-based gel that incorporated both covalently attached calmodulin (CaM) and phenothiazine, which bind to each other creating a non-covalent cross-link in the gel. When the competing ligand chlorpromazine (CPZ) is present in the surrounding solution, it binds to CaM and displaces phenothiazine, thereby disrupting the cross-link, which induces gel swelling. The swelling response changes the curvature of the microdome as well as the refractive index. Therefore, the focal length changes and a focused image becomes defocused.

Another microlens construct was developed using a hydrogel ring to manipulate the curvature of a water–oil interface.⁵⁸ This construct was made to possess pH and temperature responsivity by incorporating acid or amino groups and thermo-responsive monomers into the hydrogel ring, respectively. In the temperature responsive system the hydrogel ring was polymerized with NIPAm. Similarly, in the pH-responsive case the gel contained acrylic acid or 2-(dimethylamino)ethyl methacrylate (DMAEMA). These microlenses were fabricated by sandwiching a hydrogel ring between a solid glass surface and a surface containing a hole aligned with the hole in the ring. Water was incorporated into this region of the device, and oil was sandwiched between the top surface containing the hole and another glass surface. This created a system that had a water–oil interface in the middle of the hydrogel ring. The curvature of this interface was then tuned by swelling or deswelling the gel by inducing a pH change or change in temperature.

In our group a different type of microlens structure has been developed using well-defined microgels as the building block. These microgels are synthesized by free-radical polymerization of NIPAm and acrylic acid (AAc) with N,N′-methylene(bisacrylamide) (BIS) as a typical covalent cross-linker. The microgels are then adsorbed to a surface and have been shown to act as individual optical elements following adsorption and deformation into hemispherical structures.⁵⁹ A typical optical response is shown in Fig. 6.⁶⁰ Given the incorporation of temperature- and pH-responsive monomers, the resultant microlenses are similarly responsive to changes in temperature⁶¹ and pH.⁶⁰ The optical response shown in Fig. 6 is due to a change in pH, which changes the degree of AAc protonation and therefore the degree of Coulombic repulsion in the gel. Decreasing the pH of the surrounding media protonates the acid groups in the gel, causing deswelling that increases the refractive index, which decreases the lens focal length. In the thermo-responsive case, a rise in temperature causes a contraction of the polymer network and a subsequent decrease in focal length.

Fig. 6 Microgel microlenses: SEM image (left panel) at a grazing angle of an array of microlenses (a)–(d). DIC microscopy images at pH 3.0 (a) and 6.5 (b) with the corresponding projection images (c) and (d). Scale bar = 1 µm.⁶⁰

Following on from the examples described above, we designed microlenses that would display a change in refractive index that was induced via protein binding for use in sensing applications.^62–65 Specifically, microlenses have been designed for two different pathways for sensing: a direct binding-induced response and a displacement-induced response (Scheme 1⁶²). To illustrate each method, the small vitamin biotin was conjugated to the acrylic acid groups on the microgels. For the first binding-induced method (Scheme 1, Route A), avidin or anti-biotin (antibody) was added to the solution around the microlens, resulting in binding of the protein to the microlens surface. Since both avidin (four binding sites) and anti-biotin (two binding sites) are able to bind multiple equivalents of biotin, the protein-binding events increase the surface cross-linking of the microlens. This cross-linking induced a refractive index change, and a visual signal was observed as shown in Fig. 7.⁶³ In this example, the lenses transition from a single square projection to almost double image response while in projection mode, which is achieved by simply placing a pattern between the light source and analyzer in an inverted microscope. This method is simple and could be applied to many different protein-binding applications. A displacement-induced method can be achieved by designing a reversible antibody–antigen cross-linking construct. In this case, a photoaffinity approach is used to couple a bound antibody to the antigen-laden microlens. When the free biotin disrupts the cross-links via displacement, the microlens swells and the focal length increases accordingly. A biotin-free buffer wash removes the free biotin, allowing for re-cross-linking of the gel and regeneration of the sensor.⁶⁵ Using microgels as microlenses is attractive because of the ability to use many different solution-based bioconjugation methods and the ease of assembly with simple electrostatic adsorption. Also, unlike the bulk gels described above, post-synthesis modifications can help ensure uniformity of the lens structure between different bioconjugations. However, the microgel–microlens examples do need to be extended to more relevant antibodies or other biomolecule sensing as well as be applied to multi-sensing applications before they are demonstrably applicable to real world diagnostic tools.

Scheme 1 Scheme showing the microlens-sensing strategies: the binding-induced deswelling method (route a) and displacement-induced swelling method (route b).⁶²

Fig. 7 Microlens response to increasing amounts of avidin showing the DIC (left panels) and projection images (right panels) with particles with (left) an without (right) biotin.⁶³

6 Concluding remarks

Here we have shown examples of different hydrogel constructs that have potential for use in sensing applications. We have highlighted examples of hydrogels that give different responses due to the presence of analyte. These responses include simple physical changes in gel size or shape, colorimetric responses such as fluorescence intensity changes or diffraction wavelength change, and finally optical responses that occur due to focal length changes in microlenses. In some cases, as with the well-developed PCCA approach, the pathway to true sensing systems is well established. However, other methods are very much still in the materials discovery and development stage. Further improvements in sensitivity, specificity, and dynamic range are required for many of the sensor elements described, but are presumably attainable for a subset of these approaches through further optimization. As these sensors become more relevant to the detection of specific analytes, coupling of these approaches to commercially applicable sensor constructs will be the logical next step in device development. Indeed, the rational engineering of bioresponsive materials into robust, functional sensors is the largest challenge facing the field.

References

1. E. A. Moschou, M. J. Madou, L. G. Bachas and S. Daunert, Sens. Actuators, B, 2006, 115, 379–383.
2. E. A. Moschou, S. F. Peteu, L. G. Bachas, M. J. Madou and S. Daunert, Chem. Mater., 2004, 16, 2499–2502.
3. D. Seliktar, A. H. Zisch, M. P. Lutolf, J. L. Wrana and J. A. Hubbell, J. Biomed. Mater. Res., Part A, 2004, 68, 704–716.
4. G. Tae, Y. J. Kim, W. I. Choi, M. Kim, P. S. Stayton and A. S. Hoffman, Biomacromolecules, 2007, 8, 1979–1986.
5. D. A. Wang, C. G. Williams, F. Yang, N. Cher, H. Lee and J. H. Elisseeff, Tissue Eng., 2005, 11, 201–213.
6. N. Yamaguchi, L. Zhang, B. S. Chae, C. S. Palla, E. M. Furst and K. L. Kiick, J. Am. Chem. Soc., 2007, 129, 3040.
7. D. J. Gan and L. A. Lyon, Macromolecules, 2002, 35, 9634–9639.
8. C. M. Nolan, C. D. Reyes, J. D. Debord, A. J. Garcia and L. A. Lyon, Biomacromolecules, 2005, 6, 2032–2039.
9. N. Singh, A. W. Bridges, A. J. Garcia and L. A. Lyon, Biomacromolecules, 2007, 8, 3271–3275.
10. B. G. De Geest, C. Dejugnat, G. B. Sukhorukov, K. Braeckmans, S. C. De Smedt and J. Demeester, Adv. Mater., 2005, 17, 2357.
11. Y. H. Deng, C. C. Wang, X. Z. Shen, W. L. Yang, L. An, H. Gao and S. K. Fu, Chem.–Eur. J., 2005, 11, 6006–6013.
12. J. D. Ehrick, S. K. Deo, T. W. Browning, L. G. Bachas, M. J. Madou and S. Daunert, Nat. Mater., 2005, 4, 298–302.
13. J. X. Gu, F. Xia, Y. Wu, X. Z. Qu, Z. Z. Yang and L. Jiang, J. Controlled Release, 2007, 117, 396–402.
14. K. Rao, B. V. K. Naidu, M. C. S. Subha, M. Sairam and T. M. Aminabhavi, Carbohydr. Polym., 2006, 66, 333–344.
15. K. S. Soppimath, A. R. Kulkarni and T. M. Aminabhavi, J. Controlled Release, 2001, 75, 331–345.
16. G. Tae, M. Scatena, P. S. Stayton and A. S. Hoffman, J. Biomater. Sci., Polym. Ed., 2006, 17, 187–197.
17. P. D. Thornton, R. J. Mart and R. V. Ulijn, Adv. Mater., 2007, 19, 1252.
18. I. R. Wheeldon, S. C. Barton and S. Banta, Biomacromolecules, 2007, 8, 2990–2994.
19. K. C. Wood, H. F. Chuang, R. D. Batten, D. M. Lynn and P. T. Hammond, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 10207–10212.
20. J. Zhou, G. N. Wang, L. Zou, L. P. Tang, M. Marquez and Z. B. Hu, Biomacromolecules, 2008, 9, 142–148.
21. C. D. H. Alarcon, S. Pennadam and C. Alexander, Chem. Soc. Rev., 2005, 34, 276–285.
22. I. Y. Galaev and B. Mattiasson, Trends Biotechnol., 1999, 17, 335–340.
23. A. S. Hoffman and P. S. Stayton, Prog. Polym. Sci., 2007, 32, 922–932.
24. R. J. Mart, R. D. Osborne, M. M. Stevens and R. V. Ulijn, Soft Matter, 2006, 2, 822–835.
25. J. Li, X. Li, X. P. Ni, X. Wang, H. Z. Li and K. W. Leong, Biomaterials, 2006, 27, 4132–4140.
26. X. J. Loh, S. H. Goh and J. Li, Biomaterials, 2007, 28, 4113–4123.
27. S. Deo, E. Moschou, S. Peteu, P. Eisenhardt, L. Bachas, M. Madou and S. Daunert, Anal. Chem., 2003, 75, 207A–213A.
28. A. Kikuchi and T. Okano, Adv. Drug Delivery Rev., 2002, 54, 53–77.
29. T. Miyata, T. Uragami and K. Nakamae, Adv. Drug Delivery Rev., 2002, 54, 79–98.
30. R. V. Ulijn, N. Bibi, V. Jayawarna, P. D. Thornton, S. J. Todd, R. J. Mart, A. M. Smith and J. E. Gough, Mater. Today, 2007, 10, 40–48.
31. S. Kiyonaka, K. Sada, I. Yoshimura, S. Shinkai, N. Kato and I. Hamachi, Nat. Mater., 2004, 3, 58–64.
32. Z. M. Yang, P. L. Ho, G. L. Liang, K. H. Chow, Q. G. Wang, Y. Cao, Z. H. Guo and B. Xu, J. Am. Chem. Soc., 2007, 129, 266–267.
33. Z. M. Yang and B. Xu, Chem. Commun., 2004, 2424–2425.
34. T. Miyata, A. Jikihara, K. Nakamae and A. S. Hoffman, Macromol. Chem. Phys., 1996, 197, 1135–1146.
35. T. Miyata, A. Jikihara, K. Nakamae and A. S. Hoffman, J. Biomater. Sci., Polym. Ed., 2004, 15, 1085–1098.
36. T. Miyata, N. Asami and T. Uragami, Macromolecules, 1999, 32, 2082–2084.
37. T. Miyata, N. Asami and T. Uragami, Nature, 1999, 399, 766–769.
38. T. Miyata, M. Jige, T. Nakaminami and T. Uragami, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 1190–1193.
39. L. Ye, P. A. G. Cormack and K. Mosbach, Anal. Chim. Acta, 2001, 435, 187–196.
40. R. V. Ulijn, J. Mater. Chem., 2006, 16, 2217–2225.
41. K. N. Plunkett, K. L. Berkowski and J. S. Moore, Biomacromolecules, 2005, 6, 632–637.
42. Z. J. Sui, W. J. King and W. L. Murphy, Adv. Mater., 2007, 19, 3377.
43. W. L. Murphy, W. S. Dillmore, J. Modica and M. Mrksich, Angew. Chem., Int. Ed., 2007, 46, 3066–3069.
44. E. A. Moschou, B. V. Sharma, S. K. Deo and S. Daunert, J. Fluoresc., 2004, 14, 535–547.
45. M. F. McCurley, Biosens. Bioelectron., 1994, 9, 527–533.
46. R. J. Russell, M. V. Pishko, C. C. Gefrides, M. J. McShane and G. L. Cote, Anal. Chem., 1999, 71, 3126–3132.
47. V. L. Alexeev, S. Das, D. N. Finegold and S. A. Asher, Clin. Chem., 2004, 50, 2353–2360.
48. V. L. Alexeev, A. C. Sharma, A. V. Goponenko, S. Das, I. K. Lednev, C. S. Wilcox, D. N. Finegold and S. A. Asher, Anal. Chem., 2003, 75, 2316–2323.
49. S. A. Asher, V. L. Alexeev, A. V. Goponenko, A. C. Sharma, I. K. Lednev, C. S. Wilcox and D. N. Finegold, J. Am. Chem. Soc., 2003, 125, 3322–3329.
50. M. Ben-Moshe, V. L. Alexeev and S. A. Asher, Anal. Chem., 2006, 78, 5149–5157.
51. J. P. Walker and S. A. Asher, Anal. Chem., 2005, 77, 1596–1600.
52. J. P. Walker, K. W. Kimble and S. A. Asher, Anal. Bioanal. Chem., 2007, 389, 2115–2124.
53. J. H. Holtz and S. A. Asher, Nature, 1997, 389, 829–832.
54. K. W. Kimble, J. P. Walker, D. N. Finegold and S. A. Asher, Anal. Bioanal. Chem., 2006, 385, 678–685.
55. A. C. Sharma, T. Jana, R. Kesavamoorthy, L. J. Shi, M. A. Virji, D. N. Finegold and S. A. Asher, J. Am. Chem. Soc., 2004, 126, 2971–2977.
56. V. Lapeyre, I. Gosse, S. Chevreux and V. Ravaine, Biomacromolecules, 2006, 7, 3356–3363.
57. J. D. Ehrick, S. Stokes, S. Bachas-Daunert, E. A. Moschou, S. K. Deo, L. G. Bachas and S. Daunert, Adv. Mater., 2007, 19, 4024.
58. L. Dong, A. K. Agarwal, D. J. Beebe and H. R. Jiang, Nature, 2006, 442, 551–554.
59. M. J. Serpe, J. Kim and L. A. Lyon, Adv. Mater., 2004, 16, 184.
60. J. Kim, M. J. Serpe and L. A. Lyon, J. Am. Chem. Soc., 2004, 126, 9512–9513.
61. J. Kim, M. J. Serpe and L. A. Lyon, Angew. Chem., Int. Ed., 2005, 44, 1333–1336.
62. J. Kim, N. Singh and L. A. Lyon, Biomacromolecules, 2007, 8, 1157–1161.
63. J. Kim, S. Nayak and L. A. Lyon, J. Am. Chem. Soc., 2005, 127, 9588–9592.
64. J. S. Kim, N. Singh and L. A. Lyon, Angew. Chem., Int. Ed., 2006, 45, 1446–1449.
65. J. S. Kim, N. Singh and L. A. Lyon, Chem. Mater., 2007, 19, 2527–2532.

---