Immobilization and Release of the Redox Mediator Ferrocene

Biomacromolecules 2009, 10, 2135–2143

2135

Immobilization and Release of the Redox Mediator Ferrocene Monocarboxylic Acid from within Cross-Linked p(HEMA-co-PEGMA-co-HMMA) Hydrogels Ali Ozgur Boztas† and Anthony Guiseppi-Elie*,†,‡,§ Center for Bioelectronics, Biosensors and Biochips (C3B), Clemson University Advanced Materials Center, 100 Technology Drive, Anderson, South Carolina, 29625, Department of Chemical and Biomolecular Engineering and Department of Bioengineering, Clemson University, Clemson, South Carolina 29634 Received March 12, 2009; Revised Manuscript Received June 4, 2009

Cross-linked hydrogels are synthesized from hydroxyethylmethacrylate (HEMA), polyethyleneglycol methacrylate (PEGMA), and N-[tris(hydroxymethyl)methyl]-acrylamide (HMMA) [p(HEMA-co-PEGMA-co-HMMA) hydrogels] containing 1, 3, 5, 7, 9, or 12 M % of the cross-linker tetraethyleneglycol diacrylate (TEGDA) and are loaded during synthesis with the well-known redox mediator, ferrocene monocarboxylic acid (FcCOOH). In the absence of FcCOOH, the M% TEGDA in deionized (DI) water (48%; 1 mol % TEGDA to 32%; 12 mol % TEGDA) scales with the cross-link density in accordance with Flory-Huggins-Rehner theory. The release profiles of FcCOOH from hydrogel slabs (43.0 mM) into 0.1 M HEPES/0.1 M KCl buffer are determined from the oxidation peak current of FcCOOH via cyclic voltammetry (100 mV/s) and are decidedly Fickian with overall diffusion coefficients that range from 2.64 × 10-8 cm2/s (1 mol % TEGDA) to 4.87 × 10-9 cm2/s (12 mol % TEGDA) and with n parameters that approximated 0.5 but nonetheless linearly declined from 0.49 (1 mol %) to 0.42 (12 mol %). Diffusion coefficients, like hydration, strongly correlate with the M% TEGDA and hence with the cross-link density or the molecular weight (MW) between cross-links. The temperature dependence of the release profiles measured at 10, 15, 20, 25, 30, 35, 40, and 45 °C reveal thermally activated transport with activation energies that are 30 kJ/mol (3 mol %), 36 kJ/mol (5 mol %), 45 kJ/mol (7 mol %), 47 kJ/mol (9 mol %) and 57 kJ/mol (12 mol %). Covalent tethering of the FcCOOH via the UV-polymerizable monomers ferrocene monomethacrylate (Fc-AEMA) and ferrocene polyethylene glycol monomethacrylate (Fc-PEG(3500)-AEMA) to produce pendant redox moieties is shown to eliminate or attenuate release of Fc. While, Fc-AEMA showed no evidence of release (0%) from the hydrogel, its PEG-conjugated equivalent, Fc-PEG(3500)-AEMA, shows release of 16% Fc after 5 days of immersion. These hydrogels will serve as the immobilization matrix for oxidoreductase enzymes of biosensors and the parameters obtained used in the modeling of such systems.

1. Introduction Bioactive hydrogels reflect an emerging paradigm in the development of responsive,1 multifunctional,2 biorecognition membrane layers for implantable biosensors and deep brain stimulation devices.3 These polymeric materials may be molecularly engineered to possess hydration levels, mechanical properties, surface chemistries, and micro/nanotopologies that render them similar to and thus mimetic of the tissue bed within which they are to be implanted. Hydrogels are usually formed by the covalent cross-linking of hydrophilic monomers and prepolymers to form a three-dimensional network with the ability to swell in the presence of water yet remain an insoluble pseudosolid.4 The design and molecular engineering of bioactive hydrogels as the recognition membrane layer of biosensors requires that bioactive molecules such as enzymes, their cofactors, redox mediators, and biomimetic moieties be purposefully incorporated at biofunctionally relevant levels within the hydrogel.5 Retaining these bioactive and or functional entities, either by physical entrapment or covalent tethering, * Corresponding author. E-mail: [email protected]. Tel: 864 656 1712. Fax: 864 656 1713. † Center for Bioelectronics, Biosensers and Biochips (3B), Clemson University, Advanced Materials Center. ‡ Department of Chemical and Biomolecular Engineering, Clemson University. § Department of Bioengineering, Clemson University.

Figure 1. Structures of monomers of hydrogel precursor mixture.

necessitates an understanding of the transport characteristic of these entities within the hydrogel matrix. Hydroxyethylmethacryate (HEMA) is a water-soluble monomer that may be UV polymerized at low temperature (-20 to +10 °C) and may be readily copolymerized with other acrylate, methacrylate, and acrylamide monomers, e.g., polyethyleneglycol methacrylate (PEGMA) and N-[tris(hydroxymethyl)methyl]-acrylamide (HMMA) to yield hydrogels of varying physical and chemical properties.6 Such hydrogel membrane layers and gel pads may serve as microbioreactors for the

10.1021/bm900299b CCC: $40.75  2009 American Chemical Society Published on Web 07/14/2009

2136

Biomacromolecules, Vol. 10, No. 8, 2009

Boztas and Guiseppi-Elie

Table 1. Components and the Composition of Various Cross-Linked p(HEMA-co-PEGMA-co-HMMA) Hydrogel Formulationsa hydrogel components 1 mol % 3 mol % 5 mol % 7 mol % 9 mol % 12 mol % HEMA TEGDA PEG(200) MMA HMMA p(HEMA) DMPA ethylene glycol water

85.4 1.0 5.0

83.4 3.0 5.0

81.4 5.0 5.0

79.4 7.0 5.0

77.4 9.0 5.0

74.4 12.0 5.0

4.6 2.1 1.9 20 wt %

4.6 2.1 1.9 20 wt %

4.6 2.1 1.9 20 wt %

4.6 2.1 1.9 20 wt %

4.6 2.1 1.9 20 wt %

4.6 2.1 1.9 20 wt %

20 wt % 20 wt % 20 wt % 20 wt % 20 wt % 20 wt %

a

All numbers in formulations are mol % of monomers or of repeat units.).

Figure 2. Preparation of cross-linked p(HEMA-co-PEGMA-co-HMMA) hydrogel slabs for hydration and transport studies. The p(HEMA-coPEGMA-co-HMMA) hydrogel is colorless, and the p(HEMA-coPEGMA-co-HMMA) hydrogel with FcCOOH is light yellow.

stabilization of bioactive entities in several forms of enzymelinked antibody biosensors and DNA biochips.7-11 There are multiple studies involving hydrogels that investigate their biocompatibility,12 biodegradability,13 bioadhesion,14 dielectric relaxation,15 and mass transport properties.16 Since p(HEMA)based hydrogels are hydrolytically stable, exhibit good in Vitro and in ViVo biocompatibility, and may be made similar to body tissues in their high water contents,17,18 they have now become one of the most widely researched and successfully commercialized biomedical hydrogels.19 Numerous studies that aim to modify the properties of p(HEMA) have been conducted to improve the mechanical properties,20 permeability,21 temperature responsive characteristics,22 and degree of hydration or extent of swelling of the hydrogel network and to elicit more favorable biological responses.23,24 The degree of hydration and/or swelling is one of the important properties that allows for an understanding of the transport of small molecule solutes through the hydrogel matrix. However, this property also influences mechanical properties and surface properties such as wettability and protein adsorption and so is strongly correlated with in Vitro and in ViVo biocompatibility.25 The presence of PEGMA and HMMA confers the antiprotein fouling properties of the pendant polyethylene glycol (PEG) chains and the temperature responsive properties of the acrylamide to the characteristics of the p(HEMA-co-PEGMA-co-HMMA) hydrogels.26,27 When immersed and equilibrated in aqueous medium, crosslinked hydrogels assume their final hydrated network structure. The final structure is the result of a balance of forces arising from the solvation of the repeat units of the macromolecular chains that leads to an expansion of the network (the swelling force) and the counter balancing elastic force of the cross-linked

structure (the retractive force).28 This equilibrated structure can support transport that is dependent upon the cross-link density and the molecular weight (MW) between cross-links. There are three well-established models that seek to describe the relative rates of solute transport and polymer chain relaxation across the diverse range of possible responses of hydrophilic polymer networks.29,30 Among these are (A) Fickian diffusion (case I diffusion) wherein the diffusion is significantly slower than the rate of relaxation of the polymer chains; (B) case II diffusion wherein the rate of solute diffusion is greater than the rate of relaxation of the polymer chains; and (C) anomalous diffusion (case III diffusion) wherein the rate of solute diffusion is comparable to that of the polymer relaxation and shows transport characteristics that are between the extremes of the case I and case II models. In the design and development of intramuscularly implantable oxidoreductase-based biosensors to allow for the continuous monitoring of glucose and lactate during trauma-induced hemorrhage,31 a small molecule mediator may be incorporated within the hydrogel membrane. The mediator serves as a surrogate for molecular oxygen in the regeneration of the enzyme’s cofactors and is necessitated by the low oxygen tension within the hypoxic tissue bed. The diffusion of that mediator, like the diffusion of glucose and lactate into and within the hydrogel biorecognition membrane layer, is important to the proper functioning of the enzyme-based biosensor. This paper reports on the transport properties of the mediator molecule, ferrocene monocarboxylic acid (FcCOOH), within p(HEMA-co-PEGMA-co-HMMA) hydrogels. FcCOOH is a well-known and widely characterized mediator of oxidoreductase enzymes in biosensor configurations.32,33 FcCOOH containing hydrogel slabs were prepared by first dissolving the mediator in hydrogel precursor mixture and then casting the precursor mixture into molds. Using the release rate of FcCOOH from within the hydrogel slabs, the transport characteristics were studied as a function of the cross-link density (1, 3, 5, 7, 9, and 12 mol % cross-linker) and as a function of temperature (10, 15, 20, 25, 30, 35, 40, and 45 °C) in physiologically relevant buffer media [0.1 M HEPES/0.1 M KCl (pH ) 7.4)] and measured by cyclic voltammetry (CV). In addition, methacryloyl functionalized ferrocene monomers were synthesized and similarly incorporated into the hydrogel slabs, resulting in covalent tethering of the mediator. Covalent tethering of the FcCOOH via UV-polymerizable monomers, ferrocene monomethacrylate (Fc-AEMA) and ferrocene polyethylene glycol monomethacrylate (Fc-PEG(3500)-AEMA), resulted in pendant redox moieties that were shown to eliminate the release of Fc.

2. Experimental Section 2.1. Reagents. The monomers HEMA and HMMA (93%), the crosslinker tetraethyleneglycol diacrylate (TEGDA, technical grade), the photoinitiator 2,2-dimethoxy-2-phenylacetophenone (DMPA, 99%), and the reagents 3-aminopropyltrimethoxysilane (γ-APS, 97%) and 3-mercapto-1-propanol (95%) were purchased from Sigma-Aldrich Co. (St. Louis, MO). Polyethylene glycol(200)monomethacrylate (PEG(200)MA) was purchased from PolySciences, (Warrington, PA), and the heterobifunctional PEG derivative acryloyl(polyethyleneglycol)-N-hydroxysuccinamide (Acrl-PEG(3500)-NHS) (MW 3500) was purchased from Jenkem Technology, USA (Allen, TX) or Nektar Therapeutics, (Huntsville, AL). The diacrylate and methacrylate reagents were separately passed over an inhibitor removal column (Sigma-Aldrich) for removal of hydroquinone and monomethyl ether hydroquinone polymerization inhibitors, before use. FcCOOH (97%, Sigma-Aldrich) was used as received. A 0.1 M Tris buffer solution (adjusted to pH )

FcCOOH Release from p(HEMA-co-PEGMA-co-HMMA) Hydrogels


2137

Scheme 1. Synthesis of Fc-AEMA

Scheme 2. Synthesis of Fc-PEG(3500)-AEMA

Table 2. Comparison of the Hydrogel Slabs for the Study of the Release of Ferrocene Monocarboxylic Acid from p(HEMA-co-PEGMA-co-HMMA) Hydrogels A

B

C

mediator loaded into p(HEMA-co-PEGMA-co-HMMA) hydrogel

FcCOOH: 43.0 mM in 70 µL of hydrogel

Fc-AEMA (immobilized): 10.8 mM in 70 µL of hydrogel

Fc-PEG-AEMA (immobilized): 10.8 mM in 220 µL of hydrogel

solution for release study

5 mL of 0.1 M HEPES - 0.1 M KCl

2.5 mL of 0.1 M HEPES - 0.1 M KCl

4 mL of 0.1 M HEPES - 0.1 M KCl

releasing % of Fc from hydrogel

98.6%

0%

16%

7.2 using 1.0 M HCl) was made from Tris(hydroxymethyl)aminomethane or Trizma (99.8+%, A.C.S., Sigma-Aldrich). A 0.1 M HEPES buffer solution (adjusted to pH ) 7.4 using 1.0 M HCl) was made from N-(2-hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid) sodium salt (99.5+%, Sigma-Aldrich). Trichloroethylene (spectrophotometric grade, g99.5%, Sigma-Aldrich), acetone (g99.9%, Sigma-Aldrich), 2-propanol (g99.8%, Sigma-Aldrich), ammonium hydroxide solution (ACS reagent, 28.0-30.0% as NH3, Sigma-Aldrich), hydrogen peroxide solution (ACS reagent, 30 wt % in H2O, Sigma-Aldrich), ethanol (CHROMASOLV, Sigma-Aldrich), and toluene (anhydrous, 99.8%, Acros) were used as received. 2.2. Synthesis and Preparation of p(HEMA-co-PEGMA-coHMMA) Hydrogel Slabs. The structures of the monomers used in the preparation of the p(HEMA-co-PEGMA-co-HMMA) hydrogel are

presented in Figure 1. Prior to polymerization of the monomer precursor mixture, the base monomers HEMA, PEG(200)MA, HMMA, and the cross-linker TEGDA were separately passed over inhibitor removal columns (Aldrich Chemical Co.) to remove the methyl ether hydroquinone inhibitor. Table 1 shows the different formulations of varying TEGDA content used to prepare hydrogels of different cross-link densities. For any given formulation, the required volumes of HEMA, PEG(200)MA, and TEGDA were placed in a vial, DMPA, the photoinitiator (2 mol % monomer) HMMA, and p(HEMA) were dissolved in the resulting monomer precursor mixture, and the mixture was sparged with nitrogen to eliminate oxygen. To prepare hydrogels in the form of a slab, CoverWell perfusion chambers (Grace Bio Laboratories, Inc., Bend, OR) were used as casting molds. Glass coverslips were rendered hydrophobic by chemical modification with

2138


Boztas and Guiseppi-Elie Degree of Hydration )

Figure 3. Equilibrium degrees of hydration (%) determined gravimetrically for 1, 3, 5, 7, 9, and 12 mol % cross-linked p(HEMA-coPEGMA-co-HMMA) hydrogel slabs (n ) 2).

Figure 4. Relationship between swollen/dry gel volume ratio of p(HEMA-co-PEGMA-co-HMMA) hydrogels and the mol % of TEGDA cross-linker representing the degree of cross-linking, demonstrating accord with Flory-Huggins-Rehner theory.

octadecyltrichlorosilane (1 wt % in toluene, 30 min),34 and the adhesivebacked perfusion chambers were pressed onto the glass coverslip. Volumes (70 µL) of each hydrogel precursor mixture were then pipetted through the chamber openings to fill the chamber. The perfusion chambers containing the hydrogel precursor mixture were placed into a UV cross-linker (UVP CX-2000, CA) and immediately irradiated (365 nm) for 10 min while under nitrogen. The perfusion chamber was then removed and the resulting hydrogel slab (scalene ellipsoid; 0.60 mm thick, 19.00 mm long, × 6.35 mm wide, and 70 µL volume) was released from the coverslip. The process steps for hydrogel slab production are shown pictorially in Figure 2. 2.3. Hydration Studies. The six different formulations representing cross-link densities of 1, 3, 5, 7, 9, and 12 mol % TEGDA, shown in Table 1, of UV-polymerized and cross-linked hydrogel slabs were subjected to sequential immersion for 45 min each in ethanol/water mixtures of 75:25, 50:50, and 25:75 v/v %. Slabs were finally transferred to pure deionized (DI) water and allowed to achieve equilibrium hydration by overnight immersion at room temperature (RT). Hydrogel slabs were blotted dry with Kimwipes to remove excess water and weighed. This weight was recorded as the initial wet weight. Hydrogel slabs were subsequently dried in a closed cell concentrator (Zymark, Turbo Vap 500, Closed Cell Concentrator) at 37 °C, 5000 rpm for a period of six days during which they were removed periodically to be weighed and then returned to the concentrator for another day before again being removed and reweighed. This process was repeated until no further weight loss ((1 mg) was observed. This weight was recorded as the dry weight. Hydrogel slabs were then rehydrated in DI water for 24 h, blotted dry, and reweighed to obtain the second value of the wet weight. The degree of hydration was calculated using the average of the final weights of duplicate analyses using the formula

Wwet - Wdry × 100 Wwet

(1)

2.4. Synthesis of Fc-AEMA. Equimolar amounts of N-hydroxy succinimide (NHS) (3.26 mmol) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC) (3.26 mmol) were dissolved in dichloromethane (DCM), (Scheme 1). FcCOOH (2.17 mmol) was added to this solution. The solution was incubated for 24 h at RT and washed with water within a separatory funnel to remove unreacted EDC/NHS and isourea, which is the biproduct of EDC reaction. DCM was evaporated at RT to obtain the activated Fc-NHS. Equimolar Fc-NHS and 2-aminoethyl methacrylate (AEMA) were dissolved in ethylene glycol and water (50:50 v/v%) and incubated for 1 h in a UV-free clean room to achieve the final product, Fc-AEMA. 1H NMR spectrum of Fc-AEMA in CDCl3 was obtained using a Bruker Avance 300 and showed the characteristic ferrocene peaks between 4.0 and 5.0 ppm and gave a signal at 5.32 ppm corresponding to N-H. 2.5. Synthesis of Fc-PEG-AEMA. Equimolar amounts of Fc-NHS (0.042 mmol) synthesized in section 2.4 above, and NH2-PEG(3500)COOH (0.042 mmol) were mixed in DCM (Scheme 2) and allowed to react. Compounds were completely solubilized and incubated for 20 min. Equimolar amounts of NHS and EDC (0.063 mmol) were then added to the solution and further incubated overnight to obtain FcPEG(3500)-NHS. The Fc-PEG(3500)-NHS was isolated following evaporation of the DCM and redissolved along with AEMA in ethylene glycol and water (50:50 v/v%). The solution was then incubated for a minimum of 1 h to obtain Fc-PEG(3500)-AEMA. The 1H NMR spectrum of Fc-PEG(3500)-AEMA in CDCl3 was obtained using a Bruker Avance 300 and showed the characteristic ferrocene peaks between 4.0 and 5.0 ppm, gave a signal at 5.32 ppm corresponding to the N-H bond, and additionally showed -CH2-O-CH2- peaks at 3.46, 3.50, and 3.67 ppm. 2.6. Synthesis and Preparation of FcCOOH-loaded p(HEMAco-PEGMA-co-HMMA) Hydrogel Slabs. To study the release and transport properties of FcCOOH from within p(HEMA-co-PEGMAco-HMMA) hydrogels, hydrogel precursor ingredients were similarly prepared as described in section 2.2 above to yield the three types of FcCOOH-containing hydrogel slabs shown in Table 2 (first row). Formula A was prepared to contain a final concentration of 43 mM FcCOOH in 70 uL of hydrogel precursor mixture, Formula B was prepared to contain a final concentration of 10.8 mM Fc-AEMA in 70 µL of hydrogel precursor mixture, and Formula C was prepared to contain a final concentration of 10.8 mM Fc-PEG(3500)-AEMA in 220 µL of hydrogel precursor mixture. Each formula was uniformly dispersed by sonication and vortexing prior to sparging with nitrogen, and slabs were prepared as described in section 2.2 above. 2.7. Transport via Release Studies by CV. Transport of FcCOOH within p(HEMA-co-PEGMA-co-HMMA) hydrogels was studied by monitoring the release of the redox active mediator from the appropriately FcCOOH-loaded p(HEMA-co-PEGMA-co-HMMA) hydrogel. Release was monitored by CV performed using a three-electrode (working, counter, and reference) electrochemical setup under the control of a BAS 100 potentiostat/galvanostat electrochemical analyzer that was equipped with a low current module and controlled by BAS 100 software (Bioanalytical Systems, West Lafayette, IN). The working electrode was a periodically polished glassy carbon electrode (GCE) (3 mm dia., Bioanalytical Systems). The counter electrode was a platinum mesh electrode, and the miniature reference electrode was a RE804 Ag/AgCl, 3 M Cl- (ABTECH Scientific, Inc., Richmond, VA). Experiments were typically done in 2.5-5.0 mL solutions of 0.1 M HEPES/0.1 M potassium chloride (KCl) (pH ) 7.4) at 20 °C (Table 2, second row). Cyclic voltammograms were obtained between 0 and 600 mV/s using a scan rate of 100 mV/s, and the oxidation peak current was used to obtain the release profile and to quantify, based on a calibration curve, the concentration of FcCOOH released over time. Temperature dependence of the release of FcCOOH from hydrogel slabs was studied by incubating slabs within sealed containers followed by



2139

Table 3. Statistical Parameters (R2) of Hydrogel Formulations for Various Cross-Link Densities at Different Temperatures after Fitting the FcCOOH Release Data to the Various Release Kinetic Models Higuchi

first-order

Hixson-Crowell

Korsemeyer-Peppas

zero-order

TEGDA

°C

k

R2

k

R2

k × 105

R2

k

R2

k

R2

3%

45 40 35 30 25 20 15 10 45 40 35 30 25 20 15 10 45 40 35 30 25 20 15 10 45 40 35 30 25 20 15 10 45 40 35 30 25 20 15 10

0.0554 0.0508 0.0471 0.0444 0.0392 0.0357 0.0348 0.0320 0.0562 0.0492 0.0434 0.0411 0.0360 0.0310 0.0285 0.0268 0.0481 0.0429 0.0400 0.0341 0.0309 0.0278 0.0264 0.0230 0.0444 0.0388 0.3420 0.0325 0.0279 0.0244 0.0220 0.0194 0.0456 0.0384 0.0337 0.0278 0.0237 0.0223 0.0200 0.0183

0.9410 0.9568 0.9554 0.9464 0.9702 0.9448 0.9201 0.9336 0.9982 0.9948 0.9768 0.9708 0.9769 0.9298 0.9739 0.9863 0.9525 0.9384 0.9086 0.8889 0.9579 0.9352 0.8624 0.9054 0.9974 0.9984 0.9842 0.9645 0.9661 0.9534 0.9626 0.9741 0.9739 0.9333 0.9192 0.8491 0.9614 0.9598 0.8656 0.7565

–0.0025 –0.0031 –0.0029 –0.0021 –0.0028 –0.0019 –0.0019 –0.0026 –0.0026 –0.0030 –0.0029 –0.0026 –0.0024 –0.0017 –0.0022 –0.0021 –0.0026 –0.0027 –0.0025 –0.0018 –0.0028 –0.0021 –0.0016 –0.0025 –0.0033 –0.0032 –0.0032 –0.0027 –0.0024 –0.0018 –0.0021 0.0019 –0.0027 –0.0024 –0.0025 –0.0018 0.0028 –0.0017 –0.0016 –0.0018

0.8276 0.6881 0.7562 0.8467 0.7138 0.9206 0.8992 0.7289 0.8197 0.8349 0.7205 0.7205 0.8546 0.9238 0.9042 0.8520 0.8363 0.7475 0.7077 0.8496 0.6697 0.9119 0.9056 0.7274 0.8244 0.8660 0.7029 0.7260 0.8216 0.9195 0.8743 0.8829 0.8490 0.8292 0.6650 0.7425 0.6904 0.9366 0.9135 0.7481

–7 –8 –7 –5 –6 –4 –4 –5 –7 –7 –7 –6 –5 –4 –4 –4 –6 –6 –6 –4 –6 –4 –3 –4 –8 –7 –7 –6 –5 –4 –4 –3 –6 –6 –5 –4 –5 –3 –3 –3

0.8694 0.7618 0.8176 0.8817 0.7840 0.9426 0.9215 0.7941 0.8492 0.8789 0.7819 0.8208 0.8909 0.9405 0.9263 0.8851 0.8774 0.8071 0.7826 0.8831 0.7477 0.9610 0.9254 0.7793 0.8805 0.9082 0.7727 0.7901 0.8667 0.9407 0.9000 0.9075 0.8918 0.8700 0.7490 0.7897 0.7657 0.9579 0.9317 0.7964

0.0285 0.0157 0.0097 0.0215 0.0052 0.0042 0.0186 0.0050 0.0120 0.0074 0.0083 0.0100 0.0100 0.0207 0.0113 0.0054 0.0084 0.0241 0.0200 0.0147 0.0042 0.0080 0.0146 0.0075 0.0036 0.0055 0.0053 0.0074 0.0085 0.0110 0.0096 0.0075 0.0097 0.0067 0.0034 0.0081 0.0035 0.0119 0.0111 0.0090

0.9883 0.9972 0.9898 0.9980 0.9568 0.9899 0.9983 0.9609 0.9873 0.9945 0.9826 0.9895 0.9938 0.9982 0.9998 0.9709 0.9273 0.9998 0.9990 0.9933 0.9193 0.9857 0.9952 0.8790 0.9888 0.9979 0.9893 0.9682 0.9832 0.9789 0.9933 0.9887 0.9898 0.9339 0.8824 0.9986 0.9232 0.9710 0.9943 0.9534

–0.0020 –0.0019 –0.0017 –0.0015 –0.0015 –0.0012 –0.0012 –0.0011 –0.0020 –0.0020 –0.0017 –0.0015 –0.0013 –0.0010 –0.0011 –0.0011 –0.0017 –0.0015 –0.0013 –0.0011 –0.0011 –0.0009 –0.0080 –0.0080 –0.0019 –0.0016 –0.0014 –0.0012 –0.0010 –0.0080 –0.0080 –0.0070 –0.0017 –0.0013 –0.0011 –0.0009 –0.0009 –0.0008 –0.0006 –0.0005

0.8265 0.8297 0.8448 0.8124 0.8396 0.8016 0.7832 0.8164 0.8531 0.8954 0.8541 0.8368 0.8565 0.7837 0.8429 0.8606 0.8399 0.8249 0.7948 0.7654 0.8221 0.7934 0.7313 0.7915 0.9287 0.8244 0.8673 0.8301 0.8427 0.8198 0.8255 0.8259 0.8664 0.8252 0.7904 0.7261 0.8303 0.8275 0.7315 0.7139

5%

7%

9%

12%

Table 4. Diffusion Coefficient, Diffusion Exponent, and Half-Life for the Diffusion of FcCOOH within 1, 3, 5, 7, 9, and 12 mol % TEGDA Cross-Linked p(HEMA-co-PEGMA-co-HMMA) Hydrogels at RT hydrogel X-linker diffusion coefficient, diffusion half life, composition (TEGDA %) D (cm2/s) exponent, n h 1% 3% 5% 7% 9% 12%

2.64 × 10-8 1.13 × 10-8 8.61 × 10-9 7.21 × 10-9 6.28 × 10-9 4.87 × 10-9

0.495 0.482 0.475 0.457 0.436 0.423

1.86 4.33 5.70 6.81 7.82 10.08

temperature equilibration within a refrigerator (Hotpack Model 355381) or water bath (Precision, Reciprocal Shaking Bath, Model 25). Temperature dependence of the release was studied for hydrogels possessing 3, 7, and 12 mol % cross-linker, and these were studied at temperatures of 10, 15, 20, 25 30, 35, 40, and 45 °C in 0.1 M HEPES/ 0.1 M KCl (pH ) 7.4).

3. Results and Discussion 3.1. Hydration of p(HEMA-co-PEGMA-co-HMMA) Hydrogels. The transport of small molecule mediators such as FcCOOH that serve as surrogates for molecular oxygen in oxidoreductase biosensors is strongly influenced by the swelling state of the hydrogel within which they are

Figure 5. Release of FcCOOH from 1, 3, 5, 7, 9, and 12 mol % TEGDA cross-linked p(HEMA-co-PEGMA-co-HMMA) hydrogel slabs into 0.1 M HEPES/0.1 M KCl at 20 °C (n ) 2).

physically entrapped or covalently immobilized. Of particular interest is the transport coefficient through such hydrogels in relation to the cross-link density and the degree of hydration.20,21 In the current study, the degree of hydration was investigated for different molar concentrations of the cross-linker TEGDA (1-12 mol %), within the hydrogel.

2140



Figure 6. Plots of the diffusion coefficient, D (cm2 s-1), and the diffusion exponent, n, for FcCOOH as a function of the mol % crosslinker, TEGDA, in p(HEMA-co-PEGMA-co-HMMA) hydrogel slabs into 0.1 M HEPES/0.1 M KCl at 20 °C.

Figure 8. Arrhenius plots for the diffusion coefficient of FcCOOH within 3, 5, 7, 9, and 12 mol % cross-linked p(HEMA-co-PEGMA-coHMMA) hydrogels in 0.1 M HEPES/0.1 M KCl at 37 °C (n ) 2).

Figure 9. Activation energies of FcCOOH within 3, 5, 7, 9, and 12 mol % cross-linked p(HEMA-co-PEGMA-co-HMMA) hydrogels in 0.1 M HEPES/0.1 M KCl (n ) 2). Figure 7. Temperature dependence of the release of FcCOOH from a 3 mol % cross-linked p(HEMA-co-PEGMA-co-HMMA) hydrogel slab into 0.1 M HEPES/0.1 M KCl at various temperatures (n ) 2).

hydrogel volume and percent of cross-linking for a highly swollen hydrogel with the following equation:

( )

Figure 3 shows the equilibrium degrees of hydration (%) determined gravimetrically in duplicate for 1, 3, 5, 7, 9, and 12 mol % cross-linked p(HEMA-co-PEGMA-co-HMMA) hydrogel slabs. It is seen that the degree of hydration falls dramatically from 1 to 3 mol %, is approximately the same at 3 and 5 mol %, but decreases gradually with increasing concentrations of TEGDA. Classic Flory-Huggins-Rehner theory22 describes the equilibrium degree of swelling of crosslinked polymer networks under good solvent conditions. The greater the degree of cross-linking in a hydrogel, the stronger the elastic, retractive forces will be, thus countering more effectively the thermodynamic swelling force. FloryHuggins-Rehner theory23 shows the relationship between

V Vdry

-2/3

)

k ζZ

(2)

where V and Vdry represents the swollen and dry gel volumes, respectively, k is the gel’s elasticity, ξ is the degree of gel ionization, and Z is the volume of potential ions per dry gel volume. In this equation, k changes linearly with the degree of cross-linking. Physically, it was determined that the hydrogel possessing 1 mol % TEGDA was very loose and had very low mechanical integrity when compared with the other hydrogels. Thus, eq 2 was tested for hydrogel formulations containing 3-12 mol % TEGDA for which each

Table 5. Diffusion Half Life and the Activation Energy for Diffusion of FcCOOH within 3, 5, 7, 9, and 12 mol % Cross-Linked p(HEMA-co-PEGMA-co-HMMA) Hydrogels half life (min)

hydrogel-3% hydrogel-5% hydrogel-7% hydrogel-9% hydrogel-12%

10 °C

15 °C

20 °C

25 °C

30 °C

35 °C

40 °C

45 °C

activation energy (kJ/mol)

4.1 6.1 11.8 17.2 25.2

3.2 5.3 7.6 13.7 19.8

2.8 4.1 6.8 7.7 10.1

2.3 2.9 4.4 5.9 9.1

1.6 2.0 3.7 3.9 6.5

1.5 1.7 2.3 3.1 3.7

1.2 1.7 1.8 2.8 2.4

1.0 1.2 1.5 2.0 1.9

30.1 35.9 45.0 46.5 56.9


hydrated hydrogel density was assumed to be the same and similar to that of water. Figure 4 shows the relationship between the swollen/dry gel volume ratio of p(HEMA-coPEGMA-co-HMMA) hydrogels and the mole percent of TEGDA, which represents the degree of cross-linking. This relationship confirms that the -2/3 power of swollen/dry gel volume ratio varies linearly with the mole percent of the cross-linker within the hydrogel (r2 ) 0.9685, n ) 2). This linear relationship establishes that the subject hydrogels display hydration and swelling characteristics that are dependent upon gel elasticity, which is a function of the degree of cross-linking.23 3.2. Transport of FcCOOH within p(HEMA-co-PEGMAco-HMMA) Hydrogels. Mediator molecules such as FcCOOH must remain immobilized (physically entrapped or tethered) and available within the enzyme containing hydrogel in order to successfully function as redox mediators. The mediator must also possess the highest possible diffusivity so that it does not become transport limiting in the multistep, mediated, electron transfer process. Hence, the effect of hydrogel cross-link density on the transport coefficient of physically entrapped FcCOOH allows the design of biosensors based on these principles to achieve optimized performance. It is customary to evaluate the transport mechanism by fitting the release kinetic data to well established transport models. Release data were fitted to the following:35,36

Higuchi model: Mt ) kt

1/2

Hixson-Crowell cube root model: Wo1/3 - Wt1/3 ) kt

(3)

(4)

Korsemeyer-Peppas model: Mt/M∞ ) ktn

(5)

First-order model: ln(M0/Mt) ) kt

(6)

Zero-order kinetic model: Mo - Mt ) kt

(7)

and

where, Mo is the FcCOOH amount taken at time equal to zero, and Mt and M∞ are the concentrations of FcCOOH at a particular time, t, and at infinite time, respectively. Wo and Wt correspond to the weight of the FcCOOH taken initially and dissolved FcCOOH at time t, respectively. The term k is the release kinetic constant. Table 3shows the performance of each of the evaluated models, reflected in the coefficient of variation (r2), when evaluated with the temperature-dependent release data over the full 1-12 mol % cross-link density compositional space. The Korsemeyer-Peppas model is seen to best represent the release of FcCOOH from within p(HEMA-co-PEGMA-co-HMMA) hydrogels over all compositions and temperatures studied.4,24 In this model, Mt/M∞ is the fraction of solute (FcCOOH) released, Mt is the concentration of solute released at time t, M∞ is the concentration of solute released at equilibrium, t is the experimental time, k is the release constant that characterizes the mode of transport of the solute outside the matrix, and n is the release index, which can be obtained directly from the linear portion of the slope of the graph ln(Mt/M∞) versus ln(t) for Mt/M∞ < 0.6.24,25 For n ) 0.5, solute release follows a Fickian mechanism. When 0.5 < n < 1, the release follows


2141

anomalous diffusion behavior, and for n ) l, the release shows a case II diffusion. Figure 5 is a plot of Mt/M∞ versrus t for the release of FcCOOH from 1, 3, 5, 7, 9, and 12 mol % crosslinked p(HEMA-co-PEGMA-co-HMMA) hydrogel slabs into 0.1 M HEPES/0.1 M KCl at 20 °C (n ) 2). For experiments conducted with 1-12 mol % TEGDA, all experimentally determined values of n were found to closely approximate 0.5 (1 mol %, n ) 0.49; 3 mol % n ) 0.48; 5 mol %, n ) 0.48; 7 mol %, n ) 0.46; 9 mol %, n ) 0.44; and 12 mol %, n ) 0.42) but to decrease with an increase of cross-link density. For the case of n ) 0.5 (Fickian transport), eq 5 can be rearranged as the Higuchi square-root of time relationship or the Hixson-Crowell cube root law:

Mt/M∞ ) 4/π1/2(Dt/L2)n ; 0 e Mt/M∞ e 0.6

(8)

where D is the diffusion coefficient of the FcCOOH and L is the slab thickness. Accepting the release behavior to be via a Fickian diffusion mechanism, all calculations were performed using a value of n ) 0.5. Equation 8 may be rearranged in terms of the half-life of FcCOOH release, thus

D ) 0.00017676/t1/2

(9)

Table 4 shows the calculated diffusion coefficient, the diffusion exponent, n, and the half-life for the diffusion of FcCOOH within 1, 3, 5, 7, 9, and 12 mol % TEGDA crosslinked p(HEMA-co-PEGMA-co-HMMA) hydrogels at RT, and Figure 6 shows the trend in these parameters with the mole percent cross-linker, TEGDA. The diffusion coefficient of FcCOOH within the hydrogel can be seen to originate at 2.64 × 10-8 cm2/s for the 1 mol % TEGDA cross-linked hydrogel and to fall asymptotically to 4.87 × 10-9 cm2/s at 12 mol % TEGDA cross-linked hydrogel. At 1 mol % TEGDA, the diffusion coefficient is a full 2 orders of magnitude less than that found in buffer solutions (4.51-5.73 × 10-6 cm2 s-1).37,38 The n parameter is seen to originate at 0.5 for the 1 mol % TEGDA cross-linked hydrogel but nonetheless falls linearly to 0.42 at 12 mol % TEGDA cross-linked hydrogel. 3.3. Temperature Dependence of Transport of FcCOOH within p(HEMA-co-PEGMA-co-HMMA) Hydrogels. In addition to RT studies, hydrogels were also studied at multiple temperatures for their temperature dependence of the transport coefficient and to gain further insight into the thermally activated transport mechanism. Release profiles of FcCOOH (duplicate runs) from hydrogels with 3, 5, 7, 9, and 12 mol % cross-linker were measured at various temperatures (10, 15, 20, 25 30, 35, 40, and 45 °C) as a function of incubation time, as exemplified in Figure 7for the 3 mol % TEGDA hydrogel. From these release profiles, the measured half-life values corresponding to the various cross-linker concentrations in the hydrogel at eight different temperatures were determined and are summarized in Table 5. The temperature dependence of the diffusion coefficient of FcCOOH within 3, 5, 7, 9, and 12 mol % cross-linked p(HEMA-co-PEGMA-co-HMMA) hydrogels in 0.1 M HEPES/ 0.1 M KCl (n ) 2) displayed similar trends and consistently followed a second-order polynomial relationship with respect to temperature. As the amount of TEGDA incorporated into the hydrogel was increased, the structure became more rigid, the diffusion half-life was increased and the diffusion coefficient decreased. For each release study, the diffusion coefficient of FcCOOH in the hydrogel is dependent on a number of factors, such as the structure of the polymeric network, the polymer-

2142



solute interaction, and the degree of swelling. As expected, while the ratio of TEGDA in the hydrogel formulation was decreased, there was a corresponding increase in solute diffusion coefficient. This behavior can be explained as resulting from the greater influx of water into the less tightly cross-linked hydrogel and thus an enhanced rate of solute diffusion out of the swollen hydrogel. A plot of the diffusion coefficient versus the inverse of temperature on a semilog scale results in a straight line, as shown in Figure 8. The slope of this line allows the estimation of the activation energy for diffusive transport using the familiar Arrhenius equation:

D ) Ae-E/RT

(10)

where E is the required activation energy for diffusion, R is the molar gas constant, and A is pre-exponential collision factor. The activation energies of 30, 36, 45, 47, and 57 kJ/mol for hydrogels corresponding to 3, 5, 7, 9, and 12 mol % of TEGDA, respectively, agree very well with the thermally activated diffusive release of similar solutes from hydrogels; e.g., the calculated activation energy for the diffusion of paracetamol from a chitosan hydrogel was reported to be between 25 and 30 kJ/mol25 or the diffusion of acetic acid from p(HEMA).28 The magnitude of the activation energy for diffusion of FcCOOH within the p(HEMA-co-PEGMA-coHMMA) hydrogels correlates with the cross-link density, as indicated in Figure 9. 3.4. Immobilization of FcCOOH within p(HEMA-coPEG-co-HMMA) Hydrogels. Fc-AEMA (Scheme 1) and FcPEG(3500)-AEMA (Scheme 2) were separately synthesized, characterized, and incorporated within hydrogel precursor mixtures. Hydrogel slabs of covalently bonded Fc-AEMA and Fc-PEG(3500)-AEMA were separately prepared for release study of ferrocene into 0.1 M HEPES/0.1 M KCl (pH ) 7.4). Hydrogel slabs were visibly yellow, and ultraviolet-visible (UV-vis) spectroscopy showed clear evidence of the occlusion of the ferrocene moiety within the hydrogel slab both before and after the immersion period. The release study revealed that, unlike the unmodified FcCOOH, which resulted in the release of 98.6% of electrochemically detectable Fc, the Fc-AEMA was not released from the hydrogels to concentrations above the detection limit of the electrochemical technique following 5 days of continuous immersion (Table 2). For hydrogel slabs containing Fc-PEG(3500)-AEMA, it was found that measurable ferrocene concentration equivalent to 16% of the loaded amount was released into HEPES/KCl solution following 5 days of continuous immersion. The release of 16% of loaded Fc from Fc-PEG(3500)-AEMA containing hydrogel slabs is believed to be the result of PEG hydrolysis.29,30 Hydrolysis of PEG amide linkages is a source of some concern39 that extends to their use as pendant PEG moieties within the biorecognition of p(HEMAco-PEGMA-co-HMMA) hydrogel layers of implantable oxidoreductase enzyme-based biosensors.

4. Conclusions Hydrogels formed from HEMA, PEGMA, and HMMA [p(HEMA-co-PEGMA-co-HMMA) hydrogels] were synthesized to contain varying amounts of the cross-linker, TEGDA, and loaded with the redox mediator, FcCOOH. Hydrogels were shown to swell in accordance with established Flory-HugginsRehner theory and to release FcCOOH from these base hydrogel compositions at rates that reflect the cross-link density or the MW between cross-links. The release of FcCOOH from these

hydrogels is decidedly Fickian with overall diffusion coefficients that ranged between 4.75 × 10-8 cm2/s (1 mol % TEGDA) to 2.00 × 10-9 cm2/s (12 mol % TEGDA) and with n parameters that approximated 0.5 but nonetheless linearly declined from 0.49 (1 mol %) to 0.42 (12 mol %). The measured value of D was found to be strongly correlated with the cross-link density and the degree of hydration of the hydrogels. The temperature dependence of the diffusivity gave insight into the thermally activated energetics of transport through these gels resulting in activation energies that were 30 kJ/mol (3 mol %), 36 kJ/mol (5 mol %),45 kJ/mol (7 mol %), 47 kJ/mol (9 mol %) and 57 kJ/mol (12 mol %). UV-polymerizable monomers of FcCOOH, Fc-AEMA, and Fc-PEG(3500)-AEMA were covalently coupled within the hydrogel network and investigated. While Fc-AEMA showed no evidence of release from the hydrogel, its PEGconjugated equivalent, Fc-PEG(3500)-AEMA, showed evidence of release of redox active Fc after 5 days of immersion. This is believed to be the result of PEG-amide hydrolysis. This study is beneficial for further applications of hydrogels in controlled drug delivery devices, implantable biosensors, and several drugeluting prosthetic devices such as stents and intraocular implants. Acknowledgment. This work was supported by the US Department of Defense (DoDPRMRP) Grant PR023081/ DAMD17-03-1-0172 and by the Consortium of the Clemson University Center for Bioelectronics, Biosensors and Biochips. The authors thank Dr. Ashwin Rao for support in the synthesis of Fc derivatives.

References and Notes (1) Park, K. Park, H. Shalaby, W. S. W. Biodegradable Hydrogels for Drug DeliVery; CRC Press: Boca Raton, FL, 1993; pp 262. (2) Klomp, G.; Hashiguchi, H.; Ursell, P.; Takeda, Y.; Taguchi, T.; Dobelle, W. J. Biomed. Mater. Res. 1983, 17, 865. (3) Owino, J.; Arotiba, O.; Baker, P.; Guiseppi-Elie, A.; Iwuoha, E. React. Funct. Polym. 2008, 68 (8), 1239. (4) George, K.; Wentrup-Byrne, E.; Hill, D.; Whittaker, A. Biomacromolecules 2004, 5, 1194. (5) Guiseppi-Elie, A.; Brahim, S.; Narinesingh, D. AdV. Mater. 2002, 14 (10), 743. (6) Brahim, S.; Narinesingh, D.; Guiseppi-Elie, A. Biomacromolecules 2003, 4, 497. (7) Pellissier, M.; Zigah, D.; Barrie‘re, F.; Hapiot, P. Langmuir 2008, 24, 9089. (8) Pekel, N.; Salih, B.; Guven, O. J. Biomater. Sci., Polym. Ed. 2005, 16, 253. (9) Kim, D.-N.; Lee, W.; Koh, W.-G. Anal. Chim. Acta 2008, 609, 59. (10) Soto, C.; Patterson, C.; Charles, P.; Martin, B.; Spector, M. Biotechnol. Bioeng. 2005, 92, 934. (11) Rubina, A. Y.; Pan’kov, S. V.; Dementieva, E. I.; Pen’kov, D. N.; Butygin, A. V.; Vasiliskov, V. A.; Chudinov, A. V.; Mikheikin, A. L.; Mikhailovich, V. M.; Mirzabekov, A. D. Anal. Biochem. 2004, 325, 92. (12) Luo, Y.; Kirker, K. R.; Prestwich, G. D. J. Controlled Release 2000, 69, 169. (13) Metters, A. T.; Anseth, K. S.; Bowman, C. N. Polymer 2000, 41, 3993. (14) Peppas, N. A.; Sahlin, J. J. Biomaterials 1996, 17, 1553. (15) Kyritsis, A.; Pissis, P.; Grammatikakis, J. J. Polym. Sci., Part B: Polym. Phys. 1995, 33, 1737. (16) Pishko, M. V.; Revzin, A.; Simonian, A. L. Sensors 2002, 2, 79. (17) Abraham, S.; Brahim, S.; Ishihara, K.; Guiseppi-Elie, A. Biomaterials 2005, 26, 4767. (18) Kabra, B.; Gehrke, S.; Hwang, S.; Ritschel, W. J. Appl. Polym. Sci. 1991, 42, 2409. (19) Alfrey, T. J. Polym. Sci., Part C 1966, 12, 249. (20) Mack, E. J. Okano, T. Kim, S. W. In Hydrogels in Medicine and Pharmacy-Polymers; Peppas, N. A. , Ed.; CRC Press: Boca Raton, FL, 1988; Vol. II, p 65. (21) Elvira, C.; Mano, J.; San Roma´n, J.; Reis, R. Biomaterials 2002, 23, 1955. (22) Flory, P.; Rehner, J. J. Phys. Chem. 1943, 11, 521.

FcCOOH Release from p(HEMA-co-PEGMA-co-HMMA) Hydrogels (23) Brahim, S.; Narinesingh, D.; Guiseppi-Elie, A. Biomacromolecules 2003, 4, 1224. (24) Gayet, J.; Fortier, G. J. Controlled Release 1996, 38, 177. (25) Falk, B.; Garramone, S.; Shivkumar, S. Mater. Lett. 2004, 58, 3261. (26) Lee, W.; Lin, W. J. Polym. Res. 2002, 9, 23. (27) Sun, Y.; Gombotz, W.; Hoffman, A. J. Bioact. Compat. Polym. 1986, 1, 316. (28) Trigo, R.; Blanco, M.; Huerta, P.; Olmo, R.; Teijo´n, J. Polym. Bull. 1993, 31, 577. (29) Zalipsky, S. AdV. Drug DeliVery ReV. 1995, 16, 157. (30) Roberts, M.; Bentley, M.; Harris, J. AdV. Drug DeliVery ReV. 2002, 54, 459. (31) (a) Abraham, S.; Guiseppi-Elie, A. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2005, 4, 4099. (b) Guiseppi-Elie, A.; Brahim, S.; Slaughter, G.; Ward, K. R. IEEE Sens. J. 2005, 5 (3), 345–355. (c) Rahman, A. R. A.; Justin, G.; Guiseppi-Elie, A. Biomed. MicrodeVices: BioMEMS Biomed. Nanotechnol. 2009, 11 (1), 75.


2143

(32) Guiotto, A.; Canevari, M.; Pozzobon, M.; Moro, S.; Orsolini, P.; Veronese, F. Bioorg. Med. Chem. 2004, 12, 5031. (33) Ignatova, M.; Manolova, N.; Rashkov, I.; Sepulchre, M.; Spassky, N. Macromol. Chem. Phys. 1995, 196, 1663. (34) Brahim, S.; Narinesingh, D.; Guiseppi-Elie, A. Biosens. Bioelectron. 2002, 17, 973. (35) Ahuja, N.; Katare, O.; Singh, B. Eur. J. Pharm. Biopharm. 2007, 65, 26. (36) Sahoo, J.; Murthy, P.; Biswal, S.; Sahoo, S.; Mahapatra, A. AAPS PharmSciTech 2008, 9 (2), 577. (37) Justin, G. Rahman, A. Guiseppi-Elie, A. Electroanalysis 2009, 21 (10), 1125-1134. (38) Bartlett, P.; Pratt, K. J. Electroanal. Chem. 1995, 397, 53. (39) Buckley, J. Finn, R. Mo, J. Bass, L. Ho, S. In Process Chemistry in the Pharmaceutical Industry: Challenges in an EVer Changing Climate; CRC Press: Boca Raton, FL, 2007; Chapter 24.

BM900299B

Immobilization and Release of the Redox Mediator Ferrocene

Recommend Documents