Hydroxamic Acid-Containing Hydrogels for Nonabsorbed Iron

W. Harry Mandeville, and Pradeep K. Dhal*. Drug Discovery and Development, Genzyme Corporation, 153 Second Avenue,. Waltham, Massachusetts 02451...
0 downloads 0 Views 77KB Size
Biomacromolecules 2005, 6, 2946-2953

2946

Articles Hydroxamic Acid-Containing Hydrogels for Nonabsorbed Iron Chelation Therapy: Synthesis, Characterization, and Biological Evaluation Steven C. Polomoscanik, C. Pat Cannon, Thomas X. Neenan, S. Randall Holmes-Farley, W. Harry Mandeville, and Pradeep K. Dhal* Drug Discovery and Development, Genzyme Corporation, 153 Second Avenue, Waltham, Massachusetts 02451 Received January 17, 2005; Revised Manuscript Received August 4, 2005

Iron overload is a severe clinical condition and can be largely prevented by the use of iron-specific chelating agents. A successful iron chelator needs to be orally active, nontoxic, and selective. In this study, hydrogels containing pendant hydroxamic acid groups have been synthesized as potential nonabsorbed chelators for iron in the gastrointestinal tract. The synthetic method employed to introduce hydroxamic acid groups to polymer chains involved reaction of polymer gels based on N-acryloxysuccinimide, acryloyl chloride, and (2-hydroxyethyl)acrylate monomers with hydroxylamine. These hydroxamic acid-functionalized polymer gels swell favorably in water and effectively sequester iron. In vitro iron-binding properties of these hydrogels were evaluated from their binding isotherms by use of iron(II) alone and in the presence of other competing metal ions. These polymers bind iron over a broad pH range. The iron-binding properties of the polymers were found to depend on the concentration of hydroxamate groups on polymer chains. The in vivo ironbinding efficacy of the polymers was evaluated in rat as the animal model. The polymers prevented an increase in serum hemoglobin and hematocrit levels in the animals, thus suggesting the prevention of systemic absorption of dietary iron from the gastrointestinal tract. The animals also maintained normal body weight during the treatment period, indicating the absence of any apparent toxicity associated with these polymers. Introduction Iron overload in humans (caused by genetic disorders such as hereditary hemochromatosis and β-thalassemia major) can lead to highly toxic and lethal conditions such as cardiomyopathy and hepatic fibrosis.1 Since there is no physiological mechanism for excreting excess accumulated iron, artificial means are needed to remove the excess iron. In the case of hereditary hemochromatosis, defective genes lead to increased intestinal hyperabsorption of dietary iron, which leads to systemic iron overload. Phlebotomy has been demonstrated to prevent the serious side effects of iron overload in the case of hemochromatic patients. Although phlebotomy has been generally the method of treatment for hemochromatosis, the frequency of this treatment (in some cases twice a week for one year followed by 3-4 times a year in subsequent years) does not make it ideal from both the cost and patient compliance point of view.2 There are potentially two alternatives to phlebotomy: (i) minimize iron intake by adapting to an iron-deficient diet, and (ii) use oral iron chelators to sequester and remove dietary iron in the gastrointestinal tract, thus preventing the pool of available iron for hyperabsorption. These two approaches could be used as a preventative measure to maintain iron balance but, more importantly, could be used to achieve a negative iron balance when there is already an iron overload condition. Therefore, with a

highly efficacious iron chelator, the frequency of phlebotomy could be reduced or possibly eliminated. Chelation therapy to remove excess iron has proven its benefits for the treatment of β-thalassemia patients who do not have adequate erythropoietic reserve for phlebotomy treatment.3 The widely used agent for this purpose is desferrioxamine (DFO) (1). Although it is an excellent iron chelator, the toxicity of this compound and its parenteral mode of administration have led to poor patient compliance.4

In humans and other vertebrates, iron is obtained from dietary intake through absorption by mucosal cells in the jejunum.5 Divalent metal transporter 1 (DMT1) is a transporter that is responsible for active uptake of ferrous iron

10.1021/bm050036p CCC: $30.25 © 2005 American Chemical Society Published on Web 10/07/2005

Hydrogels for Nonabsorbed Iron Chelation Therapy

from the gut.6 The binding of ferric ions by glycine-extended gastrin at low pH is consistent with a role for progastrinderived peptides in iron uptake from the lumen of the gastrointestinal tract.7 Development of a polymeric iron chelator that is nonabsorbed by the intestinal mucosa and has sufficient capacity and strength to bind iron throughout the gastrointestinal tract over its dynamic pH range would be an ideal candidate for treating iron overload condition. Nonabsorbed polymer therapies that act by sequestering a number of undesired ionic species in the gastrointestinal tract have been successful clinically.8-10 In the present study, we have explored the possibility of using an insoluble polymeric iron chelator to sequester and remove dietary iron from the gastrointestinal tract, thereby minimizing the intestinal hyperabsorption of iron. Particularly, this polymeric approach to sequester dietary iron may lead to potential treatment for hemochromatosis. Experimental Section Materials and Methods. Unless stated otherwise, all reagents were obtained from Aldrich and were used as received. N-Acryloxysuccinimide (NAS) was synthesized following a reported procedure.11 Elemental analysis and atomic absorption measurements were carried out at QTI Laboratories. Synthesis of Poly(acrylohydroxamic acid) Gel from Polymeric Acid Chloride Precursor: (A) Synthesis of Cross-Linked Acryloyl Chloride-Divinylbenzene Copolymer Gel. In a 500 mL, three-necked, round-bottomed flask were taken 78 mL of acryloyl chloride, 5.85 mL of 80% divinylbenzene (DVB), 78 mL of toluene, and 1.3 g of 2,2′-azobis(isobutyronitrile) (AIBN). The reaction mixture was stirred and was bubbled with a gentle stream of N2 for 75 min. The reaction mixture was subsequently heated to 60 °C with stirring while under N2 atmosphere and allowed to proceed for 72 h. After cooling to room temperature, the formed solid was broken down into smaller pieces and stirred with 250 mL of tetrahydrofuran (THF) for 15 min. This suspension was poured into 500 mL of acetonitrile and stirred for 15 min. The polymer particles were filtered, stirred with 600 mL of hexane for 30 min, and filtered again. After the hexane washing was repeated one more time, the polymer was dried in a forced-air oven at 60 °C. The yield of the polymer was 74 g. (B) Synthesis of Cross-Linked Acrylohydroxamic AcidDVB Copolymer Gel. To a solution containing 87.0 g of 50% NaOH and 200 mL of deionized water was added 75.0 g of hydroxylamine hydrochloride. After cooling to room temperature, this hydroxylamine solution was slowly added to a vigorously stirred suspension containing 20.0 g of the above acryloyl chloride/DVB copolymer and 250 mL of methylene chloride. During the initial addition, the reaction mixture was maintained at or below room temperature with an ice bath. The reaction mixture was allowed to stir for 48 h at room temperature. At the end of this time period, 500 mL of 2-propanol was then added to thin out the thick suspension, which was then filtered. The polymer particles were suspended in 1 L of deionized water, stirred for 30

Biomacromolecules, Vol. 6, No. 6, 2005 2947

min, and filtered. The filtered polymer was dried in a forcedair oven at 60 °C for 24 h, yielding 7.0 g of the polymer. Synthesis of Cross-Linked Poly(acrylohydroxamic acid) Gel from Polymeric Active Ester Precursor: (A) Synthesis of Cross-Linked NAS-co-DVB Copolymer. In a 100 mL, three-necked round-bottomed flask were taken 9.76 g of NAS, 0.84 mL of 80% DVB, 104 mg of AIBN, and 25 mL of N,N-dimethylformamide (DMF), While it was stirred at room temperature, the reaction mixture was purged with a gentle stream of nitrogen for 60 min. Subsequently, the reaction mixture was warmed to 60 °C and was held at this temperature for 18 h under nitrogen. The polymer gel thus formed was broken up with a spatula, heated for an additional 2 h, and subsequently cooled to room temperature. (B) Hydroxyamidation of Poly(NAS-co-DVB) Gel. The above polymer gel (without further workup) was transferred to a 1 L, three-necked round-bottomed flask under nitrogen atmosphere. In a separate 500 mL, three-necked roundbottomed flask were taken 250 mL of DMF and 20.5 g of hydroxylamine hydrochloride. To this suspension, 41.2 mL of triethylamine was added slowly, and the resulting reaction mixture was stirred under nitrogen for 15 min. The triethylamine hydrochloride salt was quickly filtered off, and the filtrate was added to the poly(NAS-co-DVB)/DMF suspension. The reaction mixture was stirred for 60 h at room temperature under nitrogen. At the end of this time period, the reaction mixture was filtered. The polymer particles were stirred in 1 L of methanol for 30 min, filtered, suspended in 1 L of deionized water, and stirred for 30 min. After filtration, the wet polymer gel was dried at 60 °C in a forced-air oven for 18 h, yielding 3.0 g of an off-white solid. Synthesis of Cross-Linked Poly(acrylohydroxamic acid) Gel from Polymeric Ester Precursor: (A) Synthesis of Cross-Linked (2-Hydroxyethyl)acrylate-co-DVB Copolymer. In a 5 L, three-necked round-bottomed flask were taken 297 g of HEA, 17.5 g of 80% DVB, 3.15 g of AIBN, and 920 mL of ethanol. While it was stirred at room temperature, the reaction mixture was purged with a gentle stream of nitrogen for 90 min. Subsequently, the reaction mixture was warmed to 60 °C and was held at this temperature, under nitrogen, for 3 h. The temperature of the reaction mixture was raised to 75 °C and was kept at this temperature for 24 h. After cooling to room temperature, the formed gel was broken into smaller pieces, suspended in 2 L of ethanol, stirred for 2 h, and filtered. The filtered polymer gel was resuspended in 2 L of ethanol, stirred for 1 h, and filtered. The filtered particles were dried in a forced-air oven at 60 °C, yielding 225 g of an off-white soft solid. (B) Hydroxyamidation of Poly(HEA-co-DVB) Gel. In a 5 L, three-necked round-bottomed flask were taken 120 g of poly(HEA-co-DVB) gel and 1.2 L of deionized water. The polymer suspension was stirred for 30 min, and 600 mL of a 50% aqueous solution of hydoxylamine was then added. The reaction mixture was stirred at room temperature for 30 min and at 55 °C for 48 h. After the reaction mixture was cooled to room temperature, 2 L of methanol was added. After the mixture was stirred for 5 min, the polymer gel was filtered and dried at 60 °C in a forced-air oven for 12 h. The dried polymer was ground, suspended in 3 L of deionized

2948

Biomacromolecules, Vol. 6, No. 6, 2005

water, stirred for 30 min, and filtered. This water washing process was repeated four more times. The filtered polymer was dried in a forced-air oven at 60 °C, yielding 56.2 g of an off-white solid. In Vitro Metal Binding Studies. The in vitro metal binding experiments were carried out with ferrous sulfate as the iron source. The amount of iron bound to the polymer was estimated by spectrophotometry and atomic absorption spectroscopy after the polymer particles were filtered from the test solution. The results obtained from both methods agreed closely with one another. The general procedure for the in vitro iron-binding test for a single concentration is given below. For measurement of the iron-binding isotherm, the concentration of metal salt in the test solution was proportionately varied. Thus, to 25 mg of the chelating polymer gel taken in a 100 mL beaker was added 50 mL of the Fe2+ solution. This Fe2+ solution was prepared by adding ∼13.3 g of NaCl, ∼0.09 g of FeS04‚7H2O, and ∼0.06 g of citric acid to 1 L of deionized H2O. The initial pH of this solution was ∼3.8. The pH of the suspension was adjusted to appropriate values with 0.1 or 1.0 N sulfuric acid (for lower pH) and with 0.1 or 0.5 N NaOH (for higher pH). After the suspension was stirred at the desired pH for 3 h, it was filtered. To 25 mL of the filtrate was added 1.5 mL of 0.3% (w/w) aqueous phenanthroline and 0.5 mL of 10.0% (w/w) aqueous hydroxylamine hydrochloride. The pH was adjusted to 3.5, and the volume of the solution was adjusted to 30.0 mL with deionized water. The solution was stirred for 5 min and was subsequently allowed to stand for ∼3 h. The concentration of the iron in the solution was estimated by measuring its corrected absorbance at 508 nm (A508corr ) A508meas - {(A400 + A616)/2}. A standard curve produced by using known concentrations of FeSO4 solutions was used for the estimation of metal ion concentration. In Vivo Studies. Male Wistar (from Charles River Labs) rats weighing 51-75 g were housed in groups in plastic shoebox cages on Alpha Chip bedding with free access to 20 ppm iron chow or a custom-formulated iron-free diet. Hydroxamic acid-functionalized polymer was mixed with diet. The polymer was added at 0.2% and 1 wt %. The treatment animals remained on the iron-deficient diets for 2 weeks, while the control group received normal diet containing iron. At the end of 2 weeks, the iron-deficient animals were randomized into four groups. The control group continued to receive iron chow. The first treatment group continued to receive the iron-deficient diet, the second was switched from the iron-deficient to the 20 ppm iron diet, the third group received the iron diet containing 0.2% of the chelating polymer, and the fourth group received the iron diet containing 1% of the chelating polymer. Blood samples were taken weekly from each animal. Hematocrit and hemoglobin contents in the blood samples were measured. Results and Discussion Design and Syntheses of Iron Chelating Polymers. The desired nonabsorbed polymeric iron chelator needs to maintain or produce negative iron balance by competitively

Polomoscanik et al.

binding and removing dietary iron from intestinal hyperabsorption. A clinically relevant chelator needs to possess a sufficiently strong association constant for iron. Iron is present in both +2 and +3 oxidation states in the gut, and these oxidation states can interchange under physiological conditions. The pH of the gastrointestinal (GI) tract and the presence of citrate, ascorbate, and other agents in the GI tract may contribute to this redox process.12 Therefore, it may become necessary to sequester both ferrous and ferric iron. Furthermore, specificity of the chelator toward iron over other physiologically important metal ions such as zinc, copper, and calcium is also necessary to avoid undesired depletion of these important metal ions. In addition to these primary ligand design issues, other physiochemical properties of the polymer gels need to be considered. To accomplish maximum utilization of the polymer-bound ligands, adequate swelling of the polymer gel in the gastrointestinal tract and favorable steric factors need to be considered. Furthermore, the polymer needs to be biocompatible. Among various iron-chelating ligands known in the literature, hydroxamic acid derivatives are particularly appealing. This ligand has a high affinity for both Fe2+ and Fe3+. For example, the therapeutically relevant hydroxamatebased iron chelator DFO (1) binds iron very strongly. The binding strength of DFO toward Fe(III) is very strong with an affinity constant of ∼29, but it binds Fe(II) somewhat modestly with an affinity constant value ∼7.13 The proposed mode of chelation of iron ions by hydroxamic acid-containing ligands involves the positively charged iron being coordinated by two hydroxamic acid groups in the case of Fe(II) and by three hydroxamic acid groups in the case of Fe(III).14 The coordination geometry of these metal complexes is illustrated in Figure 1. Unlike other chelators such as polyamines that retain coordinating ability under basic conditions, hydroxamic acid groups preserve their coordinating property over a broader pH range. The high affinity of DFO toward iron has been attributed to its multidentate ligand character resulting in all donors in a single molecular structure, which maximizes thermodynamic stability of the metal complex.15 The above finding suggests that preparation of an insoluble polymer bearing multiple hydroxamic acid groups would mimic the iron-binding properties of DFO. Being nonabsorbed, the resulting polymeric chelator would not present systemic side effects that are associated with DFO. Design of polymeric ligands based on low molecular weight coordination chemistry has led to a number of polymeric metal chelators over the years that have found applications ranging from separation to catalysis. These polymer-bound metal chelators maintain coordination geometries similar to their low molecular weight counterparts.16 For example, polymeric resins with modest concentrations of hydroxamic acid groups have been prepared in the past and have been used as ion-exchange resins.17 Covalent conjugates of DFO with various biocompatible polysaccharides such as dextran and hydroxyethyl starch have been synthesized and evaluated for their iron-binding properties under in vitro and in vivo conditions.15b The in vitro iron-binding property of DFO was maintained after it was linked to polysaccharides. These

Hydrogels for Nonabsorbed Iron Chelation Therapy

Biomacromolecules, Vol. 6, No. 6, 2005 2949

Figure 1. Schematic representation of complexation of Fe(II) and Fe(III) ions by hydroxamate ligands. Scheme 1. Synthesis of Polymeric Hydroxamic Acid Derivatives from Acid Chloride and Active Ester Precursors

findings suggest that iron-binding properties of polymerbound hydroxamic acids will be similar to those of low molecular weight multidentate ligands. Since hydroxamic acid groups act as free radical inhibitors, direct polymerizations of hydroxamic acid-containing vinyl monomers has met with little success. Therefore, we adopted a postpolymerization chemical modification procedure (using appropriate precursor polymer) to prepare the desired polymeric hydroxamic acid. Since the concentration of hydroxamic acid groups is critical for enhanced binding of iron, the choice of appropriate chemical modification condition(s) is key to obtaining polymers with high ligand density. Preparation of low molecular weight hydroxamic acid derivatives involves the reaction of different carboxylic acid derivatives (e.g., acid chloride, alkyl esters, active esters, etc.) with hydroxylamine.18 These methods have been evaluated in the present case to prepare the target hydroxamic acidcontaining hydrogels. At first, lightly cross-linked gels of acryloyl chloride and NAS were prepared by polymerizing these functional monomers with DVB (cross-linking monomer). The resulting polymer gels were subsequently treated with hydroxylamine to obtain the desired poly(acrylohydroxamic acid) gels. Preparation of hydroxamic acid-containing polymers from these two precursors is shown in Scheme 1. Neither of the methods was found to be suitable. In the case of poly(acryloyl chloride) precursor, in addition to the undesired smell of the monomer (which makes the workup of the polymer unpleasant), the reaction of this precursor polymer with hydroxylamine brought about a low degree of conversion. The maximum degree of conversion of the acid chloride groups to hydroxamic acid groups in the polymer

was about 35%. The remaining acid chloride groups were hydrolyzed into carboxylic acid groups. Such a low degree of functionalization produces low-capacity chelating resins. Although efficient incorporation of hydroxamic acid groups into the polymer backbone was possible with the poly(NAS) precursor, incorporation of a high level of hydroxamic acid groups required anhydrous experimental conditions. The reaction of ester groups with hydroxylamine is a classical way to prepare hydroxamic acid derivatives. However, implementation of this approach for polymeric systems, such as transformation of poly(alkyl acrylate) to the corresponding poly(acryloyl hydroxamate), has been reported to involve harsher reaction conditions and result in a lower degree of substitution.19 The hydrophobicity of alkyl acrylate polymers makes them insoluble or nonswellable in water. The latter is the preferred medium to carry out this ester-hydroxamate exchange reaction. To overcome this shortcoming, we considered using polymers derived from hydrophilic acrylic monomers as the precursors. Toward this end, we used (2-hydroxyethyl)acrylate (HEA) as the functional monomer. Gels derived from this monomer sufficiently swell in water. Polymers based on this monomer are also known to be biocompatible.20 Copolymerization of HEA with a cross-linking monomer, such as DVB or ethylene glycol dimethacrylate (EGDMA), produced insoluble gels. These polymer gels swelled quite well in water. The transformation of hydroxyethyl ester groups to hydroxamic acid groups in the polymer gels occurred at milder temperature (45-55 °C) without the need for any base. Thus, by heating an aqueous suspension of the poly(HEA) gels with hydroxylamine at 55°C, a hydrogel containing over 80% hydroxamic acid groups along the polymer chain was formed. The reaction pathway for this synthesis is presented in Scheme 2. Earlier reports reveal the need for very high temperatures (>100 °C) as well as strong bases such as sodium methoxide or sodium hydroxide to transform polyacrylates to poly(acrylohydroxamic acid) derivatives.19,21 On the other hand, in the present system a significant degree of substitution was observed even at 45 °C. Increasing the reaction temperature

2950

Biomacromolecules, Vol. 6, No. 6, 2005

Polomoscanik et al.

Scheme 2. Synthesis of Polymeric Hydroxamic Acid Hydrogel from Poly[(2-hydroxyethyl)acrylate] as Precursor Polymer

Table 2. Iron-Binding Properties of Different Polymers (Single-Point Iron-Binding Experiments)a polymerb

Table 1. Compositions of Different Poly(hydroxamic acid)-Based Hydrogels Obtained by Chemical Modifications of Different Polymeric Precursorsa polymer

precursor monomer

Cross-linker

C/N

P1 P2 P3 P4 P5 P6

acryloyl chloride NAS NAS HEA HEA HEA

5% DVB 5% DVB 5% EGDMA 5% EGDMA 5% DVB 10% DVB

13.04 5.11 4.42 4.34 5.69 7.99

a NAS, N-acryloxysuccinimide; DVB, divinylbenzene; EGDMA, ethylene glycol dimethacrylate; HEA, (2-hydroxyethyl)acrylate.

to 55 °C led to a faster reaction, and a very high conversion was achieved within a short period of time. Transformation of the poly(HEA) to poly(acrylohydroxamic acid) was evident from the infrared (IR) spectral analysis of the precursor polymer and the product polymer. The characteristic bands in the IR spectrum of poly(HEA) are a sharp band at 1725 cm-1 corresponding to ester carbonyl stretching and a broad band at 3600-3400 cm-1 due to hydroxyl stretching. In the IR spectrum of the hydroxylamine-modified polymer, the 1725 cm-1 peak has completely disappeared. On the other hand, a new peak at 1633 cm-1 was evident that corresponds to the carbonyl stretching vibration of the hydroxamic acid group. Similarly, a new and intense broad band was evident from 3400 to 3200 cm-1 that corresponds to the NH stretching. No peak corresponding to acid carbonyl was evident in the IR spectrum, thus suggesting selective transformation of the ester group to the hydroxamic acid without the occurrence of any ester hydrolysis reaction. The latter is a problem with polymeric acid chloride and active ester precursor. The high degree of substitution was also evident from the elemental analysis of the resulting poly(acrylohydroxamic acid) gel. The physical characteristics of the modified polymer changed with progress of the reaction. For example, the reaction mixture became highly swollen in water due to increasing hydrophilicity of the poly(acrylohydroxamic acid) network. Increasing the cross-linking density of the poly(HEA) hydrogel was found to affect the reactivity of the ester groups toward hydroxylamine. For example, by increasing the DVB content of the polymer gel from 5 to 10 mol %, it became necessary to increase the reaction temperature to 75 °C to achieve any appreciable conversion of ester groups to the hydroxamic acid groups. The decrease in reactivity of functional groups in insoluble polymer matrices, with increasing cross-linking density, is known in the literature.22 Results on the extent of incorporation of hydroxamic acid groups in polymers by the above three methods of synthesis are summarized in Table 1. Evaluation of In Vitro Iron-Binding Properties: (A) Single Metal Ion Binding. The ability of these hydroxamic acid based hydrogels to chelate iron in vitro was assessed

binding pH

P1 P3 P4 P5

Higher pH 5.6 5.32 5.13 5.3

P1 P2 P3 P5

Lower pH 2.36 2.35 2.22 2.22

% free iron 0.2 0 0.3 3.0 20 3 10 13

a The concentration of iron in the binding solution is 0.32 mM. b See Table 1 for polymer compositions.

by an equilibrium-binding study with ferrous iron. While the literature is rich with regard to evaluation of hydroxamate ligands for binding Fe(III), very few references are available for binding studies involving these types of ligands with Fe(II). Since it may be necessary to bind both Fe(II) and Fe(III) in the GI tract, we felt it necessary to evaluate the binding of Fe(II) by these polymeric ligands. Polymeric ligands containing a modest number of hydroxamic acid groups have been reported to bind Fe(III) very strongly.23 These polymeric hydroxamates have been considered to maintain a similar coordination geometry as the low molecular weight hydroxamate ligands. Furthermore, the flexibility of the acrylate polymer backbone as well as local high concentration of the ligand group is likely to provide an appropriate number of ligands that would coordinate to metal ions with binding constants similar to those of low molecular weight polydentate ligands such as DFO. In a preliminary study to evaluate the Fe(III)-sequestering capability of our polymers, 25 mg of the polymeric hydroxamic gel (prepared from HEA precursor) was treated with an aqueous solution containing 175 ppm of Fe(III) solution and stirred for 5 min. The polymer gel was transformed into an intense red-colored resi, indicating the formation of the polymeric hydroxamateFe(III) complex. Filtration followed by addition of more polymer gel to the solution did not result in any change in color of the polymer resin. This suggests complete sequestration of Fe(III) by the polymeric resin at the first treatment. All subsequent experiments were carried out with Fe(II), with the high binding affinity of these ligands for Fe(III) and relevance of binding Fe(II) in biological systems kept in mind. At first, binding affinities of these polymers toward Fe(II) at a single concentration of the metal ion was determined to compare the relative potencies of polymeric hydroxamic gels prepared by the different methods. Subsequently, equilibrium-binding isotherms were obtained for the most promising polymers to obtain detailed information about their binding strengths and capacities. Measurements were carried out at different pH values to evaluate the effect of pH on the binding strength of the polymer. This may have relevance to the pH profile of the gastrointestinal tract. The single-point iron-binding affinities of different polymers measured at pH values of ∼5.5 and ∼2.2 are summarized in Table 2. In these experiments the amount of the polymer ligand and the concentration of metal ion in the

Biomacromolecules, Vol. 6, No. 6, 2005 2951

Hydrogels for Nonabsorbed Iron Chelation Therapy

Figure 2. Iron-binding isotherms of polymeric hydroxamic acid resins derived from cross-linked (a) polymeric acid chloride precursor and (b) polymeric hydroxyethyl ester precursor.

experimental solutions were kept closely similar for direct comparison, which would enable us to compare different polymers in terms of their relative iron-binding affinities. As can be seen from Table 2, all of the polymers showed quite similar affinities for iron at pH ∼5.5. This implies most polymer types bound nearly all the iron available in solution at this concentration. Metal binding studies carried out at pH 2.2 (Table 2) suggest that, although somewhat reduced, these polymeric hydroxamic acid derivatives maintain their binding affinity toward iron under acidic conditions. Thus, retaining their iron-binding properties over a broad pH range could make these iron-chelating hydrogels potential physiologically relevant nonabsorbed chelators for binding dietary iron in the gastrointestinal tract. While the poly(acrylohydroxamic acid) hydrogels prepared by the acid chloride route and ester route do not show any apparent differences in these single-point studies, their binding isotherms are clearly different. The latter study enabled us to distinguish their relative strengths as iron chelators. Binding isotherms obtained at pH ∼5.7 with these two poly(acrylohydroxamic acid) gels are shown in Figure 2. As can be seen from this figure, the poly(acrylohydroxamic acid) gel derived from the HEA polymer precursor is more effective as an iron chelator than that obtained from the acryloyl chloride polymer precursor. The dissociation constant (Kd) value (a measure of binding strength: the lower the Kd value, the stronger the binding strength) of HEAderived polymer was found to be 3.9 × 10-7 M-1 and that obtained from acid chloride precursor was 4 × 10-6 M-1, thus making the former a nearly 10 times stronger chelator than the latter. The binding strength of the HEA-derived polyhydroxamic acid toward Fe(II) is quite similar to that of DFO toward Fe(II).13 On the other hand, the polymeric ligand derived from the acid chloride precursor is an order of magnitude weaker. Furthermore, the maximum binding capacity (Bmax) of the former was found to be 81 mmol/g, and for the latter it was 48 mmol/g. Since the acid chloride route produces a polymer with ∼35 mol % hydroxamic acid groups and the ester precursor produces a polymer with ∼85 mol % hydroxamic acid groups, it appears that the ironbinding strength depends on the concentration of the coordinating ligands along the polymer chain. This finding supports the earlier proposed structure of iron-hydroxamate

Figure 3. Competitive iron-binding isotherm of polymeric hydroxamic acid resin in the presence of competing metal ions: (a) Fe2+ only, (b) with Ca2+, (c) with Zn2+, and (d) with Cu2+. Table 3. Iron-Binding Properties of Polymeric Hydroxamic Acid Resin in the Presence of Competing Metal Ions entry 1 2 3 4

competing metal ion none Ca2+ Zn2+ Cu2+

Kd (M-1) 10-7

3.9 × 1.7 × 10-7 3.4 × 10-7 2.6 × 10-6

Bmax (mmol/g) 0.46 0.12 0.21 0.29

complex, where divalent positively charged iron is stabilized with two negatively charged hydroxamic acid moieties forming a spirobicyclic structure around the metal center [2:1 ligand:Fe(II) complex]. In the case of polymer with a high concentration of hydroxamic acid groups, the probability of two hydroxamic acid groups at closer proximity to one another increases. Thus, polymers with higher concentrations of hydroxamic acid groups would favor the formation of the desired polymeric metal complex structure. This leads to polymeric chelators with higher capacity and increased binding strength. (B) Competitive Metal Ion Binding. To be a clinically effective chelator for treating iron-overload conditions, the polymeric ligand should possess selective affinity toward iron in the presence of other physiologically important metal ions. Toward this end, the iron-binding properties of the polymeric ligands were evaluated in the presence of some of the relevant metal ions. Physiologically important metal ions that could potentially compete with iron for binding with the polymeric chelator could include calcium, zinc, copper, manganese, etc. The poly(acrylohydroxamic acid) hydrogel prepared from poly(HEA) precursor was selected as the polymeric ligand for this competitive metal ion binding study. Equilibrium binding experiments were set up to measure the affinity of this polymer toward iron in the presence of other metal ions. Thus, solutions of binary metal ion pairs Fe2+/ Zn2+, Fe2+/Cu2+, Fe2+/Ca2+, and a mixture containing all of the above-mentioned metal ions were allowed to equilibrate with the polymer gel. The concentrations of the metal ions in the experimental solutions were chosen on the basis of the U.S. recommended daily allowance (RDA) data on the daily dietary uptake of these metal ions present in a normal meal.24 The binding capacity of this polymer toward iron present in the binary metal salt solutions containing these metal ions were measured at pH ∼5.7. The results are presented in Figure 3. The results on the binding parameters Kd and Bmax for these experiments are summarized in Table

2952

Biomacromolecules, Vol. 6, No. 6, 2005

Polomoscanik et al.

Table 4. In Vivo Iron-Binding Properties of Polymeric Hydroxamic Acid Resinsa

control animals with 20 ppm iron in diet control animals with iron-free diet 20ppm iron diet fed after iron-free diet treatment 20 ppm dietary iron + 0.2% (w/w) polymeric chelator 20 ppm dietary iron + 1% (w/w) polymeric chelator 20 ppm dietary iron + 2% (w/w) polymeric chelator a

initialb hematocrit, %

finalc hematocrit, %

initialb hemoglobin, g/dL

finalc hemoglobin, g/dL

43.2 43.2 32.9 32.9 32.9 32.9

47.75 25.6 40.0 39.1 36.5 33.2

11.45 11.45 8.9 8.9 8.9 8.9

13.5 6.75 10.9 10.75 9.38 6.5

In all cases p value is at least