Bioconjugate Chem. 2007, 18, 477−483
477
A Novel Affinity-Based Controlled Release System Involving Derivatives of Dextran with Enhanced Osmotic Activity Jyothi Devakumar and Vijayalakshmi Mookambeswaran* LIMTechS, Deptartment Ge´nie Biologique, Universite´ de Technologie de Compie`gne, 60205 Compiegne, France. Received March 1, 2006; Revised Manuscript Received October 5, 2006
Dextran is a highly biocompatible molecule with osmotic activity. We synthesized histidine derivatives of dextran, DexH (dextran histidine), to test the feasibility of an IMAC-based controlled release system. DexH was synthesized by the periodate oxidation method. The effect of periodate oxidation and histidine conjugation on osmotic activity was tested. The oxidized intermediate itself exhibited higher osmotic activity than native dextran. Conjugation with histidine further increased the osmotic activity, and the resulting DexH exhibited nine times more osmotic activity than native dextran. A positive correlation was observed between the extent of derivatization with histidine and osmotic activity. Association of DexH was tested on two of the matrices, namely, Cu-IDA-Novarose and Zn-IDA-Novarose. DexH bound to both these matrices, and only partial elution was achieved with stepwise lowering of pH, and complete elution was possible only with EDTA. Interestingly, it was found that DexH in its bound state (DexH-Cu-IDA-Novarose and DexH-Zn-IDA-Novarose) exhibited lesser osmotic activity than the eluted soluble form. The IDA-Cu and IDA-Zn-based solid supports bound strongly to DexH in a speciesdependent manner, as the IDA-Zn matrix selectively bound DexH with clustered histidine. Further, this DexH with clustered histidine shows higher osmotic activity. A controlled release system is proposed on the basis of this difference in the osmotic activity between the bound and eluted forms of DexH with EDTA as the external trigger to induce this transition in osmotic activity.
INTRODUCTION Osmotically active compounds swell in the presence of aqueous solutions as they imbibe water. The rate of swelling has been studied for a large number of synthetic as well as naturally occurring substances, as these compounds find a wide range of biomedical applications that include their usage as monomers in the preparation of hydrogel networks that are used in drug delivery, entrapment of cells, and so forth (1-3). In their monomeric state, they are used as tablet disintegrants wherein they control the delivery of the pharmaceutical agent in a spatially and temporally controlled manner, resulting in what are known as controlled release systems (4). Molecules such as glucose, hydroxyethylcellulose, glycerol, poly(ethylene glycol)s, and dextrans of different molecular weights are extensively employed osmotic agents (4-7). In controlled release systems, the osmotic agents usually form the expandable core contained within a semipermeable compartment. They expand and bring about a large volume increase in the biological milieu, releasing the pharmaceutical agent held within. Another interesting application solely based on the ability of these osmotically active agents to bring about large volume changes is illustrated in the patent of ref 8. In the proposed concept, the osmotically active agent forms the expandable core surrounded by a semipermeable membrane in the implanted endoprosthesis. Here, both induction and rate of expansion are controlled using a proposed affinity-based release system. Dextran is a naturally occurring, osmotically active, high molecular weight polymer. It contains repeating glucose units comprising mainly R-(1-6)-linked D-glucose units and some short R-(1-3)-linked D-glucose branch units. It is water-soluble, * Corresponding author. Vijayalakshmi Mookambeswaran, LIMTechS, Dept. Ge´nie Biologique, Universite´ de Technologie de Compie`gne, Centre de Recherche de Royallieu, BP 20529 - 60205 Compiegne. Te´l.: 03 44 23 44 04. Fax: 03 44 20 39 10.
[email protected];
[email protected].
inert in biological systems and does not affect cell viability, and hence highly biocompatible. Dextran has been used extensively as an osmotically active colloid in re-establishing the plasmatic volume, in postoperative prophylaxis of thromboembolic risk, for the acute treatment of ischemic cerebral attacks, for the prevention of anaphylactic reactions (9, 10), for other applications. Several derivatives of dextran, such as carboxymethyldextran, [2-(diethylamino)ethyl]dextran (11), dexamethasone succinate dextran (12), and lauroyldextran (13), have been synthesized and studied for their potential use in tablet formulations. Periodate oxidation is a well-known method for preparation of dextran derivatives such as streptomycin-tagged dextran (14), amino dextran, and protein derivation to enhance their half-life (15, 16). We exploited this method to couple the amino acid histidine to dextran. Histidine-containing proteins and peptides have been extensively studied for their metal-binding ability (17). Poly(histidine) tags have been used to modify recombinant proteins to aid in their effective purification (18, 19). In view of this, it was proposed that histidine coupled to dextran would act as an anchor that can be effectively retained on the immobilized metal affinity matrix. Also, the presence of histidine groups was found to enhance the osmotic activity of native dextran. Here, we report the study of a novel histidine derivative of dextran with enhanced osmotic properties along with its retention and release from a metal affinity system. The dextranhistidine (DexH1) and its oxidized intermediate were evaluated for their osmotic activity in solution, with reference to native dextran also in solution. In parallel, the binding properties of the DexH derivative to the well-known IDA-Cu and IDA-Zn 1 Abbreviations: DexH, dextran-histidine conjugate; EDTA, ethylenediaminetetraacetic acid; Zn-IDA-Novarose, Novarose matrix coupled to IDA and in turn complexed with Zn; Cu-IDA-Novarose, Novarose matrix coupled to IDA and in turn complexed with Cu.
10.1021/bc060050e CCC: $37.00 © 2007 American Chemical Society Published on Web 01/26/2007
478 Bioconjugate Chem., Vol. 18, No. 2, 2007
Devakumar and Mookambeswaran
matrices were exploited for retaining rather strongly the derivative to these matrices, and then eluted by dissociation with EDTA. The released DexH exhibited tremendous enhancement of osmotic activity. This was measured by the entrapment of the matrix-bound DexH as well as the free eluted form in a semipermeable membrane. On the basis of the differences between the osmotic properties of DexH in its matrix-bound form and freely soluble form, the concept of a novel controlled release system is proposed.
MATERIALS AND METHODS Dextran, sodium meta periodate, sodium borohydride, potassium iodide, ethylene glycol, histidine, copper sulfate, zinc sulfate, IDA, MOPS, sodium acetate, sodium chloride, sodium hydroxide, and EDTA were obtained from Sigma, Dialysis tubing (10 000 MW cutoff) was from Spectrophor, and AN69S membranes were from Biospal. Novarose High Act 100/ 40 was obtained from Inovata. Commercially available epoxy glue Poxymatic Rapide was used to glue the cut AN-69 membranes. Oxidation of Dextran. The procedure was adopted from Guisan et al. (20). Briefly, a 25 mL solution containing 0.5 g of dextran (40 000 Da) in M/Q water was prepared. Then, 1.3 g of sodium periodate was added, and the mixture was allowed to react at room temperature. After 24 h, the solution was dialyzed against 50 volumes of distilled water to eliminate the formaldehyde produced during oxidation. Oxidation of dextran by periodate was measured by two different methods, viz, by following the consumption of periodate in the reaction mixture and second by measuring the production of formic acid as the reaction byproduct, as a function of time (16). Briefly, for the estimation of periodate, 10 µL aliquots were withdrawn from the reaction mixture at regular intervals and mixed with 500 µL of 1.5% KI solution. The volume was brought up to 1 mL with M/Q water, and absorbance of the resulting solution was measured at 400 nm. Formic acid released was measured by titrating an aliquot of the mixture against 0.1 N NaOH to pH 7.0. Calculations were based on the hypothesis that each molecule of glucose consumes 2 molecules of periodate to produce a dialdehyde product, in turn releasing 2 molecules of formic acid. To prepare dextrans of varying degree of oxidation, the reaction was stopped at the end of 1, 2, 3, and 24 h by the addition of ethylene glycol, and aliquots withdrawn were dialyzed to remove unreacted periodate and concentrated by lyophilization. Synthesis of DexH. To the solution containing 95% oxidized dextran, known concentrations of histidine were added. The amount of ligand added was calculated on the basis of the assumption that each dialdehyde glucose moiety in dextran can potentially bind to one histidine. Thus, on the basis of the extent of derivatization required, histidine is added and allowed to react at 4 °C for 16 h based on the optimized experimental conditions. The pH of the resulting solution is increased to 8.5 with sodium carbonate (100 mM), followed by addition of 1 mg/mL sodium borohydride to reduce the unreacted aldehyde groups as well as to stabilize the aldehyde-amine bonds. The mixture is stirred for 15 min, and the sodium borohydride step is repeated. Finally, the reaction mixture is dialyzed extensively against M/Q water, to remove the unbound ligand and other small molecular weight reactants, and lyophilized. Concentrations of histidine and dextran in the resulting compound were measured using Pauly’s reagent and by the phenol-sulfuric acid method (21), respectively.
Figure 1. The AN-69 membrane setup used for measuring the osmotic activity.
Measurement of Osmotic Activity. To measure the waterabsorbing capacities of different preparations the following setup was used (Figure 1). AN-69 membrane flat sheets were taken out of the hemeodialyser (Biospal, AN69S) and cut in to 5 × 5 cm2 pieces. They were glued onto the open end of 50 mL Falcon tubes (using the epoxy glue Poxymatic Rapide). The pointed ends of the tubes were cut open to facilitate sample insertion. The tubes with the membranes were stored overnight at 4 °C, and the membrane was further secured with elastic. These tubes were then wetted in water and checked for leaks. For dialysis, the sample was inserted directly onto the membrane from the open conical end of the tube. The tubes were then weighed and positioned into a 1-L glass beaker containing water so that the membrane portion is completely immersed in water. In all the experiments, the volume of the test solution is kept to a minimum so as to maximize the membrane surface area available for exchange. The tubes were taken out at specified time intervals and weighed to measure the volume increase. Effective volume increase is calculated using the formula
EV (effective volume) ) final weight of the setup - initial weight of the setup initial weight of the setup All the experiments were carried out at least in triplicate, typically yielding an experimental error of less than 5%. Affinity of DexH to Matrix-IDA. Novarose IDA was prepared as described earlier (22). 1 mL of Novarose (Novarose High Act, Inovata, 100/40) matrix coupled to IDA was saturated with Cu2+ or Zn2+ and was used as the matrix to test the affinity of DexH. The column was equilibrated with 25 mM MOPSacetate buffer pH 7.0, and a known concentration of DexH was loaded onto the column. Elution was carried out with stepwise lowering of the pH (25 mM MOPS-acetate of pH 6.0, 5.5, 5.0, and 4.0) and finally with 100 mM EDTA. The flow rate was maintained at 0.5 mL/min, and the fractions were analyzed for histidine and dextran content. Measurement of Water Uptake by DexH in Bound and Free Eluted Forms. A 3 mL Novarose column was packed and equilibrated with copper. From this, 1 mL was removed and kept aside as a negative control. A known concentration of DexH (of specific DexH stoichiometry) was loaded onto the remaining 2 mL column material and extensively washed until DexH was no longer detected in the wash fractions. 1 mL of this gel was used to test the osmotic activity of DexH in its bound form. The gel samples were maintained in 25 mM MOPsacetate buffer pH 7.0. The remaining 1 mL was washed with 100 mM EDTA to elute all the bound DexH. The eluted fraction was collected and dialyzed initially against water and finally against 25 mM MOPs-acetate buffer pH 7.0 to remove copper ions and 100 mM EDTA, and then tested in comparison with the other two gel fractions for its osmotic activity in the AN69 setup at different time intervals.
Bioconjugate Chem., Vol. 18, No. 2, 2007 479
Controlled Release System of Dextran Derivatives
Table 2. Analysis of Osmotic Activity of DexH derivatives tested % oxidation
% histidine derivatization
0a
0a
96b 96c
0b 70c
increase in volume (mL)
24 h
48 h
72 h
2.7 10.8 26.4
4.5 12.0 33.7
5.0 12.4 37.0
a 0% oxidation and 0% histidine derivatization represents native dextran. 96% oxidation and 0% His derivatization represents the oxidized intermediate. c 96% oxidation and 70% derivatization represents the oxidized intermediate wherein 70% of the glucose residues were derivatized with histidine.
b
Figure 2. Time course of periodate oxidation. From the reaction mix consisting of 0.5 g of Dex-40000 and 1.3 g of periodate in a total volume of 25 mL of water, 10 and 100 µL aliquots were withdrawn and assayed for periodate consumption and release of formic acid, respectively, as described in Materials and Methods. Table 1. Effect of Oxidation on the Osmotic Activity of Dextran oxidation state in % (reaction time in h) 86 (1 h) 89 (2 h) 91 (3 h) 96 (24 h) 0 (0 h)
osmotic activity (effective increase in volume in mL) 24 h 48 h 72 h 4.5 6.6 8.1 10.8 2.7
5.5 8.2 9.1 12.0 4.5
8.0 9.0 9.9 12.4 5.0
RESULTS Oxidation of Dextran. Oxidation of dextrans is usually carried out using the periodate method. The reaction is highly specific and results in the cleavage of the ring structure to form an acyclic backbone with intact glycosidic linkages (23, 24). The degree of oxidation is correlated to the amount of periodate used in the reaction mixture, which in turn determines the nature of the reaction end product that could be either a monoaldehyde or a dialdehyde. A dialdehyde product is ideally suited for further derivatization with an amino group-containing compound. Also, most of the oxidized dextrans reported in the literature range from 10% to 40% oxidized product based on the estimation of the liberated formic acid as well as consumption of periodate in the reaction mixture (16). We followed the oxidation reaction kinetics using both of these methods to obtain dextrans with different degrees of oxidation. These dextrans were tested for their osmotic activity to understand the effect of excess of OH groups generated by oxidation on the osmotic potential. On the basis of this kinetic data, we could also obtain a >90% oxidized product that was further used to obtain optimally derivatized DexH. As indicated in Figure 2, the reaction profile obtained is in agreement with the earlier reports (16). All the estimations were carried out using the phenol-sulfuric acid method (21). Results depicted in Figure 2 show rapid initial kinetics reaching more than 80% completion at the end of 1 h. The reaction was allowed to proceed for a total of 24 h. We could achieve a maximum of 96% oxidation under the conditions employed. Effect of Oxidation on the Osmotic Activity. The oxidation reaction was stopped at different time intervals with the help of ethylene glycol, and the resulting product was subjected to dialysis and concentrated by lyophilization. The extent of oxidation was estimated, and the results are shown in Table 1. The various oxidized dextrans obtained were dissolved in water (1 wt %/vol) and comparatively analyzed for their osmotic activity with native dextran. On the basis of the results obtained, it can be inferred that there is a positive correlation between the degree of oxidation
Table 3. Effect of Ligand Concentration on the Osmotic Activity concn of DexH in % (wt/vol) 1 2.5 5 Native Dextran 1 2.5 5
increase in volume in mL 24 h 48 h 24.9 27.9 33.2
32.3 34.2 47.8
2.7 4.0 5.2
4.5 5.3 7.3
and osmotic activity (Table 1). There is more than a twofold increase in the osmotic activity as measured by the increase in volume at the end of 72 h. It is possible that generation of excess OH groups is responsible for this increase. It can also be observed here that oxidation affects the swelling time. Dextrans with lower oxidation levels (95
3.5 4.2 4.7 7.8 10.8 16.5 26.4 41.8
4.6 5.3 5.6 9.3 12.8 19.8 33.7 56.5
dextran. At a concentration of 1%, there is about a ninefold increase, whereas at a concentration of 2.5%, only about a sixfold increase in the activity is seen (Figure 3). No increase in the activity was observed from 2.5% to 5% concentration. This discrepancy could arise from the pressure exerted in the opposite direction by the rising column of water, interfering with the in-flow in the open setup (tube of 50 mL capacity) used for testing, indicating the limitation of the system used. It is possible that the figures indicated here could represent an underestimation of the actual osmotic potential of DexH. Effect of Concentration of Histidine on the Osmotic Activity of DexH. As can be seen from the data presented so far, the presence of histidine enhances the osmotic activity of dextran, indicating that histidine alters the osmotic property of the native dextran molecule. Thus, to understand the effect of histidine further, dextran was derivatized with varying quantities of histidine, and the volume changes of the resulting preparations were studied. As indicated in Figure 4 and Table 4, there is a steep increase in the osmotic activity as the percent derivatization with histidine increases. The correlation between enthalpic changes associated with periodate oxidation have been reported in literature. It is proposed that cleavage of the pyranose ring of dextran due to
Figure 5. The envisioned affinity-based controlled release system.
oxidation removes the enthalpic constraint on the elasticity of the molecule, making it shorter and stiffer (23). Due to these changes in the elasticity, the molecules may shift to a more extended conformation with an excess of exposed OH groups that might positively affect the osmotic activity. In conjunction with this, the presence of histidine could increase the repulsion between individual dextran molecules, enhancing the osmotic activity further. Another contributing factor could be the hydrogen-bonding abilities of excessive His groups present on dextran. Association and Dissociation of DexH to Metal Chelate Affinity Matrix. Once the histidine derivatization was successfully carried out and the resulting compound was found to exhibit a large increase in osmotic activity, due to oxidation as well as an increasing number of histidine groups on the dextran molecule, it was of interest to test the validity of the proposed controlled release systems. The presently envisioned system would comprise a matrix capable of binding IDA that would in turn bind to the metal ion of interest at very high concentrations. This matrix would anchor DexH. The binding and release of this osmotically active DexH should be controlled by an external trigger such as changes in pH or use of EDTA. The released molecule would produce a change in osmotic activity leading to the influx of water and thereby a volume change. A schematic representation of the envisioned concept is provided below (Figure 5). Such a system would be well-suited either as a controlled release system in tablet formulations or as an affinity-based in vivo expandable system described by Cinquin et al. (8). It is also required that the ligand in its bound form should be osmotically inactive or should exhibit less osmotic activity than the free form so that the system would be inert in the absence of the external trigger. After its release from the matrix in the presence of the trigger, it would bring about a large volume change and thereby behave as an ideal controlled release system. For example, the endoprosthesis would expand only in presence of an external trigger and not on immediate contact with the biological milieu, which would be the case if the osmotically active ligand alone were to be placed instead of the osmotically inactive complex. The extent of volume increase can thus be controlled in a spatial and temporal manner. Novarose was chosen as the model matrix, as it can be coupled to high IDA and hence IDA-metal concentration. Two of the metal ions were tested for their possible affinity to DexH, viz., Cu and Zn. Association Studies of DexH with Cu-IDA-Novarose. Association of DexH was tested on Cu-IDA-Novarose as described in Materials and Methods. Elution was initially carried out with stepwise lowering of the pH, followed by elution with 100 mM EDTA. As indicated in Figure 6, DexH binds to the Cu-IDA matrix rather strongly. Complete elution of the bound material is possible only on treatment with 100 mM EDTA. Small peaks
Bioconjugate Chem., Vol. 18, No. 2, 2007 481
Controlled Release System of Dextran Derivatives
Figure 6. Chromatogram of DexH on Cu-IDA-Novarose. 1 mL of Novarose-IDA-Cu2+ was mixed with 25 mM MOPS-acetate buffer pH 7.0 and a known concentration of DexH was loaded onto the column. Elution was carried out with stepwise lowering of pH (25 mM MOPS acetate of pH 6.0, 5.5, 5.0, and 4.0) and finally with 100 mM EDTA. The flow rate was maintained at 0.5 mL/min, and the fractions were analyzed for histidine and dextran content. Table 5. Chromatographic Data of DexH on Cu-IDA-Novarose
Figure 7. Chromatogram of DexH on Zn-IDA-Novarose. 1 mL of Novarose-IDA-Zn2+ was mixed with 25 mM MOPS-acetate buffer pH 7.0 and a known concentration of DexH was loaded onto the column. Elution was carried out with stepwise lowering of pH (25 mM MOPS acetate of pH 6.0, 5.5, 5.0, and 4.0) and finally with 100 mM EDTA. The flow rate was maintained at 0.5 mL/min, and the fractions were analyzed for histidine and dextran content. Table 7. Chromatographic Data of DexH on Zn-IDA-Novarose concentration (based on dextran equivalents in mg)
concentration (in mg equiv of dextran) total quantity loaded
load unbound
16.824 mg
14.000 8.250
Bound pH 6.0 pH 5.5 pH 5.0 pH 4.0 EDTA total bound
Bound
0.08 0.250 0.573 0.575 6.876 8.36
eluted at pH 5.0 eluted with EDTA total bound
Table 8. Osmotic Activity of DexH in Its Bound (with Zn-IDA) and Free Eluted Form
Table 6. Osmotic Activity of DexH in Its Bound and Free Eluted Forms
Novarose-IDA-Cu Novarose-IDA-Cu-DexH eluted DexH
0.728 2.200 3.788
24 h
48 h
7.36 10.03 31.67
10.771 14.724 42.3
are visible at each of the pH values, possibly indicating heterogeneity in the polymer preparation. 1 mL of the Cu-IDA matrix was found to bind a total of 8-9 mg of DexH (dextran equivalents). Osmotic Activity of DexH-Cu-IDA Novarose. On the basis of our initial hypothesis, DexH should bind to the matrix of interest at very high concentrations and should be effectively released only on application of an external stimulus. Also, the osmotic effect exhibited by the compound in its bound form should be less than that exhibited in its eluted free form. The following experiment was carried out to test this possibility. DexH was tested bound to Novarose-IDA-Cu as well as in its free soluble from after elution. Novarose-IDA-Cu without any bound DexH was also tested for comparison. Data represented in Table 6 clearly indicate that DexH exhibits a large activity difference between its bound and eluted soluble form. The Novarose-IDA-Cu complex was also found to contribute to the osmotic activity. The DexH-Cu-IDANovarose complex exhibits osmotic activity almost equal to that exhibited by Novarose-IDA-Cu, indicating that DexH has very little osmotic activity in its bound form and fulfils one of the prerequisites of the proposed hypothesis. Also, it should be noted here that 1 mL Novarose-Cu-IDA can effectively bind and release about 8 mg of DexH preparation (Table 5), which should, in principle, exhibit osmotic activity comparable to 1% (wt/ vol) DexH solution (Table 3). On the contrary, DexH in its eluted form exhibits activity equivalent to a 5% solution. It could be reasoned that Novarose-Cu-IDA is able to selectively bind DexH with high histidine content from a heterogeneous pool. Association Studies of DexH with Zn-IDA-Novarose. Zn is relatively less toxic than Cu, and also, histidine binding to Zn-IDA is known to be less strong than that between Cu-IDA (25). Thus, we tested the affinity of DexH on an IDA-Zn matrix.
volume increase in mL matrix
24 h
48 h
Novarose-IDA-Zn Novarose-IDA-Zn-DexH DexH eluted
4.235 17.2 25.2
4.3 19.4 40.4
Similar chromatographic conditions were employed as in the case of the Cu-IDA matrix. The chromatogram obtained with Zn is very similar to that of Cu (Figure 7 and Table 7), with a major elution peak only with 100 mM EDTA. It was also found that Zn exhibits a much lower binding capacity then Cu. 1 mL matrix binds to about 3.8 mg of DexH as against 8 mg in the case of Cu-IDANovarose (Tables 7 and 5). This could be attributed to the fact that Zn requires adjacent histidine residues for effective binding (25, 26) and is thus able to bind only a small fraction of DexH molecules from the heterogeneous pool. Osmotic Activity of DexH-Zn-IDA-Novarose. We tested the osmotic activity of DexH as a complex with Zn-IDANovarose under similar conditions as in the case of Cu complexes. The results shown in Table 8 indicate that Novarose-IDAZn exhibits less osmotic activity than Cu-IDA-Novarose, whereas the complex of Novarose-IDA-Zn-DexH exhibits significant osmotic activity. It is of interest to note here that eluted free DexH from the Zn-IDA matrix amounts to only about 0.3% (wt/vol) solution in terms of concentration, but the osmotic activity exhibited is almost equivalent to 5% free DexH preparation. This again indicates that the ability of the Zn-IDA matrix to concentrate DexH with higher histidine content may be due to its ability to bind to adjacent histidines, which is significantly higher when compared to DexH-Cu-IDA-Novarose.
DISCUSSION We report here a novel derivative of dextran, DexH, that exhibits enhanced osmotic properties. Oxidized intermediates of dextran are prepared using the well-known periodate method and are subsequently derivatized with histidine. An enhancement in osmotic activity is observed at the oxidized intermediate stage
482 Bioconjugate Chem., Vol. 18, No. 2, 2007
itself. Significant changes are also observed with respect to the swelling time. There seems to be an inverse correlation between the degree of oxidation and the swelling time. This behavior can be explained by taking into account the structural features of native dextran as well as the changes that have been reported to occur during oxidation. Dextran has, in its structure, a ring-oxygen atom and a bridgeoxygen atom with hydrogen-acceptor properties and a hydroxyl group with hydrogen-donor and -acceptor properties (27). Dextrans consists of glucose molecules bonded together mostly by R-(l-6) linkages with about 5% R-(1-3) linkages serving as branch points. The branches are short and are mainly one or two glucose monomers in length (28). The behavior of dextran in aqueous solution has been the subject of several investigations (29-32) Periodate-mediated oxidation is known to break the bond between the third and fourth carbon atoms, leading to formation of the dialdehyde product. This cleaves the pyranose ring, in the process generating an excess of CHO groups. This event increases the elasticity of the molecule as well as the hydrogen-bonding ability, depending on the extent of oxidation. Also, dextran solutions are known to exhibit intermolecular associations (32), and periodate oxidation has been shown to decrease these molecular interactions proportionally to the degree of oxidation (33). All these structural changes could contribute to the enhanced osmotic activity observed. Further, the osmotic activity was found to increase on conjugation with histidine. A positive correlation was observed between the extent of derivatization and the osmotic activity. It has previously been documented that, at lower concentrations, the osmotic potential of dextran is comparable to that of albumin. But, at higher concentrations, there is a marked deviation, with dextrans exhibiting a large increase in osmotic potential with increasing concentration. This behavior is attributed to greater interaction between dextran and water, and the properties conferred by the random coil structure. Also, the coiled nature is known to change from a compact one to an expanded coil in aqueous solutions. It is possible that the introduction of a polar group such as histidine enhances the hydrogen-bonding ability of dextran, and the presence of multiple histidines can ease the conformational constraints, further enhancing the hydrodynamic interactions leading to a significant increase in osmotic activity. It would be of interest to test the effect of histidine conjugation on the intermolecular associations of dextran. Another interesting feature observed is the selectivity observed in the case of Cu- and Zn-bound IDA matrices. When DexH preparations are loaded onto these columns, the eluted soluble fractions exhibit higher osmotic activity. It can be suggested here that the polymer preparation contains a heterogeneous pool of DexH molecules with varying degrees of derivatization with histidine, and both the IMAC systems tested could bind selectively to the DexH species with higher content of histidine. The effect is more pronounced in the case of the Zn-IDA matrix, as eluted DexH at a concentration of 0.38% exhibits a tenfold increase in osmotic activity compared to native dextran tested at a concentration of 1%. It is known that, in the case of proteins, multiplicity, density, and localization of histidine residues on the surface determines the strength of interaction with the metal-IDA matrix (26). Also, Cu-IDA is known to efficiently bind proteins even with a single surface histidine, whereas metals in the lower end of the series, such as Zn, bind only when multiple histidines are present on the surface. For example, as reported by Hamden et al., (26) myoglobin from two different species exhibits radically different affinities to the Zn-IDA matrix. Dog myoglobin with one histidine exhibits very little affinity, whereas sperm whale myoglobin with two surface histidine exhibits strong binding. Histidines separated by two amino acid residues, forming a part
Devakumar and Mookambeswaran
of an R helix, are known to be favorably positioned to form strong coordination with IDA-Zn. The same analogy can be extended to explain the behavior of DexH. Here, in DexH with a high degree of derivatization, histidines possibly exist as clusters and are readily exposed on the surface due to the expanded coil structure. The Zn-IDA matrix is able to selectively bind to these molecules with histidine clusters. Thus, the ZnIDA matrix can be effectively used to concentrate the DexH species with high osmotic activity. DexH with its enhanced osmotic properties would find many applications apart from the two discussed in the above sections, such as in contrast enhancers, preparation of novel hydrogels with higher swelling capacity, and so forth. Also, it would be interesting to test the effect of molecular weight on the osmotic activity. It is possible that higher molecular weight dextrans can potentially conjugate with histidine to a larger extent and thus can exhibit more osmotic activity than the one tested in this paper. This approach could also provide a method to engineer DexH with the tailor-made osmotic properties to suit specific requirements.
ACKNOWLEDGMENT This work was carried out as a part of the project SURGETIQUE 2 within the framework of the National Network of Technology for Health (RNTS) France, from the French Ministry of Industry. We wish to thank RNTS for the financial support. We particularly thank Prof. P. Cinquin and the Societe PRAXIM, for their valuable discussions.
LITERATURE CITED (1) Peppas, N. A., Bures, P., Leobandung, W., and Ichikawa, H. (2000) Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 50, 27-46. (2) Hoffman, A. S. (2002) Hydrogels for biomedical applications. AdV. Drug DeliVery ReV. 54, 3-12. (3) Drury, J. L., and Mooney, J. D. (2003) Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials 24, 4337-4351. (4) Bolhuis, G. K., Van Kamp, H. V., Lerk, C. F., and Sessink, F. G. M. (1982) On the mechanism of action of modern disintegrants. Acta Pharm. Technol. 28, 111-114. (5) Chang, R., Shinwari, M., Leonzio, M., Wu, L., Pang, J., and Hussain, M. A. (1998) Evaluation of the disintegrant properties for an experimental, crosslinked polyalkylammonium polymer. Int. J. Pharm. 173, 7-92. (6) Santus, G., and Baker, R. W. (1995) Osmotic drug delivery: a review of the patent literature. J. Controlled Release 35, 1-21. (7) Verma, R. K., Krishna, D. M., and Garg, S. (2002) Formulation aspects in the development of osmotically controlled oral drug delivery systems. J. Controlled Release 79, 7-27. (8) Cinquin, P., Cinquin, O., Favier, D., Orgeas, L., Pecher, M., and Pujol, S. (2001) Micro-muscle en milieu biologique, brevet no. 01/ 09526, depose par l’Universite Joseph Fourier le 17 Juillet 2001. (9) Mofenson, H. C., and Greensher, J. (1970) The nontoxic ingestion. Pediatr. Clin. North Am. 17, 583-90. (10) Mofenson, H. C., Greensher, J., and Caraccio, T. R. (1984) Ingestions considered nontoxic. Clin. Lab. Med. 4, 587-602. (11) Miyazaki, Y., Yakou, S., Nagai, T., and Takayama, K. (2003) The effect of polyion complex formation on in vitro/in vivo correlation of hydrophilic matrix tablets. J. Controlled Release 91, 315-26. (12) Pang, Y. N., Zhang, Y., and Zhang, Z. R. (2002) Synthesis of an enzyme-dependent prodrug and evaluation of its potential for colon targeting. World J. Gastroenterol. 8, 913-7. (13) Hirsch, S., Binder, V., Schehlmann, V., Kolter, K., and Bauer, K. H. (1999) Lauroyldextran and crosslinked galactomannan as coating materials for site-specific drug delivery to the colon. Eur. J. Pharm. Biopharm. 47, 61-71.
Bioconjugate Chem., Vol. 18, No. 2, 2007 483
Controlled Release System of Dextran Derivatives (14) Roseeuw, E., Coessens, V., Schacht, E., Vrooman, B., Domurado, D., and Marchal, G. (1999) Polymeric prodrugs of antibiotics with improved efficiency. J. Mater. Sci. Mater. Med. 10, 743-6. (15) Betancor, L., Lo´pez-Gallego, F., Hidalgo, A., Alonso-Morales, N., Fuentes, M., Ferna´ndez-Lafuente, R., and Guisa´n, J. M. (2004) Prevention of interfacial inactivation of enzymes by coating the enzyme surface with dextran-aldehyde. J. Biotechnol. 110, 201207. (16) Battersby, J., Clark, R., Hancock, W., Puchulu-Campanella, E., Haggarty, N., Poll, D., and Harding, H. (1996) Sustained release of recombinant human growth hormone from dextran via hydrolysis of an imine bond. J. Controlled Release 42, 143-156. (17) Lu, Z., DiBlasio-Smith, E. A., Grant, K. L., Warne, N. W., LaVallie, E. R., Collins-Racie, L. A., Follettie, M. T., Williamson, M. J., and McCoy, J. M. (1996) Histidine patch thioredoxins. Mutant forms of thioredoxin with metal chelating affinity that provide for convenient purifications of thioredoxin fusion proteins. J. Biol. Chem. 271, 5059-5065. (18) Mateo, C., Fernandez-Lorente, G., Pessela, B. C., Vian, A., Carrascosa, A. V., Garcia, J. L., Fernandez-Lafuente, R., and Guisan, J. M. (2001) Affinity chromatography of polyhistidine tagged enzymes. New dextran-coated immobilized metal ion affinity chromatography matrices for prevention of undesired multipoint adsorptions. J. Chromatogr., A 915, 97-106. (19) Mohanty, A. K., and Wiener, M. C. (2004) Membrane protein expression and production: effects of polyhistidine tag length and position. Protein Expr. Purif. 33, 311-325. (20) Guisa´n, J. M., Penzol, G., Armisen, P., Bastida, A., Blanco, R. M., Fernandez-Lafuente, R., and Garcı´a-Junceda, E. (1997) Immobilization of enzymes acting on macromolecular substrates: reduction of steric problems. In Immobilization of Enzymes and Cells (Bickerstaff, G. F., Ed.) Methods in Biotechnology Series, Vol. 1, pp 261-275, The Humana Press, Totowa, NJ. (21) Dubois, M., Gilles, K., Hamilton, J. K., Rebers, P. A., and Smith, F. (1951) A colorimetric method for the determination of sugars. Nature (London) 168, 167. (22) Berna, P. P., Mrabet, N. T., Beeumen, J. V., Devreese, B., Porath, J., and Vijayalakshmi, M. A. (1997) Residue accessibility, hydrogen bonding, and molecular recognition: metal-chelate probing of active site histidines in chymotrypsins. Biochemistry 36, 6896-6905.
(23) Marszalek, P. E., Oberhauser, A. F., Pang, Y. P., and Fernandez, J. M. (1998) Polysaccharide elasticity governed by chair-boat transitions of the glucopyranose ring. Nature (London) 396, 661664. (24) Bobbitt, J. M. (1956) In AdVances in Carbohydrate Chemistry (Wolfram, M. L., and Tipson, R. S., Eds.), Academic, New York. (25) Sulkowski, E. (1989) The saga of IMAC and MIT. BioEssays 10, 170-175. (26) Hemdan, E. S., Zhao, Y. J., Sulkowski, E., and Porath, J. (1989) Surface topography of histidine residues: a facile probe by immobilized metal ion affinity chromatography. Proc. Natl. Acad. Sci. U.S.A. 86, 1811-1815. (27) Vinogradov, S. N., and Linnell, R. H. (1971) The hydrogen bond in biological materials. In Hydrogen Bonding, p 250, Von Nostrand Reinhold, New York. (28) Larm, O., Lindberg, B., and Svensson, S. (1971) Studies on the length of the side chains of the dextran elaborated by Leuconostoc mesenteroides NRRL B-512. Carbohydr. Res. 20, 39-48. (29) Preston, B. N., Comper, W. D., Hughes, A. E., Snook, I. K., and van Megen, W. (1982) Diffusion of dextran at intermediate concentrations. J. Chem. Soc., Faraday Trans. 78, 1209-1221. (30) Laurent, T. C., Sundelof, L. O., Wik, K. V., and Warmegard, B. (1976) Diffusion of dextran in concentrated solutions. Eur. J. Biochem. 68, 95-102. (31) Brown, W., Stilbs, P., and Johnsen, R. M. (1982) Self-diffusion and sedimentation of dextran in concentrated solutions. J. Polym. Sci.: Polym. Phys. 20, 1771-1780. (32) Callaghan, P. T., and Pinder, D. N. (1983) A pulsed field gradient NMR study of self-diffusion in a polydisperse polymer system: dextran in water. Macromolecules 16, 968-973. (33) Murphy, P. T., and Whistler, R. L. (1973) Industrial Gums Polysaccharides and Their DeriVatiVes, 2nd ed. (Whistler, R. L., and De Miller, J. N., Eds.) p 513, Academic Press, New York. (34) Uraz, H., and Giber, A. (1997) Comparison of molecular association of dextran and periodate-oxidized dextran in aqueous solutions Carbohydr. Polym. 34, 127-130. BC060050E
