Aqueous and solid complexes of iron(III) with hyaluronic acid

Journal of Inorganic Biochemistry 89 (2002) 212–218 www.elsevier.com / locate / jinorgbio

Aqueous and solid complexes of iron(III) with hyaluronic acid Potentiometric titrations and infrared spectroscopy studies Ana Lucia Ramalho Merceˆ*, Luiz Carlos Marques Carrera, Lilian Kelly Santos Romanholi, ´ ´ Angeles Marıa Lobo Recio ´ ´ , Universidade Federal do Parana´ , CP 19081, Centro Politecnico , Curitiba, 81531 -990 Parana´ , Brazil Departamento de Quımica Received 17 July 2001; received in revised form 19 November 2001; accepted 28 November 2001

Abstract The coordination of iron(III) ion to hyaluronic acid (Hyal) in aqueous solutions and solid state was accomplished by potentiometric titrations and infrared spectroscopy. The potentiometric titration studies provided the binding constants for the complexes found in the systems and the speciation of these species according to the variation of pH values. The complexes found presented a complexing ability through both the chelating moieties of Hyal (via the N-glucosamine and D-glucoronic acid), showing no special preference for either one while in solid state, but when in aqueous solution the complexation via the N-glucosamine moiety was the preferred, forming two complexed species, ML and ML 2 (log KML 58.2 and log KML2 57.9). The presence of a m -oxo complex via the D-glucoronic acid was also detected in both aqueous (log K56.7) and solid states via the N-glucosamine and D-glucoronic acid simultaneously linked to two Hyal chains. A structure for this latter complex was suggested. The results indicated that these complexes could be used in eliminating the excess iron(III) in living organisms.  2002 Elsevier Science Inc. All rights reserved. Keywords: Iron complexes; Hyaluronic acid; Potentiometric titration; Infrared; Cross-linked; Metal ion coordination

1. Introduction Biological, solid state and solution properties of water soluble polymeric and oligomeric glycosaminoglycans such as heparin, keratan sulfate, dermatan sulfate and especially hyaluronic acid (Hyal) have been a topic of active research in recent years [1–3]. Among the above mentioned glycopolysaccharides, hyaluronic acid has attracted much attention for its unique properties and biological importance. Hyaluronic acid is a naturally occurring high molecular weight hetero-polysaccharide with a linear chain consisting of regularly alternating units of N-glucosamine and D-glucoronic acid (Fig. 1). Hyaluronic acid is the most abundant glycosaminoglycan in animal connective tissues. In these living substrates, Hyal occurs normally in the more stable salt form as sodium hyaluronate (Na-Hyal) intimately associated with proteins [4,5]. High concentrations of hyaluronate can be found in skin, vitreous humour, cartilage and umbilical cord where it controls the hydration level of tissues and

*Corresponding author. Fax: 155-41-263-3399. ˆ E-mail address: [email protected] (A.L.R. Merce).

performs important structural and mechanical functions [6]. In sinusoidal fluid for example [7], the concentration of Hyal can reach 1.4–3.6 mg / ml, being essential for normal joint function, not only acting as a filter for the molecules diffusing through the extracellular matrix, but also conferring the necessary rheological properties of space filling, viscoelasticity, shock-absorbing and optimal lubricity between joints. Interest in hyaluronic acid has grown steadily because recent work [8] has demonstrated that the presence of Hyal is not only important from the mechanical point of view but can also greatly influence diverse biological processes, such as cellular migration, differentiation, recognition and adhesion. These remarkable characteristics associated with its biocompatibility and immunogenicity make hyaluronic acid a very useful biomedical polymer. In recent years, low cost and high purity Hyal has been produced industrially from bacterial resources [1]. Bacterial Hyal has found application in varied fields such as cosmetic products [9], ophthalmic viscosupplementation surgery [10], osteoarthritis therapy [11], cancer metastasis control [12], biomedical hydrogels and biocompatible materials for artificial skin and bone filling [13], and novel drug carriers [14]. In these

0162-0134 / 02 / $ – see front matter  2002 Elsevier Science Inc. All rights reserved. PII: S0162-0134( 01 )00422-6

A.L.R. Merceˆ et al. / Journal of Inorganic Biochemistry 89 (2002) 212 – 218

213

Fig. 1. Structure of hyaluronic acid (Hyal).

applications the solution properties of Hyal are very important and have been studied extensively [7,15–20]. In contrast, Hyal ability to complex transition metallic ions has been studied to a lesser extent. Hyal metallic complexes with copper [21] silver, gold, cerium, tungsten [22], platinum [23] and zinc [24] have been isolated or detected. Hyal complexes with heavy metals have microbial activity; gold complex can be used in antiarthritis treatments, and the antitumor activity of complex trans(racemic)-1,2-cyclohexanediammineplatinum(II) conjugated to Hyal was successfully tested. The presence, sequentially distributed, of carboxylate groups in the glucoronic moieties of the Hyal chain, can induce the complexation of Hyal with trivalent transition metallic ions [25,26]. Additional interactions via hydroxylic oxygens [26,27] bridging oxygen atoms between monosaccharides [28] or via amide group of N-glucosamine moieties [29] must also be considered. Although potentiometric titration results cannot provide an insight to the complex structure, by looking at the binding constants which show the strength of the interaction between a ligand and a metal ion, one can infer which basic site is more likely to interact with the Lewis acid; the speciation according to the change in pH values can also be calculated. In the literature there are few works [24,30–32] reporting the complexing ability of biopoly` et al. mers and metal ions in aqueous solutions. Verchere [33] reviewed the complexing abilities of monosaccharides and metal ions. This work refers to the characterization and study of properties, in solid state as well as in solution, of complexes of Hyal with iron(III), with the ultimate objective of finding new biocompatible and immunogenic biomaterials with potential biomedical applications. Complexation of Hyal with Fe 31 ions can yield highly cross-linked biomaterials, which would facilitate the formation of polymeric gels or films of high stability, which could be used as implants in surgery to avoid the frequent adhesion formation in postoperative periods [34]. The influence on crosslinked degree of some factors, such as stoichiometrical ratio Fe 31 / Hyal disaccharide and reaction medium pH, were studied. On the other hand, the stability constants of the complexes in solution were determined by potentiometric titrations in order to assess the ability of decreasing

through complexation the concentration of excessive iron absorption in the gastrointestinal tract when this product is orally administered [35].

2. Experimental

2.1. Materials Sodium hyaluronate sample (Na-HA) was obtained from a bacterial strain of Streptococcus zooepidemicus, and was kindly donated by Professor Dusan Bakos (Slovak Technical University) and was used without further purification (MW 1,500,000). Hyal was obtained by dispersion of Na-HA in ethanol / 0.1 M HCl aqueous solution (70 / 30, v / v), filtered off, washed with ethanol / HCl and ethanol and dried in vacuo, following a previously described procedure [36]. All other chemicals were of analytical grade and were used without further purification.

2.2. Methods 2.2.1. Potentiometric titrations The metal ion solution was standardised using EDTA (Reagen) titrations according to the literature [37]. The acid content of solution was determined by Gran’s Plot [38]. KOH titrant agent was standardised against potassium hydrogenophthalate (Carlo Erba). A Sigma Techware Digitrate piston manual burette was used to deliver the titrant (0.02 0.01 ml) and the pH (2log of the molarity of H 1 ) was directly measured with an Orion model 420-A research grade pH meter (USA) fitted with a blue-glass and Ag /AgCl reference electrodes (Switzerland) calibrated to the following reproducibility: 0.005 pH units in buffer regions, absolute pH accuracy 0.005 pH in acidic values and 0.015 pH in basic values, following procedures described in the literature [38]. All titrations were performed in duplicate in the absence and in the presence of different ligand to metal ratios (0.1 mmol Hyal / 0.1 mmol Fe 31 , 0.1 mmol Hyal / 0.05 mmol Fe 31 , 0.1 mmol Hyal / 0.03 mmol Fe 31 ) under inert atmosphere (N 2 , washed through 5% m / v pyrogallol in 50% m / v KOH; and aqueous KOH 0.1 mol l 21 ; Merck), controlled temperature (25.060.1 8C) and ionic strength

214

A.L.R. Merceˆ et al. / Journal of Inorganic Biochemistry 89 (2002) 212 – 218

(I50.100 M, KNO 3 ; Merck). The aqueous (bidistilled, deionized and freshly boiled) solutions of Hyal were 1.03 10 22 M, initial concentration, and acidic (HNO 3 ; Merck) aqueous solution of Fe 31 (0.01 M) was made from p.a. salt ¨ All titrations in the of Fe(NO 3 ) 3 ?9H 2 O (Riedel-de Haen). presence of iron(III) were direct titrations, but starting at pH values below 1.8 to ensure that the necessary free metal ion concentration was present at the beginning of all titrations. Best7 [38] and SPE microcomputer programs were employed in the calculations performed in obtaining the binding constants for the complexed species and for the distribution diagrams, respectively. All mathematical aspects of those programs are described elsewhere [30,39].

2.2.2. Other techniques C, H and N elemental analyses were performed on a Perkin-Elmer model 240 analyzer. Iron analyses were performed on a Varian model 250 PLUS atomic absorption spectrophotometer. Aqueous solutions (10.0 ml) were prepared for each analysis using 10 mg of the corresponding complex treated with 1.5 ml of concentrated HNO 3 . FTIR spectra (4000–400 cm 21 ) were taken using KBr pellets (1 mg sample and 99 mg of spectrometry grade KBr), and recorded on a Bomen FTIR model MB100 spectrophotometer. The thermogravimetric (TG) analyses were recorded in a Netzsch simultaneous Thermal Analysis STA 409 EP (Germany), under air, from 21 to 520 8C, 2 8C / min, using open cylindrical aluminum sample pans, 4 mm in diameter and 2 mm high. 2.2.3. Syntheses of complexes 2.2.3.1. Complex I To an aliquot of 0.85 ml of an 0.1 M aqueous solution of Fe(NO 3 ) 3 ?9H 2 O, was added 0.1 g of Hyal dissolved in 33 ml of water (ratio disaccharide of Fe 31 / Hyal, 3 / 1). The scarce yellow precipitate instantaneously formed was eliminated by filtration, and the complex was obtained from the solution by addition of 70 ml of ethanol as a yellow powder. The solid was filtered off, washed with ethanol and dried in vacuo (yield 67 mg).

2.2.3.3. Complex III To a solution of 100 mg of Hyal in 115 ml of water (pH 2.68), 2,4,6-trimethylpyridine was added, drop by drop, until pH 4.5. Then, 40.7 mg of Fe(NO 3 ) 3 ?9H 2 O (80%) were added (ratio Fe 31 / disaccharide of Hyal, 1 / 3), and more 2,4,6-trimethylpyridine was added until pH 3.00. A yellow solid was obtained after addition of 400 ml of ethanol, 1.2 g of KNO 3 and centrifugation (20 min, 500 rpm). It was washed with ethanol, centrifuged and dried in vacuo (yield 126 mg). 2.2.3.4. Complex IV A solution of 0.06 g of Hyal in 20 ml of water and 1.53 ml of a 0.1 M Fe(NO 3 ) 3 ?9H 2 O aqueous solution were mixed and stirred (ratio Fe 31 / disaccharide of Hyal, 1 / 1). The scarce yellow precipitate instantaneously formed was eliminated by filtration. An orange microgel was quantitatively obtained by addition to the solution of 230 ml of ethanol. The complex was isolated by centrifugation (20 min, 500 rpm), washed with ethanol and dried in vacuo (yield 60 mg).

3. Results and discussion The available sodium hyaluronate was transformed into hyaluronic acid to fit in the potentiometric titration methodology and to avoid potential competition between sodium and iron ions during the precipitation of the complexes. The short disaccharide unit composed of the hyaluronate and glucosamide monosaccharides of the polyssaccharide is referred to throughout this work as L. The calculated protonation constant values of the polysaccharide (Fig. 1) titrated in the absence of the metal ions are assigned to the hydroxyl (referred to as O) in C-6 of the glucosamide monosaccharide portion (Eq. (1)) and to the carboxylate (referred to as C) of the hyaluronate group of the second monosaccharide portion (Eq. (2)) according to the equilibria below: 2

(1) 2

2.2.3.2. Complex II A 0.0339-g sample of solid Fe(NO 3 ) 3 ?9H 2 O was added to a solution of 0.1 g of Hyal in 115 ml of water (ratio Fe 31 / disaccharide of Hyal, 1 / 3). The yellow solution (pH 2.68) was stirred for 2 h and simultaneously aqueous NaOH was added to stabilize the pH at 4.50. Then 1.062 g of NaNO 3 dissolved in 8.0 ml of water were added. The complex was quantitatively precipitated as a yellow microgel by addition of 400 ml of ethanol. It was filtered off, washed with ethanol and dried in vacuo, resulting a dark orange pasty solid (yield 112 mg).

→ 2 OOC–L–OH log Ka1 5 8.23 OOC–L–O 2 1 H 1 ←

→ HOOC–L–OH log Ka2 5 2.81 OOC–L–OH 1 H 1 ← (2)

Some mathematical models were employed to treat the experimental data to overcome the drawback of BEST7 in dealing with polymeric substances. The model which best fitted the data was that in which the biopolymer was treated as a series of the two monosaccharides that compose the disaccharide units of the hyaluronic acid (Hyal) [40,41]. It was employed for the calculations of the binding constants of Hyal and Fe 31 , the two protonation

A.L.R. Merceˆ et al. / Journal of Inorganic Biochemistry 89 (2002) 212 – 218

Fig. 2. Potentiometric pH profiles of Hyal and Fe 31 . T525.0 8C, I5 0.100 M (KNO 3 ).

constants for L, the proper hydrolysis constants [42] and the values of pH in each potentiometric titration. It was also used the number of millimole of Fe 31 and of the polysaccharide taken as a function of the molecular weight of the two repeating monosaccharides. The binding constants calculated for this system refer to Fe 31 (referred to as M) bound either to the carboxylate group (C) or to the hydroxyl in C-6 (O) of the monosaccharides. When the concentration of M was high enough, the equilibria presented a m -oxo complex. These equilibria are represented by the following equations: → FeO log b1 5 [FeO] / [Fe 31 ] 3 [O] Fe 31 1 O ←

(3)

→ FeO 2 Fe 31 1 2O ←

(4)

→ Fe 2 C 2 2Fe 31 1 2C ←

log b2 5 [FeO 2 ] / [Fe 31 ] 3 [O] 2

log Km 5 [Fe 2 C 2 ] / [Fe 31 ] 2 3 [C] 2 (5)

The potentiometric pH profiles of the system Hyal–Fe 31 are represented in Fig. 2 and the calculated logarithms of the binding constants are shown in Table 1. The complexed system profiles steadily increased the obtainable titration points due to less formation of insoluble products in the equilibria as ligand to metal ratio increased.

215

Fig. 3. Distribution species diagram of Hyal and Fe 31 , ratio ligand to metal of 2:1, metal % set at 100%. Major species are represented by: A5M; B5M 2 C; D5CH; E5ML 2 ; F5M(OH) 2 ; G5HL; H5M(OH) 3 ; I5M(OH) 4 ; J5ML; K5M(OH); and C and L stand for themselves (see text for further explanation).

The values of the binding constants in Table 1 showed that the complexes are very stable and suggested a chelation preferentially through the hydroxyl group of the C-6 group, in a tridimensional structure. The complexation via the carboxylic group is also thermodynamically viable and stable, but probably also made through some m -oxo linkages. The species distribution curves in Fig. 3 showed the presence of stable complexed species from the very acidic pH 2.0 until the very basic pH 10.0, being a very versatile system. At the beginning of the diagram the concentration of free iron(III) is |5.0%, but at pH near 1.8 (data not shown), it is higher than 40%. In very acidic pH values the major complexed species are those in which iron(III) coordinated to Hyal via carboxylate groups (represented in the species distribution diagram as C) forming m -oxo bridges (Fig. 4). As the pH value increases there is an increase in the formation of species with iron(III) coordinated to Hyal via hydroxyl groups (Fig. 3, represented as

Table 1 Logarithms of the binding constants of hyaluronic acid and iron(III) complexes Log K [FeO] / [Fe]3[O] (b 1 ) [FeO 2 ] / [Fe]3[O] 2 (b 2 ) [Fe 2 C 2 ] / [Fe] 2 3[C] 2 (Km ) Refer to Eqs. 3–5; T525.0 8C, I50.100 M, KNO 3 .

Fe(III) 8.260.1 (8.2) 7.960.1 (16.1) 6.760.1 6.7 Fig. 4. Proposed structure for one complex species of Fe 31 -Hyal.

A.L.R. Merceˆ et al. / Journal of Inorganic Biochemistry 89 (2002) 212 – 218

216

Table 2 Analytical data [found(calc.)] of Hyal and complexes %C

%H

%N

%Fe

Hyal a

41.7 (41.3)

5.9 (6.1)

3.7 (3.4)

–

Complex I

37.3 (37.7)

6.1 (5.9)

3.3 (3.1)

4.6 (4.2)

–

Complex II

32.4 (33.2)

5.5 (5.9)

2.8 (2.8)

4.4 (4.4)

3.1 (3.2)

Complex III

34.7 (35.8)

5.8 (5.9)

3.0 (3.0)

5.7 (5.9)

–

9.0 (9.6)

Complex IV

27.0 (25.8)

4.5 (4.8)

3.1 (3.2)

17.5 (17.1)

–

14.6 (13.8)

a b

%Na

%H 2 O b

Compound

–

7.1 (6.6) 8.7 (8.1) 15.9 (15.0)

Analytical data correspond to 1.5 H 2 O by disaccharide unity. Determined by TG.

L), although the metal hydrolysis competition also increases. After potentiometric studies, Hyal–Fe complexes were isolated to compare the analogies or differences between dissolved and solid or microgel species. To study the influence of various factors, such as stoichiometric ratio iron(III) / Hyal disaccharide, reaction mean pH, and base used, in those complexations, different reaction conditions were used and four different complexes were isolated. Complexes I–III were water soluble and obtained from stoichiometric ratio iron(III) / Hyal disaccharide of 1 / 3. Complex IV was water insoluble and obtained from stoichiometric ratio iron(III) / Hyal disaccharide of 1 / 1. Fe / Hyal isolated compounds were characterized by elemental analyses of C, H, N, atomic absorption spectrometry and TG analyses (Table 2) and FTIR spectroscopy (Table 3). Complex I was obtained by mixing aqueous solutions of iron(III) and Hyal and precipitating with ethanol. Analytical data showed a relation of three disaccharide units for each Fe 31 ion, but the water solubility of complex I seems to eliminate the possibility of a simultaneous complexation of three disaccharides with one iron(III) ion, because this would imply a highly cross-linked polymer, which should not be water soluble. Complex I IR spectrum showed a band at 1734 cm 21 , of lower intensity than in the Hyal IR spectrum, assigned to the stretching vibration of non-complexed carboxylic groups. A broad unresolved

band at 1634 cm 21 was tentatively assigned to nas (OCO) of bonded carboxylate groups plus Amide I band, because it is an intermediate value of those corresponding to Na-Hyal. From the Dn value of 214 cm 21 (Dn 5 nas (OCO)2 nsym (OCO)), a monodentate coordination of carboxylate group to the Fe 31 ion can be deduced [43]. The Amide II band (1573 cm 21 ) was shifted 13 cm 21 to higher frequencies with respect to the IR of the Hyal spectrum, which would be indicative of participation of –NH–CO– group, in the complexation of iron(III) [29]. On the other hand, possible additional coordination of polymer to the metallic ion through the oxygen bridging the monosaccharide unity [28] or the ring oxygen [26] have also been suggested. Coordination through hydroxylic groups seems less probable, since they are not deprotonated in the low pH value of the reaction mean [44] and due to sterical reasons. Precipitation can occur probably by binding of additional OH 2 groups even in low pH [44], which would conserve the 31 oxidation formal state of iron ion. Analytical data are coherent with a brutto formula [Fe(Hyal 2)(Hyal) 2 (OH) 2 (H 2 O) 6 ] n . Thus, it can be suggested that of each three disaccharide units, only one would be bonded via monodentate carboxylate group to the Fe(III) in the complex. Possible additional interactions between Fe 31 ion and Hyal, via amide group, and / or via ring, bridge or hydroxylic oxygens, can not be ignored. The number of molecules of water bonded to the Fe(III) needed to maintain its hexacoordination sphere will depend on the number of these additional bonds. The remaining water would be bonded to the organic moiety of the complex. With the aim of providing the dissociation of carboxylic hydrogens of Hyal for favoring the complexation with iron(III) ion and avoiding precipitation of iron hydroxides, the pH of the mixture reaction was increased to 4.5 (refer to potentiometric studies) using aqueous NaOH, which yielded complex II. However, a lower iron(III) content in respect to complex I was found and the IR spectrum of complex II was similar to that of Na-Hyal. No evidence of unlinked carboxylic groups or of complexation via amide groups was found in the IR spectrum. Analytical data were consistent with a formulation [Fe 4 Na 7 (Hyal 2) 9 (Hyal 22)(OH) 8 (H 2 O) 42 ] n . From the above data it can be deduced that there is competition in the precipitation between iron and sodium ions. So, for

Table 3 IR spectral data (KBr, cm 21 ) of Hyal, Na-Hyal and complexes Compound

n (COO)

nas (OCO)

nsym (OCO)

Amide I

Amide II

Amide III

n (Fe–O)

NO 3 –

Hyal Na-Hyal Complex I Complex II Complex III Complex IV

1734 vs – 1734 s – 1732 m –

– 1617 vs 1634 vs,b 1623 vs 1637 vs,b 1633 vs,b

– 1413 1420 1412 1417 1416

1647 vs 1652 vs 1634 vs,b 1660 vs 1637 vs,b 1633 vs,b

1560 1562 1573 1563 1576 1570

1321 1323 1324 1321 1323 1321

– – – – – 688 m, 472 m

– – – – – 1724 m, 1709 m

s s s s sh

b, broad; m, medium; s, strong; sh, shoulder; vs, very strong; w, weak.

vs sh sh sh vs sh

w w w w m w

A.L.R. Merceˆ et al. / Journal of Inorganic Biochemistry 89 (2002) 212 – 218

each ten carboxylate groups of ten disaccharide units, seven would be ionically bonded to seven Na 1 ions and three monodentate complexed to three Fe 31 ions. The fourth iron(III) ion could be bonded to a hydroxylic oxygen, in accordance with the above potentiometric data. The water molecules would be present as crystalization molecules in the organic moiety of polymer as well as coordination bounded to iron(III) ions. To avoid the competition bonding of Fe 31 and Na 1 ions with Hyal, a new reaction at pH 4.5 was carried out, but this time using the bulky 2,4,6-trimethylpyridine as the base. The iron content of the resultant complex III was slightly higher than those of complexes I and II. Complex III IR spectrum was similar to that of complex I, suggesting the existence of some unlinked carboxylic groups and interactions via amide group. No bands assigned to the 2,4,6-trimethylpyridine appear in the spectrum. Analytical data were coherent with a formulation [Fe(Hyal 2)(Hyal)(OH 2 )(H 2 O) 5 ] n . It can be seen that when comparing complexes I and III, one of each three and one of each two disaccharides respectively, are bonded to one Fe 31 ion. Therefore, an increase in the ability of Hyal to complex iron(III) ions when pH increases can be deduced. Complex IV was obtained by mixing iron(III) and Hyal aqueous solutions in stoichiometric ratio iron(III) / Hyal disaccharide of 1 / 1, and showed higher iron(III) and lower carbon contents than the above complexes. Complex IV IR spectrum showed the typical pattern previously noted for monodentate carboxylate groups and interactions via amide groups. In addition, two bands at 688 and 472 cm 21 , absent in the spectra of the other complexes, would be indicative of the presence of m -oxo groups between Fe 31 ions [43], in accordance with the potentiometric results in this present work and in other literature data [45]. Two relatively weak bands at 1724 and 1709 cm 21 , along with nitrogen content, could indicate the presence of monodentate nitrate (Dn 525–5 cm 21 ) [43]. Other characteristic bands of NO 2 3 group would be masked by the Hyal bands. Analytical data were coherent with a formulation [Fe 8 (Hyal 2) 4 (O) 8 (NO 3 ) 2 (OH) 2 (H 2 O) 20 ] n . On the basis of the above data, a cross-linked structure is tentatively suggested for complex IV, in which the iron(III) ions would be simultaneously bonded to two different chains of Hyal. On the basis of steric requirements, the existence of two different types of ferric ions is postulated: one bonded through monodentate carboxylate groups of one chain and, consistently with potentiometric studies, hydroxylic oxygen groups of the other chain, and another bonded through amide groups. The m -oxo groups between iron(III) ions, and OH 2 , NO 32 and H 2 O bonded to the metal would be in the coordination spheres of Fe(III) ions. Fig. 4 illustrates this possibility. Thus, all potentially coordination atoms of the organic moiety of the polymer are bonded to iron(III) ions. Such a high cross-linked structure would justify the insolubility in water of complex IV after its isolation. Later assays of reaction by increasing the pH or the

217

iron(III) proportion yielded precipitation of iron hydroxides. The high ability of Hyal to complex iron(III) ions is enhanced as the pH value is increased when a bulky organic base is used, and mainly when the iron(III) / Hyal relative ratio is increased. These results indicate that these complexes could be used in eliminating the excess iron(III) in living organisms. At present, Fe 31 -Hyal films obtained in different experimental conditions and complexation of chromium(III) and gadolinium(III) ions by Hyal are being studied and the results will be reported later.

Acknowledgements We kindly thank Professor Judith Felcman, PUC-RJ, for the elemental analysis and Funpar-UFPR for financial support.

References [1] V. Crescenzi, Trends Polym. Sci. 2 (1994) 104–107. [2] E. Fuissac, M. Milas, M. Rinaudo, Macromolecules 6 (1993) 6945– 6951. [3] M. Mensitieri, L. Ambrosio, L. Nicolais, J. Mater. Sci. Mater. Med. 7 (1996) 695–698. [4] M. Tomioka, G. Matsumura, Carbohydr. Pol. 27 (1995) 197–203. [5] J.E. Christiner, M.L. Brown, D.D. Dziewiatkowski, Biochemistry 167 (1977) 711–718. [6] E.A. Turley, S. Roth, Nature 283 (1980) 268–271. [7] P. Gosh, N. Hutadilok, A. Lentini, Heterocycles 36 (1994) 1757– 1774. [8] S.M.A. Holmbeck, P.A. Petillo, L.E. Lerner, Biochemistry 33 (1994) 14246–14255. [9] E.A. Balazs, P. Band, Cosmet. Toil. 99 (1984) 65–82. [10] E.A. Balazs, J.L. Delinger, Ciba Found. Symp. 143 (1989) 265– 272. [11] M. Mensitieri, L. Ambrosio, S. Ianace, L. Nicolais, A. Perbellini, J. Mater. Sci. Mater. Med. 6 (1995) 130–137. [12] E.A. Turley, Cancer Metastasis Rev. 3 (1984) 325–339. [13] P. Giusti, L. Lazzeri, L. Lelli, Trends Polym. Sci. 1 (1993) 261–262. [14] T. Pouyani, G.D. Prestwich, Bioconjug. Chem. 5 (1994) 339–347. [15] T. Odijk, A.C. Houwaart, J. Polym. Sci. B Polym. Phys. 16 (1978) 627–631. [16] T. Pouyani, J.W. Kuo, G.S. Harbison, G.D. Prestwich, J. Am. Chem. Soc. 114 (1992) 5972–5976. [17] D.A. Gibbs, E.W. Merrill, K.A. Smith, Biopolymer 6 (1968) 777– 783. [18] N. Berriaud, M. Milas, M. Rinaudo, Int. J. Biol. Macromol. 16 (1994) 137–142. [19] E. Fouissac, M. Milas, M. Rinaudo, R. Borsali, Macromolecules 25 (1992) 5613–5617. [20] M. Rinaudo, M. Milas, N. Jouon, R. Borsali, Polymer 34 (1992) 3710–3715. [21] H. Sterk, M. Braun, O. Schmmut, H. Fleichtinger, Carbohydr. Res. 145 (1985) 1–11. [22] A. Nimrod, B. Greenman, United States Patent, 4,784,991, Nov. 15, 1988. [23] M. Maeda, N. Takasura, T. Suga, N. Uehara, A. Hoshi, Anti-Cancer Drugs 4 (2) (1993) 167–171.

218

A.L.R. Merceˆ et al. / Journal of Inorganic Biochemistry 89 (2002) 212 – 218

´ B. Gyurcsik, M. Gazdag, E. Forrai, I. Dekany, ´ ´ [24] K. Burger, J. Illes, K. ´ Mihalyfi, Carbohydr. Res. 332 (2001) 197–207. [25] H. Nolte, S. John, O. Smidsrod, B.T. Stokke, Carbohydr. Pol. 18 (1992) 243–251. [26] R. Aruga, Bull. Chem. Soc. Jpn. 54 (1981) 1233–1235. [27] S. de A. Soares, J.F. Rodrigues, A. Alario, J. Pol. Sci. Phys. 31 (1993) 1611–1616. [28] H. Sterk, M. Braun, O. Schmut, H. Feichtinger, Carbohydr. Res. 145 (1985) 1–11. [29] J.M. Nieto, C. Peniche-Covas, J. Del Bosque, Carbohydr. Pol. 18 (1992) 221–224. ˆ S.C. Lombardi, A.S. Mangrich, F. Reicher, B. [30] A.L.R. Merce, Szpoganicz, M.R. Sierakowski, Carbohydr. Pol. 35 (1) (1998) 13– 20. ˆ E. Fernandes, A.S. Mangrich, M.R. Sierakowski, J. [31] A.L.R. Merce, Braz. Chem. Soc. 11 (3) (2000) 224–231. ˆ J.S. Landaluze, A.S. Mangrich, B. Szpoganicz, M.R. [32] A.L.R. Merce, Sierakowski, Bioresour. Technol. 76 (2001) 29–37. ` S. Chapelle, F. Xin, D.C. Crans, Prog. Inorg. Chem. [33] J.-F. Verchere, 47 (1998) 837–945. [34] D.B. Johns, K.E. Rodgers, W.D. Donahue, T.C. Kiorpes, G.S. di Zerega, Fertil. Steril. 68 (1) (1997) 37–42.

[35] W. Mandeville III, H. Holmes-Farley, S. Randall, United States Patent, 5,702,696, Dec. 30, 1997. ` [36] N. Jouon, M. Rinaudo, M. Milas, J. Desbrieres, Carbohydr. Pol. 26 (1995) 69–73. [37] G. Schwarzenbach, H. Flaschka, in: Complexometric Titrations, Methuen, Bungay, UK, 1969, p. 241. [38] A. Martell, R. Motekaitis, The Determination and Use of Stability Constants, VCH, New York, 1992. ˆ B. Szpoganicz, M.A. Khan, X. Do Thanh, G. Bouet, [39] A.L.R. Merce, J. Inorg. Biochem. 73 (1999) 167–172. [40] A.E. Martell, R.D. Hancock, in: J.P. FacklerJr. (Ed.), Metal Complexes in Aqueous Solutions, Plenum, New York, 1996, p. 205. ¨ [41] F. Ciardelli, E. Tsuchida, in: D. Wohrle (Ed.), Macromolecule– Metal Complexes, Springer, Berlin, 1995. [42] C.F. Baes Jr., R.E. Mesmer, The Hydrolysis of Cations, Wiley, New York, 1976. [43] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, 5th Edition, Wiley, New York, 1997. [44] P. Sipos, T.G. St. Pierre, E. Tombacz, J. Webb, J. Inorg. Biochem. 58 (1995) 129–138. [45] M. Tonkovic, O. Hadzija, B. Ladesic, B. Klaic, S. Music, Inorg. Chim. Acta 161 (1989) 81–85.