Carbohydrate Research 371 (2013) 8–15
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research journal homepage: www.elsevier.com/locate/carres
Preparation of hyaluronan polyaldehyde—a precursor of biopolymer conjugates Petra Šedová ⇑, Radovan Buffa, Sofiane Kettou, Gloria Huerta-Angeles, Martina Hermannová, Veronika Leierová, Daniela Šmejkalová, Martina Moravcová, Vladimír Velebny´ Biotech Contipro, 561 02 Dolní Dobroucˇ 401, Czech Republic
a r t i c l e
i n f o
Article history: Received 26 November 2012 Received in revised form 9 January 2013 Accepted 11 January 2013 Available online 8 February 2013 Keywords: Cell viability Dess–Martin periodinane Hyaluronan polyaldehyde Oxidation TEMPO
a b s t r a c t Native hyaluronan (HA) has been oxidized to polyaldehyde polymers with a degree of substitution (DS) of up to 50%. Two different procedures enabling the control of the degree of substitution were followed in this study. Selective oxidation of primary hydroxyl groups of N-acetyl-D-glucosamine of hyaluronan was performed either in an aqueous solution containing AcNH-TEMPO/NaBr/NaOCl or in an aprotic solvent containing Dess–Martin periodinane (DMP). It was found that a change of reaction parameters (reaction time and temperature, type of catalyst, oxidant-to-HA ratio, presence of nitrogen, buffer type, and concentration) had an influence on the degree of substitution and molecular weight. The derivatives were characterized by MS, NMR spectroscopy, and SEC-MALLS. Degradation of hyaluronic acid by the oxidant was observed and confirmed by SEC. The effect of oxidized derivatives of hyaluronan on cells was studied by means of NIH 3T3 fibroblast viability, which indicates that prepared hyaluronan polyaldehydes are biocompatible and suitable for medical applications and tissue engineering. The function of polyaldehyde as precursor for other modification was illustrated in the reaction with lysine. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Naturally occurring hyaluronan (HA) is a high molecular weight linear polysaccharide consisting of poly[(1?3)-2-acetamido-2deoxy-b-D-glucopyranosyl-(1?4)-b-D-glucopyranosyluronic acid].1 Due to the excellent properties, mainly its rheological properties, biocompatibility, biodegradability,2 and non toxicity,3 HAbased materials are often applied in cosmetics and medicine. Derivatized or crosslinked form of HA is very useful in viscosupplementation, ophthalmic surgery aids, postsurgical adhesion prevention, tissue augmentation, tissue engineering,4 and for the preparation of drug release systems5 with prolonged effect.2 Native HA suffers from two major drawbacks: low resistance against enzymatic degradation and short half-life after internal administration. In order to overcome these disadvantages, contemporary research is currently focused on modification of HA (HA), which would prolong its degradation time6 in human body. Such modification may be achieved by introducing aldehyde groups into polysaccharide backbone. The aldehyde-functionalized polysaccharides may react with various amine-containing compounds7 to form hydrogel tissue adhesives and sealants8 and to fix a biological tissue.9 It can also be used for medical applications such as wound closure, supplementing or replacing sutures or staples in internal surgical procedures, tissue repair, and encapsula⇑ Corresponding author. Tel.: +420 465 519 573; fax: +420 465 543 793. E-mail address:
[email protected] (P. Šedová). 0008-6215/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.carres.2013.01.025
tion of cells,10 preventing leakage of fluids such as blood, bile, gastrointestinal fluid, and cerebrospinal fluid, ophthalmic procedures, drug delivery,11 and preventing post-surgical adhesions.12 The common aldehyde modification ways of HA are reported in Scheme 1. In the first method, hyaluronan was oxidized by sodium periodate. While periodate oxidation produces a large number of aldehyde groups, the disadvantage is the loss of the native backbone structure by cleavage of the hyaluronan molecule in position C2–C3 of glucuronic acid (GlcA).4 The second method is based on a conversion of HA-carboxyl to aldehyde group by a two-step procedure, in which the intermediate HA-triazine ester is attacked by a nucleophilic amino compound (aminoacetaldehyde dimethylacetal) to form an acetal HA derivative, which is hydrolyzed by an acid to aldehyde.13 Other authors14 converted a hydroxyl group of unmodified dextran into aldehyde by synthesis of allyloxy dextran and its ozonolysis which leads to cleavage of the allyl double bond resulting in formation of a terminal aldehyde group. The third method uses 9-borabicyclo[3,3,1]nonan to reduce the carboxylic group in hyaluronan directly into the aldehyde function.15 The drawback here is the loss of polyelectrolytic character, which influences the conformation and interaction of the molecule of hyaluronan. In the case of polysaccharides, the commonly used oxidant TEMPO (2,2,6,6-tetramethyl-1-piperidinyloxy radical) caused oxidation not to aldehydic group but to carboxylic group. TEMPO catalyzed the selective oxidation of primary alcohols in the presence of secondary ones.16 This form of nitrosonium cation is
P. Šedová et al. / Carbohydrate Research 371 (2013) 8–15
9
Scheme 1. Known formation of HA-aldehyde: (I) HA oxidation with NaIO4,4 (II) hydrolysis of acetal-HA derivatives,13 and (III) reduction of HA with 9borabicyclo[3,3,1]nonan.15
continuously regenerated by another oxidant present in the reaction mixture for example, NaClO,16 laccase/O2.17 A similar system consisting of peracetic acid/NaBr or OxoneÒ/NaBr with R-TEMPO oxidized polysaccharides such as inulin, starch, amylopectin, pullulan,18 glucan,19 chitin,20 galactomannan,21 mannan,22 carboxymethylcellulose, guar gums, carboxymethylpullulan,23 curdlan,24 and hyaluronan16 to carboxyl compounds. A recent approach employs Dess–Martin periodinane (DMP) during the introduction of aldehyde functional groups on mannan.22 Many reagents have been described, which oxidized primary alcohol groups to aldehydes. However, an equimolar use of oxidant offers a selective control of polyaldehyde hyaluronan formation, which was not described before. None of the methodologies described in Scheme 1 represent an optimum oxidation of HA into aldehyde and alternative procedure should be find. Therefore, this study focuses on the development of HA polyaldehyde (HA-OX) synthesis, where both the carboxyl group and pyranose ring will remain intact. The suggested optimized methods use two different oxidants (TEMPO/NaOCl and DMP) to give reactive polymer molecules. Detailed analytical data show that the presented methods lead to various degrees of substitution and molecular weight. The selective oxidation processes lead to the formation of aldehyde moiety in the position 6 of N-acetylglucosamine part (GlcNAc) of hyaluronic acid. The polyfunctional polymer is a precursor capable of binding compounds containing amino groups (protein, cross-linker, amino active compound) in physiological conditions, which will be demonstrated by the reaction of polyaldehyde–hyaluronan with a model compound, namely lysine. 2. Results and discussion 2.1. Preparation of hyaluronan polyaldehyde (HA-OX) There have been reported several methods for preparation of hyaluronan polyaldehyde but any of them is not produced polyaldehyde by easy way without oxidative cleavage. Two different methods were developed to introduce aldehyde groups into hyaluronic acid—TEMPO/NaClO-mediated oxidation in aqueous solution and oxidation with DMP in aprotic media. These methods of preparation were optimized.
In order to optimize the process of TEMPO-mediated oxidation (Table 1), the influence of several reaction parameters (reaction time, reaction temperature, type of catalyst, oxidant to HA ratio, presence of nitrogen, type of buffer, and its concentration) was investigated. The kinetics of TEMPO-mediated oxidation was monitored (Fig. 1) as a decrease of pH value of the reaction mixture during reaction time. The kinetic plot of AcNH-TEMPO mediated oxidation at room temperature consists of three phases described by Jiang (2000).16 The reaction rate decreases asymptotically corresponding to the decrease of primary hydroxyl groups remaining free for oxidation at the C6 N-acetylglucosamine. The oxidation at 4 °C was slower than at 25 °C and the reaction kinetics is similar at both temperatures and finished after 1 h. In comparison to other polysaccharides, the oxidation of hyaluronan with a catalytic amount of TEMPO (or AcNH-TEMPO) proceeded faster than oxidation of other polysaccharides as starch (5–12 h), pullulan (6–12 h), or inulin (5–10 h).18 While the rapid oxidation of hyluronan takes an hour, in literature was published oxidation of hydroxypropyl guar gum which takes only half hour.23 We found that reaction temperature has an effect on molecular weight but not on degree of substitution. Higher temperature of slightly alkaline reaction mixture (pH 9) caused greater degradation of the biopolymer chain during oxidation. For example, from the hyaluronan with initial molecular weight 73 kDa, derivatives with the same DS but different Mw were prepared at different temperatures (34 kDa at 0 °C and 27 kDa at room temperature). We used TEMPO-mediated oxidation for preparation of polyaldehyde hyaluronan and as well as this was used for oxidation of primary hydroxyl groups of hyaluronan to carboxylate groups. Oxidation to aldehyde or carboxylate groups depends on ratio of oxidation agent (NaOCl) and hyaluronan. The use of amount of oxidant lower than 1 mol NaOCl led to oxidation of primary hydroxyl groups to the desirable polyaldehyde polymer, while the use of 2 mol of primary oxidant NaOCl suggested by Jiang16 led to oxidation to carboxylate groups. As we expected, the products prepared by oxidation catalyzed by TEMPO or AcNH-TEMPO (HA-OX2 and HA-OX8) had the same degree of substitution. The catalytic potential of TEMPO and AcNH-TEMPO is the same and had no effect on the degree of substitution or molecular weight.
10
P. Šedová et al. / Carbohydrate Research 371 (2013) 8–15
Table 1 Reaction conditions of R-TEMPO-mediated oxidation Sample
n(HA) (mmol)
Mw1 (kDa)
n(NaHCO3) (mmol)
n(Na2HPO4) (mmol)
Radical
n(NaOCl) (equiv)
T (°C)
t (h)
DS (%)
Mw2 (kDa)
HA-OX1 HA-OX2 HA-OX3 HA-OX4 HA-OX5 HA-OX6 HA-OX7 HA-OX8 HA-OX9 HA-OX10 HA-OX11 HA-OX12 HA-OX13 HA-OX14 HA-OX15 HA-OX16 HA-COOH
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
73 73 73 73 73 73 73 73 73 670 670 670 202 202 1111 1111 1500
30 30 30 3 1.5 0.9 30 30 30 — — — — — — — —
— — — — — — — — — 5.6 5.6 5.6 4.3 4.3 5.6 5.6 5.6
AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO TEMPO TEMPO TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO AcNH-TEMPO
1 0.5 0.3 1 0.5 0.3 1 0.5 0.3 0.2 0.5 0.7 0.5 0.5 0.9 0.9 4
20 20 20 0 0 0 0 0 0 5 5 5 25 4 5 5 5
7 7 7 7 7 7 7 7 7 1 1 1 2 2 1 1 24
11.6 8.9 6.8 11.9 9.2 7.2 12.4 8.9 7.0 3.2 9.5 10.7 9.1 8.5 11.6 12.9 0
27 — — 34 — — — — — 475 278 226 — — 133 113 —
N2 N2 N2 N2 N2 N2
Figure 1. The temperature effect on kinetics of oxidation (HA-OX13, HA-OX14).
We also investigated the influence of various salts on the reaction outcome. The presence of salts has an effect on tertiary structure of polymer chains. As the concentrations of salts increase, the repulsion between negatively charged disaccharide units decrease and the polymer changes from a rod-like structure to the flexible random coil conformation.25 It was found that the presence of phosphate or carbonate buffering salts (Table 1) in concentrations from 0.9 to 30 equiv did not influence the degree of oxidation of hyaluronan. Oxidation in carbonate buffer (30 equiv; initial pH 9) gave a product with DS = 8.9% and the same reaction performed under the same conditions using 1.5 equiv of the carbonate buffering salt gave a product with similar DS (9.2%). Our results suggest that the effect of pH was more important than the effect of ionic strength. All experiments were carried out at pH 9, where HA exists in the form of sodium salt. The influence of the presence of inert gas (nitrogen) is shown in Table 1 (HA-OX15 and HA-OX16). Saturating the reaction mixture with an inert gas slowed down both the oxidation and degradation rates. These results can be explained by lower concentration of oxygen in the reaction mixture, which influenced the concentration of TEMPO oxidation form.
Another way how to prepare a polyaldehydic polymer from hyaluronan makes use of Dess Martin periodinane (DMP), a relatively mild oxidation reagent. Since the oxidant is sensitive to hydrolysis, it is better to perform the reaction in an aprotic organic solvent. The solubility of HA in such a solvent is a limiting factor in this reaction. Hyaluronic acid is more soluble in aprotic solvents (e.g., DMSO) than sodium hyaluronate and therefore the acidic form of hyaluronan was used in this oxidation. The degradation of hyaluronan is caused by two factors. One of them is the oxidation process itself and the other is the acid hydrolysis caused by the acidic form of hyaluronan. The degradation rate of the acidic form at room temperature is very high (8 109 h1) compared to sodium hyaluronate.26 The degree of substitution depends on the oxidant-to-HA ratio and on reaction time (Table 2). The kinetics of reaction was characterized by derivatives HA-OX21-25. The oxidation rate of hyaluronan by DMP is lower than reaction rate of oxidation by TEMPO/NaOCl. The reaction was finished after 30 h of reaction. DS was determined by NMR analysis and increases with time and growing oxidant-to-HA ratio. Small amount of water (DMSO without molecular sieve) in the reaction mixture led to
11
P. Šedová et al. / Carbohydrate Research 371 (2013) 8–15 Table 2 The effect of reaction conditions on oxidation process using DMP Sample
n(HA:DMP) (mol:mol)
Solvent
T (°C)
t (h)
DS (%)
Mw1 (kDa)
Mw2 (kDa)
HA-OX17 HA-OX18 HA-OX19 HA-OX20 HA-OX21 HA-OX22 HA-OX23 HA-OX24 HA-OX25
1:0.7 1:1 1:1 1:1.2 1:1 1:1 1:1 1:1 1:1
DMSOdry DMSOdry DMSO DMSOdry DMSOdry DMSOdry DMSOdry DMSOdry DMSOdry
20 20 20 20 20 20 20 20 20
9 6 6 24 0.5 1 3 9 24
21 43 39 48 7.32 12.8 19.9 31.8 37.8
329.9 55 55 540 114.6 114.6 114.6 114.6 114.6
25 10 — 13 43 36 24 14 8
decrease of DS (e.g., HA-OX18, HA-OX19) because of hydrolysis of DMP. The oxidation of hyaluronan with DMP has one advantage over that using NaOCl/TEMPO—high conversion to polyaldehyde molecule. This method can be used for preparation of derivatives with higher DS (approx. 50%) but with molecular weight only about 10 kDa. Oxidation to aldehyde in a protic solvent is not so efficient because of the existence of geminal diol, which can be easily transformed to carboxylic acid. This process is suppressed in an aprotic solvent, where the concentration of the geminal diol form is significantly lower. 2.2. Structure analysis Changes of chemical structure of hyaluronan after oxidation to polyaldehyde (HA-OX) or percarboxylated hyaluronan (HA-COOH) was indicated by changes in 1H NMR spectrum (Fig. 2A). The structures were proved by 2D NMR spectroscopy. NMR spectroscopy proved that the oxidation proceeded in position 6 of the glucosamine part of the polysaccharide. It was also confirmed that in water more than 95% of the aldehyde groups was hydrated and exist in the form of geminal diol. The assignment of protons for each monosaccharide residue was based on advanced NMR methodologies. A new signal (N6a in Fig. 2B) in 1H NMR spectra of hyaluronan polyaldehyde was observed at 5.23 ppm and it was assigned to H6 of hydrated form hyaluronan polyaldehyde. HSQC proved that this signal correlated with one at 88 ppm (13C) (Fig. 2B). COSY showed that the signal also correlated with skeletal proton: H5 in GlcNAc which appeared at 3.55 ppm. The intensity of the separated peak at 5.23 ppm was used for determination of degree of substitution. Oxidation of primary hydroxyl groups of hyaluronan to percarboxylated hyaluronan led to disappearance of signals N6 in GlcNAc (at 3.92 and 3.77 ppm H, 61 ppm C) as presented in HSQC spectrum of HACOOH (Fig. 2C). ESI-MS and ESI-MS/MS analyses were carried out in order to confirm the oxidation of GlcNAc in hyaluronan to aldehyde. The spectra indicated the presence of oxidized HA tetramer (m/ z = 773 and 791), HA hexamer (m/z = 1152 and 1160), HA octamer (m/z = 1531 and 1549), and HA decamer (m/z = 954.7 and 963.7) together with the unmodified ones, detected at m/z 775 (tetramer), 1154 (hexamer), 1533 (octamer), and 955 (decamer). All the detected oxidized HA oligomers were present in both the unchanged aldehyde form and in the geminal diol form. The intensity of the ESI-MS peaks of oxidized HA oligomers increased with increasing DS. For example, the intensity of m/z = 773 equals 20% in case of an oligomer with DS = 11.6% (HA-OX1), but is less than 5% for an oligomer with DS 3.2% (HA-OX10). To prove the way of oxidation, MS/MS spectra were collected for the most intensive peaks detected in ESI-MS spectra. An aldehyde in the geminal diol form was unstable and it changed into the aldehyde form at collision energies close to 20 eV. The fragmentation
patterns of both forms of oxidized tetramer are very similar (Fig. 3). The loss of 201 confirmed previous interpretation of NMR data that oxidation process occurred in the GlcNAc of hyaluronan. 2.3. The effect of hyaluronan polyaldehydes on cell viability The MTT viability test was carried out to obtain basic information about cell metabolism and proliferation influenced by oxidized hyaluronan. Cell viability was determined for derivatives prepared by both NaOCl/TEMPO oxidation (DS = 11%, Mw2 = 50 kDa) and DMP oxidation (DS = 29%, Mw2 = 15 kDa). By statistic method was found that the viability of mouse fibroblast 3T3 cell line after treatment with oxidized hyaluronan had not changed significantly in the course of the monitored interval (24–72 h after treatment). The growth of viability of HA-OX treated cells correlated with growth of viability of untreated cells or cells treated by HA. The results showed that both discussed oxidation methods produce substances, which do not show cytotoxicity (Fig. 4). This is very important for potential use in cosmetics or pharmacy. 2.4. Reductive amination of hyaluronan polyaldehyde HA-OX is polymeric substrate suitable for binding of aminocompounds, for example, bioactive aminocompounds or crosslinking agents. We attached lysine to hyaluronan to illustrate this possibility. Reductive amination of carbonyl compounds is a powerful tool used in the synthesis of structurally diverse amines. This reaction can be performed in physiological conditions which is a precondition for preparation of bioacceptable materials. Polyaldehydic hyaluronan can be used as a carrier of active aminocompounds or could be cross-linked with diaminocompounds to form hydrogels and scaffolds. Reductive amination starts as a nucleophile addition of amine to give unstable imine. The imine is converted into more hydrolytically stable amine by reduction with sodium cyanoborohydride. The character of the aminocompound influences the properties of the product. The derivative prepared in this study has DSA = 8%. NMR spectra proved that lysine reacted with hyaluronan polyaldehyde. The signals of lysine in N-lysine-HA derivative were assigned by 2D NMR spectroscopy. DOSY confirmed that the signals of lysine belong to the hyaluronan derivative because of the same translation diffusion coefficients as polymer chain. The protons of lysine were observed in the range of 1.30–1.63 and 2.61–3.07 ppm (Fig. 5). The diastereotopic protons of hyaluronan in position 6 of GlcNAc, which chemical shifts are significantly changed due to the replacement of original oxygen atom with nitrogen of lysine (Fig. 5, protons No. 7), were identified by COSY and HSQC NMR experiments and appeared at 3.09 and 2.71 ppm. The presence of this new diastereotopic pair is the most convincing evidence of HA modification by the reductive amination procedure, which is not present in the case of the ionic
12
P. Šedová et al. / Carbohydrate Research 371 (2013) 8–15
Figure 2. (A) 1H NMR spectra of HA, HA-OX prepared with AcNH-TEMPO-mediated oxidation with 0.5 equiv NaOCl, and HA-COOH prepared with 4 equiv of NaOCl, (B) HSQC of HA-OX19, (C) HSQC of HA-COOH. The assignment of protons in HSQC of hyaluronan derivatives: GlcNAc (N), GlcA (G).
Figure 3. MS/MS fragmentation of (A) m/z = 791.20 (collision energy of 40 eV), (B) m/z = 791.20 (collision energy of 15 eV), and (C) m/z = 773.20 (collision energy of 40 eV). The fragmentation pathway corresponds to one indicated structure. GA stands for glucuronic acid.
P. Šedová et al. / Carbohydrate Research 371 (2013) 8–15
13
Figure 4. Cell viability determined by MTT assay after HA-OX treatment: (A) DS = 11%, Mw2 = 50 kDa, c = 100–1000 lg mL1, (B) DS = 29%, Mw2 = 15 kDa, c = 100– 1000 lg mL1.) The data are expressed as mean ± SEM (n = 6). Statistically significant differences (Student’s t-test, p 6 0.05) to the non-treated sample group are marked with an asterisk.
Figure 5. 1H NMR spectrum of lysine-NH-derivative of hyaluronan.
interaction between HA-COO and lysine-H+. Degree of substitution (DSA 8%) was determined as the integral ratio between signals of HA-N-acetyl group (2.0 ppm) and one of the newly formed diastereotopic protons in position 6 of GlcNAc (3.09 ppm). 3. Conclusion We have shown that DMP and TEMPO/NaOCl are suitable oxidation agents for introduction of reactive aldehyde groups to hyaluronan. The formation of aldehyde is regulated by stoichiometric ratio of reagents. The prepared polyaldehyde is suitable for binding a wide range of amino-compounds and did not show any cytotoxic effect in standard viability tests up to the concentration of 1 mg mL1. The HA polyaldehydes are convenient precursors for the preparation of biocompatible materials suitable for application in tissue engineering and regenerative medicine. DMP oxidation
produces aldehydes with higher DS than oxidation with TEMPO/ NaOCl. The oxidation of hyaluronan with TEMPO/NaOCl was optimized and reaction conditions like temperature and amount of oxidant influenced parameters like DS and Mw2 the most extensively. The optimized synthesis is more convenient for the preparation of derivatives with high Mw, which are more suitable for the preparation of crosslinked hydrogels. The oxidation with DMP leads to derivatives with high DS and lower Mw which are applicable to binding of higher amount of bioactive compounds. 4. Experimental 4.1. Chemicals Bacterial hyaluronan sodium salt with molecular weight ranging from 55 to 1111 kDa was obtained from Contipro Biotech.
14
P. Šedová et al. / Carbohydrate Research 371 (2013) 8–15
Dess-Martin periodinane was purchased from Bujno Synthesis, 2,2,6,6-tetramethylpiperidine 1-oxyl, 4-acetamido-2,2,6,6-tetramethylpiperidine-1-oxyl, and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) for cell viability assay and lysine hydrochloride were purchased from Sigma–Aldrich. Sodium hypochlorite solution (Penta), sodium bromide, sodium bicarbonate, and sodium phosphate (Lach-Ner) were reagent grade commercial products. Dimethyl sulfoxide was dried using molecular sieve (4 Å). One equivalent (1 equiv) of reagents used in the study means 1 mol of reagent to 1 mol of hyaluronan dimer. 4.2. Oxidation of sodium hyaluronate with R-TEMPO/NaOCl The following is a general procedure. For molar ratios used in particular experiments see Table 1. Sodium hyaluronate (0.5 g) was dissolved in demineralized water (50 mL) containing sodium hydrogen carbonate or disodium phosphate. 0.01 mol of the catalyst (TEMPO or 4-AcNH-TEMPO) was added, the mixture was stirred briefly and then sodium hypochlorite was added. The reaction mixture was stirred for 1–7 h. The product was isolated by dialysis (cut off 12–14 kDa) against demineralized water and by freezedrying. 4.3. Preparation of the acidic form of HA Sodium hyaluronate was converted into the acidic form by ion exchange using H+ catex according to Haxaire et al.27 and isolated by freeze-drying. Hyaluronic acid cannot be stored for prolonged periods due to its quick degradation. Therefore, the following reaction steps (oxidation with DMP in DMSO) were performed immediately after freeze-drying procedure. 4.4. Dess–Martin periodinane oxidation of hyaluronan The following is a general procedure. For molar ratios used in particular experiments see Table 2. Hyaluronic acid (0.5 g) was dissolved in anhydrous dimethyl sulfoxide overnight to form a 1% (w/ v) solution. Dess–Martin periodinane was added and the resulting solution was stirred at room temperature for 0.5–24 h. Afterward, the reaction mixture was diluted with demineralized water to double the volume. The product was isolated by dialysis (cut off 6– 8 kDa) against the mixture of 0.1% NaCl and 0.1% NaHCO3 (3 5 L) and demineralized water (8 5 L). 4.5. Reductive amination with lysine A solution of lysine hydrochloride (0.46 equiv) in water (1 mL) was added to the aqueous solution of the oxidized hyaluronic acid (0.5 g in 50 mL, DS = 23%). The mixture was stirred for 48 h at 20 °C. Solid NaBH3CN (3 equiv) was then added and the mixture was stirred for another 48 h under the same conditions. The resulting solution was diluted with water to 0.1% solution and dialyzed (cut off 12–14 kDa) against the mixture of 0.1% NaCl and 0.1% NaHCO3 (3 5 L) and demineralized water (7 5 L). The product was isolated by evaporation on a rotary evaporator. 4.6. NMR analysis Solution-state NMR spectroscopy was carried out on a Bruker Avance III 500 MHz instrument operating at a proton frequency of 500.25 MHz and a carbon frequency of 125.80 MHz. The spectrometer was equipped with a 5 mm Bruker BBFO plus broadband probe with an actively shielded z-gradient coil. All of the spectra were acquired and elaborated by Bruker 2.1 Topspin software. Freeze-dried samples (7 mg) were dissolved in D2O (0.75 mL) and were transferred into 5 mm NMR quartz tubes. COSY, TOCSY, and
HSQC experiments were used for identification of protons and carbons in the derivatives. 4.7. Degree of substitution (DS) determination The aldehyde content was determined by 1H NMR spectroscopy. Samples were dissolved in D2O. It is defined as a molar ratio of aldehyde groups and dimer unit of hyaluronan. DS was calculated as a ratio of the integral of signal I(H) at 5.20–5.30 ppm (hydrated form of the aldehyde, geminal diol) and integral of one proton of a methyl in acetylamino group I(CH3) at 2.0 ppm.
DS ¼ ½3IðHÞ=IðCH3 Þ 100 ð%Þ
ð1Þ
4.8. Degree of amination (DSA) determination The content of aminoacid was determined by 1H NMR spectroscopy. Sample was dissolved in concentration 14 mg/0.8 mL D2O with addition of 20 lL of 20% NaOD (w/v). It is defined as a molar ratio of aminoacid and dimer unit of hyaluronan. DSA was calculated as a ratio of the integral of signal I(H) at 3.03–3.11 ppm (proton H6 of GlcNAc changed by binding of lysine in the structure of derivative No. 7, Fig. 5) and the integral of one proton of a methyl in acetylamino group I(CH3) at 2.0 ppm.
DSA ¼ ½3IðHÞ=IðCH3 Þ 100ð%Þ 4.9. Molecular weight determination Molecular weights (Mw1) of hyaluronic acid or hyaluronan and molecular weights of the modified products (Mw2) were assigned using SEC-MALLS (size exclusion chromatography/multiangle laser light scattering). Samples were dissolved overnight in a mobile phase to give a solution with the concentration of 2 mg cm3. The chromatographic system consisted of an LC10ADVP Shimadzu HPLC pump, SIL-10AF autosampler, CTO10AVP column oven, SCL-10AVP system controller, DGU-14A degasser, RID-10A refractive index detector, SPD-10AVVP UV–VIS detector (all from Shimadzu), and miniDAWN TREOS light scattering photometer (Wyatt Technology Corporation). The injection volume was 100 ll. Each sample was filtered through a 0.22 lm MS Nylon Syringe Filter. The mobile phase was aqueous 50 mM sodium phosphate and 0.02% sodium azide at the flow rate of 0.8 mL min1. Data acquisition and molecular weight calculations were performed using ASTRA software (version 5.3.4, Wyatt Technology Corporation, USA). The specific refractive index increment of 0.155 mL g1 was used for HA. 4.10. Mass spectrometry Powdered hyaluronan (100 mg) was dissolved in the mixture of 0.1 M sodium acetate and 0.15 M NaCl (10 mL; pH 5.3, adjusted with glacial acetic acid), and then incubated with 2000 IU of hyaluronidase (Finepharm) at 37 °C for 2 days. The enzyme was inactivated by short boiling of the solution at the end of incubation and removed by precipitation. The sample was filtered through 0.2 lm Nylon syringe filter. The filtrate (2 mL) was transferred into the Vivaspin 15R concentrator (2000 MWCO Hydrosart, Sartorius) and was centrifuged at 9000 rpm for 15 min. After preconcentration of the sample, the concentrator was filled with deionized water (10 mL) and was centrifuged at 9000 rpm for 30 min. Four wash cycles were used to remove the initial salt content. The sample was recovered from the bottom of the concentrator, diluted with 0.1% (v/v) aqueous solution of the mixture HCOOH/methanol = 1:1 to the final concentration of 1 mg mL1 and was injected directly into the mass spectrometer.
P. Šedová et al. / Carbohydrate Research 371 (2013) 8–15
Mass spectroscopic analyses of digested and desalted derivatives were performed using a Synapt HDMS mass spectrometer (Waters), equipped with an electrospray ionization source operating in negative ion mode. The effluent was introduced into an electrospray source with a syringe pump at the flow rate of 10 lL min1. Nitrogen was used as cone gas (100 L h1) and desolvation gas (800 L h1). Capillary voltage was set to 3 kV. Sampling cone was set to 100 V. Extraction cone was set to 5 V. The source block temperature was set to 100 °C, while the desolvation temperature was 250 °C. For each sample full MS and MS/MS scans from m/z 50 to 2000 were acquired for 2 min. For MS/MS measurements, argon was used as a collision gas. The collision energy was optimized to fragment the ion of interest, typically 55 eV for the ions with higher m/z and 25 eV for the ions with lower m/z. Data were collected at 1 scan s1 and elaborated using MassLynx software. 4.11. Fibroblast viability Mouse fibroblast cell line (NIH 3T3) (Sigma–Aldrich) was cultured until 20th passage. Cells of 5–20th passage were used in the following experiments. NIH 3T3 was grown in DMEM medium supplemented with 10% FBS, glutamine (0.3 mg mL1), penicillin (100 mL1) (Sigma–Aldrich), and streptomycin (0.1 mg mL1) (Sigma–Aldrich) in 7.5% CO2 at 37 °C in a culture flask as recommended by the supplier. The NIH 3T3 viability was monitored after 24, 48, and 72 h as a response of an addition of hyaluronan and its oxidized derivatives with different DS into the culture medium in concentrations of 100–1000 lg mL1. As controls were used cells cultured in medium and in medium containing HA. Solutions of derivatives were prepared using cell cultivation medium to obtain a final concentration of 1000 lg mL1, and were filtered through a sterile filtration device (0.22 lm) to produce sterile solutions to be tested for cell viability. Cell viability was measured using an MTT assay, which is based on the reduction of tetrazolium salt by viable cells to colorful formazan. The MTT assay was described previously.28 MTT stock solution was added to the cell culture medium and plates were incubated at 37 °C for 2.5 h. The supernatant was discarded and cells were lysed in lysis solution for 30 min on a shaker. The formazan product was analyzed at 570 nm using a VersaMax Microplate Reader spectrophotometer (Molecular Devices). All experiments were carried out at least in six independent
15
repeats. Student’s t-test for two samples was applied on the data, p 6 0.05 (⁄) was considered significant. References 1. Lee, H. G.; Cowman, M. K. Anal. Biochem. 1994, 219, 278–287. 2. Bulpitt, P.; Aeschlimann, D. J. Biomed. Mater. Res. 1999, 47, 152–169. 3. Ferguson, E. L.; Roberts, J. L.; Moseley, R.; Griffiths, P. C.; Thomas, D. W. Int. J. Pharm. 2011, 420, 84–92. 4. Kuo, J. W. Practical Aspects of Hyaluronan Based Medical Products; CRC Press, 2005. Chapter 5. 5. Larsen, N. E.; Balazs, E. A. Adv. Drug Delivery Rev. 1991, 7, 279–293. 6. Renier, D.; Bellato, P.; Bellini, D.; Pavesio, A.; Pressato, D.; Borrione, A. Biomaterials 2005, 26, 5368–5374. 7. Lu, H. S. M.; Shuey, S. W. US20120004194, 2012. 8. Araki, M.; Tao, H.; Nakajima, N.; Sugai, H.; Sato, T.; Hyon, S.-H.; Nagayasu, T.; Nakamura, T. J. Thorac. Cardiovasc. Surg. 2007, 134, 1241–1248. 9. Xu, Y.; Li, L.; Yu, X.; Gu, Z.; Zhang, X. Carbohydr. Polym. 2012, 87, 1589–1595. 10. Liang, Y.; Liu, W.; Han, B.; Yang, C.; Ma, Q.; Song, F.; Bi, Q. Colloids Surf., B 2011, 82, 1–7. 11. Ehrenfreund-Kleinman, T.; Azzam, T.; Falk, R.; Polacheck, I.; Golenser, J.; Domb, A. J. Biomaterials 2002, 23, 1327–1335. 12. Ito, T.; Yeo, Y.; Highley, C. B.; Bellas, E.; Kohane, D. S. Biomaterials 2007, 28, 3418–3426. 13. Bergman, K.; Engstrand, T.; Hilborn, J.; Ossipov, D.; Piskounova, S.; Bowden, T. J. Biomed. Mater. Res., Part A 2009, 91A, 1111–1118. 14. Mehta, H.; Singh, R. EP0961783, 2007. 15. Aeschlimann, D.; Bulpitt, P. US7196180 B2, 2007. 16. Jiang, B.; Drouet, E.; Milas, M.; Rinaudo, M. Carbohydr. Res. 2000, 327, 455– 461. 17. Elboutachfaiti, R.; Delattre, C.; Petit, E.; Michaud, P. Carbohydr. Polym. 2011, 84, 1–13. 18. Bragd, P. L.; Besemer, A. C.; van Bekkum, H. Carbohydr. Polym. 2002, 49, 397– 406. 19. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H. Carbohydr. Res. 1995, 269, 89– 98. 20. Muzzarelli, R. A. A.; Muzzarelli, C.; Cosani, A.; Terbojevich, M. Carbohydr. Polym. 1999, 39, 361–367. 21. Sierakowski, M. R.; Milas, M.; Desbrieres, J.; Rinaudo, M. Carbohydr. Polym. 2000, 42, 51–57. ˇ urana, R.; Lacík, I.; Paulovicˇová, E.; Bystricky´, S. Carbohydr. Polym. 2006, 63, 22. D 72–81. 23. Ding, B.; Ye, Y. q.; Cheng, J.; Wang, K.; Luo, J.; Jiang, B. Carbohydr. Res. 2008, 343, 3112–3116. 24. Delattre, C.; Rios, L.; Laroche, C.; Le, N. H. T.; Lecerf, D.; Picton, L.; Berthon, J. Y.; Michaud, P. Int. J. Biol. Macromol. 2009, 45, 458–462. 25. Furlan, S.; La Penna, G.; Perico, A.; Cesaro, A. Carbohydr. Res. 2005, 340, 959– 970. 26. Šedová, P.; Leierová, V.; Buffa, R.; Velebny´, V. Chem. Listy 2012, 106, 556. 27. Haxaire, K.; Maréchal, Y.; Milas, M.; Rinaudo, M. Biopolymers 2003, 72, 10– 20. 28. Vištejnová, L.; Dvorˇaková, J.; Hašová, M.; Muthny´, T.; Velebny´, V.; Soucˇek, K.; Kubala, L. Neuro Endocrinol. Lett. 2009, 30, 121–127.