Structure, depolymerization, and ... - Wiley Online Library

Structure, depolymerization, and cytocompatibility evaluation of glycol chitosan Darryl K. Knight, Stephen N. Shapka, Brian G. Amsden Department of Chemical Engineering, Queen’s University, Kingston, Ontario, Canada K7L 3N6 Received 8 November 2006; revised 12 February 2007; accepted 12 March 2007 Published online 8 June 2007 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.31430 Abstract: Glycol chitosan, a water soluble chitosan derivative being investigated as a new biomaterial, was fractionated via two different methods. Initial characterization of the glycol chitosan with 1H NMR spectroscopy illustrated the presence of both secondary and tertiary amine groups, contradictory to its widely accepted structure. Fractionation of glycol chitosan with nitrous acid resulted in a significant reduction in the number average molecular weight, specifically, from 170 to *7 kDa for a pH 3 and below. However, the reaction altered its chemical structure, as the secondary amine groups were converted to N-nitrosamines, which are potentially carcinogenic. An increase in the pH of the reaction limited this formation, but not entirely. Free radical degradation initiated with po-

tassium persulfate was not as effective at reducing the molecular weight as the nitrous acid approach, yielding molecular weights around 12 kDa under the same molar ratio of degrading species, but did retain the structural integrity of the glycol chitosan. Additionally, control of the molecular weight appears feasible with potassium persulfate. When assessed in vitro for cytocompatibility, the polymer exhibited no toxicity on monolayer-cultured chondrocytes, and in fact stimulated cell growth at low concentrations. Ó 2007 Wiley Periodicals, Inc. J Biomed Mater Res 83A: 787–798, 2007

INTRODUCTION

pH in organic acids; thus, a derivative soluble at physiologic conditions would greatly promote its use as a biomaterial.27 Unfortunately, most modifications to improve its water solubility also reduce the fraction of free amine containing residues. Efforts to improve chitosan’s water solubility have included adjusting its degree of acetylation to *50%,6,27,28 and through grafting of more hydrophilic components. Specifically, carboxymethyl,29 dicarboxymethyl,1 and succinyl30 functional groups, and poly (ethylene glycol)31 have all been attached to chitosan via the available amine groups. The synthesis of water soluble chitosan derivatives also included the development of glycol chitosan, whose currently accepted structure is shown in Figure 1(B).32–34 Based on this structure, the free amine groups would remain unaffected, allowing for future modification while retaining its favorable biological interactions. While studies with glycol chitosan have been conducted,30,32–34 very little characterization has been done to elucidate its actual chemical structure. Glycol chitosan owes its water solubility to the incorporation of the hydrophilic glycol group, introduced by reacting chitin with ethylene oxide followed by its deacetylation. However, if the chitin was partially deacetylated initially, its reaction with ethylene oxide would likely yield N-glycolated

Chitosan, shown in Figure 1(A), is the term given to the family of cationic polysaccharides consisting of Nacetyl D-glucosamine and D-glucosamine residues coupled through a b (1-4) (glycosidic) linkage, where the degree of N-acetylation is less than 50%.1,2 Chitosan has been demonstrated to be biocompatible,3–5 to accelerate wound healing,6,7 to possess antimicrobial properties,8–11 and to support the growth and function of osteoblasts and chondrocytes.12,13 These favorable properties are believed to be due in part to the free amine groups along its backbone. It is being examined extensively in the biomedical field for several applications, including wound dressings,14,15 scaffolds for engineering tissue such as skin16–18 and articular cartilage,19–22 and as drug delivery vehicles.23–26 While chitosan has demonstrated biocompatibility, its usage is limited by its poor water solubility. Highly deacetylated chitosan is only soluble at low Correspondence to: B.G. Amsden; e-mail: brian.amsden@ chee.queensu.ca Contract grant sponsor: Natural Sciences and Engineering Research Council of Canada ' 2007 Wiley Periodicals, Inc.

Key words: glycol chitosan; depolymerization; potassium persulfate; cytotoxicity; NMR

788

KNIGHT, SHAPKA, AND AMSDEN

methods for fractionating chitosan. Cleavage of the glycosidic bond along the glycol chitosan backbone was achieved through diazotization of the primary amine via reaction with nitrous acid,42 and free radical degradation with the thermal dissociation initiator, potassium persulfate.43 The cytocompatibility of the different molecular weight fractions was then assessed using an immortalized chondrocyte line (T/C-28a2).

EXPERIMENTAL Materials

Figure 1. Structure of chitosan (A) and currently accepted structure of glycol chitosan (B).

residues, as the epoxide would more favorably react with the amine groups than with the hydroxyl groups.35 Thus, an objective of this study was to characterize glycol chitosan to verify the currently accepted structure of solely O-glycolated chitosan. A further objective was to examine methods of depolymerizing glycol chitosan that have been commonly used for chitosan. A procedure to consistently realize a low molecular weight glycol chitosan is desirable for several reasons. Firstly, a reduction in glycol chitosan’s molecular weight could facilitate its manufacture into various biomedical devices through a reduction in its viscosity. Secondly, low molecular weight fractions have been shown to be more effective gene delivery agents. Specifically, low molecular weight chitosan has been shown to effectively complex DNA and prevent nuclease degradation,36,37 in addition to improving gene expression of luciferase plasmid.38 Finally, lower molecular weight glycol chitosan would likely be cleared more efficiently after intravenous injection. The final objective was to determine the dependence of glycol chitosan molecular weight on its cytotoxicity. There has been only one report on the cytotoxicity of glycol chitosan. Carrenõ-Go´mez and Duncan found only mild concentration-independent (up to 2.5 mg/ mL) levels of toxicity of glycol chitosan towards a murine melanoma cell line.32 They also concluded that higher molecular weight glycol chitosan (molecular weight greater than 100 kDa) displayed some degree of cytotoxicity.32 These findings, however, are in contradiction to reports on the cytotoxicity dependence on the molecular weight of chitosan of the same degree of deacetylation. It has been reported that there was no influence of chitosan molecular weight on cell viability of L929 murine fibroblasts,39 A549 lung carcinoma cells,40 and human dermal fibroblasts.41 In this study, two approaches for the fractionation of glycol chitosan were compared, based on popular Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

Diethanolamine, potassium nitrite, potassium persulfate, sodium acetate, sodium borohydride, and sodium hydroxide were all obtained from Sigma–Aldrich Canada Ltd. Commercial sources of glycol chitosan were purchased from both Sigma–Aldrich Canada Ltd. and Wako Chemicals, USA. Deuterium oxide was purchased from Cambridge Isotope Laboratories, Inc., while acetic and hydrochloric acids were obtained from Fisher Scientific, Ltd. All reagents were used as received. Water used was of Type I purity, obtained from a Millipore Milli-Q Plus Ultra-Pure Water System. Growth media for cell culture was purchased from Invitrogen Canada, Inc. Unless otherwise stated, culture reagents were from Sigma–Aldrich/ Fluka BioChemika and were of high purity grade, cell-culture tested. All media was sterile-filtered prior to use, and the pH adjusted to 7.4.

Glycol chitosan purification Glycol chitosan (1 g) was dissolved in water (75 mL) and filtered to remove insoluble impurities. The filtrate was then dialyzed against water for 8 h using molecular weight cutoff 50 kDa dialysis tubing. Both the membranes and media were replaced at 4 h. The purified high molecular weight glycol chitosan solution was frozen at 208C for a minimum of 8 h and lyophilized for at least a 48-h period.

Glycol chitosan depolymerization via nitrous acid The fractionation of glycol chitosan with nitrous acid was adapted from a procedure proposed by Allan and Peyron,44 which is illustrated in Scheme 1. Purified glycol chitosan (*0.8 g, 4 3 103 mmol) was dissolved in water (75 mL). Under magnetic stirring 1M hydrochloric acid was added followed immediately by 1M potassium nitrite (1 mL, 1 mmol) to give solutions of pH 1.6, 2.9, and 5.1. At predetermined times, sodium borohydride (100 mg, 2.6 mmol) was added and allowed to react for an additional 30 min to reduce the terminal aldehyde groups to hydroxyl groups. The pH of the low molecular weight glycol chitosan solution was neutralized and concentrated on a rotary evaporator at 358C and then dialyzed against water for 4 h using a molecular weight cutoff of 1 kDa

STRUCTURE, DEPOLYMERIZATION, AND CYTOCOMPATIBILITY OF GLYCOL CHITOSAN

Scheme 1.

789

Proposed fractionation of glycol chitosan via diazotization of the primary amine with nitrous acid.

dialysis tubing. The membranes and media were changed at 2 h. The purified low molecular weight glycol chitosan solution was neutralized prior to lyophilization.

Glycol chitosan depolymerization via potassium persulfate The fractionation of glycol chitosan with potassium persulfate was adapted from a procedure proposed by Hsu

Scheme 2.

et al.43 The reaction is illustrated in Scheme 2. Purified glycol chitosan (*0.8 g, 4 3 103 mmol) was dissolved in a 2% (v/v) hydrochloric acid solution (75 mL) followed immediately by the addition of potassium persulfate at 708C. At predetermined reaction times, sodium borohydride (100 mg, 2.6 mmol) was added and allowed to react for an additional 30 min at room temperature. The pH of the low molecular weight glycol chitosan solution was neutralized with 1M sodium hydroxide and concentrated at 358C on a rotary evaporator. The concentrated solution was then dia-

Proposed fractionation mechanism of glycol chitosan with potassium persulfate at 708C. Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

790

lyzed against water for 4 h using a molecular weight cutoff of 1 kDa dialysis tubing. The membranes and media were changed at 2 h. The purified low molecular weight glycol chitosan solution was quenched under a low flow rate of air for 24 h. The solution was again dialyzed with a molecular weight cutoff of 1 kDa dialysis tubing against water for 2 h, followed by its subsequent neutralization and lyophilization.

Nuclear magnetic resonance spectroscopy NMR spectra were conducted with a Bruker Avance-600 Ultrashield spectrometer equipped with a 5 mm TBI S3 probe with Z gradient and variable temperature capability. Samples were prepared at 20 mg/mL in deuterium oxide, preheated at 608C for 6 h, then either adjusted to pH >10 with 1M sodium hydroxide (30 lL) or adjusted to pH 10 can be seen at 3.3 ppm, while the corresponding protons on the carbon atom adjacent the terminal hydroxyl group arises at 4.25 ppm. At pH 10 can be attributed to NHCH2CH2OH. This assignment is in agreement with assignments to similar protons in Npoly(ethylene glycol) grafted chitosan reported in the literature.31,49 Furthermore, the peak at 5.6 ppm in the low pH case is assigned to the proton at carbon position 1 of an N-glycolated residue. Given that six individual peaks between 3.1 and 3.6 ppm, aside from H-2 (D), can be seen in the 1H NMR spectrum of the purified glycol chitosan, it is reasoned that both secondary and tertiary amines exist, indicating mono- and disubstitution. The protons of each methylene (CH2) group, as part of the N-glycolated groups, are diastereotopic due to the stereogenic centers at carbon position 2 along the backbone. As such, protons denoted as a and a0 , shown in Figure 4, in the secondary amine case are magnetically different and should give rise to separate peaks in the 1H NMR spectrum. This would Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a


Figure 4. Proposed structure of glycol chitosan.

produce three small peaks within the region of 3.1 to 3.6 ppm, of similar integration to each other—a, a0 and the proton at position 2. In the case of the tertiary amine, where two glycol groups are present, the protons of each methylene group are still magnetically different; however, the protons between each glycol group should be similar. The peaks attributed to b and b0 should therefore be twice that of the proton at position 2. With this knowledge, a modified structure of glycol chitosan is proposed, as shown in Figure 4, which illustrates these N-glycolated residues. Although the peaks between 3.1 and 3.6 ppm cannot be completely resolved, the degree of acetylation can be calculated from the 1H NMR spectrum of the purified glycol chitosan. Because the peaks between 5.05 and 5.23 ppm arise from the proton at position 1 for each substituent at position 2, the sum of these integrations will give an indication of the average number of residues per chain. As such, the degree of acetylation can be determined through its ratio to the methyl protons of the acetyl group, as illustrated below: Degree of acetylation ¼

ICH3 =3 IH1Total

ð1Þ

where I represents the integration of the peak indicated by the subscript. Based on Eq. (1), the degree of acetylation was calculated to range from 10 to 13% in the purified glycol chitosan samples. The remaining 87 to 90% would be the total contribution of the primary, secondary, and tertiary amine groups. Their individual contributions could not be resolved even with 2D NMR spectroscopic techniques (data not shown). Upon characterization of the structure of the purified glycol chitosan obtained from Sigma–Aldrich Canada, it was then compared to a sample purified from Wako Chemicals, USA. Offset 1H NMR spectra of the two commercially available sources are shown in Figure 5. As can be seen, the spectra are virtually identical; indicating that in all likelihood, the inherent structure of glycol chitosan is similar. In fact, the molecular weight of the purified glycol chitosan dis-


Figure 5. Offset 1H NMR spectra of purified glycol chitosan obtained from Sigma–Aldrich Canada, Ltd. and Wako Chemicals, Inc., USA.

tributed from Wako Chemicals, USA, is also comparable to that of Sigma–Aldrich Canada. Specifically, the number average and weight average molecular weights ranged from 171 to 178 and 228 to 237 kDa, respectively for a PI of *1.3. Because both the structure and molecular weight of the purified glycol chitosan samples are similar, they will be treated as such throughout the remainder of this study. Glycol chitosan depolymerization via nitrous acid One approach examined to reduce the molecular weight was diazotization of the primary amine through its reaction with nitrous acid as proposed by Allan and Peyron.44 Nitrous acid forms a number of nitrosating agents in aqueous solution, which are dependent on the pH as well as the nature of the anion present.50 The anion was not changed in this study, but the pH of the solution was altered to determine whether a change in pH would affect the fractionation of glycol chitosan. Offset 1H NMR spectra of the purified glycol chitosan along with three fractionated samples conducted at pH 1.6, 2.9, and 5.1 are shown in Figure 6. From Figure 6, it is evident that the reaction pH does affect the structural integrity of the glycol chitosan. Band broadening can be observed on the methyl protons of the acetyl group, as well as at the protons at carbon positions 1 and 2 in the reactions at pH 1.6 and 2.9. Also of note in these two reactions is the reduction in the intensity of the peaks attributed to the secondary and tertiary amine groups between 3.1 and 3.6 ppm, indicating that the electronic environment of these protons has been modified. New peaks at 3.71, 4.90, and 5.63 ppm also appear in the spectra following fractionation with nitrous acid at a

793

pH of 1.6 and 2.9, and to a much lesser extent in the pH 5.1 reaction. Allan and Peyron44 proposed the nitrous acidium ion as the nitrosating agent from kinetic data conducted with chitosan in aqueous hydrochloric acid solutions (50.0–125 mM), where a pH below 3 would be achieved. It has been noted that nitrosation of amide groups will occur below pH 3, and that a 10fold increase is observed for each whole number reduction in pH from 3 to 1.50 Nitrosation of an amide group results in the formation of an N-nitrosamide, which is unstable at physiologic pH and decomposes to a diazohydroxyl, followed by a diazonium cation, which is the same intermediate as in the primary amine case and results in the cleavage of the glycosidic bond. Differences in the 1H NMR spectra of the fractionated glycol chitosan would not arise from the formation of N-nitrosamides, as the fractionated glycol chitosan solution is neutralized to terminate the reaction, which would promote the formation of the diazonium cation. As a reduction in the degree of acetylation was not observed following nitrous acid fractionation, conversion of the amide groups to N-nitrosamides did not occur. Although N-nitrosamides are unstable at physiologic conditions, N-nitrosamines formed from the nitrosation of secondary amines are quite stable and are formed under milder conditions.50 The formation of N-nitrosamines is highly undesirable, as most Nnitroso compounds studied in the literature have demonstrated some carcinogenicity.50,51 As N-nitrosamines form readily from the reaction of secondary amines with nitrosating agents, the reduction in the intensity of the peaks between 3.1 and 3.6 ppm is attributed to their conversion to the N-nitrosamine, which would give rise to peaks further downfield. The formation of the N-nitroso

Figure 6. Offset 1H NMR spectra of fractionated glycol chitosan via nitrous acid at selected pHs. Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

794


derivatives likely accounts for the peaks around 4.90 and 5.63 ppm. Specifically, the peak seen at 4.90 ppm may be the proton at carbon position 3 in the N-nitrosamine case, while the peak observed at 5.63 ppm would be the corresponding proton at carbon position 1 in the same residue. Although nitrosation of tertiary amines is possible, yielding Nnitrosamines resulting from the cleavage of a glycol group,52 their occurrence would only be significant at elevated temperatures,52 which have not been used here. Thus, the unaffected peak arising at 3.58 ppm can be attributed to the tertiary amine case. Based on the above logic, the degrees of N-nitrosamination can be determined by applying a similar relationship to that in Eq. (1). Specifically, the degree of N-nitrosamination following the fractionation of glycol chitosan with nitrous acid can be approximated as: Degree of N nitrosamination ¼

INnsa H1 IH1Total

ð2Þ

where IH-1 Total now includes the contribution of the N-nitrosamine. The degree of N-nitrosamination was calculated to be 13, 14, and 5% for the pH 1.6, 2.9, and 5.1 trials, respectively. This indicates that fractionating glycol chitosan with nitrous acid at higher pH does inhibit the formation of the N-nitrosamines. Confirmation that nitrous acid resulted in fractionation of the glycol chitosan backbone was seen in the GPC data presented in Table I. The lowest molecular weight was achieved with the pH 2.9 reaction; however, selected fractionation was achieved in all three cases. The lower molecular weight in the pH 2.9 reaction compared with that in the pH 1.6 reaction is attributed to a greater degree of protonation in the pH 1.6 reaction, restricting the reaction between the nitrous acidium ion and an unionized primary amine group and hence the formation of the diazonium cation. In the case of the pH 5.1 reaction, where a higher proportion of the primary amine groups would be unionized, the nitrosating agent may not be the same. Allan and Peyron demonstrated that the nitrosating species was the nitrous acidium ion for pHs below 3.42 At a pH of 5.1, the nitrosating agent may be nitrosyl chloride or possibly nitrous anhydride, which are not as effective TABLE I Molecular Weights of Fractionated Glycol Chitosan via Nitrous Acid at Different pHs pH

Mn (kDa)

Mw (kDa)

PI

1.6 2.9 5.1

7.2 6.7 24.4

10.8 9.7 41.1

1.5 1.4 1.7

Average of four injections. Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

Figure 7. Offset 1H NMR spectra of fractionated glycol chitosan via nitrous acid under different reaction times.

nitrosating agents,50 hence resulting in a higher molecular weight. Although fractionation at pH 5.1 was the least effective, this reaction retained the structural integrity of the glycol chitosan to the greatest degree. To assess whether the resultant molecular weight or the formation of the N-nitrosamines were reaction time dependent, glycol chitosan was also fractionated at pH 5.1 for 1 and 2 h. The offset 1H NMR spectra of the fractionated glycol chitosan at pH 5.1 for 1 through 4 h are shown in Figure 7. Reducing the fractionation time to 1 h does not appear to inhibit the formation of N-nitrosamines as evidenced by the persistence of the peaks at 4.92 and 5.67 ppm. The molecular weight of the glycol chitosan following the fractionation at pH 5.1 for 1 to 4 h is also similar (Table II), indicating that the reaction between glycol chitosan and nitrous acid is both quick and selective. Glycol chitosan depolymerization via potassium persulfate Free radical degradation of the glycol chitosan backbone was achieved with the thermal dissociation initiator, potassium persulfate, as proposed by Hsu et al.43 As a direct comparison, an equimolar ratio of fractionating species to glycol chitosan used in the nitrous acid degradation was also used in this study. Although Hsu et al. observed a saturation effect in the reduction of the molecular weight of chitosan after 1 h, fractionation of glycol chitosan in this study was conducted up to 24 h, as the half-life of potassium persulfate is 20.9 h.43 The effect of the potassium persulfate on the structural integrity of the glycol chitosan was assessed through 1H NMR spectroscopy. Offset spectra of purified and potassium persulfate fractionated glycol chitosan are shown in Figure 8. Unlike the nitrous acid fractionated glycol


TABLE II Molecular Weights of the Fractionated Glycol Chitosan via Nitrous Acid for 1–4 h at pH 5.1

795

TABLE III Molecular Weights of the Fractionated Glycol Chitosan via Potassium Persulfate from 0.5 to 24 h

Time (h)

Mn (kDa)

Mw (kDa)

PI

Time (h)

Mn (kDa)

Mw (kDa)

PI

1 2 4

27.2 21.5 24.4

42.5 34.0 41.1

1.6 1.6 1.7

0 0.5 1 2 4 8 24

170.6 12.9 12.1 12.6 11.5 10.8 12.5

238.0 36.8 32.1 32.3 31.4 23.2 27.5

1.4 2.9 2.7 2.6 2.7 2.2 2.2

Average of four injections.

chitosan, the 1H NMR spectra are consistent, indicating that the integrity of the glycol chitosan backbone is retained following potassium persulfate fractionation. The effectiveness of potassium persulfate as a fractionating agent was assessed through the GPC data shown in Table III. There is very limited reduction in the molecular weight with reaction times longer than 1 h, confirming the results observed by Hsu et al.43 Also, the molecular weights are higher than those achieved with the nitrous acid approach, per mole of fractionating species, which may be due to the lack of specificity of the free radical degradation. This nonspecificity would also account for the higher polydispersity indices observed with the potassium persulfate fractionated glycol chitosan. Given that very little effect was observed for reaction times greater than an hour, varying amounts of the thermal free radical initiator were attempted in an effort to achieve a range of molecular weights. For an equivalent amount of glycol chitosan, the amount of potassium persulfate was varied from 0.03 to 0.5 mmol yielding a molecular weight profile as illustrated in Table IV. A rapid reduction in the molecular weight is seen initially with increasing po-

Average of four injections.

tassium persulfate; however, only a limited reduction in molecular weight is observed at higher concentrations, indicative of a saturation effect. Control of the molecular weight appears feasible with potassium persulfate up to this saturation point. Cleavage of the glycol chitosan backbone was achieved with both nitrous acid and potassium persulfate. Although nitrous acid was a more effective fractionating agent, it readily reacts with secondary amine groups to produce N-nitrosamines, which are potentially carcinogenic. Fractionation with potassium persulfate did not alter the structural integrity of glycol chitosan. Furthermore, tailoring of the molecular weight can be achieved by adjusting the molar ratio of potassium persulfate to glycol chitosan. In the only other report on the depolymerization of glycol chitosan published to date, glycol chitosan was depolymerized via acid hydrolysis using concentrated (4M) HCl at 508C.34 The depolymerization proceeded slowly in comparison to the use of persulfate, requiring 24 h to reduce the molecular weight to one-tenth the original molecular weight. This result is in agreement with the findings of Va˚rum et al.53 who examined the chemistry of acid hydrolysis of chitosan using HCl. Va˚rum et al. found that the rate of depolymerization decreased as the degree of N-deacetylation of the chitosan increased. TABLE IV Molecular Weight Profile of Glycol Chitosan Following its Fractionation with Potassium Persulfate for 2 h

Figure 8. Offset 1H NMR spectra of fractionated glycol chitosan via potassium persulfate under different reaction times.

KPS (mmol)

Mn (kDa)

Mw (kDa)

PI

0 3.0E-02 4.0E-02 7.5E-02 1.1E-01 1.4E-01 2.0E-01 3.5E-01 5.0E-01

175.4 88.0 64.2 49.2 37.8 29.6 20.2 13.6 10.7

231.4 127.7 112.2 78.0 72.6 57.6 38.0 30.3 26.1

1.3 1.5 2.2 1.6 2.1 2.0 1.9 2.2 2.4

Average of four injections. Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

796


Figure 9. Cytotoxicity of purified and potassium persulfate fractionated glycol chitosan towards chondrocytes. (*) denotes statistically significant difference from positive control and (**) denotes statistically significant difference with respect to purified and unfractionated glycol chitosan.

Cytotoxicity assay As glycol chitosan depolymerized with nitrous acid contained carcinogenic N-nitrosamines, only the cytotoxicity of potassium persulfate depolymerized glycol chitosan is reported. A chondrocyte cell line was chosen for this assay, because of the recent interest in the use of chitosan scaffolds in the engineering of articular cartilage.19,20,54–56 The glycol chitosan tested invoked no cytotoxic effects on the chondrocyte cells (Fig. 9). The measured absorbances at concentrations of 0.1 mg/mL all showed an improvement over the positive control. For the high molecular weight purified and unfractionated sample, this was particularly noticeable, and it is possible there is some stimulatory effect of the glycol chitosan on chondrocyte growth, at low concentrations. A similar stimulatory effect was noted by Howling et al. on human keratinocytes cultured with chitosan of high degrees of deacetylation.41 When the amount of glycol chitosan was increased to 0.5 mg/mL, this behavior did not appear, though absorbance levels were again close to the control, residing on the low side of the mean. At 1.0 mg/mL, there is a further slight reduction in compatibility with the cells, although all samples resembled each other regardless of their molecular weights. The glycol chitosan samples exhibited absorbances that were generally the same as, or near to, the positive control of cells in growth media alone, and these levels decreased with increased amounts of subJournal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

strate. At low concentrations of glycol chitosan, there may be a positive effect on cell growth, as evidenced by continual improvements over the positive control that was within statistical significance. This could be explained by a similarity in structure between chitosan and its derivatives to the glycosaminoglycan molecules (hyaluronan in particular) prevalent in cartilage tissues. It is possible that at sufficiently low concentrations, these molecules promote cell growth by mimicking the natural environment and signaling of chondrocytes within the extracellular matrix, while also not being so concentrated so as to have any disrupting effect on nutrient and waste transport within the media. No impact of molecular weight on sample toxicity was evident; an observation consistent with previous findings of the influence of chitosan molecular weight on cell viability, and in contrast to that reported by Carrenõ-Go´mez and Duncan. This is an important factor to consider, as one might regard the effect of lower molecular weight molecules of glycol chitosan being representative of the response to degradation products expected in vivo. Arranged in the order of decreasing molecular weight, as in Figure 9, the effect of molecule size on cytocompatibility is more or less random, and does not change with sample concentration. Upon closer statistical inspection of the data, a comparison of the highest molecular weight sample with the remaining ones does not elucidate any relationship. Further investigation of the effects of chitosan concentration in solution was


not possible, as at levels greater than 1.0 mg/mL the material was highly viscous and clear readings could not be obtained. It is possible that increased levels of chitosan caused greater complexing with either amino acids in the media or with the WST-1 reagent. The results of this assay must be assessed with caution, as there is large variability in much of the data, and, while quantitative numbers are generated, they cannot be truly extrapolated numerically to cellular response. Furthermore, the direct output of absorbance values is a further limitation on analysis, as values which might be considered outside of the normal distribution of data points cannot be explained. Nonetheless, there is clear evidence that glycol chitosan, and its lower molecular weight residues, has no harmful effect on the T/C-28a2 chondrocyte cell line.

CONCLUSIONS The water soluble derivative, glycol chitosan, was depolymerized by two separate methods, reducing its viscosity to facilitate its use in biomedical applications. Initial characterization of the purified glycol chitosan with 1H NMR spectroscopy demonstrated that some of the amine groups were glycolated indicating the presence of both secondary and tertiary amines. The presence of these N-glycolated groups had significant implications when depolymerization methods were examined. Depolymerization of glycol chitosan with nitrous acid at different pHs resulted in a reduction of molecular weight, specifically, from 170 to *7 kDa in the lower pH cases, and 25 kDa at a pH of 5.1. Unfortunately, the chemical structure of the glycol chitosan was altered, as the secondary amine groups were converted to potentially carcinogenic N-nitrosamines. Fractionation at pH 5.1 limited their formation, but not completely. Free radical degradation with the thermal dissociation initiator, potassium persulfate, was not as effective at reducing the initial molecular weight of the glycol chitosan per mole of depolymerization species. Potassium persulfate depolymerization yielded products of *12 kDa at reaction times from 0.5 to 24 h, but retained the structural character of the glycol chitosan. Moreover, control of the molecular weight could be achieved by simply adjusting the molar ratio of potassium persulfate to glycol chitosan. Cytotoxicity studies on the glycol chitosan showed cellular compatibility with immortalized chondrocytes, as assessed through 24-h cell growth and viability. Furthermore, no difference in cytotoxicity was observed with varying glycol chitosan molecular weight, which would indicate that the biocompatible nature of glycol chitosan is not lost via reduction in its molecular weight. These two

797

facts taken together imply that glycol chitosan shows promise as a biomaterial.

References 1. Le Dung P, Milas M, Rinaudo M, Desbrie`res J. Water-soluble derivatives obtained by controlled chemical modifications of chitosan. Carbohydr Polym 1994;24:209–214. 2. Rinaudo M, Milas M, Le Dung P. Characterization of chitosan—Influence of ionic-strength and degree of acetylation on chain expansion. Int J Biol Macromol 1993;15:281–285. 3. Chandy T, Sharma CP. Chitosan—As a biomaterial. Biomater Artif Cells Artif Organs 1990;18:1–24. 4. Rao SB, Sharma CP. Use of chitosan as a biomaterial: Studies on its safety and hemostatic potential. J Biomed Mater Res 1997;34:21–28. 5. Lee KY, Ha WS, Park WH. Blood compatibility and biodegradability of partially N-acylated chitosan derivatives. Biomaterials 1995;16:1211–1216. 6. Cho YW, Cho YN, Chung SH, Yoo G, Ko SW. Water-soluble chitin as a wound healing accelerator. Biomaterials 1999;20: 2139–2145. 7. Ueno H, Yamada H, Tanaka I, Kaba N, Matsuura M, Okumura M, Kadosawa T, Fujinaga T. Accelerating effects of chitosan for healing at early phase of experimental open wound in dogs. Biomaterials 1999;20:1407–1414. 8. Bae K, Jun EJ, Lee SM, Paik DI, Kim JB. Effect of watersoluble reduced chitosan on Streptococcus mutans, plaque regrowth and biofilm vitality. Clin Oral Investig 2006;10:102– 107. 9. Mi FL, Yu SH, Peng CK, Sung HW, Shyu SS, Liang HF, Huang MF, Wang CC. Synthesis and characterization of a novel glycoconjugated macromolecule. Polymer 2006;47:4348– 4358. 10. Liu N, Chen XG, Park HJ, Liu CG, Liu CS, Meng XH, Yu LJ. Effect of MW and concentration of chitosan on antibacterial activity of Escherichia coli. Carbohydr Polym 2006;64:60–65. 11. Shi ZL, Neoh KG, Kang ET, Wang W. Antibacterial and mechanical properties of bone cement impregnated with chitosan nanoparticles. Biomaterials 2006;27:2440–2449. 12. Sechriest VF, Miao YJ, Niyibizi C, Westerhausen-Larson A, Matthew HW, Evans CH, Fu FH, Suh JK. GAG-augmented polysaccharide hydrogel: A novel biocompatible and biodegradable material to support chondrogenesis. J Biomed Mater Res 1999;49:534–541. 13. Lahiji A, Sohrabi A, Hungerford DS, Frondoza CG. Chitosan supports the expression of extracellular matrix proteins in human osteoblasts and chondrocytes. J Biomed Mater Res 2000;51:586–595. 14. Chen RN, Wang GM, Chen CH, Ho HO, Sheu MT. Development of N,O-(carboxymethyl)chitosan/collagen matrixes as a wound dressing. Biomacromolecules 2006;7:1058–1064. 15. Azad AK, Sermsintham N, Chandrkrachang S, Stevens WF. Chitosan membrane as a wound-healing dressing: Characterization and clinical application. J Biomed Mater Res B 2004;69: 216–222. 16. Adekogbe I, Ghanem A. Fabrication and characterization of DTBP-crosslinked chitosan scaffolds for skin tissue engineering. Biomaterials 2005;26:7241–7250. 17. Ma L, Gao C, Mao Z, Zhou J, Shen J, Hu X, Han C. Collagen/chitosan porous scaffolds with improved biostability for skin tissue engineering. Biomaterials 2003;24:4833–4841. 18. Duplan-Perrat F, Damour O, Montrocher C, Peyrol S, Grenier G, Jacob MP, Braye F. Keratinocytes influence the maturation and organization of the elastin network in a skin equivalent. J Invest Dermatol 2000;114:365–370.

Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

798

19. Suh JKF, Matthew HWT. Application of chitosan-based polysaccharide biomaterials in cartilage tissue engineering: A review. Biomaterials 2000;21:2589–2598. 20. Lu JX, Prudhommeaux F, Meunier A, Sedel L, Guillemin G. Effects of chitosan on rat knee cartilages. Biomaterials 1999; 20:1937–1944. 21. Cui YL, Qi AD, Liu WG, Wang XH, Wang H, Ma DM, Yao KD. Biomimetic surface modification of poly(L-lactic acid) with chitosan and its effects on articular chondrocytes in vitro. Biomaterials 2003;24:3859–3868. 22. Hoemann CD, Hurtig M, Rossomacha E, Sun J, Chevrier A, Shive MS, Buschmann MD. Chitosan-glycerol phosphate/ blood implants improve hyaline cartilage repair in ovine microfracture defects. J Bone Joint Surg Am 2005;87:2671– 2686. 23. Lee JY, Nam SH, Im SY, Park YJ, Lee YM, Seol YJ, Chung CP, Lee SJ. Enhanced bone formation by controlled growth factor delivery from chitosan-based biomaterials. J Controlled Release 2002;78:187–197. 24. Park YJ, Lee YM, Lee JY, Seol YJ, Chung CP, Lee SJ. Controlled release of platelet-derived growth factor-BB from chondroitin sulfate-chitosan sponge for guided bone regeneration. J Controlled Release 2000;67:385–394. 25. Liao IC, Wan ACA, Yim EKF, Leong KW. Controlled release from fibers of polyelectrolyte complexes. J Controlled Release 2005;104:347–358. 26. Ishihara M, Obara K, Ishizuka T, Fujita M, Sato M, Masuoka K, Saito Y, Yura H, Matsui T, Hattori H, Kikuchi M, Kurita A. Controlled release of fibroblast growth factors and heparin from photocrosslinked chitosan hydrogels and subsequent effect on in vivo vascularization. J Biomed Mater Res A 2003; 64:551–559. 27. Va˚rum KM, Ottøy MH, Smidsrød O. Water-solubility of partially N-acetylated chitosans as a function of pH—Effect of chemical-composition and depolymerisation. Carbohydr Polym 1994;25:65–70. 28. Kubota N, Tatsumoto N, Sano T, Toya K. A simple preparation of half N-acetylated chitosan highly soluble in water and aqueous organic solvents. Carbohydr Res 2000;324:268–274. 29. Muzzarelli RAA, Lough C, Emanuelli M. The molecular weight of chitosans studied by laser light-scattering. Carbohydr Res 1987;164:433–442. 30. Sato M, Onishi H, Kitano M, Machida Y, Nagai T. Preparation and drug release characteristics of the conjugates of mitomycin C with glycol-chitosan and N-succinyl-chitosan. Biol Pharm Bull 1996;19:241–245. 31. Sugimoto M, Morimoto M, Sashiwa H, Saimoto H, Shigemasa Y. Preparation and characterization of water-soluble chitin and chitosan derivatives. Carbohydr Polym 1998;36:49–59. 32. Carrenõ-Go´mez B, Duncan R. Evaluation of the biological properties of soluble chitosan and chitosan microspheres. Int J Pharm 1997;148:231–240. 33. Kwon S, Park JH, Chung H, Kwon IC, Jeong SY, Kim IS. Physicochemical characteristics of self-assembled nanoparticles based on glycol chitosan bearing 5 b-cholanic acid. Langmuir 2003;19:10188–10193. 34. Wang W, McConaghy AM, Tetley L, Uchegbu IF. Controls on polymer molecular weight may be used to control the size of palmitoyl glycol chitosan polymeric vesicles. Langmuir 2001;17:631–636. 35. Shechter L, Wynstra J, Kurkjy RP. Glycidyl ether reactions with amines. Ind Eng Chem 1956;48:94–97. 36. Richardson SCW, Kolbe HVJ, Duncan R. Potential of low molecular mass chitosan as a DNA delivery system: Biocompati-

Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a


37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48. 49.

50. 51.

52.

53. 54.

55.

56.

bility, body distribution and ability to complex and protect DNA. Int J Pharm 1999;178:231–243. Lee M, Nah JW, Kwon Y, Koh JJ, Ko KS, Kim SW. Watersoluble and low molecular weight chitosan-based plasmid DNA delivery. Pharm Res 2001;18:427–431. Sato T, Ishii T, Okahata Y. In vitro gene delivery mediated by chitosan. Effect of pH, serum, and molecular mass of chitosan on the transfection efficiency. Biomaterials 2001;22:2075–2080. Mao S, Shuai X, Unger F, Simon M, Bi D, Kissel T. The depolymerization of chitosan: Effects on physicochemical and biological properties. Int J Pharm 2004;281:45–54. Huang M, Khor E, Lim LY. Uptake and cytotoxicity of chitosan molecules and nanoparticles: Effects of molecular weight and degree of deacetylation. Pharm Res 2004;21:344–353. Howling GI, Dettmar PW, Goddard PA, Hampson FC, Dornish M, Wood EJ. The effect of chitin and chitosan on the proliferation of human skin fibroblasts and keratinocytes in vitro. Biomaterials 2001;22:2959–2966. Allan GG, Peyron M. Molecular weight manipulation of chitosan. I. Kinetics of depolymerization by nitrous acid. Carbohydr Res 1995;277:257–272. Hsu SC, Don TM, Chiu WY. Free radical degradation of chitosan with potassium persulfate. Polym Degrad Stab 2002;75: 73–83. Allan GG, Peyron M. Molecular weight manipulation of chitosan. II. Prediction and control of extent of depolymerization by nitrous acid. Carbohydr Res 1995;277:273–282. Lavertu M, Xia Z, Serreqi AN, Berrada M, Rodrigues A, Wang D, Buschmann MD, Gupta A. A validated H-1 NMR method for the determination of the degree of deacetylation of chitosan. J Pharm Biomed Anal 2003;32:1149–1158. Kubota N, Eguchi Y. Facile preparation of water-soluble Nacetylated chitosan and molecular weight dependence of its water-solubility. Polymer J 1997;29:123–127. Goycoolea FM, Heras A, Aranaz I, Galed G, Fernańdez-Valle ME, Arguëlles-Monal W. Effect of chemical crosslinking on the swelling and shrinking properties of thermal and pH-responsive chitosan hydrogels. Macromol Biosci 2003;3:612–619. Amiji MM. Pyrene fluorescence study of chitosan self-association in aqueous solution. Carbohydr Polym 1995;26:211–213. Hu Y, Jiang H, Xu C, Wang Y, Zhu K. Preparation and characterization of poly(ethylene glycol)-g-chitosan with waterand organosolubility. Carbohydr Polym 2005;61:472–479. Digenis GA, Issidorides CH. Some biochemical aspects of Nnitroso compounds. Bioorg Chem 1979;8:97–137. Lijinsky W, Garcia H, Keefer L, Loo J, Ross AE. Carcinogenesis and alkylation of rat liver nucleic acids by nitrosomethylurea and nitrosoethylurea administered by intraportal injection. Cancer Res 1972;32:893–897. Lijinsky W, Keefer L, Conrad E, Van de Bogart R. Nitrosation of tertiary amines and some biologic implications. J Natl Cancer Inst 1972;49:1239–1249. Va˚rum KM, Ottøy MH, Smidsrød O. Acid hydrolysis of chitosans. Carbohydr Polym 2001;46:89–98. ¨ zogul C, DenkSenkoÿlu¨ A, Simsek A, Sahin FI, Menevse S, O bas EB, Piskin E. Interaction of cultured chondrocytes with chitosan scaffold. J Bioact Compat Polym 2001;16:136–144. Di Martino A, Sittinger M, Risbud MV. Chitosan: A versatile biopolymer for orthopaedic tissue-engineering. Biomaterials 2005;26:5983–5990. Montembault A, Tahiri K, Korwin-Zmijowska C, Chevalier X, Corvol MT, Domard A. A material decoy of biological media based on chitosan physical hydrogels: Application to cartilage tissue engineering. Biochimie 2006;88:551–564.