Depolymerization reactions of hyaluronic acid in solution

Depolymerization reactions of hyaluronic acid in solution M i l e n a Reht, kovti*, Du~,an Bako~,, M a r o t Soldtm and Katarina Viz,qrov8 Faculty of Chemical Technology, Slovak Technical University, Department of Printing Technology and Applied Photochemistry, Radlinskdho 9, 812 37 Bratislava, Slovakia

Received 10 September 1993; revised 30 January 1994 Samples of microbial sodium hyaluronate were degraded by heating, ultrasonication ultraviolet (UV) and 7-ray irradiation and enzymatic treatment. The weight-average molecular weight, /l~w, of hyaluronate in 0.15 M NaCI and 0.06 M Na2HPO4 was determined by gel filtration with UV detection. The /l~w of the degraded samples varied from 8 x 1 0 4 to 1.38×106. Depolymerization processes can be described by linear relationship (1//~w) 2 =f(t) in the case of ultrasonic treatment and by non-linear relationships in the cases of heating and UV irradiation at 257 nm. Gamma-ray irradiation and enzymatic treatment caused chemical degradation and depolymerization to oligosaccharides, respectively.

Keywords: sodium hyaluronate; gel filtration; depolymerization

Hyaluronic acid (HA) [(l~3)-O-(2-acetamido-2-deoxy/~-D-glucopyranosyl)-(1 -.4)-O-~-D-glucopyranuronosyl], (Figure 1) is a natural polysaccharide belonging to the class of non-sulfated glycosaminoglycans. HA is an important component of the extracellular matrix, synovial fluid and tissues, cartilage, umbilical cord and vitreous humour of the eye, in which its concentration varies greatly with source and age. It plays an important role in both mechanical and transport systems in the body. Solutions of sodium hyaluronate (NaHA) form hydrated polymeric matrices which display a combination of coherence, elasticity, viscosity and pseudoplasticity at low concentrations ~. These solutions have unusual rheological properties due to the stiffness of the long chain and their polyelectrolyte character z. HA may be obtained from some natural tissue, e.g. by extraction of cocks' combs 3'4, or may be produced by micro-organisms 5-7. The molecular weight and purity of HA depend on the means of production. Generally, samples with higher relative molecular weight, in the range of 1-8 x 106, can be obtained by extraction. Solutions of this type of HA are suitable as a therapeutic agent in ophthalmic surgery ('viscosurgery')8 and arthritis 9. HA produced by microbiological methods, as well as fractions with a lower molecular weight after the common purification and/or degradation procedures (0.1-1 x 106), have been used in wound repair and for cosmetological applications l°'t ~. The molecular chain of the polysaccharide can be depolymerized quite easily under the influence of various physical and chemical treatments, such * To whom correspondence should be addressed

0141-8130/94/030121-04 © 1994Butterworth-HeinemannLimited

as mechanical influence, pressure, heating, radiation, oxidation and enzymatic hydrolysis. This is useful for controlled depolymerization of original high-molecularweight HA substances. This paper shows the results of a depolymerization study in a particular microbial HA and compares the influence of various depolymerization agents on relative molecular weight and molecular weight distribution, We have used size-exclusion chromatography, which is an excellent method for determination of molecular weight distribution, as well as individually defined averages of the molecular weight of hyaluronan in solution ~2-14.

Experimental Materials The following materials were used: • sodium hyaluronate isolated from rooster combs (Movis, Holi~, Slovakia) • sodium hyaluronate produced by Streptococcus zooepidermicus(Contipro, 13sti n. Orlici, Czech Republic)

~H=OH

ioo o R ~H~~ H

H

O

O

O H

H Figure 1

NHCOCH 3

OH

Structural formula

of HA

Int. J. Biol. Macmmol. Volume 16 Number 3 1994 121

Depolymerization reactions of hyaluronic acid in solution: M. Rehdkovd et am. • sodium hyaluronate (Meiji Seika Kaisha Ltd, Japan) as a reference material with A3, = 1.75 x 106 • protein standards with relative molecular weight / ~ , = 1.58 x 105, 2.32 x 105, 4.4 x 105, 6.69 x 105 and 2 x 106, used for calibration of the chromatographic system • bovine testicular hyaluronidase type I-S, 290 IU mg- 1 specific activity (purchased from Sigma, St Louis, Missouri, USA) • sodium chloride (NaC1) and sodium hydrogen phosphate (Na2HPO4) (Lachema, Brno, Czech Republic), of analytical grade.

samples were calculated using the equation: g.v =(Vo- Vo)/(V, - Vo)

where K=v is the coefficient of availability, V, is the elution volume for the sample, Vo is the column void volume (equal to the elution volume for Blue Dextran 2000) and Vt is the total bed volume. The K=v value (on a linear scale) was plotted against the corresponding molecular weight (on a logarithmic scale) (Figure 2) and the straight line which best fitted the points on the graph was drawn. The relationship between molecular weight and K,v is expressed as: K,, = 0.49 - 3.6 × 10 -~ A]',

Instruments The experiments were performed using liquid chromatograph fast-protein liquid chromatography (FPLC; Pharmacia, Uppsala, Sweden) involving of a peristaltic pump P-500, a monitor UV-M, an LCC-500 PLUS system controller, and a 100/A sample loop. Analyses were carried out at room temperature using a Superose 6 (10 x 300 mm) column, and the ultraviolet (UV) detector was set at a wavelength of 214nm. The mobile phase consisted of 0.15 M NaCI and 0.06 M Na2HPO4 aqueous solutions in a 1:1 ratio. The flow rate of the eluent was 0.2 ml min- 1. All solutions were filtered through cellulose acetate filters, pore size 0.2/tm (Pharmacia). The samples were dissolved in the mobile phase and injected as 0.05% (w/v) solutions. The protein standards with the parameters described in Table 1 were used for calibration of the FPLC system. Samples dissolved in the mobile phase were injected at concentrations of: Blue Dextran, 1 m g m l - t ; aldolase, 2 mg ml- 1; catalase, 5 mg ml- t; ferritin, 0.5 mg ml- 1; thyroglobulin, 2 mg ml-1. K,, values for each of the

(1)

(2)

Hence, by substitution of Vo = 6.668 ml and Vt = 24 ml: Mw=2.445 × 106-0.16 x l06 lie

(3)

The linearity of equation (3) is guaranteed in the range of A~, from 8 x 104 to 2 x 10 6. The UV/visual spectra of hyaluronate solutions were recorded using a Philips PU-8800 instrument in the range of 190--400 nm.

Depolymerization by heating The solution of sodium hyaluronate, 0.05% w/v, was heated at 60, 70, 80 or 90°C for 1 h. The samples (1 ml) were cooled to room temperature and analysed.

Ultrasonic depolymerization Samples (5 ml) of NaHA solution (0.05% w/v) were sonicated at 25°C using a UGA 206 13 ultrasonic generator (20 kHz, 800 W; Tesla, Czechoslovakia). Samples (1 ml) were withdrawn at defined time periods (15, 30, 45 and 60s) and their molecular characteristics were determined.

Table 1 Molecularcharacteristicsof the proteinstandards Standard

Mw

Stokes' radius (/am)

AIdolase Catalase Ferritin " Thyroglobulin

158000 232000 440000 669000

4.81 5.22 6.10 8.50

U V-exposure depolymerization Samples (2ml) of 0.05% (w/v) NaHA solution were exposed using an RWL middle-pressure mercury discharge tube (400 W) with emission lines at 297, 305, 335, 368, 409, 441, 545 and 580 nm (Applied Photophysics, UK) for 0.5, 1, 2, 3 or 4 h, or a monochromatic UV lamp with an emission line at 257 nm for the same durations.

Gamma-ray depolymerization

0.5

Solutions of NaHA sealed in tubes were exposed using RCH-GAMA-30 (radiator = 6°Co) with doses of 6, 12, 18 and 24 kGy.

Enzymatic depolymerization 0.4

Enzymatic reactions were carried out in phosphate buffer (pH 5.0) containing 0.15 M sodium chloride at a constant temperature of 37°C. Each reaction tube contained 3.3 mg NaHA and 0.2 mg bovine testicular hyaluronidase (58 IU) in a total volume of 1 ml buffer. The enzymatic reactions were stopped by heating at 100°C for 5 min. The tubes were cooled in an ice-bath for 10 rain and centrifuged at 2500 rev min- 1 for 5 min. The supematant was filtered and injected into the gel filtration system.

age

0.3

thyro~lobulin •

Results and discussion

0.2 100000

10oooo0 M,,

Figure 2 Calibration of the Supcrosc6 columnwith proteinstandards

122

Int. J. Biol. Macromol. Volume 16 Number 3 1994

The molecular weights and UV spectra of both types of NaHA sample (i.e. prepared by extraction (NaHA-E1, NaHA-E2) and microbially produced (NaHA-M) were

Depolymerization reactions of hyaluronic acid in solution: M. Rehdkovd et al. compared before treatment to allow comparison of our results with the previous degradation study is. The molecular weight of sample E (1.42 x 106) was practically equal to that of sample M (1.40 x 106) but the polydispersity of sample M was lower. Moreover, impurities of protein and/or nucleic acids at 258 nm were observed in the UV spectrum of sample E (Figure 3). As these impurities can influence degradation reactions, we decided to study the depolymerization of the microbial hyaluronic acid instead. It seems that the higher purity of microbial HA could be more appropriate for use in medical and pharmacological applications. Therefore, from a practical viewpoint, the degradation studies are useful in determining conditions for processing, e.g. sterilization. Treatment at a higher temperature is the most common degradation process. Treatment at temperatures from 60 to 90°C for 1 h results in only mild depolymerization, which corresponds to a decrease in molecular weight to about 2 x l0 S and a small increase in polydispersity at 90°C. A considerable fraction with a /~,, of 8 x 105 was also observed. The decrease in molecular weight and changes of the polydispersity (D) with increasing temperature can be seen in Table 2, which shows the effects of different treatments on depolymerization of HA. Ultrasound treatment was performed over 15 to 60 s before analysing the samples. Relatively short times were used to complement the results in reference 15, where authors using longer treatment times found a linear decrease of molecular weight with time. The decrease in M,~ is rather low but polydispersity increases (Table 2). The dependence of molecular weight on the time of degradative treatment can be expressed by the relationship (1//~'w) 2 -f(t) 16. In agreement with previous results 15, our results confirmed the linearity in the case of ultrasonic treatment at shorter lengths of time (Figure 4). The relationship is expressed as: (.h~o/.~t,,)2 - 1 = 0.00287 + 0.0035t

(4)

where -~o is the weight-average molecular weight of

Influencesof various treatments on the depolymerizationof hyaluronic acid Table 2

Mwx 10-6

D

1.40 1.26 1.28 1.25 1.18

1.001 1.008 1.004 1.004 1.016

1.40 1.38 1.37 1.35 1.29 1.15

1.001 1.005 1.008 1.025 1.046 1.052

1.40 1.36 1.33 1.30 1.27

1.001 1.004 1.012 1.030 1.028

Temperature (°C)

25 60 70 80 90 Duration of irradiation at 257 nm (h)

0 0.5 l 2 3 4

Duration of sonication (s)

0 15 30 45 60

Duration of enzyme treatment (h)

0 1 3 6 24 30

1.40 1.001 0.08 1.013 Oligosaccharides Oligosaccharides Oligosaccharides Oligosaccharides Chemical degradation Chemical degradation

UV irradiation (Hg, 400 W) •3,-rays (6 kGy)

0.2

7 I~,_. 0.1

1.5 - HA-M ......... H A - E 1 ..... HA-E2

1.0 o.o

¢J

zb

'

time

,a

'

of

4'0

sonieation,

60 s

Figure 4 Degradation of NaHA by ultrasound: the inverse value of (./~,,)2 is plotted as a function of time

o

II

a

,i b~'

o.5

0.0

. . . . 190

/

,

,,.,

. . . .

240 wavelength,

, -;' 290 nm

,

,'7 ,

,-840

Figure3 UV spectraof hyaluronatesamples:HA-M is microbiologically produced HA; HA-E1 is the sample obtained by extraction of cock combs, unpurified; HA-E2 is the sample obtained by extraction of cock combs, purified

original NaHA and ~ , , is the weight-average molecular weight of hyaluronate fractions after treatment. UV irradiation often results in photochemical reactions of HA in solutions and/or in the solid state 1~. The exposure of the N a H A solution to the large-spectrum Hg lamp caused degradation of the origin sample to other products which were unidentified by the chromatographic Superose 6 column at short treatment times. These products come from radical cleavage and consequent radical recombination t8.19. On the other

Int. J. Biol. M a c r o m o l .

Volume 16 Number 3 1994

123

Depolymerization reactions of hyaluronic acid in solution: M. Reh6kov6 et al.

--OkGy ....... 6 kGy, powder . . . . 6 kGy, solution

a~ M 0

.................

c¢

0

..........._

12

elution volume, m l

Figure 5 The influenceof ?-rays on the chromatograms of N a H A samples

El a-Oh b- 1 h

d II ft

e-6h

d - 24h

cl

II It

¢.2

e

i I/',

t..

/ - \

O ,',~7

ca

0

"" "'" . . . . . .-. .-.- x

References

l~b

1

. <- ' f ' ~~q / C /l~

12 elution volume, ml

and 1440min. Fractions with Atw corresponding to oligosaccharides of HA were observed after only I h of contact with the enzyme. The amount of these fractions increases with the duration of enzymatic treatment. The elution volume decreases moderately with the duration of enzymatic reaction. This may be caused by condensation reactions between oligosaccharides units to give products with higher Mw. Different methods of treatment can be used for HA depolymerization according to the requirement for a controlled depolymerization or preparation of HA fractions with a desired AIw. Temperature and ultrasonic treatments for short times allow mild depolymerization, and they are the most appropriate for preparation of controlled fractions. Irradiation by UV or ~,-rays causes a profound decrease in AIw, creating new products. These means of degradation cannot be used for controlled depolymerization of HA. The enzymatic method has advantages when fractions with a large range of AI, are required. This may be useful for pharmacology, dermatology and medical research 2°.

24

Figure 6 Chromatogramsof original NaHA and fractionsprepared by enzymatictreatment

2 3 4 5 6 7

hand, it is possible to control this kind of depolymerization by irradiation at a chosen wavelength. HA samples were exposed at 257 nm for different lengths of time. Considerable changes in M , were observed after 4 h of irradiation, but polydispersity increased after a shorter treatment time (1 h). For the effects of heating and UV irradiation at 257 nm, the relationships (1/~r,)2 =f(t) are not linear. The degradation of NaHA is slow at lower temperatures and shorter exposure times. Then the course of degradation is accelerated. The depolymerization processes can be described by polynomial relationships. In Figure 5, we can see the influence of ~,-ray irradiation at the lowest dose necessary for medical sterilization. The distributions of molecular weight of the solution NaHA and NaHA in the solid state after treatment by 7-rays at the same dose are comparable. The distribution shows a profound degradation of HA; nevertheless, some high-molecular-weight products of radical recombination are observed (similar to the distribution after irradiation with the large-spectrum Hg lamp). The chromatograms in Figure 6 illustrate the effects of hyaluronidase digestion of NaHA for between 60

124

Int. J. Biol. Macromol. Volume 16 Number 3 1994

8 9 10 11 12 13 14 15 16 17 18 19 20

Saettone, M.F., Giannaccini, B., Chetoni, P., Torracea, M.T. and Monti, D. Int. J. Pharmacol. 1991, 72, 131 Cleland, R. L. Biopolymers 1968, 6, 1519 Galatik, A., Kul~na, K. and Bla~j, A. CS Patent 264,719 (C.A. 114, P88722), 1989 Chabre~k, P., Spit&, L., Hradec, H., Filip, J. and Orvisk~,, E. Collect. Czech. Chem. Comraun. 1992, 57, 2151 Akasaka, H., Seto, S., Yanagi, M., Fukushima, S. and Mitsui, T. J. Soc. Cosmet. Chem. Jpn 1988, 22, 35 Balazs, E.A. US Patent 4,141,973 (C.A. 90, 174663), 1979 Kjeims, E. and Lebech, K. Acta Pathol. Microbiol. Scand. 1976, 883, 1962 Balazs, E.A. in 'Heaion (sodium hyaluronate) - A Guide to Its Use in Ophthalmic Surgery' (Eds D. Miller and R. Stegman), Wiley, New York, 1983, p 5 Leardini, G., Mattara, L., Franceschini, M. and Perbellini, A. Clin. Exp. Rheumatol. 1991, 9, 375 Nagai, M., Nosizaki. S. and Kanb¢, N. JP Patent 158,203 (C.A. 107, 204949), 1986 Cortivo, R., Brun, P., Rastrelli, A. and Abatangelo, G. Biomaterials 1991, 12(8), 727 Brun, P., DeGalateo, A., Camporese, A., Cortivo, R. and Abatangelo, G. J. Chromato#r. 1990, 526, 530 Fedarko, N.S., Termine, J.D. and Robey, P.G. Anal. Biochem. 1990, 188, 398 Beaty, N.B., Tew, W.P. and Mello, R.J. Anal. Biochem. 1985, 147, 387 Chabreeek, P., ~;olt6s, L. and Orvisk~,, E. d. Appl. Polym. Sci. Appl. Polym. Syrup. 1991, 48, 233 Tikhonova, Z.A., Semchikov, D. and Trubina, I.V. Vysokomolek. Sped. 1987, 29, 2392 Minotti, G. and Aust, S.D. Chem. Biol. Interact. 1989, 71, 1 Lap~:ik,L'. Jr, Omelka, L., Kub~na, K., Galatik, A. and KelI6, V. Gen. Physiol. Biophys. 1990, 9, 419 Deeble, D.J., Phillips, G.O., Botke, E., Schuchmann, H.P. and Von Sonntag, C. Radiat. Phys. Chem. 199t, 37, 115 Ouchi, T., Banba, T., Huang, T.Z. and Ohya, Y. Polym. Prep. 1990, 31(2), 262