ELSEVIER
`
`0141-3910(95)00041-0
`
`48 (I 995) 269-273
`
`Polymer Degradacion and Stability
`
`© 1995 Elsevier
`
`Science Limited
`
`
`
`Printed in Northern Ireland. All rights reserved
`
`0141-3910/95 /$09.50
`
`
`
`Hydrolytic degradation of hyaluronic acid
`
`
`
`
`
`Y.Tokita & A. Okamoto
`
`
`
`
`
`Research Center, Denki Kagaku Ko?,yo Co. Ltd., 3-5-1 Asahimachi, Machida, Tokyo 194, Japan
`
`
`
`
`
`
`
`
`
`(Received 24 January 1995; accepted 8 February 1995)
`
`
`
`
`
`The hydrolytic degradation of hyaluronic acid (HA) was investigated through
`
`
`
`
`
`
`
`
`
`kinetic measurements. The first-order rate constants were obtained under
`
`
`
`
`different pH conditions on the basis of the decrease in the molecular weight.
`
`
`
`
`The mechanism of the hydrolytic degradation of HA was also investigated by
`
`
`
`theoretical calculations (MNDO-MO) on model compounds of HA.
`
`
`
`
`It was speculated that hydrolysis occurs in acid solution on the glucuronic
`
`
`
`
`acid residue and the hemiacetal ring remains, while the destruction of the
`
`
`N-acetylglucosamine residue takes place in basic solution.
`
`
`
`
`These speculations are consistent with the structurnl analysis of the
`
`
`degradation products by 13C-NMR.
`
`1.INTRODUCTION
`
`2.1. Materials and methods
`
`NMR structure analysis of degradation products,
`
`
`
`
`
`we have predicted the reaction sites and
`
`mechanisms of the hydrolysis.
`
`
`Hyaluronic acid (HA) is an unbranched high
`
`
`
`molecular weight polysaccharide. It consists of
`
`
`2-acetamide-2-deoxy-,B-D-glucose and /3-D­
`
`
`glucuronic acid residues linked 1-3 and 1-4,
`2.EXPERIMENTAL
`
`
`alternately. It is widely distributed in the
`
`
`
`connecting tissuest of the human body as well as
`
`
`
`2 One of the interesting features is
`in bacteria.
`High molecular weight sodium hyaluronate
`
`
`
`that HA has an extremely high molecular weight
`3 Aqueous solutions
`
`
`
`samples were extracted from the culture broth of
`among glucosaminoglycans.
`
`
`by Denki Kagaku Streptococus equi and purified
`
`of HA show very high viscosity due to its high
`
`Kogyo Co. Ltd. One weight percent aqueous
`molecular weight.
`
`
`
`solutions of HA were prepared by stirring for 3 h
`
`
`On the other hand, taking advantage of some
`
`
`
`
`
`
`
`
`useful features (keeping humidity, high viscosity to dissolve completely. The HA solution was
`
`
`
`and the nature of decomposition etc.), there have
`
`divided into equal parts and sealed in 1 ml
`
`been developments of new drugs using HA.
`
`
`
`ampoules followed by heating in an oil bath at
`
`
`
`constant temperature. The number-average mol­
`
`
`Therefore, the basic investigation of HA
`
`
`ecular weight was determined by gel-permeation
`
`
`
`degradation is very important including the
`
`
`chromatography (column: Shodex OHpak SB806
`
`
`
`kinetics of degradation, degradation product
`
`X2, solvent: 0.1M NaNO3 aq, flow rate:
`
`
`
`analysis and reaction mechanism. It has been
`
`
`1.2 ml/min, detector: J ASCO830RI, column
`
`
`reported that the degradation of HA is caused by
`
`
`
`with standard HA temperature: 40°C, calibrated
`4 and active oxygen in the form of
`hydrolysis
`
`hydroxyl radicals.
`
`
`
`
`s;imples of different intrinsic viscosity). The pH
`
`5 7 Recently, the energy barrier
`
`
`conditions were controlled by a wide buffer9
`
`
`
`of glycoside bond cleavage with hyaluronane in
`
`
`
`acidic condition has been calculated by Pratt et
`
`
`which was prepared as follows: A-Solution
`8 In the
`al. by means of the MNDO-MO method.
`
`(tris-(hydroxylmethyl)-aminomethane 0.1M, KCI
`
`
`
`0.lM, potassium phosphate monobasic 0.lM,
`
`
`
`present study, we have investigated the rate of
`
`
`
`pH HA hydrolytic degradation under different
`
`
`citric acid anhydrous 0.lM, and sodium tetrabor­
`ate 0.lM); B-Solution (HCI aq 0.4M or NaOH aq
`
`
`
`conditions. Furthermore, on the ba_sis of quantum
`
`chemical calculations (MNDO-MO)10 and 13C-
`
`
`0.4M). pH solutions were prepared by adding the
`
`
`269
`
`ALL 2038
`PROLLENIUM V. ALLERGAN
`IPR2019-01505 et al.
`
`

`

`270
`
`
`
`Y.Tokitu, A. Okumoto
`
`
`
`
`
`Integrating eqn ( 4), we obtain
`
`A-Solution diluted 4 times to the appropriate
`
`
`
`amount of B-Solution. The samples for 11C-NMR
`
`
`
`analysis of degradation products were prepared
`
`as follows: the acidic sample (pH 1.44) was
`According to eqn (5), l/[DP]n-1/[DP]no plotted
`
`
`
`
`prepared using aqueous HCl solution and heated
`
`
`against t gives a straight line as shown in Fig. 1.
`at 60°C for 10 days, the basic sample (pH 13.30)
`
`The values of the first-order rate constant
`kh
`
`
`was prepared using NaOII solution and heated at
`
`
`were evaluated from the slopes of the straight
`60°C for 7 days.
`
`
`
`
`lines obtained under different conditions. These
`
`
`linear plots show that the degradation reaction of
`
`
`2.2. Rate constant of the hydrolysis reactions of
`
`HA is due to random chain scission and
`
`
`accordingly obeys first-order kinetics.
`
`1/[DP],, -1/[DP],,0 = kh · t
`
`(5)
`
`
`HA
`
`The rate of the hydrolytic degradation of HA in
`
`
`
`
`aqueous solution was followed by the measure­
`
`
`
`ment of the number-average molecular weight.
`
`
`Assuming the rate of the main chain scission is
`13C normal and DEPT11•12 spectra were measured
`
`
`
`proportional to the number of glycoside bonds
`on a modified JNM-ALPHA 500 (JEOL).
`
`(first-order), we obtain
`
`2.3. NMR measurement
`
`(1)
`2.4. Calculation
`
`Nb: number of main chain scissions, M0: total
`
`number of monomer units, P: number of polymer
`The semi-empirical SCF MO method, MNDO
`
`
`molecules, I' = I'0 I Nb (I'0: number of polymers
`
`was employed using the MOPAC 5.01 package."
`at t = 0), M0 - P: number of glycosidc
`bonds, kh:
`
`
`first-order rate constant of main chain scission.
`
`The number of main chain scissions at time t is
`
`expressed by the number average degree of
`3.RESULTS AND DISCUSSION
`zation [DP]" at time t and [DP]no at
`polymeri
`t =0.
`
`The rate of the hydrolytic degradation of HA in
`
`
`Nb= M0/[DP],, - M0/[DP],,
`(2)
`
`
`
`aqueous solution is strongly dependent on pH.
`of kt, at 40°C
`Table 1 shows the pH dependence
`From eqns (1) and (2):
`
`and 60°C. It is clear that HA is most stable at pH
`d(l/[DP],,)/(1 - 1/[DP],,) =kb· dt (3)
`
`
`
`values around neutrality and more labile in acidic
`
`
`
`conditions than basic conditions. In addition, HA
`Further, 1 « /[DP],, for HA with high molecu­
`
`is less stable at higher temperature.
`lar weight,
`
`
`
`For the theoretical molecular orbital calcula­
`(4)
`
`tion of the HA hydrolysis, two model compounds
`were chosen as shown in Fig. 2. In order to
`
`
`reduce end-group effects, we focused only upon
`
`
`
`the central residue, a glucuronic acid residue in
`
`the model A and a N-acethylglucosamine residue
`de bonds are
`
`in model B, and terminal glycosi
`
`substituted by methoxy groups in the model
`
`0
`
`•
`
`6
`
`�
`
`5
`
`4
`
`•
`
`3
`
`0
`
`-1
`0.0 0.2 0.4 0.6 0.8
`Time (h)
`plot al pH 11. 80°C.
`Fig. 1. 1/(DP]n·l/[DP]no vs time (h)
`
`
`
`
`
`Table 1. Rate constant (kb) VS pH at 40°C and 60°C
`
`Temperature
`(degree)
`pH3
`
`----·-··-------
`
`k, (s ')
`
`pH7
`
`pHll
`
`40
`
`bU
`
`.S.OX 10 10 :'i.9 X 11 II 'i.3X 10-'"
`"
`10 7.6 X I()
`b.l X 10 g 2.5 X 10
`
`

`

`
`
`
`
`Hydrolytic degradation of hyaluronic acid
`
`271
`
`A
`
`
`
`
`
`3.1. Acidic hydrolysis of hyaluronic acid
`
`The results of the calculation clearly show that
`
`
`
`
`the electron density of LUMO on the glucuronic
`
`acid (HA units) is localized and that the
`
`
`hydrolytically active sites are on the Cl, C4 and
`
`
`
`
`carbonyl carbons. Among these, hydrolysis at Cl
`
`
`or C4 contributes to chain scission. In addition,
`
`
`hydrolysis at Cl might involve 2 reaction
`
`
`
`processes, glycoside bond cleavage 1-3 and ring
`
`
`
`opening, while hydrolysis at C4 results in a one
`
`
`
`step reaction for glycoside bond cleavage (1-4)
`
`(Fig. 3). On the other hand, from the results of
`
`
`13C-NMR (Fig. 5), the hemiacetal ring structure is
`
`
`maintained during the degradation of HA. Thus,
`for MNDO-MO calculations.
`
`
`the reaction mechanisms predicted from MNDO
`
`
`calculation were supported experimentally.
`
`NHCOCHs
`o�OH
`B xooc O
`0
`0 OH HO
`,,c,0{;)
`0 ."°(v""'
`CH2OH COOX
`OH
`Fig. 2. Model compounds
`
`3.2. Basic hydrolysis of sodium hyaluronate
`
`compounds. The molecular orbital calculation
`
`
`
`
`
`shows that the lowest unoccupied molecular
`
`
`
`orbital (LUMO) is localized on a glucuronic acid
`From the results of the calculation, it was clear
`
`
`
`
`
`residue and the highest occupied molecular
`
`
`that the electron density of HOMO on the
`
`orbital (HOMO) on a N-acethylglucosamine
`
`
`N-acethylglucosamine (HA units) was localized
`
`residue irrespective of the two model
`
`
`and the hydrolytically active sites are on the Cl,
`compounds.
`
`
`0 (ring) and amide nitrogen. Among them, only
`
`
`This result implies that HA hydrolysis under
`
`
`the Cl position contributes to the lowering of the
`
`
`acid conditions occurs on a glucuronic acid
`
`molecular weight of HA. In this case, the
`
`
`
`residue because of the nucleophilic nature of
`
`
`
`glycoside bond cleavage reaction on the Cl might
`
`acidic hydrolysis.
`On the other hand, the basic hydrolysis of HA
`
`
`is anticipated to start on a N-acethylglucosamine
`A
`
`
`
`residue because of the electrophilicity of basic
`
`
`hydrolysis. The mechanisms of HA hydrolysis
`
`under acidic and basic conditions will be
`
`discussed in detail.
`
`y
`
`(2)
`
`HOQOHHO 0---..._
`'-oH\ HO O
`
`B
`
`Fig. 3. Hydrolysis
`
`Fig. 4. Hydrolysis
`
`reaction mechanism of HA in the basic
`
`
`reaction mechanism of HA in acidic
`
`
`
`
`
`
`condition: figures are electron densities of HOMO.
`
`
`
`conditions: figures arc electron densities of LCMO.
`
`

`

`272
`
`
`
`Y. Tokita, A. Okamoto
`
`B
`
`',rH,. 1'
`
`A
`
`200
`
`150
`
`50
`JOO
`Chemical shift (ppm)
`
`[)
`
`
`
`
`
`
`
`products of hyaluronic acid (pH 1 ·44. aflcr 10 days). acid. (B) degradation Fig. 5. "C-NMR spectra (vs TSP): (A) hyaluronic
`
`involve two steps (contrary to acidic hydrolysis):
`
`
`
`pendent functional groups (Fig. 4), and this was
`
`
`by 13C-NMR (Fig. 6).
`
`
`the first step is the cleavage between Cl and C2,
`also confirmed
`
`and the second step is glycoside bond cleavage
`
`(1-4) (Fig. 4). In the first reaction, the production
`
`of methylene groups is to be expected.
`4.CONCLUSIONS
`
`The 13C-NMR spectrum of the basic hydrolysis
`
`
`products is shown in Fig. 6. In order to
`(1)The hydrolytic degradation of HA can be
`
`
`
`
`attributed to random chain scission and
`
`
`discriminate carbon species, the samples were
`
`measured by means of the DEPT method,ir>-
`
`
`accordingly obeys first-order kinetics.
`
`which confirmed the existence ot methylene
`
`(2)From the MNDO calculation on model
`
`
`compounds of HA, it was speculated that
`
`
`
`carbon. In addition, the production of sodium
`
`
`acetate was expected from the hydrolysis of
`
`
`acid and base catalysed hydrolyses occur
`
`12
`
`B
`
`0
`
`jl.
`
`A
`
`
`
`
`
`
`
`Fig. 6.
`
`
`(0) CH,COONa, (0) methylene carbon.
`
`13C-NMR spectra (vs TSP): (A) hyaluronic acid, (B) degradation products of hyaluronic acid (pH 1 �-30. after 7 days),
`
`0
`
`

`

`
`
`
`
`Hydrolytic degradation of hyaluronic acid
`
`
`
`273
`
`
`
`
`5.Weissmann, B., Rappurt, M. M., Linker, A. & Meyer,
`on glucuronic acid and N-acethyl­
`
`K., J. Biol. Chem., (1953)
`205.
`
`
`
`glucosamine moieties, respectively. These
`
`
`
`
`
`speculations were supported by the results
`121.
`
`
`
`7.Sakashita, H., Maeda, K., Tokita, Y. & Miyoshi, T.. J.
`of 13C-NMR spectroscopy.
`
`
`
`Action Oxygen Free Radicals, 5 (1994) 211.
`
`
`8.Pratt, L. M. & Chu, C. C.. I Com put. Chem , 15 (1994)
`241.
`
`
`I.Casadaban, M. J. & Cohen, S. N., Proc. Natl Acad. Sci.
`
`
`(1977) 4899,4907.
`
`
`U.S.A., 76 (1979) 4530.
`10.Davies, M., Analyst, 84 (1959) 248.
`
`
`
`659.2.Babior, B. M., New Engl. J. Med., 298 (1978)
`
`
`
`
`Phys., 77 (1982) 2745.
`
`
`
`3.Cleland, R. L., Biopolymers, 23 (1984) 647.
`
`
`
`4.Inoue, Y. & Nagasawa, K., Carbohydr. Res., 141 (1985)
`
`99.
`
`
`Reson., 48 (1982) 323.
`
`999.Dewar, M. J. S. & Thiel, W., J. Am. Chem. Soc.,
`
`6.Jeanloz, R. W. & Jeanloz, D. A., Biochemiwy, 3 (1964)
`
`11.David, T. P., David, M. D. & Robbin, B., J. Chem.
`
`12.David, M. D., David, T. P. & Robbin, B., J. Magn.
`
`REFERENCES
`
`