Carbohydrure Reseurch, 141 (1985) 99-110 Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands PREPARATION, BY CHEMICAL DEGRADATION OF HYALURONIC ACID, OF A SERIES OF EVEN- AND ODD-NUMBERED OLIGOSAC- CHARIDES HAVING A 2-ACETAMIDO-2-DEOXY-D-GLUCOSE AND A D- GLUCURONIC ACID RESIDUE, RESPECTIVELY, AT THE REDUCING END YUKOINOUEAND KINZONAGASAWA*
`
`School of Pharmaceutical Sciences, Kitasato University, 9-1, Shirokane 5 chome, Minato-ku, Tokyo 108
`(Japan)
`
`(Received January 25th, 1984; accepted for publication in revised form, November 5th, 1984) ABSTRACT A series of even-numbered hyaluronate oligosaccharides (di- to octadeca-) having a 2-acetamido-2-deoxy-D-glucose residue at the reducing end was prepared by treatment of sodium hyaluronate with dimethyl sulfoxide containing 10% of 0.1~ hydrochloric acid for 16 h at 95”. The mixture of the even-numbered oligosac- charides obtained was converted with 0.15~ sodium carbonate for 6 h at 40” into a series of odd-numbered oligosaccharides (mono- to pentadeca-) having a D- glucuronic acid residue at the reducing end. Reaction of the hyaluronate tetrasac- charide with saturated calcium hydroxide gave P-D-GlcpA-(1+3)-P-D-GlcpNAc- (1+3)-D-a&o-trihydroxyglutaric acid besides the expected trisaccharide. INTRODUCTION Hyaluronic acid is a high-molecular-weight glycosaminoglycuronan which consists of alternating 2-acetamido-2-deoxy-P_D-glucose and pD-glucuronic acid residues linked (1+3) and (l-44, respectively’. It has been usually depolymerized to even-numbered oligosaccharides .by enzymes. Testicular hyaluronidase gave a series of oligosaccharides consisting of the disaccharide unit, P-D-GlcpA-( 1+3)-p- D-GlcpNAc, and leech hyaluronidase a second group of oligosaccharides consist- ing of the reversed sequence ~-~-GlcpNAc-(1+4)-/3-~-GlcpA~~~. Chemical de- polymerization of the polysaccharide by acid hydrolysis or methanolysis gave hyalobiuronic acid or its methyl glycoside 4,5. In our previous papers, it was reported that a sulfated glycosaminoglycuronan, chondroitin 6-sulfate (free acid or pyridinium salt) was desulfated and depolymerized to form a series of nonsulfated di- to octadeca-saccharides having a 2-acetamido-2-deoxy-D-galactose residue at the reducing end by treatment with dimethyl sulfoxide containing 10% of watefi, *To whom correspondence should be addressed. 0008-6215/85/$03.30 @ 1985 Elsevier Science Publishers B.V.
`
`ALL 2036
`PROLLENIUM V. ALLERGAN
`IPR2019-01505 et al.
`
`

`

`100 Y. INOUE. K. NAGASAWA or methanol’. Furthermore, a nonsulfated glycosaminoglycuronan, hyaluronic acid (free acid), was depolymerized to some extent with the dimethyl sulfoxide reagents, and the addition of either pyridinium sulfate or pyridinium chloride greatly acceler- ated the reaction. Since the presence of pyridine was found not to be essential for the reaction, the addition of a small amount of inorganic acid was considered to be responsible for the enhancement of the reaction rate. The present report describes the formation of a series of even-numbered oligosaccharides from hyaluronic acid by solvolysis with aqueous dimethyl sulfoxide containing a small amount of hydrochloric acid, the transformation of the oligo- saccharides into a series of odd-numbered oligosaccharides by treatment with alkali, and a discussion of the alkali treatment. RESULTS AND DISCUSSION The reaction products, obtained by hydrolysis of sodium hyaluronate with 1OmM hydrochloric acid or with dimethyl sulfoxide containing 10% of 0.1~ hydro- chloric acid for 16 h at 105”, were chromatographed on AG l-X2 (Cl-) anion- exchange resin *. Peaks 2, 4, 6, and 8 (Figs. la,b) were identified as di-, tetra-, 0.6- 2 (a) ’ 4 1 1 50 100 RO I 2 10.4 aJ 0.4- (b) __-- --* : x 4 __-- ,_-- r_-- k , 9 -02- 202- I , b I #’ 6 II ,’ , I 8 /’ JL J A- 50 Tube number 100 Fig. 1. Anion-exchange chromatography, on AG l-X2 (Cl-) resin, of the reaction products of hyaluronic acid treated for 16 h at 105” with: (a) 1OmM hydrochloric acid, and (b) 1:9 (v/v) 0.1~ hyd- rochloric acid-dimethyl sulfoxide; (-) carbazole reaction (absorbance at 530 nm), (. . . . . .) Morgan- Elson reaction (absorbance at 585 nm), and (------) concentration of lithium chlorrde. Unless otherwise noted, the concentration of the sample subjected to the carbazole reaction was the same as that sub- jected to the Morgan-Elson reaction
`
`

`

`101
`
`DEPOLYMERIZATION
`
`OF HYALURONIC
`
`ACID (HA) WITH DIMETHYL SULFOXIDE CONTAINING
`
`10% OF ACID
`
`Acid
`
`Reaction conditions
`
`Reaction productp (%)
`
`Cont. of HA
`(mglmL)
`
`Temp. (‘)
`
`Time (h)
`
`pHb
`
`Disaccharide
`
`Higher mol.-W.
`oligosaccharides
`
`0.2M
`HCI 0.1~ HCI
`50mM
`
`HCI 0.1~ HCI 0.1~ HCI 0.1~ HCI 0.1~ H,SO, 50mM H,SO, 105 15.5 2.54 95 5 105 15.5 2.93 48 52 105 30 3.66 3 97 95 15.5 2.96 11 89 95 15.5 3.36 1 99 90 15.5 3.01 4 96 105 15.5 2.79 56 44 105 15.5 3.30 4 96 4Proportion of each fraction-size of the reaction products separated on Sephadex G-25 (based on uranic acid determination). me pH was measured after an equal volume of water had been added to the reaction mixture. hexa-, and octa-saccharides having a 2-acetamido-2-deoxy-o-glucose residue at the reducing end. A free 2-amino-2-deoxyhexose residue was not detected in any fraction (Figs. la,b) by the trinitrophenylation method9. Accordingly, it is suggested that the peaks positive with the Morgan-Elson reagent, except those containing the even-numbered oligosaccharides (see Fig. la), contain odd-num- bered oligosaccharides having a 2-acetamido-2-deoxy-o-glucose residue at the reducing end, and that the peaks negative with the reagent contain odd-numbered oligosaccharides having a D-glucuronic acid residue at the reducing end. The results illustrated in Figs. la,b suggest that, in dimethyl sulfoxide containing 10% of 0.1~ hydrochloric acid, the rate of hydrolysis of the 2-acetamido-2-deoxy-D-glucosyl linkage in the polysaccharide is accelerated, and that of the D-glucuronosyl linkage is retarded probably by the solvent effect of dimethyl sulfoxide. The results of the depolymerization performed under various conditions indicate that the reaction primarily depends on the pH and temperature (Table I). Sodium hyaluronate was treated, on a preparative scale, with dimethyl sulfoxide containing 10%
`
`of
`
`hydrochloric acid for
`
`16 h at 95”. Gel chromatog- raphy of an aliquot of the reaction product did not reveal the formation of free 2-acetamido-2-deoxy-D-glucose, which may have been liberated from the reducing end of the even-numbered oligosaccharides formed. A major part of the reaction product was fractionated on AG l-X2 (Cl-) anion-exchange resin into di- to octadeca-saccharide fractions (Fig. 2). The analytical data, summarized in Table II, are in good agreement with the values calculated for the respective even-numbered oligosaccharides. It is knownlo that the reducing terminal 2-acetamido-2-deoxy-D-glucose residue of hyaluronate oligosaccharides is decomposed under alkaline conditions to form chromogens having a maximum absorption at 230 nm, and the stability of
`
`EVEN- AND ODD-NUMBERED OLIGOSACCHARIDES FROM HYALURONIC ACID
`TABLE I
`0.1~
`

`

`50
`
`100 Tube number
`
`150
`
`Fig. 2. Anion-exchange chromatography, on AG l-X2 (Cl-) resin, of the reaction product of hyaluronic acid treated with 1:9 (v/v) 0.1~ hydrochlonc acld-dimethyl sulfoxide for 16 h at 95”: (-_) carbazole reaction, and (------) concentration of lithium chloride. the even-numbered hyaluronate oligosaccharides was examined. Hyaluronate hexasaccharide was heated at 60” in buffer solutions of pH 6.0-9.0 and the absorb- ance at 235 nm measured at various time intervals. The results (see Fig. 3a) indicated that the rate of decomposition decreased with reduction of the pH of the solution and that decomposition proceeded slowly, even at pH 6.5. Therefore, the mixture obtained after depolymerization had to be treated at pH 5.5-6.0 in order TABLE II ANALYTICALDATA
`
`SALTS)
`
`BYTREATMENT~F
`
`OF
`
`HYALURONIC
`AT
`
`ACID WITH DIMETHYL
`
`SULFOXIDE
`
`CONTAINING
`
`10% OF o.lM HYDROCHLORIC
`
`AClD FOR
`
`Oligo-
`saccharides”
`
`Yieldh
`(mg)
`
`2-Amino-
`
`(Ironic
`acid
`(%J
`
`Ratro of 2-amino-2-
`deoxyhexose to
`uranic acid residues
`
`Ratio of reducrng
`2-acetamido-2-
`deoxyhexose to
`crcid residues’,d
`
`41.76
`67.3
`43.23
`91.8
`0.50 (0 50)
`0.50 (0.50)
`38.64 1.00 0.05 (0) 1.00 (1.00) 4
`80.0
`44.39
`1.00
`0.67 (0.67)
`0.33 (0.33)
`64 7
`46 97
`0.74 (0.75)
`0.24 (0.25)
`1.01
`46.6
`46.75
`0.79 (0.80)
`34.7
`46.20
`0.19 (0 20) 12
`25.0
`47.47
`43.12 1 .Ol 0.83 (0.83) 0.16 (0.17) 14
`14.0
`47.40
`44.06 1.01 0.88 (0.86) 0.13 (0.14) 16
`8.2
`47.25
`44.00 1.01 0.89(0.88) 0.11 (0.13) 18
`25.8
`47.71
`42.74 0.98 0.88(0.89) 0.10(0.11) >20
`
`40.45 1.01
`
`43.43 1.00
`
`44.33 1.01 0 93 (>0.90) 0.08 (<RIO)
`
`“Number of monosaccharide umts. “Amount of product obtained from 640 mg of hyaluronic acid. ‘Ex- pressed relative to the molar ratio of 2-acetamido-2-deoxyhexose to uromc acid residue in N-acetyl- hyalobiuronic acid. din parentheses, calculated value for each oligosaccharide.
`
`(%I 2
`
`Before
`reduction
`
`After
`reductiond
`
`Y. INOUE. K. NAGASAWA
`EVEN-NUMBEREDOLIG~~A~~HARIDES(LITHIUM
`PREPARED
`16 h
`95”
`2-deoxyhexose
`6
`40.87
`8
`10
`43.48
`uranic
`

`

`EVEN- AND ODD-NUMBERED OLIGOSACCHARIDES FROM HYALURONIC ACID 103 (a) al 1 ^ I 2 4 6120 Reaction time (b) A -Gi- (h) Fig. 3. Determination of absorbance of hyahuonate- (a) and chondroitin-hexa~ccha~de (b), in buffer solutions having various pH values at 60’: (@) pH 9.0, (0) pH 8.0, (A) pH 7.0,
`
`(A)
`pH 6.5, (0) pH 6.0, and
`(x)
`
`pH 5.5. P 8 + u Loi $ ES c $ z:; 1.0
`
`+JO
`5c
`
`$3
`
`2%
`bz
`
`2Jg
`
`2 6 Reaction4 time(h) Fig. 4. Determination of absorbance of hyaluronate di- to tetradeca-saccharides in a buffer solution of pH 8.0 at 60”: (0) di-, (0) tefra-, (A) hexa-, (A) deca-, and (x) tetradeca-saccharide. to obtain the even-numbered hyaluronate oligosaccharides. On the contrary, the chondroitin hexasaccharide was much more stable than the hyaluronate hexa- saccharide at pH 7.0-9.0 (Fig. 3b). To compare the stabilities in a buffer solution of pH 8.0, hyal~ro~ate di- to tetradeca-saccharides were heated at 60”. The results (Fig. 4) showed that the smaller the oligosaccharide was, the more unstable it was in buffer solution. In order to find the best conditions for preparing odd-numbered hyaluronate oligosaccharides from the parent even-numbered oligosaccharides by alkaline degradation, the hyaluronate tetrasaccharide was treated with various alkaline reagents. Treatment with 2OmkI carbonate buffer (pH 10.0) for 5 min at lOO*, or 0.15~ sodium carbonate” for 2 h at 37”, gave, on AG l-X2 (Cl-) anion-exchange column chromatography (elution diagrams not shown), a single peak that was posi-
`
`

`

`C-l c-2 c-3 C-4 C-5 C-6 c=o CH, 104.1 102.9 103.7 92.0 95.7 74.0 56.5 (55.5b) 73.7 53.7 56.6 76.6 75.0 (84.ob) 76.4 81.6 84.0 72.9 71.0 (70.0b) 72.6 69.8 69.8 77.0 76.9 76.6 72.3 76.4 61.8 61.7 61.7 176.7 175.5 176.4 175.5 175.8 23.2 23.1 23.3 103.9 73.7 76.3 72.6 76.3 176.3 101.4 55.3 84.1 69.9 76.5 62.2 175.9 23.5 72.9 or 74.6 81.9 72.9 or 74.6 178.8, 179.6 “Values reported by Bociek et
`
`al.“. hIn
`
`glutaric acid
`D-arabo-Trihydroxy-
`
`/3-D-G~cQNAc
`
`P-D-GlcpA
`
`_
`
`-___
`
`Trmaccharide-II
`
`j3-~-GlcpA D-GlqNAc
`
`hyalobiuronic acid
`N-Acetyl-
`
`P-D-GlcpNAc
`
`-~-GlcpA~
`
`P
`
`a
`
`TABLE III W-N M R. DATA (@FOR N-ACETYLHYALOBIURONICACIDANDTRISACCHARIDE-II p
`
`parentheses, predicted values due to additional substitution at C-3.
`

`

`EVEN- AND ODD-NUMBERED OLIGOSACCHARIDES FROM HYALURONIC ACID 105 t’ c x 2 % 9 ’ 4 6 8 10 ,-===I04 t I ,’ i’: I; /’ ii ,’ :: ,I :: .: I I :,’ :r! ,: .’ ..--. . . . . . . . . f 50 100 150 Tube number Fig. 5. Anion-exchange chromatography, on AG l-X2 (Cl-) resin, of the reaction products obtained by treatment of hyaluronic acid with 1:9 (v/v) 0.1~ hydrochloric acid-dimethyl sulfoxide for 16 h at 95”, followed by treatment with 0.15~ sodium carbonate for 6 h at 40”: (--) carbazole reaction (absorbance at 530 nm) ; (. . . . . .) Morgan-Elson reaction (absorbance at 585 nm), and (------) concentration of lithium chloride. Each of the fractions was subjected to the Morgan-Elson reaction without prior dilution. For uranic acid assay, each fraction was diluted twenty times with water, then subjected to the carbazole reaction. The arrows indicate the elution positions of hyaluronate di- to deca-saccharides. tive to the carbazole reaction. The peak material was isolated and identified as the expected trisaccharide, /~-D-G~c~A+~-D-G~c~NAc+D-G~cPA (Trisaccharide-I). Prolonged reaction (-30 min) under the former conditions or reaction at elevated temperature (-60“) under the latter conditions resulted in a decrease of the tri- saccharide peak and appearance of small peaks at an earlier elution time, indicating that the trisaccharide
`is unstable to alkali. Treatment with M ammonium hydroxide for 2 h at 37” or with Dowex l-X2 (HO-) anion-exchange resin’* for 24 h at room temperature was found to be unsuitable for degradation because of formation of by-products. Treatment with a saturated calcium hydroxide solution13 afforded, on AG l-X2 (Cl-) anion-exchange column chromatography, two peaks positive to the carbazole reaction, one of which corresponded to Trisaccharide-I (elution diagram not shown). The second peak (Trisaccharide-II), which was eluted at a more retarded position, slightly increased under an oxygen atmosphere or under air free from carbon dioxide, and disappeared under a nitrogen atmosphere. Trisaccharide-I was converted into Trisaccharide-II, in 14% yield based on the carbazole reaction (experiments not described), upon treatment with saturated aqueous calcium hydroxide for 4 h at room temperature. Trisaccharide-II did not show reducing power by the Park-Johnson method14. The 13C-n.m.r. spectrum showed five signals at 6 72.9, 74.6,81.9, 178.8, and 179.6, in addition to those due to the N-acetylhyalobiuronic acid unit; these signals (Table III) were assigned on the basis of the assignment reported by Bociek et aZ.15. The ‘H-n.m.r. spectrum of the methyl ester of Trisaccharide-II showed an intensity of the methyl ester signal (6 3.79) corresponding to 3.27 moles per mole of acetamido signal (6 2.00), which was probably over-estimated because of over-
`
`per se
`
`

`

`106
`
`Y. INOUE, K. NAGASAWA TABLE IV
`
`ANALYTICAL DATA OF ODD-NUMBERED OLIGOSACCHARIDES
`
`~~~~IUM
`
`SALTS) PREPARED BY TRE,ATMENT OF
`
`HYALURONIC ACID WITH DIMETHYL SULFOXIDE CONTAINING 10% OF o.lM HYDROCHLORIC
`
`ACID FOR
`
`AT
`
`FOLLOWED BY ALKALINE DEGRADATION WITH 0.15M SODIUM CARBONATE
`~-_~~~~~~~~_~~~~_~~
`
`FOR
`
`AT 40”
`
`Oligo-
`saccharides”
`
`Yieldb
`(mg)
`
`Uronic
`acid
`&)
`
`2-Amino-
`Sdeoxyhexose
`%f
`
`Ratio of 2-ammo-2-
`deoxyhexose to
`uranic acid residues
`__I-
`
`Ratio of reducmg
`2-acetamido-2-
`de~xy~exose to 1000
`urontc actd residues
`at nonreducing end
`
`__II
`
`--
`
`78.82
`57.86
`51.09
`50.41
`
`1.26
`26.00
`30.61
`34.22
`
`Before
`reduction’
`
`After
`reduction
`
`a.o2(o)
`0.49 (0.50)
`0.65 (0.66)
`0.74
`
`0
`0.84
`0.77
`
`17
`21
`19
`
`3
`5
`I
`9
`I1
`13
`15
`>17
`
`34.2
`58 5
`63.9
`44.9
`43.5
`32.9
`20.6
`7.3
`41.2
`
`(0.75) 0.75 I9 51.12 37.70 0.80 (0.80) 0.88 21 51.14 39 3.5 0.83 (0.83) 0.83 21 49.23 39.15 0.86 (0.86) 0.89 22 49.40 39.40 0.86 (0.88) 0.90 20 51.65 43.77 0.92 (>0.89) 0.95 17 “Number of monosaccharide units. bAmount of product obtained from 640 mg of hyaluronic acid. <In parentheses, calculated value for each oligosaccharide. lapping with the signals of the ring protons. This suggested that Trisaccharide-II contains three carboxyl groups. Trisaccharide-II was treated with l-(3-di- methylaminopropyl)-3-ethylcarbodiimide hydrochloride and then reduced with sodium borohyd~de. Reduced Trisaccha~de-II was hydrolyzed and the product reduced with sodium borohydride to give the corresponding alditols, which were analyzed by g.1.c. after trimethylsilylation. The retention times of the two peaks obtained were identical with those of standard D-arabinitol and D-glucitol. Accord- ingly, the structure of Trisaccharide-II was estabhshed as j?-D-Glc~A-(1--+3)-/3-~- GlcpNAc-( l-+3)-D-a&o-trihydroxyglutaric acid. On the basis of the aforementioned results, a series of odd-numbered oligosaccharides (mono- to pentadeca-) having a D-glucuronic acid residue at the reducing end was prepared from sodium hyaluronate. The elution diagram of the reaction product on AG l-X2 (Cl-) anion-exchange chromatography and the analytical data of the material from each peak are given in Fig. 5 and Table IV, respectively. Each peak corresponding to the mono- to pentadeca-saccharide was weakly positive with the Morgan-Elson reagent, showing -0.02 mmol of 2- acetamido-2-deoxyhexos~ at the reducing end per mol of uranic acid residue at the nonreducing end. This is probably not due to the presence of unreacted even- numbered oligosaccharides, but to the known peeling-reaction under the alkaline conditions of the Morgan-EIson assay 16. As seen in Fig. 5, each even-numbered oligosaccharide was eluted earlier than the odd-numbered oligosaccharide from which it derived. The molar ratios of 2-amino-Z-deoxyhexose to uranic acid residue before reduction with sodium borohydride were in good agreement with the values
`
`16 h
`95”.
`6 h
`I
`

`

`AND ODD-NUMBERED OLIGOSACCHARIDES FROM HYALURONIC ACID
`
`107 calculated for the corresponding odd-numbered oligosaccharide, but those ob- tained after reduction were not (see Table IV). This discrepancy among the values obtained is probably due to the slightly visible coloration given, in the carbazole reaction, by the L-gulonic acid residue resulting from the reduction of the reducing D-glucuronic acid residue.
`
`EXPERIMENTAL
`
`Materials. -
`
`(M,
`
`2 000 000, determined by viscometry) was obtained from Seikagaku Kogyo Co., Tokyo. Chondroitin hexasaccharide was obtained by solvolytic depolymerization of chondroitin 6- sulfate by the method previously reported6.
`
`Analytical methods. -
`
`The methods for determination of uranic acid, 2- amino-Zdeoxyhexose, and 2-acetamido-2-deoxy-hexose, and for reduction of oligosaccharides with NaBH, were described previously6. 13C-N.m.r. spectra were recorded at 22.50 MHz with a JEOL FX90Q n.m.r. spectrometer operated in the f.t. mode. Chemical shifts (6) are expressed downfield from the signal of tetra- methylsilane, calculated from the signal of 1,4-dioxane (S 67.6). lH-N.m.r. spectra were recorded with a Varian EM-90 spectrometer at 22” for solutions in D,O (99.95%). Chemical shifts (6) are expressed downfield from the signal of sodium 4,4-dimethyl-4-silapentane-1-sulfonate, calculated from the signal of HOD (64.65). G.1.c. was performed with a Shimadzu GC4BM gas chromatograph, equipped with a flame-ionization detector, on a glass column (0.3
`
`x 200
`
`cm) packed with 3% SE-30 on Chromosorb W (80-100 mesh). Column temperature was programmed from 120 to 200” at 6”/min. N, was used as carrier gas at a flow rate of -50 mUmin. Analytical gel chromatography on Sephadex G-25 was carried out as described pre- viously7.
`
`Solvent effect of dimethyl sulfoxide on acid hydrolysis of sodium hyaluronate.
`
`-
`
`in
`
`vacua
`
`0-0.2M
`
`A solution of Na hyaluronate (160 mg for each experiment) in 1OmM HCl or dimethyl sulfoxide containing 10% of 0.1~ HCl (80 mL) was heated for 16 h at 105”. Each reaction mixture was cooled and diluted with an equal volume of water, and the pH was adjusted to 6.0 with 0.1~ NaOH. The solution was evaporated
`and the residue dissolved in water (5 mL) was applied to a column (2 X 90 cm) of AG l-X2 (Cl-, 200-400 mesh) anion-exchange resin. It was eluted at room temperature with linear gradients of
`(0.6 L) and 0.2-0.4~ LiCl(2.0 L) (Figs. la,b). The eluate was collected in 16-mL fractions and each fraction was analyzed by the carbazole and Morgan-Elson reactions, and by the trinitrophenylation method9.
`
`Depolymerization of sodium hyaluronate with dimethyl sulfoxide containing
`10% of water in the presence
`HCl or H$O,.
`-
`
`A solution of Na hyaluronate (4 mg for each experiment) in dimethyl sulfoxide containing 10% of water (2 mL) and various amounts of HCl or H,SO, was heated under various conditions, as shown in Table I, cooled, and diluted with an equal volume of water, and the pH adjusted
`
`EVEN-
`Rooster comb hyaluronic acid
`of
`

`

`108
`
`Y. INOUE, K. NAGASAWA to 6.0 by the addition of 1OmM NaOH. The solution was evaporated
`and the residue, dissolved in 0.1~ NaCl (1 .O mL), was examined by analytical gel chromatography on Sephadex G-25.
`
`oligosaccharides
`Preparation of a series of even-numbered
`uronate with dimethyl sulfoxide containing IO% of 0.1~ HCE. -
`
`from Na hyal-
`
`in vacua
`
`A solution of Na hyaluronate (640 mg) in dimethyl sulfoxide containing 10% of 0.1~ HCl (320 mL) was heated for 16 h at 95”, cooled, and diluted with an equal volume of water, and the pH was adjusted to 6.0 with 0.5~ NaOH. The solution was evaporated
`
`in vacua
`
`and the residue dissolved in water (5 mL). An aliquot (50 pL), diluted with 0.1~ NaCl (1
`mL) ,
`
`in the presence of P,O,. The analytical data of the isolated oligosaccharides are reported in Table II.
`Determination of the absorbance of hyaluronate hexasaccharide
`in buffer solu-
`tions having various pH values. -
`
`in vacua
`
`was examined by analytical gel chromatography on Sephadex G-25. The remaining was applied to a column (2 x 90 cm) of AG l-X2 (Cl-, 200-400 mesh) anion-exchange resin and the column eluted at room temperature with linear gradients of 0-0.2~ (0.6 L), 0.2-0.4~ (2.0 L), and 0.4-0.6~ LiCl (1.0 L) (Fig. 2). The eluate was collected in 15.9-mL fractions and each fraction analyzed for uranic acid content. The material from each peak was pooled, lyophilized, and desalted on a column (2.5 x 70 cm) of Sephadex G-15 by elution with 10% ethanol. The desalted solution was lyophilized and the residue dissolved in a minimum volume of methanol, or methanol containing a few drops of water. The solution was filtered and the filtrate poured into cooled acetone with stirring to give a white precipitate. The precipitate was collected by centrifugation and dried in air, and then
`Hyaluronate hexasaccharide (1.5 mg for each experiment) was dissolved in 2 mL of 0.1~ KH,PO,-Na,B,O, buffers of various pHs (5.5-9.0). The solutions were heated at 60”. At intervals, the solutions were cooled and the absorbance at 235 nm was measured (Fig. 3a). If necessary, the solution of pH 8 or 9 was diluted appropriately with the same buffer. The absorb- ance of chondroitin hexasaccharide was determined in the same manner (Fig. 3b).
`Di-, tetra-, hexa-, deca-, and tetradeca-saccharide (1.5 mg for each sample) were heated at 60” in 0.1~ KH,PO,-Na,B,O, buffer (2.0 mL, pH 8.0). At intervals, the solution was cooled and the absorbance at 235 nm measured (Fig. 4). The uranic acid content of each oligosaccharide solution was determined by the carbazole reaction.
`Hyaluronate tetra- saccharide (4 mg for each experiment) dissolved in the alkaline medium (1 mL) was stirred in a stoppered glass tube under the following conditions: (a) 20mM Na,CO, buffer (pH 10.0) for 5, 10, 30, and 45 min at 100”; (b) freshly prepared 0.15M Na,CO, for 2, 4, and 6 h at 37”; (c) Dowex l-X2 (HO-, 0.2mL) anion- exchange resin for 24 h at room temperature; (d) M NH,OH for 2 h at 37”; and (e) saturated Ca(OH), solution for 2 and 4 h at room temperature, under N, or 0, atmosphere, or under CO,-free air. In (a) and (b), the mixture was made neutral by
`
`Determination of the absorbance of even-numbered hyaluronate oligosacchar-
`ides (di- to tetradeca-) in a buffer solution of pH 8.0. -
`
`Alkaline degradation of hyaluronate
`
`tetrasaccharide.
`
`-
`
`

`

`EVEN- AND ODD-NUMBERED OLIGOSACCHARIDES FROM HYALURONIC ACID
`
`109 the addition of Dowex 5OW-X2 (H+) cation-exchange resin. In (c), the mixture was filtered and the resin was washed with a small volume of 0.1~ HCl; the combined filtrate and washing were made neutral with 0.1~ NaOH. In (d), the mixture was evaporated to remove NH,. In (e), the mixture was made neutral by bubbling CO,. Each of the solutions was applied to a column (1 x 26 cm) of AG l-X2 (Cl-, 200-400 mesh) anion-exchange resin and eluted with a linear gradient of 0.2-0.3~ LiCl(0.3 L) . Each fraction (4 mL) was analyzed by the Morgan-Elson and carbazole reactions.
`
`Degradation of hyaluronate tetrasaccharide with saturated Ca(OH), on a
`preparative scale. -
`
`Hyaluronate tetrasaccharide (Li salt, 100 mg), dissolved in a saturated Ca(OH), solution (25 mL), was stirred for 2 h at room temperature under CO,-free atmosphere, and the mixture was made neutral with gaseous CO,. The precipitate was filtered off and the filtrate evaporated
`
`in vacua. The
`
`Preparation of the methyl ester of Trisaccharide-II.
`
`-
`
`Preparation of alditols from Trisaccharide-II.
`
`-
`
`residue was applied to a column (1.4 x 40 cm) of AG l-X2 (Cl-, 200-400 mesh) anion-exchange resin and the column eluted at room temperature with a linear gradient of 0.2-0.3~ LiCl (0.8 L). Each fraction (10 mL) was analyzed by the Morgan-Elson and carbazole reactions. The material from each peak reacting with carbazole was pooled, lyophilized, and desalted on a column (1.5 x 75 cm) of Sephadex G-15. Each of the desalted solutions was lyophilized to give Trisaccharide-I and -11 in yields of 25 and 23 mg, respectively.
`A solution of diazomethane in dietbyl ether (2 mL), prepared from 0.3 g of N-methyl-ZV-nitroso- 4-toluenesulfonamide, was added to Trisaccharide-II (8.5 mg) dissolved in methanol (1.1 mL), and the mixture stirred for 20 min at 5”. After evaporation of the solvent, the residue was dissolved in a small volume of water and applied to a column (1 x 4 cm) of AG l-X2 (Cl-, 200-400 mesh) anion-exchange resin, which was washed with water. The combined effluent and washing were lyophilized to give the methyl ester of Trisaccharide-II (6.0 mg).
`Trisaccharide-II (4.4 mg) was dissolved in water (1 mL), and the pH of the solution was adjusted to 4.75 by the addition of 0.1~ HCl. To this solution was added 1-(3-dimethylaminopropyl)-3- ethylcarbodiimide hydrochloride (35 mg) in several portions at room temperature. The pH was kept at 4.5-5.0 by the addition of 0.1~ HCl. After the hydrogen-ion uptake had ceased, 2~ NaBH, (3 mL) was added to the solution in three portions. During the reaction (1 h), the pH was kept at -8.5 by the addition of
`
`M
`HCl. The mixture was made acidic with M HCl, then neutral with
`M
`
`NaOH, and applied to a column (1.5 x 75 cm) of Sephadex G-15 for desalting. Fractions reacting with anthrone were collected and evaporated to dryness. The residue, dissolved in
`
`M
`
`HCl (1.5 mL), was heated for 2 h at 100” and the mixture evaporated to re- move HCl. To the residue was added 1% NaBH, (5 mL) and the mixture kept for 30 min at room temperature. Excess NaBH, was removed by the addition of Dowex 5OW-X2 (H+) cation-exchange resin. The resin was filtered off, and the filtrate and washing were combined and repeatedly evaporated to dryness in the
`
`

`

`110 Y.
`
`INOUE. K. NAGASAWA
`
`presence of methanol. l,l,l-Tris(hydroxymethyl)ethane (0.95 mg) was added to the residue and the mixture was dried in vucuo overnight at room temperature. After the addition of 1-(trimethylsilyl)imidazole (0.2 mL), the mixture was heated for 30 min at 80” and analyzed by g.1.c.
`
`hyaluronate oligosaccharides having
`Preparation of a series of odd-numbered
`acid residue at the reducing end. -
`a D-ghcuronic
`
`Sodium hyaluronate (640 mg) was depolymerized to give a series of even-numbered oligosaccharides according to the conditions described earlier. The mixture of even-numbered oligosaccharides was dissolved in freshly prepared 0.15~ Na,CO, (64 mL) and the solution kept for 6 h at 40”. The solution was made neutral by the addition of Dowex 5OW-X2 (H+) cation-exchange resin. The resin was filtered off, and the combined filtrate and washing were desalted on a column (2.5 x
`
`75
`
`cm) of Sephadex G-15. The desalted solution was lyophilized and the residue, dissolved in water (10 mL), was applied to a column (2 x
`90
`cm) of AG l-X2 (Cl-, 200-400 mesh) anion-exchange resin. The column was eluted at room temperature with linear gradients of 0-0.2~
`(0.6
`
`L) and 0.2-0.4~ LiCl (2 L). The eluate was collected in 16-mL fractions, which were analyzed by the carbazole and Morgan-Elson reactions (Fig. 5). The materials from each peak reacting with carbazole were isolated by the same procedure as described in the section of preparation of even-numbered oligosaccharides. Analytical data of the isolated oligosaccharides are reported in Table IV.
`
`ACKNOWLEDGMENTS
`
`The authors thank Dr. Y. Inoue, University of Tsukuba for determination of the i3C-n.m.r. spectra and its helpful discussion, and Misses K. Ryuzaki and A. Tsuchimochi for technical assistance.
`
`REFERENCES
`
`Biol.. 17 (1958) 1075-1077.
`1 K. MEYER,
`Chem., 208 (1954) 417429.
`2 B. WEISSMANN, K. MEYER, P. SAMPSON, AND A. LINKER,
`3 A. LINKER, K. MEYER, AND P. HOFFMAN,
`B. WEISSMANN, M. M. RAPPORT, A. LINKER, AND K. MEYER,
`J. Btol. Chem., 235 (1960) 924-927. 4
`5 R. W. JEANLOZ AND D. A. JEANLOZ, Biochemistry, 3 (1964) 121-123.
`6 Y.
`K. NAGASAWA,
`Carbohydr. Res., 97 (1981) 26S278. 7 Y.
`INOUE AND K. NAGASAWA,
`L.-A. FRANSSON, L. ROD~N, AND M. L. SPACH,
`Carbohydr. Res., 85 (1980) 107-119. 8
`YOSHIZAWA. T. KOTOKU. F. YAMAUCHI, AND M. MATSUNO,
`Anal. Biochem., 3 (1968) 317-330. 9 2.
`Biochim. Blophys. Acta, 141 (1967) 358-365.
`89 (1956) 1473-1486.
`10 R. KUHN AND G. KRUGER,
`11 J. E. CHRISTNER, M. L. BROWN,AND D. D. DZIEWIATKOWSKI,
`27 (1963) 134-149.
`12 K. ONODERA AND S. HIRANO,
`13 M. K COWMAN, E. A. BALAZS, C. W. BERGMANN. AND K. MEYER,
`
`Chem., 205 (1953) 205-211.
`
`Chem., 254 (1979) 4624-4630.
`
`Biochemistry, 20 (1981) 1379- 1385. 14
`
`181 (1949) 149-1.51.
`J. T.
`M. J. JOHNSON,
`15 S. M. BOCIEK, A. H. DARKE, D. WEI.TI, AND D. A REES.
`R. L. WHISTLER AND J. N. BEMILLER.
`Eur. J. Blochem., 109 (1980) 447-456 16
`
`Fed. Proc., Fed. Am. Sot. Exp.
`J. Biol.
`J. Biol.
`INOUE AND
`Chem. Ber.,
`J. Biol.
`Agr. Biol. Chem.,
`PARK AND
`J. Biof. Gem.,
`Adv. Carbohydr. Chem., 13 (1958) 289-329.
`