The molecular weight-dependent distribution of urinary glycosaminoglycans in Werner's syndrome

BIOCHEMICAL

MEDICINE

34,

251-258

(1985)

The Molecular Weight-Dependent Distribution of Urinary Glycosaminoglycans in Werner’s Syndrome KATSUMI Depurtment

of

Medicine

MURATA

and Physical T/zerapy. HO~RO. BunXyo-krr. Received

University TohTo (113).

of

Tokw Jupan

Sclwol

of

Medicine.

7-3-I

July 7, 1983

Werner’s syndrome has scleroderma-like features and unique clinical manifestations resembling those observed in the aging process of normal subjects (I3). It appears to affect mesodermal connective tissues throughout the body. Acidic glycosaminoglycans (AGAG) are one component of connective tissue that may be specifically affected in patients with Werner’s syndrome (4). Degraded components of AGAG was partly excreted in urine as a catabolic product. The main urinary AGAG in normal subjects consist of chondroitin 4-sulfate (C-4S), chondroitin 6-sulfate (C-6S), and heparan sulfates (HS) with other minor components of chondroitin sulfate (CS) isomers (5-7). Hyaluronic acid (HA) in normal human urine is either possibly absent or may be present in very small amounts (5, 7). However, recent investigations showed that HA was present in the urine of patients with Werner’s syndrome (8, 9). Compositional changes of other urinary AGAG were also indicated in the syndrome (9). It is necessary, therefore, to know whether, and to what extent, urinary AGAG of patients with Werner’s syndrome depend on molecular weight. The components of AGAG can be determined by use of an enzymatic assay which employs chondroitinases and hyaluronidase (10-12). An attempt was made to ascertain in more detail the molecular weight-dependent distribution of urinary AGAG in the patients with Werner’s syndrome by employing enzymatic approaches after fractionation by gel filtration, together with electrophoretic characterization and other chemical methods. MATERIALS

AND METHODS

Approximately 20 liters of pooled urine was collected from each of four patients with Werner’s syndrome. The two male patients were 48 (Case 1) and 50 years old (Case 2). The two females were 38 (Case 3) and 42 years old (Case 4) (13). Simultaneously, urine was prepared from healthy age- and sex-matched subjects. Urinary AGAG were prepared by a modification of the method reported previously (9, 14). After the pH of the urine was adjusted to 5.0,5% cetylpyridinium chloride 251 0006-2944/85

$3.00

Copyright 0 1985 by Academic Press. Inc. All rights of reproduction in any form reserved.

252

KATSUMI MUKAT.2

(CPC) was added to the urine at a rate of 15 ml/liter of urine foiiowcd by genrjc agitation. After the solution was kept at 4°C overnight to precipitate the AGAGCPC complex. it was centrifuged at 4°C at 2000 rpm for 30 min. The supernatant was decanted and the residual AGAG were reprecipitated completely with the CPC solution by the same manner mentioned above. Both AGAG-CPC compieu fractions were combined and then dissociated by stirring with 109 sodium acctatc and precipitated by adding 4 voi of ethanol. The specimens were digested three times with pronase (i,OOO.OOO tyrosine unitsig; Kakenyakukako Co., Tokyo) al 20 mg/liter of urine in l/15 M phosphate buffer, pH 7.8. at 8-hr intervals. Sodium hydroxide was added to a final concentration of 0.5 M and the mixture kept at 5°C overnight. Trichioroacetic acid was then added to a final concentration ot 10% and the mixture was centrifuged. The supernatant\ were diaiyred agams~ water in dialysis tubing (V&king Co.) in which the pore size had been reduced by acetyiation (14). The nondiaiyzable AGAG which remained in the tube were concentrated with a flash evaporator and 4 vol of ethanol was added to precipitate the AGAG. The crude AGAG were applied to Dowcx l-X2 columns i 1 I 45 cm, Cl form, 200-400 mesh) and eiuted with 0.25 and 3.0 M NaCI. The AGA(; in the latter eluate were desalted using a Sephadex G-IO column I I .X 40 cm). condensed, and applied to a Sephadex G-100 column (1 J 77 cm). The AGAG were eluted with 0.1 M acetic acid at a rate of 3 ml115 min by gel filtration through a Sephadex G-100 column which had been eyuilibrxted with the same solution and were divided into five fractions according to the molecular weight distribution. The AGAG were then subjected to an enzymatic assay as described belou. Chondroitinase ABC (EC 4.2.2.4) and chondroitinase AC (EC 4.7.1.51 mere prepared from Proteus 1~u1gari.s and Arthwhr~t~tc~r UIII’~.\(‘CFI.S. respectively I I t _ 15, 16). Both enzymes were prepared and purified at the Tokyo Institute of Seikagaku Kogyo Company (Tokyo). o-Glucuronic acid and I -iduronic acid 11~ the CS component were determined by an enzymatic assay using chondroitinase3 (11, 15, 16). Chondroitinase-AC digests only r)-giucuronic acid in CS chains and converts it to the corresponding unsaturated disaccharide unit ( 1 I. 15). In addition to D-glucuronic acid, chondroitinase-ABC also converts I.-iduronic acid to the corresponding unsaturated disaccharide unit ( 1 1. 16). Strvptott~yccs hyaiuronidahc was extracted from Streptonzyces hyalutwrzkrrs tzo\‘. ( 12) and it was purchased from the Tokyo Institute of Seikagaku Kogyo Company. This enzyme specificall) digests only HA and did not attack other AGAG (1’1. The AGAG thus prepared, approximately 400 kg a> uranic acid. were dlge\tctl with chondroitinase ABC (1.0 unit) in 0.1 M Tris buffer pH 8.0. for I20 min and the same amount of the AGAG with chondroitinase AC (I,2 units) in 0, I \I acetate buffer, pH 6.0, at 37°C for 120 min (9. 1 I). The unsaturated disaccharide3 were then applied to Whatman No. 1 filter paper and separated in a 1-butyric acid/O.5 M ammonia solution (513, v/v) for 60 hr by descending chromatography. The separated unsaturated disaccharides as well as the origin were eluted for measurement by the borate carbazole reaction ( 17). In this process the urinar) AGAG (400 ,ug as uranic acid) were degraded approximately 6X and 60% with chondroitinases ABC and AC. respectively. The macromolecular AGA(i i?--5 pg as uranic acid) undigested by either enzyme were characterized by cicctro-

URINARY

GLYCOSA.~INOGLYCANS

IN WERNER’S

SYNDROME

253

phoresis on cellulose acetate membranes (IS), as follows: In 0.1 M pyridineformic acid, pH 3.0, at 0.5 mA/cm for 60 min; in 0.1 M calcium acetate, pH 7.0, at 0.5 mA/cm (19); and in 0.1 M barium acetate, pH 7.2, at 5 V/cm for 180 min (20).

For the standard AGAG, C-4s and C-6s were prepared at the Tokyo Institute of Seikagaku Kogyo Company. Standard HA, dermatan sulfate (DS), HS, and heparin were generous gifts from Dr. M. B. Mathews (University of Chicago, Chicago, Ill.). All chemicals used were of analytical grade unless otherwise! specified. Thin-layer chromatographic separation of hexosamines was performed by the method described by Moczar rt al. (21). RESULTS

The average amounts of AGAG excreted in urine of the patients with Werner’s syndrome were 4.1, 5.4, 3.5, and 3.8 mg/24-hr urine in Cases I, 2, 3, and 4. respectively. corresponding to 0.09-o. 14 mg/24-hr urine/kg body wt. Although the urinary excretion of AGAG in the patient during a 24-hr period was within normal limits, the daily urinary excretion of AGAG per kilogram of body weight was somewhat higher than that of a normal adult (O.OS-0.07 mgi24 hr urine/kg body wt). The urinary AGAG obtained from the patients were then fractionated by Sephadex G-100 column chromatography. A representative elution profile of the AGAG showed two main peaks which were fractionated fractions (Fr.) I to V (Fig. 1) and the AGAG components were characterized by electrophoresis and an enzymatic assay (5, 11). The electrophoretic mobility of the AGAG in the higher molecular fractions (Fr. I and II) shows that the faster moving band,

10 Tube No. Fr No.

20 ,

30 1 II

1 Ill

40 1 IV

1

50 v

FIG. 1. A represenstative gel filtration profile of the urinary AGAG of a patient with Werner’s syndrome (Case 1) on a Sephadex C-100 column (I Y 77 cm) eluted with 0. I M acetic acid.

254

KATSUMI

MURATA

corresponding to standard CS isomers, HS, and DS, was stained more dense11 than the slower moving HA band. Since the AGAG in the slower band was hyaluronidase. it was judged to be madigested completely with Streptomyces cromolecular HA. The presence of HA and HS in the higher molecular fractions was supported by electrophoretic characterization using three buffer systems. In contrast, urinary AGAG in the smaller molecular weight fraction of Werner’5 patients mainly consisted of CS isomers, including C-4S, C-6S, and chondroitin. which are the main components of urinary AGAG of healthy persons (5-7). The paper chromatographic profile of the separated unsaturated disaccharide5 from the CS isomers of urinary AGAG of the patients with Werner’s syndrome after digestion with chondroitinase AC is shown in Fig. 2. Under these condition%. HA was degraded by the treatment. This enzyme gave two spols of unsaturated nonsulfated disaccharides derived from HA and chondroitin ( 11,22). Unsaturated nonsulfated, 4-sulfated, and 6-sulfated disaccharides were present ah the ma.jor constitutional units. In addition, the appreciable amounts of unsaturated nonsulfated disaccharide derived from HA were detected in the higher molecular weight fractions in the urine of Werner’s patients. Unsaturated disulfated disaccharide was prominently detected in the lower molecular weight fractions. though the amount was small. The data for the unsaturated disaccharides originating from the urmary AcJAC; of four patients with Werner’s syndrome and the normal subjects are summarized in Table I. The major AGAG in the higher molecular weight fractions were HA and HS. The proportions of HS or HA to the total AGAG in each fraction decreased with decreasing molecular weight. Conversely, CS ixomers. such as C-4S, C-6s. and chondroitin, were rather predominant in the lower molecular

thsaturateo di-sulfated disaccharide Unsaturated &sulfated disaccharide Unsaturated I-sulfated disaccharide

Unsaturated nonsulfated disaccharide (chondraitin) Unsaturated rmnsulfated disaccharide (HA)

FIG. 2. Paper chromatographic pattern of unsaturated Werner’s patient (Case I) fractionated with a Sephadex the faster spot of unsaturated nonsulfated disaccharide molecular weight fractions.

disaccharides of the urmary AGAb 4 ;I G-100 chromatography column. Note thal derived from HA is dominant in the higher

Q-3) (O-3) (3-8)

1.7 5.4

5.6 14.8

8.4 12.1

0.7 5.1

1.4 2.9

63.7 57.1

0.70 (0.6-0.9) 1.23 (1.1-1.4)

20.1 10.7

Chondroitin

Chondroitin 4-sulfate

Chondroitin 6-sulfate

Dermatan sulfate

Oversulfated chondroitin sulfate

Heparan sulfates

Chondroitin 4-sulfate/ chondroitin 6sulfate

Recovery (%)

(32-39) (31-33)

(4-6)

(l-5)

G-4) U-3)

(16-27) (16-25)

(13-16) (23-27)

(10-13) (3-4) (5-17) (8-16)

15.3 13.0

(1 l-22) (10-16)

0.67 (0.4-I .O) 1.20 (1.1-1.4)

35.9 31.9

3.3 5.1

2.8 2.9

20.1 20.8

15.0 24.8

11.6 11.2

11.4 3.1

II

(17-35) (19-22)

(3-5) (4-5)

(3-4)

(l-6)

(14-24) (14-23)

(16-20) (27-30)

(9-36) (20-29)

(6-12)

22.8 23.6

(14-29) (16-30)

1.00 (0.7-1.2) 1.59 (1.3-2.0)

26.7 20.1

4.2 4.5

4.1 3.9

18.3 18.1

17.6 28.9

20.5 24.6

8.6 0

III

Fraction

~~

(18-34) (26-33)

(4-7) (6-8)

(3-6)

(2-5)

(13-18) (13-18)

(14-24) (29-30)

(16-32) (12-18)

(7-11)

25.2 42.0

(23-30) (38-45)

1.17 (0.7-1.5) 1.92 (1.6-2.2)

27.1 29.6

5.3 6.5

3.0 4.7

16.4 15.3

18.5 29.4

21.4 14.9

8.4 0

IV

(27-34) (14-16)

(6-W

(4-10)

(1-2)

(l-4)

(18-19) (18-22)

(16-24) (33-34)

(12-18) (20-25)

U-8)

16.6 10.7

(11-31) (7-15)

1.10 (0.9-1.3) 1.69 (1.5-1.9)

30.9 15.7

6.8 7.0

1.9 1.6

18.5 19.8

19.8 33.4

14.7 22.6

7.7 0

V

(33-39) (26-33)

(4-7)

O-6)

(3-5)

C-4

(15-20) (13-20)

(13-17) (22-31)

(10-18) (13-20)

(10-13) (O-1)

100.0 100.1

0.95 (0.7-1.1) 1.69 (1.5-1.9)

35.7 29.0

4.0 5.5

2.8 4.0

17.1 16.8

15.7 27.5

13.9 16.6

10.8 0.7

Total

TABLE I of Urinary AGAG of the Patients with Werner’s Syndrome and the Normal Subjects

Note. All values are expressed as percentages of uranic acid determined by the carbazole reaction (17) after fractionation by gel filtration on Sephadex G-100. Upper and lower rows are the means (and ranges) of urinary AGAG of four Werner’s patients and four age-matched healthy subjects, respectively. Hyaluronic acid, chondroitin, chondroitin 6-sulfate, and oversulfated chondroitin sulfate were calculated from the mean values of the unsaturated nonsulfated, 6-sulfated, and disulfated disaccharides by digestion with chondroitinases ABC and AC. Dermatan sulfate was estimated by subtraction of the value for the unsaturated 4-sulfated disaccharide from digestion with chondroitinase AC from that with the ABC iyase. Heparan sulfate was calculated from the value of AGAG not digested with chondroitinase ABC (9, 11).

(17-25) (8-14)

(50-78) (55-60)

(I-2) Q-4)

(3-W

(O-5)

(3-13) (10-14)

(O-12) (14-15)

(17-19)

17.6 2.8

Hyaluronic acid

Distribution

I

Weight-Dependent

AGAG

Molecular

256

KATSUMI

MUKA’I

A

weight fractions. The ratio of C-4s to C-6S increased with decreasing molecular weight. In the urine of Werner’s patients. HA accounted for 10.8% and HS for 35.7% of the AGAG. whereas CS isomers such as chondroitin. C-4s. and C-M accounted for 14-18s of the total AGAG. The distribution of AGAG in males and females with the disease was similar. In all patients. the distribution of C6s was similar to that of normal sub.jects whereas there was less C-4s. especialIF in the lower molecular weight fractions. in Werner’s patients than in normals. Thus, the ratios of C-4s to C-6s in all fractions were consistently less in the patients with Werner’s syndrome than normal persons and the ratios increased with decreasing molecular weight. AGAG in the urine of Werner’s patients were applied to a Dowe. l-X2 chrumatography column and the AGAG in the stepwise eluates at NaCl differentiation were characterized by electrophoresis (Fig. 3). The AGAG of the 0.5 M NaCI eluate. which specifically consisted of HA, was collected and subjected to moicculat weight measurement by viscometry (33). The average molecular weight 01’ the HA was found to be 32.000. The HA could be exhaustively digested with S’twp tom~c~.s hyaluronidase and the hexosamine moiety was found to be glucosaminc (21). DISCUSSION Previous studies indicated the presence of HA excreted in the urine of patients with Werner’s al. reported the presence of HA in the urine of (8). In the urinary AGAG of Werner’s patients.

ah a specific AGAG component syndrome (8. 9). Tokunaga et patients of Werner’\ \yndrornc an elevated excretion of other

c 6s c-4s DS 0.5M 0.7M

N

1.25M 3.OM HA

HOP

FIG. 3. Electrophoretic characterization of the urinary (Case I) fractionated with a Dowex l-X2 chromatography 0.5 M NaCl corresponds to the standard HA.

AGAG of a patient with Werner’\ \yndrorrw column. Note that the \pol of AGAG .II

URINARY

GLYCOSAMINOGLYCANS

IN WERNER’S

SYNDROME

257

AGAG components such as HS was found in addition to that of HA (9). On the other hand, the u~nary AGAG components of healthy subjects showed changes with aging (24). The characteristic distribution of the urinary AGAG components on a molecular weight basis was reported in normal (14) and pathological states (25). The question whether the distribution of AGAG components in the urine of patients with Werner’s syndrome is dependent on molecular weight has yet to be answered. Previous studies reported that the major AGAG components in normal human urine are C-4s and C-6s (5, 6). Since the ratio of C-4s to C-&S decreased to less than 1.O in the urine of adults after adolescence (26) and the proportion of HS increased with aging (27). the present data indicated that such changes in the urinary AGAG in Werner’s syndrome patients reflect a process of accelerated aging. Nevertheless, a proportion of HA, as high as 10% has only been detected the urine of Werner’s syndrome (8, 9). In addition, the presence of HA in the urinary AGAG of Werner’s syndrome may indicate a new type of mucopolysaccharidosis. HA is known as the major AGAG in renal connective tissue (28), though very little of the substance is excreted into normal human urine (5). Since macromolecular HA was excreted to an appreciable extent in the urine of Werner’s patients, it can be assumed that there may be some metabolic disorder in the renal connective tissue. A certain renal dysfunction has been observed in laboratory examinations (3). On the other hand, HA was found to present in human plasma (29) and the molecular weight of HA was estimated as 25,000 in urine after HA was injected intravenously (30). Therefore. the urinary AGAG composition of Werner’s syndrome may reflect not only a degenerative process in the renal tissue but also a metabolic disorder of the connective tissues throughout the whole body. SUMMARY The macromolecular AGAG in the urine of patients with Werner’s syndrome were analyzed by enzymatic methods after digestion with chondroitinases and ~f~e~ru~~e~.~ hyaluronidase. The molecular weight-dependent distribution of the urinary AGAG has been determined by gel filtration on a Sephadex G-100 column. The distribution of HA and HS was predominant in the macromolecular fractions. Chondroitin sulfate isomers were prominent in the low molecular weight fractions but the ratio of the 4-type to the 6-type increased with decreasing molecular weight. These observations indicated that Werner’s syndrome is a metabolic disorder of the molecular weight-dependent AGAG composition. ACKNOWLEDGMENTS The present study was supported by grants in aid from the Ministry of Education, Culture and Science. Japan, for an aging research project from Institute of Physical and Chemical Research and Adult Disease Research Memorial Foundation, Tokyo. The author expresses his thanks to Prof. T. M~yamoto for his advice and to Miss R. Abe. Miss R. Yamaguchi and Dr. Y. Yamada for sampling urinary specimens and analysing AGAG.

REFERENCES I. Werner, C. W. O., in doctoral dissertation, Schmidt and Klauning. Kiel, 1904. 2. Epstein, C. J., Martin, G. M.. Schultz, A. L., and Motulsky, A, G., Medic+ne 45, 177 (1966).

258 3. 4. 5. 6. 7. 8. 9. 10. I I. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25, 26. 27. 28. 29. 30.

KATSUMI

MURAI‘A

Murata. K.. and Nakashima, H.. J. Amer. Gcrkrr. SCX,. 30, 103 t 19821. Fleischmajer. R., and Nedwich, A., Amer. J. Med. 54. Ii 1 (1973). Varadi, D. P., Cifonelli, J. A.. and Dorfman A.. Bioc~him. Biophy.\. Ac:to 141, 103 (1967). Murata. K., Harada. T.. Fujiwara. T., and Furuhashi. T.. Biochim. Bioplzy,t. Ado 230, 483 (1971). Wessler. E.. Bioc/7em. J. 112, 373 (1971). Tokunaga, M.. Futami. T.. Wakamatsu. E.. Endo. M.. and Yoshirawa. %.. ~‘lrrf. (‘i7im. ,Ac,~u 62, 89 (1975). Murata. K.. Exp~r-irntitr 38, 313 (1982). Saito. H., Yamagata, T.. and Suzuki. S., J. Bid. Cl7rnr. 243, IL36 11968). Murata. K.. in “Methods in Carbohydrate Chemistry” (R. L. Whistler and J. N. BeMiller Ed>.). Vol. 8, p. 81. Academic Press, New York, 1980. Ohya. T.. and Kaneko. Y.. Biochitn. Bi0phy.s. Ac,to 198. 607 I IY70). Murata. K.. Gerontology 29, 131 (1983). Murata, K., and Takeda. M.. C/in. Chin7. AL.ICI 108, 49 (I9XO). Yamagata, T.. Saito. H.. Habuchi. 0.. and Suzuki, S.. .I. Bid C‘hrm. 243. 1523 (lY6.X). Hiyama, K., and Okada. S., J. Bid. Chrw7. 250, 1824 t 197s). Bitter. J.. and Muir. H. M.. Anul. Biochru7. 4, 330 (1967). Murata, K.. Ogura. T.. and Okuyama, T.. C‘onn~c~t. Tirsrrc Kc\. 2, 101 (1974). Seno. N.. Anno. K., Kondo, K.. Nagase. S.. and Saito. S.. .drrtr/. Bioc~l7~~m. 37. 197 ( 19701. Wessler, E.. Ano/. Bioc~hrn7. 26. 439 f 1968). Moczar, E., Moczar, M.. Schillinger. G.. and Robert, L., J. (‘lrr~~tmt~~~~.. 31, .56l t 1967). Murata. K.. and Nakazawa. K.. Athe~o.scle~c,sis 25, 31 t 1976). Shimada, E., and Matsumura. G.. J. Biochem. 78. 513 (1975). Taniguchi. N.. Clir7. Cl7ir77. Ac,fcr 37, 725 (1972). Murata. K.. Yukiyama. Y.. and Horiuchi, Y.. Clirl. C’kirt7. Ac,rtr 49. 13Y (1973). Murata. K.. Ishikawa. T.. and Ninomiya. H.. Bioc~hc~n7. Med. 8, 472 lIY73). Allalouf, D.. and Ber. A.. Biochim. Bioph~.\. Acfrr 201, 61 I 1970). Murata. K.. Rrr7ol. P/~ysio/. 2. 346 ( 1979 I. Murata. K.. and Horiuchi, Y.. C/i/7. Clri777. Ac,frr 75, SY f lY771. Fraser. J. R. E.. Laurent. T. C.. Pertoft. H.. and Baxter-. E.. flioc~hc,tri. J. 200, 415 f 1981 1,