Free Radical Biology & Medicine, Vol. 13, pp. 205-210, 1992 Printed in the USA. All rights reserved.

0891-5849/92 $5.00 + .00 Copyright © 1992 PergamonPress Ltd.

Original Contribution THE SITE OF NONENZYMIC GLYCATION OF HUMAN EXTRACELLULAR-SUPEROXIDE DISMUTASE IN VITRO

TETSUO ADACHI, HIDEKI OHTA, KYOZO HAYASHI, KAZUYUKI HIRANO, and STEFAN L. MARKLUND* Department of Pharmaceutics, Gifu Pharmaceutical University, Gifu 502, Japan; and *Department of Clinical Chemistry, Umeh University Hospital, S-901 85 Ume& Sweden (Received 6 January 1992; Revised 12 March 1992; Accepted 31 March 1992)

Abstract--The secretory enzyme extracellular-superoxide dismutase (EC-SOD) has affinity for heparin and some other sulfated glycosaminoglycansand is in vivo bound to heparan sulfate proteoglycan. Nonenzymic glycation of EC-SOD, both in vivo and in vitro, is associated with a reduction in heparin affinity, whereas the enzymic activity is not affected. The glycation sites in EC-SOD are further studied in the present article. It is shown that modification of a few of the five lysyl residues of the subunits of the enzyme with trinitrobenzene sulfonic acid nearly abolishes the in vitro glycation susceptibility. From a chymotryptic digest of in vitro glycated EC-SOD, two peptides with affinity for boronate could be isolated. Amino acid sequence analysis showed that both encompassed the carboxyterminal end. ~-Glucitol lysine was identified in both peptides at positions 211 and 212. The primary glycation sites in EC-SOD are thus lysine-211 and lysine-212 in the putative heparin-binding domain in the carboxyterminal end. Keywords--Extracellular-superoxide dismutase, Nonenzymic glycation, Diabetes, Heparin, Free radicals

INTRODUCTION

Nonenzymic glycation of protein, a factor possibly contributing to the development of some complications in diabetes mellitus, has been extensively modeled by exposure of proteins to glucose in vitro, i Increased glycation and subsequent reactions of proteins are thought to be involved in not only structural but also functional changes in proteins. 2,3 Extracellular-superoxide dismutase (EC-SOD, EC 1.15. I. l), a secretory tetrameric Cu and Zn-containing glycoprotein,4'5 is the major isozyme in extracellular fluids such as plasma, l y m p h , 6 and synovial fluid 7 but also occurs in tissues. 8'9 EC-SOD is heterogeneous with regard to heparin affinity and can be divided into three fractions: A, which lacks affinity; B, with intermediate affinity; and C, with high affinity.l°'ll The C fraction forms an equilibrium between the extracellular fluid phase and heparan sulfate proteoglycan in the glycocalyx of cell surfaces.l°-13 Previous studies in our laboratory ~4indicated that the proportion of gly-

cated EC-SOD in serum of diabetic patients was significantly higher than in normal subjects. Of the subfractions, EC-SOD B was by far most glycated followed by EC-SOD A. EC-SOD C was glycated only to a minor extent. Recombinant EC-SOD C (r-EC-SOD C) could also be glycated in vitro. 14 The enzymic activity was not affected in glycated EC-SOD, but the high heparin affinity was lost in about half of the glycated preparation. We speculated that the glycation sites were localized rather far away from the active site and may occur on lysines in the putative heparin-binding domain in the carboxyterminal end of the enzyme. In the present investigation, the glycated sites in glycated EC-SOD in vitro are identified by the determination of amino acid sequences of chymotryptic peptides which have affinity for boronate column. MATERIALS AND METHODS

Materials

Human r-EC-SOD C (courtesy SYMBICOM AB, Umefi, Sweden) was prepared as previously described. 4 Glyco-gel B (boronate affinity gel) was pur-

Address correspondence to: Tetsuo Adachi, PhD, Department of Pharmaceutics, Gifu Pharmaceutical University, 5-6-1 Mitahorahigashi, Gifu 502, Japan. 205

206

T. ADACHIel al.

chased from Pierce (Rockford, IL). TLCK-chymotrypsin (type VII) was obtained from Sigma Chemical Co. (St. Louis, MO). o-glucose, trinitrobenzene sulfonic acid (TNBS), and poly-L-lysine hydrobromide were obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). All other chemicals used were of analytical grade.

TNBS modfhcation of r-EC-SOD C A two-volume portion of 3 mM TNBS in 0.2 M NaHCO3 was added to one volume of r-EC-SOD C solution (450 #g/mL) in distilled water and the mixture was incubated at 25°C. At indicated times, aliquots (80 uL) were removed and mixed with 720 uL of ice-cold 50 mM Tris-HC1, pH 7.5, containing 0.1 M NaC1 (to stop the reaction~5). For lysyl modification determination, the absorbance at 420 nm of final reaction mixture was assayedJ 6 Modified r-EC-SOD C was applied to Sephadex G-25 column ( 1 × 60 cm) equilibrated with 0.1 M potassium phosphate, pH 7.3, to remove excess TNBS.

Nonenzymic glycation of TNBS-modflied r-EC-SOD C in vitro Three milliliters of TNBS-modified r-EC-SOD C (about 3 ug/mL) was incubated with 100 mM of Dglucose in 0.1 M potassium phosphate, pH 7.3, at 37°C for 1 week under sterile conditions, followed by extensive dialysis against 25 mM potassium phosphate, pH 7.5.

Boronate column (Glyco-gel B column) chromatography The samples were applied to a Glyco-gel B column (vol = 0.5 mL) equilibrated with 1 M sodium acetate, pH 8.5, containing 10 ug/mL bovine serum albumin (BSA) and washed with the same buffer. The bound fraction was eluted with 1 M sodium acetate, pH 5.5, containing 10 ~g/mL BSA. EC-SOD concentration in each fraction was assayed by enzyme-linked immunosorbent assay (ELISA)J 4

Chymotrypsin digestion of in vitro glycated r-EC-SOD C One milliliter of r-EC-SOD C (4.5 mg/mL) was nonenzymically glycated by incubation with 100 mM D-glucose in 0.1 M potassium phosphate, pH 7.3, at 37°C for 1 week under sterile conditions. The reaction product was dialyzed against 10 mM sodium acetate, pH 6.0, followed by rotary evaporation to dry-

ness. The residue was dissolved in 1 mL of 0.5 M Tris-HCl, pH 8.5, containing 7 M guanidinium HC1 and 10 mM ethylenediaminetetraacetic acid (EDTA), and flushed with nitrogen. This sample was thereafter reduced with 4 mg of dithiothreitol, flushed with nitrogen, and incubated for 2 h at room temperature. After the addition ofiodoacetic acid (16 mg), the reaction mixture was incubated in the dark for I h, tbllowed by the extensive dialysis against 10 mM sodium acetate, pH 6.0. After rotary evaporation, the residue was dissolved in 200 uL of I0 mM Tris-HCl, pH 8.6, and then it was digested with 10 uk of TLCK-chvmotrypsin (1 mg/mL of 1 mM HCI) at 30°C for 1 h.

Isolation ~?[in vitro gl),valed peplides Separation of glycated from unmodified peptides was achieved by boronate column (Glyco-gel B column) chromatography method. ~7,~s The chymotryptic digest of glycated r-EC-SOD C was applied to a Glyeo-gel B column (vol - l mL) equilibrated with 0. l M sodium acetate, pH 8.5. After washing with the same buffer, the bound peptides were eluted with 0. l M sodium acetate, pH 8.5, containing 0.1 M sorbitol, and they were then lyophilized. In vitro glycated peptides were separated by high-pressure liquid chromatography (HPLC) on a C I 8 reverse-phase column (ubondasphere 5~, Nihon Waters Ltd., Tokyo, Japan) operated at a flow rate of 0.7 mL/min. A gradient system formed between solvent A (0.1% trifluoroacetic acid in Milli Q water) and solvent B (0.1% trifluoroacetic acid in acetonitrile) was used. Peptides were eluted with a linear gradient up to 60% of solvent B for 60 rain, and continuously up to 100% of solvent B for 10 min. The peptides were detected on the basis of the absorbance at 210 nm.

Preparation of glycated poly-L-lysine Glycated poly-L-lysine was prepared method of Garlick and Mazer. ~7

by the

Amino acid sequence determination Amino acid sequences were determined with an Applied Biosystems Protein Sequencer Model 473A (Foster City, CA). The retention time of ~-glucitol lysine was identified by assay of glycated poly-L-lysine. RESULTS

Effect of TNBS mod~lication on in vitro glycation ~?[ r-EC-SOD C The r-EC-SOD C subunit contains five lysine residues, of which three occur in the carboxyterminal

Glycated site of EC-SOD

100

of:

8O

So

o~

5 4.0e

tryptic digest is shown, whereas Figure 2B depicts the chromatogram of material bound to and desorbed from the Glyco-gel B column. Two peptides, termed A-1 and A-2, were recovered in the Glyco-gel-bound fraction, and their amino acid sequences were assayed. In accord with the primary structure of r-ECSOD C, 19 their peptides correspond to residues Ser208-Ala-222 (A-l) and Glu-201-Ala-222 (A-2) (Fig. 3). In both peptides, an unknown peak was detected in addition to lysine in cycles corresponding to positions 211 and 212 (Fig. 4A). This peak was identified as ~-glucitol lysine by the assay of glycated poly-L-lysine (Fig. 4C). At position 220, (-glucitol lysine was not detected (Fig. 4B).

~

6o i/o

_~o 20 0

0 ~-

0

10

20

30

40

50

207

60

Time (min) Fig. 1. Effect of TNBS modification on in vitro glycation of r-ECSOD C. r-EC-SOD was exposed to an excess of TNBS as described in Materials and Methods. After indicated times, aliquots were removed for analysis of the number of modified lysyl residues (O). TNBS-modified r-EC-SOD C were glycated in vitro, followed by assay of glycated form with Glyco-gel B column chromatography (e). The proportions of glycated form are presented as % vs. control, which was not incubated with TNBS.

DISCUSSION

Recently, we reported that r-EC-SOD C glycated in vitro lost its high affinity for heparin, whereas the enzymic activity was retained. '4 This result suggested that the heparin-binding domain of the enzyme should be a glycation site. In addition, we have shown previously that the modification of only a few lysine residues per subunit with TNBS causes loss ofheparin affinity without affecting the enzymic activity. 2° In the present investigation, we show that the subtle modification of only a few lysyl residues is sufficient to decrease significantly the in vitro glycation susceptibility, similar to the loss of high heparin-binding ability. 2° These results suggest that heparin, glucose, and TNBS primarily react with the same lysine residues. The C-terminal end of r-EC-SOD C is highly hydrophilic, forms an a-helix according to computer pre-

end. r-EC-SOD C exposed to 100 mM glucose for a week was glycated, and about 20% of r-EC-SOD C was bound to Glyco-gel B column (similar to our previous resultl4). Pretreatment with the lysine-modifying reagent TNBS reduced the glycation, and after modification of a few residues the susceptibility to glucose was mostly eliminated (Fig. 1).

Identification of glycated sites in EC-SOD The chymotryptic digestion of in vitro glycated rEC-SOD C was separated by reverse-phase HPLC. In Figure 2A, the chromatogram of the whole chymo-

A

B

i 0

10

20

3~0

460

I

I

50 0

1'0

i,

t

,o

,',

Retention time (rain) Fig. 2. Elution profiles of chymotryptic digested peptides of glycated r-EC-SOD C in vitro on reverse-phase HPLC. (A) An aliquot of chymotryptic digested peptides of glycated r-EC-SOD C was applied to the C I 8-reverse-phase HPLC column. (B) Chymotryptic digested peptides bound to Glyco-gel B were eluted and followed by application on the reverse-phase HPLC column.

208

T. ADACHIel al. 1

10

20

30

40

WTGEDSAEPNSDSAEWIRDMYAKVTEIWQEVMQRRDDDGT A 50

60

70

80

LHAACQVQPSATLDAAQPRVTGVVLFRQLAPRAKLDAFFA A 90

100

110

120

LEGFPTEPNSSSRAIHVHQFGDLSQGCESTGPHYNPLAVP 130

140

150

160

HPQHPGDFGNFAVRDGSLWRYRAGLAASLAGPHSIVGRAV 170

180

190

200

VVHAGEDDLGRGGNQASVENGNAGRRLACCVVGVCGPGLW 210

220

ERQAREH,SERKKRRRESECKAA, ,

AA

'

I !

I I

A

,=

, !

I

, I |

A-1

=°!

A-2

=I

I

I

I~

Fig. 3. Amino acid sequence of r-EC-SOD C and glycation sites in vitro. Peptides A-I and A-2 were the chymot~ptic digested peptides recovered ~om the bound ~action on Glyeo-gel B column (Fig. 2B). Lysyl residues are indicated by triangles.

diction (S. L. Marklund, unpublished data), and should extend into the solvent. The three lysines localized there should be easily accessible for the reagents and are likely candidates for early modification, r-ECSOD C digested with trypsin or lysylendoproteinase, which lost three or one lysine per subunit in the C-terminal end, had no or weak affinities for heparinHPLC column, respectively (Adachi et al., in preparation). This suggests that the heparin-binding site pc-

A

curs on the cluster of basic amino acids in the carboxyterminal end. In this article, we have shown the predominant site of nonenzymic glycation of r-EC-SOD C in vitro. Boronate affinity column method is widely used for clinical and biochemical research to separate glycated from unmodified protein. Garlick and Mazer 17 and Watkins et al. 18 used this method to isolate the glycated peptides. The sequence analyses of isolated chy-

C

B

Lys

DPTU

Lys

DPTU ¢-glucitol-Lys

-gluciloI-Lys

Lys

DPTU I

I

0

I

I

10

20

I

I

30 0

I

I

I

I

I

I

10

20

3C

10

20

3(

Retention

time

(min)

Fig. 4. Sequence analyses of isolated chymotryptic digested peptides and glycated poly-L-lysine.(A) Cycle 12 from peptide A-2, which corresponds to Lys-212. (B) Cycle 20 from peptide A-2, which corresponds to Lys-220. (C) Cycle 2 from glycated poly-L-lysine prepared as described in Materials and Methods. Dimethylphenylthiourea (DMPTU) and diphenylthiourea (DPTU) are byproducts in amino acid sequence analysis.

Glycated site of EC-SOD

motryptic peptides, A-1 and A-2, showed that both encompassed the carboxyterminal end and had been glycated, because ~-glucitol lysine was detected at positions 211 and 212 (Fig. 4A) but not at position 220 (Fig. 4B). However, it is impossible to estimate the ratios of ~-glucitol lysine to unmodified lysine from peak height, because the recoveries of amino acids are various. Lys-21 1 a n d Lys-212 seem to be glycated alm o s t equally. T h e presence o f u n m o d i f i e d lysine besides ~-glucitol lysine at positions 21 1 and 212 m a y show that the peptides glycated at Lys-21 1 or Lys-212 have affinity for b o r o n a t e c o l u m n . T h e two lysine residues, Lys-21 1 and Lys-212, are enclosed with four arginine residues, Arg-Lys-Lys-Arg-Arg-Arg, and constitute a positively charged cluster. It has been reported that lysine residues located in a sequence o f basic a m i n o acids are m o r e likely to be glycated. 2~ T h e authors hypothesized that appropriately located positively charged a m i n o acid residues could afford local acid-base catalysis o f the A m a d o r i r e a r r a n g e m e n t to f o r m a stable, practically irreversible k e t o a m i n e structure in the course o f in vitro glycation. 22 O n the other hand, Lys-220 is adjacent to neutral a m i n o acids, Cys-219 and Ala-221, which m a y not stimulate the a f o r e m e n t i o n e d reaction. W e could not detect any glycated c h y m o t r y p t i c peptides containing the other two lysine residues, Lys23 and Lys-74. T h e reasons m a y be that they are less distinctly exteriorized in the protein and that their vicinities are less basic (S. L. M a r k l u n d , unpublished data). However, the b a c k g r o u n d to the heterogeneity in heparin affinity o f p l a s m a E C - S O D is still unresolved; E C - S O D is primarily synthesized as type C in the body. 4'23 T h e binding o f E C - S O D C to endothelial cellular surfaces might be an especially efficient way o f protecting cells against external superoxide anion, rE C - S O D C has been shown to be m o r e efficient t h a n c o p p e r , z i n c - S O D ( C u , Z n - S O D ) in diminishing ischemia-reperfusion d a m a g e in the h a m s t e r cheek pouch. 24 Moreover, it has been reported that heparinbinding S O D (HB-SOD), a fusion gene p r o d u c t consisting h u m a n C u , Z n - S O D and a C - t e r m i n a l basic peptide with high affinity for heparin, 25 significantly suppressed p a w e d e m a at a m a r k e d l y lower close t h a n that o f C u , Z n - S O D . 26 G l y c a t i o n o f relatively few a m i n o groups in proteins m a y have p r o n o u n c e d effects on their functions. 2'3'~4Although the glycation o f a protein is a slow reaction, glucose functions as a site-specific reagent. In E C - S O D C, glycation decreases the affinity o f the e n z y m e for heparin.14 T h e in vivo consequence o f the reduced heparin affinity is weakened, binding to heparan sulfate proteoglycan in the glycocalyx o f e n d o t h e -

209

lial cells 12and other cell types.13 The glycation of ECSOD is increased in diabetes,14 and one may speculate that reduced protection of the exterior of cells against superoxide radicals may contribute to vascular and other complications associated with diabetes. Acknowledgement - - This work was supported in part by grants-in-

aid for the encouragement of young scientists from the Ministry of Education, Science, and Culture of Japan.

REFERENCES 1. Cerami, A. Aging of proteins and nucleic acids; what is the role of glucose? Trends Biochem. Sci. 11:311-314; 1986.

2. Ahmed, M. U.; Thorpe, S. R.; Baynes, J. W. Identification of N'-carboxymethyllysine as a degradation product of fructoselysine in glycated protein. Z Biol. Chem. 261:4889-4894; 1986. 3. Arai, K.; lizuka, S.; Tada, Y.; Oikawa, K.; Taniguchi, N. Increase in the glucosylated form oferythrocyte Cu-Zn-superoxide dismutase in diabetes and close association of the nonenzymatic glucosylation with the enzyme activity. Biochim. Biophys. Acta 924:292-296; 1987. 4. Tibell, L.; Hjalmarsson, K.; Edlund, T.; Skogman, G.; EngstrOm, A.; Marklund, S. L. Expression of human extracellular superoxide dismutase in Chinese hamster ovary cells and characterization of the product. Proc. Natl. Acad. Sci. USA 84:6634-6638; 1987. 5. Marklund, S. L. Human copper-containing superoxide dismutase of high molecular weight. Proc. Natl. Acad. Sci. USA 79:7634-7638; 1982. 6. Marklund, S. L.; Holme, E.; Hellner, L. Superoxide dismutase in extraceUular fluids. Clin. Chim. Acta 126:41-5 l; 1982. 7. Marklund S. L.; Bjelle, A.; Elmqvist, L. G. Superoxide dismutase isozymes of the synovial fluid in rheumatoid arthritis and in reactive arthritides. Ann. Rheum. Dis. 45:847-851; 1986. 8. Marklund S. L. Extracellular superoxide dismutase in human tissues and human cell lines. Z Clin. Invest. 74:1398-1403; 1984. 9. Marklund S. L. Extracellular superoxide dismutase and other superoxide dismutase isoenzymes in tissues from nine mammalian species. Biochem. Z 222:649-655; 1984. 10. Karlsson K.; Marklund, S. L. Heparin-induced release ofextracellular superoxide dismutase to human blood plasma. Biochem. J. 242:55-59; 1987. ll. Karlsson K.; Marklund, S. L. Extracellular superoxide dismutase in the vascular system of mammals. Biochem. J. 255:223228; 1988. 12. Karlsson K.; Marklund, S. L. Plasma clearance of human extracellular-superoxide dismutase C in rabbits. J. Clin. Invest. 82:762-766; 1988. 13. Karlsson K.; Marklund, S. L. Binding of human extracellularsuperoxide dismutase C to cultured cell lines and to blood cells. Lab. Invest. 60:659-666; 1989. 14. Adachi, T.; Ohta, H.; Hirano, K.; Hayashi, K.; Marklund, S. L. Non-enzymic glycation of human extracellular superoxide dismutase. Biochem. J. 279:263-267; 1991. 15. Jorgensen, A. M.; Borders, C. L., Jr.; Fish, W. W. Arginine residues are critical for the heparin-cofactor activity of antithrombin IIl. Biochem. ,L 231:59-63; 1985. 16. Liu, C.-S.; Chang, J.-Y. The heparin binding site of human antithrombin III. J. Biol. Chem. 262:17356-17361; 1987. 17. Garlick, R. L.; Mazer, J. S. The principle site of nonenzymatic glycosylation of human serum albumin in vivo. J. Biol. Chem. 258:6142-6146; 1983. 18. Watkins, N. G.; Thorpe, S. R.; Baynes, J. W. Glycation of amino groups in protein. Studies on the specificity of modification of RNase by glucose. J. Biol. Chem. 260:10629-10636; 1985.

210

rl. ADACHI el al.

19, Hjalmarsson, K.; Marklund, S. L.; Engst6m, A.; Edlund, T. Isolation and sequence of complementary DNA encoding human extracellular superoxide dismutase. Proc. Natl. Acad. &'i USA 84:6340-6344: 1987. 20, Adachi, T.; Marklund, S, L. Interactions between human extracellular superoxide dismutase C and sulfated polysaccharides. J, Biol. Chem. 264:8537-8541: 1989. 21, Iberg, N.; Fltickiger, R. Nonenzymatic glycosylation of albumin in vivo. J. Biol. Chem. 261:13542-13545: 1986. 22, lsbell, H. S.; Frush, H. L. Mutarotation, hydrolysis, and rearrangement reaction ofglycosylamines. J. Org. Chem. 23:13091319" 1958. 23. Marklund, S. L. Expression ofextracellular superoxide dismutase by human cell lines. Biochem. J. 266:213-219: 1990. 24. Johnasson, M.; Deinum, J.: Marklund, S. L.: SjOquist, P. O. Recombinant human extracellular superoxide dismutase reduces concentration of oxygen free radicals in the reperfused rat heart. Cardiovasc. Res. 24:500-503, 1990. 25. Inoue, M., Watanabe, N.: Matsuno, K., Sasaki, J.: Tanaka, Y."

Hatanaka, H.: Amachi, T. Expression of a hybrid Cu/Zn-typc superoxide dismutase which has high affinity for heparin-like proteoglycans on vascular endothelial cells..1. BioL Chem. 266:16409-16414; 199 l. 26. Oyanagui, Y.; Sato, S., lnoue, M, Inhibition ofcarrageenan-induced paw edema by superoxide dismutase that binds to heparan sulfated on vascular endothelial cells. Biochem. Pharmacol. 42:991-995: 1991. ABBREVIATIONS

BSA--bovine serum albumin Cu,Zn-SOD--copper,zinc-superoxide dismutase EC-SOD--extracellular-superoxide dismutase ELISA--enzyme-linked immunosorbent assay HB-SOD--heparin-binding superoxide dismutase TNBS--trinitrobenzene sulfonic acid