Age-related changes in human skin collagen galactosyltransferase and collagen glucosyltransferase activities

95

Clinica Chimica Acta, 76 (1977) 95-101 @ Elsevier/North-Holland Biomedical Press

CCA 8414

AGE-RELATED CHANGES IN HUMAN SKIN COLLAGEN GALACTOSYLTRANSFERASE AND COLLAGEN GLUCOSYLTRANSFERASE ACTIVITIES

HENRIK

ANTTINEN,

Department (Received

AARNE

OIKARINEN

of Medical Biochemistry, October

28th,

and KARI

University

I. KIVIRIKKO

*

of Oulu, Oulu (Finland)

1976)

Summary Collagen galactosyltransferase and collagen glucosyltransferase activities were assayed in human skin specimens of about 100 mg wet weight. The assay of the glucosyltransferase activity was found to be highly specific. The assay of the galactosyltransferase activity was somewhat less specific, but there was no difference in specificity between the foetal and adult human skin samples. The activities of the two collagen glycosyltransferases in human skin extract were found to vary with age, being highest in foetal skin, and higher in the skin of young children than in that of adults. The galactosyltransferase and glucosyltransferase activities in foetal skin were respectively about 4 times and 6 times those in adult skin. The magnitudes of the changes with age in the two collagen glycosyltransferase activities were smaller than those occurring in the activities of the two other intracellular enzymes of collagen biosynthesis, namely prolyl and lysyl hydroxylase. This difference suggests that the four intracellular enzyme activities of collagen biosynthesis are not regulated in an identical manner.

Introduction The biosynthesis of collagen involves a number of post-translational modifications of the initial polypeptide chains (for reviews, see refs. l-3). The intracellular modifications include the hydroxylation of certain prolyl and lysyl residues and the glycosylation of certain hydroxylysyl residues. These reactions are catalyzed by four separate enzymes : prolyl hydroxylase, lysyl hydroxylase, collagen galactosyltransferase and collagen glucosyltransferase. Collagen galac* Correspondence try, University

should be addressed to: Prof. Kari 1. Kivirikko, Department of Oulu, Kajaanintie 52 A. SF-90220 Oulu 22. Finland.

of Medical Biochemis-

tosyltransferase catalyzes thrl transfer of galactose to hydroxylysine and c,~llagen glucosyltransferase that of glucosc~ to galactosylhydroxylysine [ 3,5]. Th(J sugars are derived from UDP-galactosr> and UDP-glucose, and both reactions require the presence of manganf’sr’ or certain other bivalrnt cations [ lm-8 1. Both enzymes are probably located within the cisterna of the cndoplasmic reticulum of cells synthesizing collagc~n [ S--l3 I. Gli~cosyltransferascl has rtlct’ntly been purified to homogeneity 1141 and galactosylt,ransf’~~rase about lOOO-fold [ 151 from chick embryo extract. Several properties of these enzymes and reactions have been studied [4,5,7,8,11--181, including the mechanism of thcl collagen glucosylation reaction [ 17 1 and factors affecting the rate and (axtent of these reactions [8,11-13,18 J. A number of studies have dt>alt with changes in plolyl hydroxylasc> activity in animal and human tissues under various experimental and pathological conditions [l-3], whereas little is known about changes occurring in the two collagen glycosyltransferase activities. In particular, no study has been rcportcd on changes in thesta enzyme activities in human t,issues. 11ge-dependent variations have been reported in prolyl [ 19,20 ] and lysyl [ 211 hydroxylase activities in human skin, and some studies suggc’st that agerelated changes also take place in the two collagen glycosyltransfclrasc activities in animal tissues [ 22-241. In the present work collagen galactosyltransferas(s and collagen glucosyltransferase activities were assayed in human skin samples. The reliability of the assay methods developed for animal tissues was studicld with human skin extracts, and age-related changes in the two enzyme activities were measured. Material and methods Material Skin specimens, minimum wet weight about 80 mg, were ohtaincd cith(>r at surgical operations on patients without any active disease that could be t>xpected to affect collagen metabolism or from foetuses obtained in legal abortions by abdominal hysterotomies. The specimens were stored at -20°C until used. UDP-[ ‘%I galactose and UDP-[ 14C] glucose were purchased from NPW England Nuclear Corp., l3oston, Mass., U.S.A. and the unlabc~ll~~d compounds from Sigma Chemical Co., St. Louis, MO., U.S.A. The source’s of th(l other matc)rials were the same as specified clscwhere 16,7,16], Preparation of skin samples All procedures were carried out at 0--4°C. The skin specimens were freed from the subcutaneous tissue. The specimens were minced with scissors for 5 min in a solution consisting of 0.2 M NaCl, 50 PM dithiothreitol, 0.1% (w/v) Triton X-100 and 20 mM Tris/HCl buffer, adjusted to pH 7.5 at 4°C (400 ~1 of solution per 100 mg of skin) [7]. The minced specimens were then homogenized for 5 set with an Ultra-Turrax homogenizer, and the homogenate incubated for 60 min at 4°C and then centrifuged at 15 000 X g for 30 min. Aliyuots of the supernatant were used for assays of the enzyme activities anal protein content.

97

Assays In most experiments collagen galactosyltransferase activity was assayed in a final volume of 100 ~1 containing lo-40 ~1 of the 15 000 X g supernatant of the skin homogenate, 40 mg of calf skin gelatin substrate/ml, 32 I_IM UDP-[ 14C]galactose (23 Ci/mol), 10 mM MnCl*, 1 mM dithiothreitol and 50 mM Tris/ HCl buffer adjusted to pH 7.4 at 20°C [6,8]. The assay mixture for collagen glucosyltransferase was similar, except that 62 I.IM UDP-[ “C]glucose (13 Ci/mol) was used [6,7]. The gelatin substrate was heated to 100°C for 10 min and rapidly cooled to 0°C immediately before addition to the incubation mixture [ 8,111. The samples were incubated at 37°C for 45 min, and the reactions terminated by adding 2 ml of 1% (w/v) phosphotungstic acid in 0.5 M HCl. The products were purified and assayed as reported previously [6], including the ion-exchange chromatography step, except that the last step in the purification of the reaction products, paper electrophoresis, was omitted [ 13,241. The assays were performed in duplicates or in triplicates. One unit of collagen galactosyltransferase or collagen glucosyltransferase activity was defined as the activity required to synthesize an amount of the radioactive product in dpm corresponding to 1 pmol in 1 h at 37°C under conditions in which a saturating concentration of denatured purified citrate-soluble collagen is used as a substrate and the concentrations of the other reactants are those indicated previously [ 7,8]. The protein content of the 15 000 Xg supernatant was assayed [25] using bovine serum albumin as standard. All “C-radioactivity counting was performed in a Wallac liquid-scintillation spectrometer with an efficiency of 80% and a background of 25 cpm using the scintillants reported in the assay procedures [6,12]. Results Specificity and linearity of the assays with human skin extract The assay procedures for collagen galactosyltransferase and collagen glucosyltransferase activities described by MyllylS et al. [6] include purification of the reaction products in several steps, including brief ion-exchange chromatography and, as the last step, paper electrophoresis. Since this last step limits the number of assays which can be performed in one day, and since the analysis of radioactivity in several fractions which are then summed introduces some inaccuracy, a test was performed to ascertain whether the assay of collagen glycosyltransferase activities in human skin extract can be achieved without it [ 13,241. The specificity of assays carried out omitting the paper electrophoresis was studied using an amino acid analyzer (Jeol, JLC-5AH, 63 cm column) and a program which clearly separates the hydroxylysine glycosides [ 12,261, fractions being collected every 2 min. Radioactivity was measured in a liquidscintillation counter [ 121. The specificity of the assay in extract of foetal human skin was found to be about 75% for collagen galactosyltransferase and 95% for collagen glucosyltransferase activity (Fig. l), similar figures also being obtained with samples of adult human skin. It thus seems that the procedures can be used for the study of age-dependent variation in the enzyme activities omitting the paper electrophoresis.

20

I

*

15

10

P. 0 x

5

0 B

: D 40 -:;:

R

30 20 10 0 :-.:. 0

-L 40

uu TIME

160 120 (~-II”)

200

Fig. 1. Specificity of assays of collagen galactosyltransferase (A) and glucosyltransferase (B) activities in human skin extract. The products of the glycosylation reactions were chromatographed in an amino acid analyzer, fractions were collected every 2 min. and the radioactivity was measured. The dashed lines represent the elution positions of the glucosylgalactosylhydroxylysine (first dashed line, at 54 min) and galactosylhydroxylysine standards (second dashed line, at 89 min).

The formation of products of both reactions amount of the skin extract added (Fig. 2).

was directly

dependent

on the

Age-dependent changes in collagen glycosyltransferase activities The activities of both enzymes were highest in foetal skin and higher in the

0

10

20 SKIN

30

EXTRACT

40 ( ~1)

50

0

0

20

10 SKIN

30

EXTRACT

co

50

(~1)

Fig. 2. Linearity of the assays of collagen galactosyltransferase (A) and glucosyltranferase (B) activities human skin extract. The formation of products was measured as dpm of [ 14C]galactosylhydroxylysine (A) or [‘4Clglucosylgalactosylhydroxylysine (B).

in

99

60 (foctal months)

AGE

(years

)

60

0

36 0 ( foetal

20

40 AGE

(years

60

60

)

months)

Fig. 3. Effect of age on the activity of collagen galactosyltransferase activity is expressed as units per g of skin extract protein.

in human skin extract.

The enzyme

Fig. 4. Effect of age on the activity of collagen glucosyltransferase activity is expressed as units per g of skin extract protein.

in human skin extract.

The enzyme

skin of 0 to 1 year-old children than in adult skin (Figs. 3 and 4). The mean value for collagen galactosyltransferase activity was 6.55 U/g skin extract protein in four foetuses of 3 to 4 months gestation, 3.70 U/g in three children aged 0 to 1 year, 1.92 U/g in five children aged 7 to 14 years, and 1.60 U/g in fifteen adult subjects aged 20 to 78 years (Fig. 3). The corresponding values for the glucosyltransferase activity were 2.77, 1.29, 0.40 and 0.45 U/g (Fig. 4). The values are expressed in Figs. 3 and 4 in relation to extract protein, but the effect of age was similar when the values were expressed in relation to wet weight of skin sample. Discussion Collagen is subject to many age-related changes occurring not only in the rates of collagen deposition and removal, but also in the ratios of the various types of collagen in the tissues and the structure of the protein itself [2]. Foetal human skin has a much higher ratio of type III collagen to type I collagen than adult skin [27], and it has been suggested that this difference is due to a decline in the synthesis of type III collagen before birth [28]. The level of hydroxylation of lysyl residues in type I collagen decreases with age in several tissues [ 291, and recently age-related variations have been reported in the level of glycosylation of hydroxylysyl residues in human skin collagen [30]. Some other molecular properties of collagen have also been reported to change with age, particularly the amount and nature of the cross-links [31]. It has been proposed that such age-related alterations may play an aetiological role in the disturbances which accompany the physiological ageing process [2]. It seems that physiological changes with age can be used as a model in studying certain general responses and adaptations of collagen biosynthesis. The biological significance of the level of collagen glycosylation is not com-

pletely understood, but it may affect collagen fibro formation 11.21. The extent of collagen glycosylation seems to be regulatebd by the rate of triplt> helix formation [8,11,12,18], and by the activities of the collag,rc‘n glycosyltransferascs present in the tissues. A method is proposed hcrc for the assay of collagen gala~tos3lltransft~r~lse and glucosyltransferase activities in human skin specimens. The assay of the glucosyltransferase activity was found to bc highly specific, and while that of the galactosyltransferase activity was less so there was at least no diffcrcncc~ in specificity between the results from the foetal and adult skin samples. llowever, the specificity of the assay of galactosyltransfc~rnst activity should bc studied in instances, when the samples differ in tht>ir physyiological, clinical or experimental properties. The activities of the two collagen glycosyltransferases in human skin were found to vary with age. They were highest in foetal skin and higher in the skin of young children than in that of adults. Age-related changes in the two other intracellular enzymes catalyzing post-translational modifications in collagen biosynthesis, namely prolyl and lysyl hydroxylases, have been studied prcviously in human skin [19-211. The activities of all these four enzymes decline with age, but in differing proportions. The activity in foctal skin is about ‘10 times that in adult skin in the case of lysyl hydroxylase [al], about 10 times in that of prolyl hydroxylase [19,20], about 6 times for collagen glucosyltransferase and about 4 times for collagen galactosyltransferase. It has been reported recently that the two collagen hydroxylase activities also decrease more with age than the two collagen glycosyltransferase activities in rat liver [ 241, although the magnitudes of all the changes were smaller than those found in the human skin. The age-related changes in the ratios between the four enzyme activities indicate that the decreases with age cannot be explained simply by a decline in the number of collagen producing cells in human skin, but a certain regulation of the enzyme activities must exist to enable certain enzyme activities to decrease more than others. Acknowledgements This work was supported in part by Grants from the Medical Research Council of the Academy of Finland. The authors gratefully acknowledge the expert technical assistance of Miss Raija Leinonen and Mrs Raija Harju. They are also grateful to Dr. Johanncs Anttinen, M.D., Chief Surgeon at the Iisalmi District Hospital, Professor Pentti Jarvinen, M.D., of the Department of Obstetrics and Gynaecology, University of Oulu, and Associate Professor Erkki Heikkinen, M.D., of the Department of Surgery, University of Oulu, for permission to obtain the skin specimens. References 1

Bornstein,

P. (1974)

2

Kivirikko,

K.I.

3

Prockop. dran.

4

Spiro.

G.N. R.G.

D.J.. and and

and

Annu. Risteli.

Berg, Reddi, Spiro.

R.A.,

Rev.

Biochem.

L. (1976)

Med.

Kivirikko,

A.H..

cds.),

M.J.

(1971)

K.I.

pp.

43.

567

Viol. and

163-273,

J. Biol.

Chem.

-603

54,

159-186

Uitto,

J. (1976)

Plenum 246,

Press, 4899-4909

Biochemistry New

York

of Collagen

(Ramachan-

5

Spiro.

6

hlvllyl&

M.J. R.,

and

Risteli.

L. and

Kivirikko.

K.1.

(1975)

Eur.

.J. Biochem.

52,

,101

7

Mvllyl;i,

R.,

Risteli,

I,. and

Kivirikko,

K.I.

(1976)

Eur.

,J. Biochem.

61.

5947

8

Risteli.

Myllyl8’.

R. and

K.I.

(1976)

Biochem.

9

Brownell,

A.(;.

10

JIarwood.

R..

L..

11

.Myllyl%,

12

Oikarincn.

R.,

13

Anttinrn.

14

Myllyli,

Spiro.

and

R.(i.

Vris,

(;rant,

If. R.,

L. and

.Jackson,

and

Biochem.

K.J.

I,.

(1975)

4910

J.

Res.

Eur.

K.I.

~4918

155.

29-

35

and

Kivlrlkko.

J.

(1977)

371-m377

291-302

58.

Biochrm. K.I.

63,

.I. 152.

.J. Biochrm.

(1976)

410

145-153

Commun.

Binchem.

(1975)

Kivirikko.

Risteli,

246.

Diophys.

D.S.

.J. 160.

H.,

Chem.

Biorhem.

K1virikko.

Ii.

Anttinrn,

Biol.

Kivirikko. and

Anttinen, (1976)

J.

A. (1975)

M.E.

Risteli. A.,

(1971)

517-m521

156.

545

Hiochim.

--551 Biophvs.

Acta

480.

113

121 15

Ristrli.

16

Antlinen.

L..

17

blyllyla.

18

Oikarincn.

19

Ditto.

J.,

Invest.

23,

hlyllyl~, II. and

R.

R. and

(197fi) A.,

Ki\,irikk<,,

Ki\,irikko, Eur.

II. and

J..

K.I.

(1976)

,J. Biochcm.

Anttincn.

Ilalme.

K.I.

(1976)

70.

.J. Biochvm.

Biophys.

67.

Acta

429.

197-202 750

-75X

225-231

Kivirikko.

IIannuksela,

Eur.

Blochim. K.I.

(1976)

iX1.. Prltokallio.

Biochem. P.

and

.I. 160,

Kivirikko.

839445

K.I.

(1969)

Stand.

J.

Clin.

241-247

20

Tudrrman.

21

Anttincn,

L. and

22

Spiro.

23

Iienkel,

24

Risteli,

J.

25

Lowry.

D.H.,

26

Askenasi,

II.. R.G.

and

W. and and

Kivirikko.

Orava.

S..

Spiro.

Epstein.

28

Shuttleworth,

E.II.

29

Barnes,

30

Murai.

A.,

31

Bailey.

A.,J.,

E. K.I.

Rosebrough, J. and

Constable,

Miyahara. Robins,

S.P.

and

Blol. Hoppe A.L. 249.

Shiorawa, Balian,

Invest., K.1.

246,

in press

(1973)

%. Physiol.

J.

361-367

158,

Randall,

304.

Clin.

Chim.

Arta

47,

298--294

4919~~4925

Sryler’s and

Acta

R.J.

Chem. (1951)

356, .J. Biol.

921--928 Chpm.

199,

265-275

375-383

3225-3231

L. (1975) Morton,

Clin.

Chem.

Biochrm.

Far,

J.

Kivirikko.

(1975)

Chem.

Forrest,

T. and

J.

Biophys.

Biol. B.J.,

Eur.

L. and

(1976)

N.J.,

Biochim.

(1974)

(1977)

(1971)

Buddecke,

C.A.

M.-J.,

M.J.

Kivirikko,

H. (1973)

27

K.l.

Ryhinen,

Eur.

L.F.

and

S. (1975) G.

(1974)

.I. Biochem.

55,

Royce.

(1974)

Biochim. Nature

P.M.

391--395

Biophys. (London)

Biochem. Acta 251.

404.

105-109

J.

139,

345-348

461--468

I,,,\,.