CATABOLISM OF HYALURONAN IN RABBIT SKIN TAKES PLACE LOCALLY, IN LYMPH NODES AND LIVER

Transcription

1 Experimental Physiology (1991), 76, Printed in Great Britain CATABOLISM OF HYALURONAN IN RABBIT SKIN TAKES PLACE LOCALLY, IN LYMPH NODES AND LIVER ULLA B. G. LAURENT*t, LAURITZ B. DAHL$ AND ROLF K. REED* * Department of Medical and Physiological Chemistry and t Department of Ophthalmology, University of Uppsala, Sweden, $ Department of Pediatrics, University of Tromso, Norway and Department of Physiology, University of Bergen, Norway (MANUSCRIPT RECEIVED 16 NOVEMBER 1990, ACCEPTED 13 MARCH 1991) SUMMARY The catabolism of hyaluronan has been studied by injecting hyaluronan, labelled with tyramine cellobiose (1251-TC), subcutaneously into the hindpaw of rabbits. Following endocytosis, 125I-TC remains in the cells at the site of uptake, allowing localization of the site of catabolism. At 6 h after subcutaneous injection, 65 % of the injected radioactivity was recovered. The skin at the injection site contained 47%, the popliteal gland at the side of injection 10 %, and the liver 8 % of the injected dose. At 48 h the three organs contained 40 % of the injected dose with 17 % in the skin, 10% in the lymph node and 13 % in the liver. The decline in recovery could be accounted for by urinary excretion of the tracer, implying that some tracer had been released from the cells after endocytosis. Chromatography revealed that over 85 % of 125I-TC-hyaluronan in the lymph nodes and liver was of low molecular mass throughout the experiment. In skin, 4 % of the injected tracer was recovered with low molecular mass at 6 h, increasing to 12% of injected dose at 24 and 48 h. Thus, a minimum of 12 % of the injected tracer was catabolized per 24 h at the skin injection site. If cells in skin are responsible for the subsequent release of tracer, as seen from the decrease in recovery of the injected dose, another % of the tracer could have been catabolized locally in the skin per day. The major part of the hyaluronan injected in the skin was, however, catabolized by lymphatic removal and subsequent degradation in local lymph nodes and liver. INTRODUCTION The two major glycosaminoglycans in skin are hyaluronan and dermatan sulphate, present in about equal amounts (Pearce, 1965). Of these dermatan sulphate is, as a proteoglycan, closely associated with collagen fibres (Pearce, 1965; Scott, 1988) and hyaluronan is the main polysaccharide in the interfibrillar matrix. Hyaluronan has a molecular mass of several million Daltons and seems to have a role in fluid homeostasis of the tissues (Comper & Laurent, 1978). In skin we may envisage that it exists in pericellular coats bound to fibroblasts (Clarris & Fraser, 1968), but presumably also partly in a 'free' or unbound pool. This 'free' pool will eventually drain into lymph, and the hyaluronan concentration in prenodal lymph from skin is 5-10,tg ml-' (Reed, Laurent & Taylor, 1990b). Hyaluronan is catabolized in lymph nodes (Fraser, Kimpton, Laurent, Cahill & Vakakis, 1988) or eventually, when reaching blood, by the endothelial cells of the liver (Smedsr6d, Pertoft, Eriksson, Fraser & Laurent, 1984; Smedsr6d, Pertoft, Gustafsson & Laurent, 1990) so that the half-life of hyaluronan in the blood stream is 2-5 min (Fraser, Laurent, Pertoft & Baxter, 1981). Despite the rapid clearance from blood, human plasma concentrations of * Address at which experiments were carried out.

2 696 U. B. G. LAURENT, L. B. DAHL AND R. K. REED hyaluronan are still in the order of ng ml-' (Engstr6m-Laurent, Laurent, Lilja & Laurent, 1985). Recent work indicates that the 'free' or unbound pool of hyaluronan constitutes about 25 % of the total hyaluronan in the tissue and that it is turning over with a half-life of about 16 h (Reed, Laurent, Fraser & Laurent, 1990a). Furthermore, the hyaluronan concentration in pre-nodal lymph from skin is not altered with increasing lymph flow (Reed et al b), implying that interstitial fluid flux and lymph flow will influence the interstitial removal and turnover of hyaluronan. Similar observations have been made on lung lymph (Lebel, Smith, Risberg, Gerdin & Laurent, 1988). While it thus is known that local lymph nodes and liver endothelial cells will catabolize hyaluronan, and that interstitial fluid flux will influence the turnover of hyaluronan, it is presently unknown whether catabolism also occurs locally in skin (Laurent & Fraser 1991). The present study was designed to address whether there is a local catabolism of hyaluronan in skin. The experiments were carried out by using hyaluronan labelled with 1251-tyramine cellobiose (1251-TC-HYA) (Dahl, Laurent & Smedsr6d, 1988). When this compound is endocytosed it will be rapidly degraded in the lysosomes and the 1251-tyramine cellobiose (125I-TC) will be retained intralysosomally in the cells at the site of uptake (Pittman, Carew, Glass, Green, Taylor & Attie, 1983; Dahl et al. 1988). The use of 1251 to label tyramine-cellobiose allows easy and rapid localization of the tissues where hyaluronan is endocytosed as well as their relative importance in this process. Subsequent extraction of the tracer from the tissues and chromatography is used to reveal whether it has been degraded. Briefly, the present experiments showed that at most 25 % of the catabolism is taking place in the skin. METHODS Sixteen male New Zealand White rabbits were used in the experiments. The rabbits weighed kg at the start of the experiments and had free access to food and water throughout the experimental period. Tracer preparation Hydrazinolytically modified hyaluronan substituted with tyramine-cellobiose (TC) was the same as described by Dahl et al. (1988). The product contained mg hyaluronan per ml and 1 out of 130 disaccharide units was substituted with TC. The weight average molecular mass at the time of injection was Da as determined by chromatography (Laurent & Granath, 1983; Lebel, Smith, Risberg, Laurent & Gerdin, 1989). TC-hyaluronan (100,ug; 84,ul) was mixed with Bq 1251 (9,dl) in a microreaction vessel coated with 10,ug 1,3,4,6-tetrachloro-3a,6a-diphenylglycouril (lodogen) (Sigma, St Louis, MO, USA) which is a water-insoluble oxidizing agent. The incubation was stopped after 2 h by adding 10 mg of sodium iodide (NaI in Dulbecco phosphate-buffered saline, 100 mg ml-'). The mixture was then applied to a Sephadex G-25 column (PD-10 column, Pharmacia, Uppsala, Sweden) equilibrated in phosphatebuffered saline and eluted with the same solvent into 0 5 ml fractions. On the day of the animal experiments (1-4 days after labelling), the 125I-TC-HYA was again applied to a PD-10 column and the first peak collected. In this way radioactive ligand which had been spontaneously released from the polymer was removed from the solution that was to be injected. The released ligand is adsorbed to the gel and eluted late in the chromatogram, and is therefore called 'late peak' by Dahl et al. (1988). Injection of labelled hyaluronan The tracer was injected subcutaneously in the dorsal part of the right hindpaw of the rabbit using a tuberculin syringe and a needle of 04 mm outer diameter. The injected volume ( ,tl) was divided into doses of 25,u1 which were injected at least 1 cm apart. The amount of radioactivity injected in each rabbit was x 103 Bq ( x 106 d.p.m.). The animals were returned to their cages after the injections.

3 CATABOLISM OF HYALURONAN IN SKIN 697 Sampling of tissue The animals were killed by intravenous injections of pentobarbitone (about 100 mg kg-' given in the marginal ear vein) at 6-48 h after the injection of 125I-TC-HYA. The skin at the site of injection as well as underlying subcutaneous tissue and tendons were collected for analysis. Furthermore, regional lymph glands in the popliteal and inguinal regions were collected as well as abdominal lymph glands, skeletal muscle, heart, spleen, kidneys, liver and lungs. Radioactivity was measured, if possible, in the whole organ. Alternatively, radioactivity was determined in randomly selected and weighed samples from the organ. The total radioactivity was calculated as the product of radioactivity per gram tissue and the total organ weight. For comparison the skin from the dorsal part of the left paw as well as lymph glands from the left popliteal and inguinal regions were also taken out and radioactivity determined. Extraction of TC-HYA from the tissues Extraction of the 125I-TC-HYA was performed after determination of 1251 in samples from skin at the injection site, the regional popliteal lymph gland and the liver. The tissue samples were homogenized by cutting the tissue into small pieces with a pair of scissors. The homogenized tissue was then submerged in 0 25 M-phosphate buffer, 0 5 M sodium chloride, ph 7 0, with detergent (1 % Tween 20, Merck, Germany) and 002 % sodium azide. The vials were rotated end-over-end at + 4 C for h. The preparations were kept frozen until chromatography when they were thawed and centrifuged, first at 2000 r.p.m. and then by ultracentrifugation in a Beckman L8-M Ultracentrifuge at r.p.m. for 45 min. Chromatography Chromatography was carried out on pooled tissue extracts (skin, lymph node and liver) from two of the experiments at 6, 12, 24 and 48 h. Three millilitre portions of the soluble extract from each tissue were applied to a 1 6 x 84 cm Sephacryl S-300 column (Pharmacia, Uppsala, Sweden). The column was eluted with phosphate-buffered saline containing 1 % Tween 20 and 0 02 % sodium azide. The flow rate was maintained at 18 ml h-' and 3 ml fractions were collected. Void volume (VJ) and total volume (V) were determined from the elution of the bacteria Serratia marcescens and 3H20, respectively. The radioactivity was essentially eluted in three peaks; one high molecular peak at V0, one low molecular peak at V, and one distinctly retarded peak ('late peak'). The late peak corresponds to spontaneously released ligand formed during the processing of the material and was supposed to have been released from both the high and low molecular portions with equal rate. This was investigated in another experiment where chromatography was performed twice on the same tissue extracts (rabbit knee joint tissue, U. B. G. Laurent, unpublished observation) with a 5 week interval. (The extracts were stored at -20 C between the two chromatographies.) Although the late peak had increased after 5 weeks, the relative magnitude of the high and low molecular mass fractions were unchanged. The proportions of high and low molecular mass hyaluronan were therefore calculated from the V0 and V, peaks. Radioactivity measurements The radioactivity was determined in a Packard 5166 Auto-Gamma counter. Results are presented as means+standard deviations (S.D.). RESULTS Distribution of 12JI-TC-HYA On average 65 % of the injected radioactivity could be accounted for at the injection site, in local lymph nodes and liver at 6 and 12 h after injection. This fell to 40 % at 48 h (Table 1). The amount recovered in skin fell from 46 7 % at 6 h to 16 8 % at 48 h (Table 1, Fig. 1). In the same time interval the amount recovered in each of liver and popliteal lymph node, increased to between 10 and 15 % of the injected dose. Less than 0.5 % of the injected dose was found in each of spleen, lungs and kidneys. Based on radioactivity in urine,

4 698 U. B. G. LAURENT, L. B. DAHL AND R. K. REED Table 1. Percentage of injected dose regained in different organs and at different times after injection of the tracer. Mean + S.D.; n = 4 for each time period Skin Lymph node Liver Total 6h h h h ) 0 I.) la c) 0 u 4) 0 4) Liver Node 20 Skin High MM Skin High MM I I Time (h) Fig. 1. Percentage of injected dose recovered in the liver, the popliteal lymph node (Node) and the skin at 6, 12, 24 and 48 h after injection of l25i-tyramine-cellobiose hyaluronan subcutaneously in the paw. In skin, the amount of high and low molecular mass fractions are also shown. In the lymph node at 12 h one outlying point (31-9) has been excluded. (The value in Table 1 should therefore be ) measured at 24 h after injection in two rabbits, and assuming a urine production of ml per 24 h (Kaplan & Timmons, 1979) the radioactivity excreted in urine will correspond to 4-13 % of the injected dose being excreted by this route per 24 h (see also below). Chromatograms Skin. The relative amount of low molecular mass 125I-TC-HYA increased from 8 4 % at 6 h to 73 % at 48 h (Figs 1 and 2). These figures indicate that 4% of the injected dose was metabolized to low molecular mass hyaluronan at 6 h, 7% at 12 h and 12 % at 24 h and 48 h (Figs 1 and 2). Lymph node. The TC-HYA recovered in the popliteal lymph node was of low molecular mass throughout the experimental period. At 6 h 81 % of the total 1251I-TC-HYA in the node was of low molecular mass, and this increased to 94% at 48 h (Fig. 3 shows chromatogram at 24 h). 48

5 CATABOLISM OF HYALURONAN IN SKIN h VO Vt h VO Vt 48 h Eluate (ml) Fig. 2. Chromatograms of skin extracts at 6, 12, 24 and 48 h after injection of radioactively labelled hyaluronan. The peak eluted at void volume (V0) (from 50 to 120 ml) represents high molecular mass material and the peak at total volume (V,) (from 120 to 210 ml) low molecular mass material. See Methods for the retarded peak eluted after Vt, 'late peak', ml. The ordinate has been adjusted so that the area under each chromatogram represents the total activity recovered from the tissue.

6 700 U. B. G. LAURENT, L. B. DAHL AND R. K. REED VO it Lymph node v o V _ Liver Eluate (ml) Fig. 3. Chromatograms of extracts from popliteal lymph node and liver at 24 h after injection of radioactively labelled hyaluronan. For definitions of high and low molecular mass hyaluronan and also for adjustment of the ordinate, see Fig. 2. Liver. The pattern seen in liver was similar to that observed in the popliteal lymph node with at least 85 %0 of the 125I-TC-HYA in the liver being of low molecular mass (Fig. 3). Recovery of injected dose Because the recovery of tracer was only 65 %0 at 6 h after injection, two additional experiments were carried out. In these two experiments all urine produced during the first 6 h after injection of tracer was collected. The amount of radioactivity found in urine was 9 1 and of the injected dose in the two experiments. Radioactivity at 6 h after injection found in the tissues of the paw (including the metatarsal bones) and not obtained in the previous experiments, accounted for another 5 5 and 124 of the injected dose, respectively. Finally, from the radioactivity in plasma and assuming an extracellular volume of 20 0 of bodyweight, another 8 0 was accounted for in the total extracellular volume. Thus, at 6 h after injection another %0 of the injected dose was located elsewhere than in the skin at the site of injection, the local lymph node and the liver. Including this radioactivity the recovery increases to about 90 0 of the injected dose.

7 CATABOLISM OF HYALURONAN IN SKIN 701 Until a decade ago it DISCUSSION was taken for granted that hyaluronan and other connective tissue matrix components were catabolized locally in the peripheral tissues. Lysosomal hyaluronidase had been demonstrated in many tissues (Bollet, Bonner & Nance, 1963) including skin (Cashman, Laryea & Weissman, 1969), although it is notable that it seemed to be absent in human skin fibroblasts (Arbogast, Hopwood & Dorfman, 1975). The exoglycosidaseș8-n-acetyl-d-hexosaminidase and /3-D-glucoronidase, that are required for the degradation, had also been found widely distributed in the tissues (Roden, Campbell, Fraser, Laurent, Pertoft & Thompson, 1989). The discovery that hyaluronan is removed from the tissues in significant amounts by lymph (Laurent & Laurent, 1981) and subsequently catabolized in lymph nodes (Fraser et al. 1988; Fraser & Laurent, 1989) and liver endothelial cells (Smedsrod et al. 1984, 1990) suddenly demonstrated an alternative catabolic pathway, which was of considerable physiological significance, since the rate of turnover could be controlled by the water flux through the tissues. Until now we have lacked information concerning the relative magnitude of the two different catabolic pathways. To localize the sites of degradation of hyaluronan and their relative importance we have now applied a technique modified after Pittman et al. (1983). They showed that125i-tc bound to protein is retained intralysomally in the cells at the site of uptake even after degradation of the polymer. This is also the case with tyramine-cellobiose bound to hyaluronan (Dahl et al. 1988). The chromatography performed on 125I-TC-HYA in a specific tissue determines to what extent the polysaccharide has been catabolized to smaller fragments. In the popliteal lymph node and liver, degraded hyaluronan accounted for about 90% of the radioactivity while in skin, the amount of hyaluronan of low molecular mass increased from about 4% of the injected dose at 6 h to about 12% at 24 and 48 h, indicating that high molecular mass hyaluronan present extracellularly had been continuously endocytosed and degraded. In order to allow exact quantification of importance of the three tissues in the degradation of hyaluronan, it is necessary that the125i-tc complex really is retained in the cells at the site of uptake for the whole duration of the experiment. The data in Fig. 1 only describes the local catabolism of hyaluronan correctly if the complex is not released from the cells after endocytosis. If some of the complex is released, the above figures represent minimal values of the catabolism in each tissue. However, there are good reasons to believe that such a release is small during the 48 h of the experiment. According to Pittman et al. (1983), the rate of leakage from fibroblasts is only 5 % in 24 h. Thus, provided that the complex is retained in the skin once it is endocytosed, 12% of the injected dose is catabolized locally in skin in 24 h. Conversely, the falling recovery of the injected dose does suggest that some of the tyramine-cellobiose complex is released from the cells at which it was endocytosed. This release would amount to % of the injected dose per 24 h. The present experiments do not allow a conclusion to be made as to which of the three sites would be responsible for such a release, but the maximal limit for the catabolism at each site will be the sum of the low molecular mass hyaluronan found in that location plus the amount that leaves by the urine in the same time period. For skin this means that the maximum value that is catabolized locally in 24 h amounts to 25 % of the injected dose. This is based on the assumption that all release of low molecular mass 125I-TC-HYA recovered in urine, and 25 ~~~~~~~~~~~~~~~~~~~~~~EPH EPH 76

8 U. B. G. LAURENT, L. B. DAHL AND R. K. REED visualized in Fig. 1 as a gradual fall in recovery of the injected radioactivity, originates from release from cells in the skin. The substitution of hyaluronan with tyramine-cellobiose results in a polymer with molecular weight around which is considerably lower than for endogenous hyaluronan. The use of hyaluronan of low molecular mass could be a possible source of error in validating the turnover in tissues if it is catabolized in a different way to the high molecular mass hyaluronan. The results show clearly that125i-tc-hya is retained in the liver as well as in the local popliteal lymph node, as is the case for high molecular mass hyaluronan (Fraser et al. 1988). More importantly, however,'25i-tc-hya is cleared from the skin at a rate which corresponds to a half-life of somewhat less than 24 h since the total radioactivity in skin was reduced from 47 to 25 % of the injected dose between 6 and 24 h after injection (cf. Table1 and Fig. 1). This rapid removal rate of 'free' or unbound hyaluronan is similar to that which has recently been obtained in rabbits by Reed etal. (1990 a) when using subcutaneous injections of tritiated hyaluronan with a molecular mass of 2-4 million Da. The similar removal rate for the two different hyaluronan molecules suggests that they are comparable for evaluation of the turnover of hyaluronan in tissues. The studies by Dahl et al. (1988) showed that the 125I-TC-labelled hyaluronan is taken up by the liver endothelial cells and that the uptake can be inhibited by excess unlabelled hyaluronan. Thus, this labelled hyaluronan is recognized by the specific hyaluronan receptor (Dahl etal. 1988). Taken together these data suggest that the tyramine-cellobioselabelled hyaluronan reflects hyaluronan. the turnover and normal metabolic pathway for endogenous This project has been supported by the Swedish Medical Research Council (grant 03x-4), V's 80 years fund and Pharmacia AB. REFERENCES Gustaf ARBOGAST, B., HoPWOOD, J. J. & DORFMAN, A. (1975). Absence of hyaluronidase in cultured human skin fibroblasts. Biochemical and Biophysical Research Communications 67, BOLLET, A. J., BONNER, W. M. JR & NANCE, J. L. (1963). The presence of hyaluronidase in various mammalian tissues. Journal of Biological Chemistry 238, CASHMAN, D. C., LARYEA, J. U. & WEISSMANN, B. (1969). The hyaluronidase of rat skin. Archives of Biochemistry and Biophysics 135, CLARRIS, B. J. & FRASER, J. R.E. (1968). On the pericellular zone of some mammalian cells in vitro. Experimental Cell Research 49, COMPER, W. D. & LAURENT, T. C. (1978). Physiological function of connective tissue polysaccharides. Physiological Reviews 58, DAHL, L. B., LAURENT, T. C. & SMEDSROD, B. (1988). Preparation of biologically intact radioiodinated hyaluronan of high specific radioactivity: Coupling of1251-tyramine-cellobiose to amino groups after partial N-deacetylation. Analytical Biochemistry 175, ENGSTR6M-LAURENT, A., LAURENT, U. B. G., LILJA, K. & LAURENT, T. C. (1985). Concentration of sodium hyaluronate in serum. Scandinavian Journal of Clinical and Laboratory Investigation 45, FRASER, J. R. E., KIMPTON, W. G., LAURENT, T. C., CAHILL, R. N. P. & VAKAKIS, N. (1988). Uptake and degradation of hyaluronan in lymphatic tissue. Biochemical Journal 256, FRASER, J. R. E. & LAURENT, T. C. (1989). Turnover and metabolism of hyaluronan. In The Biology of Hyaluronan, Ciba Foundation Symposium 143, pp Wiley, Chichester. FRASER, J. R. E., LAURENT, T. C., PERTOFT, H. & BAXTER, E. (1981). Plasma clearance, tissue distribution and metabolism of hyaluronic acid injected intravenously in the rabbit. Biochemical Journal 200, KAPLAN, H. M. & TIMMONS, E. H. (1979). The Rabbit. A Model for the Principles of Mammalian Physiology and Surgery, pp Academic Press, New York.

9 CATABOLISM OF HYALURONAN IN SKIN LAURENT, T. C. & FRASER, J. R. E. (1991). Catabolism of hyaluronan. In Degradation of Bioactive Substances. Physiology, Location and Pathophysiology, ed. HENRIKSEN, J. H., pp CRC Press, Boca Raton, FL, USA. LAURENT, U. B. G. & GRANATH, K. A. (1983). The molecular weight of hyaluronate in the aqueous humour and vitreous body of rabbit and cattle eyes. Experimental Eye Research 36, LAURENT, U. B. G. & LAURENT, T. C. (1981). On the origin of hyaluronate in blood. Biochemistry International 2, LEBEL, L., SMITH, L., RISBERG, B., GERDIN, B. & LAURENT, T. C. (1988). Effect of increased hydrostatic pressure on lymphatic elimination of hyaluronan from sheep lung. Journal of Applied Physiology 64, LEBEL, L., SMITH, L., RISBERG, B., LAURENT, T. C. & GERDIN, B. (1989). Increased lymphatic elimination of interstitial hyaluronan during E. coli sepsis in sheep. American Journal ofphysiology 256, H PEARCE, R. H. (1965). Glycosaminoglycans and glycoproteins in skin. In The Amino Sugars, vol. IIA, ed. BALAZS, E. A. & JEANLOZ, R. W., pp Academic Press, New York. PITTMAN, R. C., CAREW, T. E., GLASS, C. K., GREEN, S. R., TAYLOR, C. A. & ATTIE, A. D. (1983). A radioiodinated, intracellularly trapped ligand for determining the sites of plasma protein degradation in vivo. Biochemical Journal 212, REED, R. K., LAURENT, T. C. & TAYLOR, A. E. (1990b). Hyaluronan flux in prenodal lymph from skin: Changes with lymph flow. American Journal of Physiology 259, H REED, R. K., LAURENT, U. B. G., FRASER, J. R. E. & LAURENT, T. C. (1990a). Removal rate of [3H]hyaluronan injected subcutaneously in the rabbit. American Journal of Physiology 259, H RODEN, L., CAMPBELL, P., FRASER, J. R. E., LAURENT, T. C., PERTOFT, H. & THOMPSON, J. N. (1989). Enzymic pathways of hyaluronan catabolism. In The Biology of Hyaluronan, Ciba Foundation Symposium 143, pp Wiley, Chichester. SCOTT, J. E. (1988). Proteoglycan-fibrillar collagen interactions. Biochemical Journal 252, SMEDSROD, B., PERTOFT, H., ERIKSSON, S., FRASER, J. R. E. & LAURENT, T. C. (1984). Studies in vitro on the uptake of sodium hyaluronate in rat liver endothelial cells. Biochemical Journal 223, SMEDSR6D, B., PERTOFT, H., GUSTAFSSON, S. & LAURENT, T. C. (1990). Scavenger functions of the liver endothelial cell. Biochemical Journal 266,