JOURNAL OF PATHOLOGY, VOL. 178: 30-35 (1996)

TENASCIN EXPRESSION DURING WOUND HEALING IN HUMAN SKIN

MIEKE A. LATIJNHOUWERS, MIEKE BERGERS, BERT H. VAN BERGEN, KARIN 1. SPRUIJT, MONIQUE P. ANDRIESSEN AND JOOST SCHALKWIJK

Department of Dermatology, University Hospital Nijmegen, P. 0. Box 9101, 6500 H B Nijmegen, The Netherlands

SUMMARY In adult human skin, the expression of the extracellular matrix glycoprotein tenascin is limited. Under hyperproliferative conditions

such as psoriasis and epidermal tumours, dermal tenascin expression is strongly upregulated. The aim of this study was to investigate the pattern and kinetics of tenascin expression in human skin during wound healing and to address the question of whether keratinocytes can directly interact with tenascin during re-epithelialization. Tenascin expression was investigated in excisional wounds in normal human skin, in explants of normal human skin, and in chronic venous ulcers, using immunohistochemistry. No tenascin staining was found directly underneath the leading edge of the sheet of migrating keratinocytes in the excisional wounds and explants. In the excisional wounds and the ulcers, dermal tenascin was strongly upregulated in areas adjacent to hyperproliferative epidermis. These hyperproliferative areas are located approximately 10-50 cells behind the leading edge, as assessed by staining for the Ki-67 antigen and the proliferating cell nuclear antigen (PCNA). At the later stages of normal wound healing and in the chronic ulcers, tenascin was also detected in the wound bed. In these areas, the dermal-epidermal junction stained positive for laminin but was negative for heparan sulphate. The absence of the latter basement membrane component suggests that the formation of a new basement membrane is not completed in these wounds. These findings suggest that tenascin is not a substrate for migrating keratinocytes; that the rapid induction of tenascin expression in the papillary dermis during wound healing results from interaction with the hyperproliferative epidermis; and that in the later stages of wound healing, keratinocytes can potentially interact with tenascin in the wound bed, because the basement membrane of the neo-epidermis is incomplete.

KEY WORDS-tenascin; cytotactin; extracellular matrix; basement membrane; wound healing; skin; venous ulcers

INTRODUCTION

Wound healing is characterized by complex changes at the inter- and intracellular level, including inflam- mation, cell migration, and proliferation and reconstruc- tion of the extracellular matrix (ECM).'-' One of the ECM molecules whose expression is altered during wound healing is tenascin (TN-C).2,',7 l 2 This large hexameric glycoprotein was first detected in developing embryos, where it was found predominantly at epithelial-mesenchymal interaction sites.I3-l7 In adult tissues, expression of tenascin is rather limited. In nor- mal human skin, tenascin is present in the papillary dermis and near the basement membranes surrounding blood vessels and epidermal adnexa. 1231s,19 We and others have shown that the expression of tenascin in the papillary dermis of normal skin varies from almost absent to a patchwise distribution underneath the basement membrane, but is strongly upregulated in conditions of epidermal hyper roliferation such as

Although in vitvo studies have suggested various func- tions for tenascin, including modulation of cell

and the immune r e s p ~ n s e , ~ ~ , ~ " its in vivo role is still a matter of debate."

Tenascin expression during wound healin has been studied in various animal model^.'^^.'^^^^^^^^ The

psoriasis,'* epidermal tumours, ps-21 and injury.339322

Addressee for correspondence: Joost Schalkwijk, Department of Dermatology. University Hospital Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands.

$2 1996 by John Wiley & Sons, Ltd. CCC 0022-34 17/96/0 10030-06

data on wound healing in humans, however, are quite limited: tenascin expression was studied previously in surgical wounds,' chronic venous ulcers,32 and in the tape-stripping model for epidermal injury.22 On the basis of the findings, it was postulated that tenascin might stimulate keratinocyte m i g r a t i ~ n ~ ~ ~ - ~ ~ " and that tenascin expression correlates with epidermal proliferation.

We now provide new immunohistochemical data on the pattern and kinetics of tenascin expression in human skin wounds. Experimental excisional wounding of nor- mal human skin was performed to obtain standardized wounds. This experimental set-up allowed the compari- son of tenascin expression at various stages of wound healing in the same subject. Tenascin expression in the excisional wounds was compared with expression in explant cultures of normal human skin. These explants are useful as a model for some aspects of normal wound healing, including keratinocyte migration. In addition, tenascin expression was investigated in chronic venous ulcers, which represent inadequately healing wounds. In parallel with the tenascin staining, immunostaining of proliferation markers was performed to see whether tenascin expression correlates with keratinocyte prolif- eration, as we suggested previously for hyperprolifera- tive skin disorders and the tape-stripping modeL2* Finally, heparan sulphate and laminin, two major com- ponents of basement membranes, were stained to estab- lish whether during wound healing keratinocytes are separated from dermal tenascin by a complete basement membrane, as in normal intact skin.

Received 14 February 1995 Accepted 4 July 1995

TENASCIN IN HEALING SKIN WOUNDS 31

MATERIALS AND METHODS Biopsies

Experimental wounds were made by taking superficial punch biopsies from the upper arm of eight healthy volunteers; four biopsies with a 3 mm diameter and a 2-3 mm depth were taken from each volunteer. These primary biopsies were prepared for immunohistochem- istry or used for explant culturing. Full-thickness punch biopsies with a 4 mm diameter were taken from the different wounds after 2, 4, 7, and 14 days. These secondary biopsies, which included the healing wounds and some of the adjacent normal skin, were prepared for immunohistochemistry .

In addition to the healthy volunteers, eight patients with chronic venous leg ulcers participated in this study. Rectangles of approximately 10mm long and 3 mm wide were excised from the margins of their ulcers, to include the surrounding intact skin, the ulcer edge, and the ulcer base. All biopsies were taken under local anaesthesia and with the informed consent of the sub- jects. Permission for experiments on humans was obtained from the Medical Ethical Committee of the University Hospital Nijmegen.

Explant cultures

Primary biopsies of healthy volunteers were placed immediately in modified Eagle’s medium (MEM; Life Technologies, Breda, The Netherlands) supplemented with penicillin (1000 U/ml), streptomycin (1000 pg/ml), and fungisone (2.5 U/ml) (Life Technologies, Breda, The Netherlands) and were incubated for 1 h at room tem- perature to disinfect them. The biopsies were subse- quently placed in Dulbecco’s modified Eagle’s medium (DMEM; Flow Laboratories, Irvine, Scotland, U.K.) supplemented with penicillin (1 00 U/ml), streptomycin (100 pg/ml), and 5 per cent newborn calf serum (Flow Laboratories, Irvine, Scotland, U.K.) floating with the epidermis upwards. The explants were cultured at 37”C, 95 per cent relative humidity, and 8 per cent CO, for 1 4 days.

Zmmuno~istoc~emist~y

Biopsies and explants were fixed in 3.8 per cent phosphate-buffered formalin (J. T. Baker, Deventer, The Netherlands) and embedded in paraffin. Five- micrometre-thick sections were stained. To detect tenascin, a rabbit polyclonal antiserum against human tenascin was used (Life Technologies, Breda, The Netherlands). To detect proliferation markers, we used a mouse monoclonal antiserum against human proliferat- ing cell nuclear antigen (PCNA) (Dakopatts, Glostrup, Denmark) and a mouse monoclonal antiserum against human Ki-67 antigen (Clone MIB-1) (Immunotech S.A., Marseille, France). Immunostaining of heparan sulphate and laminin was performed using a monoclonal anti- serum (JM-403) against the glycosaminoglyan side-chain of human heparan sulphate proteoglycan (a kind gift from J. van den Born, Department of Nephrology, University Hospital Nijmegen33) and a polyclonal rabbit

Fig. 1-Tenascin expression (arrow-heads) in an excisional wound 2 days after wounding. Note that under the leading edge of the sheet of migrated keratinocytes no tenascin is present, indicating that kerati- noyte migration precedes tenascin expression. Tenascin is also detected around a blood vessel (arrows). No counterstaining; e=epidermis; d = dermis; le = leading edge

antiserum against mouse laminin (ICN Biomedicals, Zoetermeer, The Netherlands). Pretreatment of sections included enzymatic digestions [15 min at 37°C with 0.1 per cent trypsin in 0.1 per cent CaCl, (pH 7.8) for heparan sulphate staining or 20 min at room tempera- ture with 0.1 per cent pronase in demineralized water for laminin staining], microwave irradiation [2 x 5 min in 1.8 mM citric acid monohydrate, 8.2 mM trisodium citrate 2-hydrate (pH 6.0) in a microwave oven (Miele Model M720) at 450 W for Ki-67 antigen staining], and preincubations with normal sera [20 per cent in phosphate-buffered saline (PBS) for all staining pro- cedures except the PCNA staining]. Immunostaining for PCNA and tenascin was performed according to standard indirect immunoperoxidase methods. Peroxidase-conjugated antisera to mouse and rabbit immunoglobulins were obtained from Dakopatts (Glostrup, Denmark). Staining for laminin and Ki-67 antigen was performed with the Vectastain Elite ABC kit, according to the manufacturer’s instructions (Vector Laboratories, Burlingame, U.S.A.). After immuno- staining, sections were either counterstained with haematoxylin or directly mounted.

RESULTS Tenascin expression

Tenascin expression was considerably altered in all conditions when compared with the expression in nor- mal skin. In the excisional wounds from healthy volun- teers, keratinocytes had already started to migrate over the wound bed at 2 days after wounding. In half of the biopsies of these wounds, tenascin was stained as a continuous band in the papillary dermis adjacent to the wound (Fig. 1). This tenascin-positive area was located at some distance (corresponding to approximately 20 basal keratinocytes) from the leading edge of the sheet of migrated keratinocytes. Such continuous tenascin stain- ing was further seen in all excisional wounds at 4-14 days after wounding (Figs 2a and 3a) and in all explant cultures (Fig. 4), even those that had been cultured for

32 M. A. LATIJNHOUWERS ET A L .

Fig. 2-Serial sections of an excisional wound 7 days after wounding. Section a shows the distribution of tenascin (arrow-heads). In section b. proliferating cells are detected by means of immunostaining of the nuclear antigen Ki-67. Note that the epidermis adjacent to the tenascin-positive dermis shows a high density of Ki-67-positive nuclei. Both sections are counterstained with haematoxylin. The wound margin is indicated by arrows; e=epidermis; d=dermis

Fig. 3-Serial sections of an excisional wound 14 days after wounding. In a, the staining pattern of tcnascin is shown. Tenascin is discretely present at the dermal-cpidermal junction and diffusely throughout the wound bed (arrow-heads). The section shown in b is stained with an antiserum against the glycosaminoglycan side-chain of heparan sulphate proteo- glycan. Note that staining ceases abruptly near the wound margins (arrows). In contrast, laminin continuously lines the neo-epidermis as shown in c. Section a was counterstained with haematoxylin; no counterstainiug was performed with sections b and c. e=Epidermis; d=dermis

only 1 day. Although tenascin was sometimes also diffusely stained beneath migrated keratinocytes in the excisional wounds and the explants, it never appeared beneath the leading edge of the sheet of migrating keratinocytes. By the time the excisional wounds were entirely covered by a neo-epidermis (day 7 or 14), the wound bed had become positive for tenascin. Tenascin expression in the excisional wounds did not correlate with the presence of cellular infiltrate or granulation tissue. The results of tenascin staining in excisional wounds and explants are summarized in Table I.

In the chronic venous ulcers, the pattern of tenascin staining showed considerable variation. In six ulcers, tenascin expression in the papillary dermis adjacent to

the ulcer base was markedly upregulated in comparison with normal skin. Staining was most intense in the papillary dermis adjacent to the ulcer bed and penetrated more deeply into the dermis than was the case in the excisional wounds (Fig. 5). In two other ulcers, tenascin was not clearly detectable in the papil- lary dermis adjacent to the ulcer bed. Tenascin was also observed in the wound bed of all ulcers and beneath the keratinocytes covering the edge of the ulcer. Here, tenascin staining was low in areas with much cellular infiltrate, while in granulation tissue, staining of tenascin was intense and diffusely distributed. In contrast with the excisional wounds, tenascin was detected in the ulcers beneath the tip of the neo-epidermis (not shown).

TENASCIN IN HEALING SKIN WOUNDS 33

Fig. &Tenaxin expression in a skin explant that has been cultured for 4 days (hamatoxylin counterstained). Tenascin continuously lines the demal-epidemal junction of the biopsy and also a blood vessel located in the biopsy (arrow-heads). Beneath migrated keratinocytes, tenascin is stained diffusely (arrows), e=Epidermis; d=dermis

Fig. 5-Section of a chronic venous leg ulcer, showing tenascin expression in skin adjacent to the ulcer bed (haematoxylin counter- stained). The darker cells in the epidermis (asterisk) are cells of the granular layer that pick up the counterstaining more intensely than the other cells: they are not tenascin-positive. e=Epidermis; d=dermis

Proliferation markers To study whether tenascin expression correlates with

keratinocyte proliferation as has been suggested pre- viously, tenascin expression was compared with Ki-67

and PCNA staining patterns. In general, more cells were found to be positive for Ki-67 than for PCNA, but the patterns of Ki-67 and PCNA staining were similar. In normal skin, typically 4-10 per cent of the basal

Table I-Summary of the immunohistochemical data of tenascin (TN-C) expression in (A) excisional wounds and (B) skin explant cultures. Tenascin in the papillary dermis of intact skin and tenascin staining beneath migrated keratinocytes (KC) are mentioned separately. The number of biopsies showing the indicated expression and the total number of biopsies studied are given in parentheses

(A) Wound age (days) TN-C in papillary dermis TN-C under migrated KC

0 2 4 7

14

Absent (2/4) or patchy (2/4) Patchy (215) or continuous (315)

Continuous (7/7) Continuous (8/8) Continuous (8/8) Diffuse (8/8)

-

Diffuse (U7) or absent (617) Diffuse (2/8) or absent (6/8) Diffuse (4/8) or absent (4/8)

(B) Culture TN-C in papillary dermis TN-C under migrated KC (days)

0 1 2 3 4 days

Absent (2/4) or patchy (2/4) Continuous (4/4) Continuous (2/2) Diffuse (2/2) Continuous (3 /3) Diffuse (3 /3) Continuous (3 /3) Diffuse (3 /3)

-

Diffuse (2/4) or absent (2/4)

34 M. A. LATIJNHOUWERS ET AL.

keratinocytes stained positive for both proliferation markers. In all excisional wounds and ulcers, but not in the explants, proliferating cells were markedly increased compared with normal skin. With respect to the excisional wounds, the staining results can be described as follows. At days 2,4, and 7, proliferating keratinocytes were rarely detected in the neo-epidermis but mainly at some distance from the wound margins. An area with almost 100 per cent proliferating keratinocytes was located approxi- mately 10-15 cells away from the leading edge at day 2. This intervening area of 10-15 cells contained about 20 per cent proliferating cells. The flattened cells of the leading edge itself were never Ki-67- or PCNA-positive. At days 4 and 7 after wounding, an area with almost 100 per cent proliferating keratinocytes was separated by some 50 cells from the leading edge of migrating keratinocytes. Of these 50 cells, approximately 20 per cent stained positive. When the wounds were closed, at day 7 or 14, hyperproliferation at these sites was less pro- nounced. Basal keratinocytes directly adjacent to the former wound margins and in the newly formed epidermis had become proliferative instead (Fig. 2b).

In the skin explants, proliferation rates did not alter markedly during the 4-day culture period. No PCNA- positive cells could be detected in the explants and fewer than 5 per cent of the basal cells stained positive for the Ki-67 antigen.

In the sections of chronic venous ulcers, PCNA- and Ki-67-positive cells were abundant in the basal layers of the neo-epidermis and the adjacent epidermis. Consist- ent with our findings concerning the excisional wound model, hyperproliferative areas with 75-100 per cent PCNA- or Ki-67-positive keratinocytes were seen at some distance from the ulcer edges, typically 50-150 basal keratinocytes away. In this area, 20-50 per cent of the basal keratinocytes were Ki-67- or PCNA-positive (not shown).

Basal membrane staining Heparan sulphate and laminin were stained to address

the question of whether keratinocytes are separated from dermal tenascin by an intact basement membrane during wound healing. Positive staining for heparan sulphate was found in the basement membranes of normal skin before and after explant culture, and in intact skin adjacent to the excisional wounds and the ulcers. Remarkably, heparan sulphate staining ceased abruptly at the margins of the wounds and the cultured biopsies and was never detected under the neo-epidermis at any time in any of the models used (Fig. 3b).

The staining pattern of laminin was similar to that of heparan sulphate in normal skin and in all the explants, but only up to 4 days after wounding in the excisional wounds. Seven days after wounding in half of the excisional wounds, laminin was detectable beneath the migrated keratinocytes near the wound margins, but not beneath the leading edge of migrated keratinocytes; in the other excisional wounds of day 7, laminin staining still resembled heparan sulphate staining. In the 14-day- old excisional wounds, laminin was found beneath the entire neo-epidermis of all biopsies (Fig. 3c).

In one of five ulcers, the staining pattern of laminin resembled that of heparan sulphate. In the other four ulcers studied, laminin was detected beneath the entire neo-epidermis.

DISCUSSION

We investigated tenascin expression during normal and pathological wound healing in vivo and in an in vitro model for keratinocyte migration using skin explant culturing. The results indicate that tenascin is markedly upregulated under all these conditions, but with different patterns and kinetics of upregulation. Tenascin is dis- continuously distributed near the basement membrane in the papillary dermis of normal skin. In contrast, in the papillary dermis bordering excisional wounds and ulcers, or in the papillary dermis of the original biopsies that were used for explant culturing, tenascin continu- ously lined the basement membrane and to a variable extent penetrated more deeply into the dermis. In the dermal parts that were covered by migrated keratino- cytes during wound healing or explant culturing, tenascin was clearly detected in all ulcers and explants, but only after wound closure in the excisional wounds. These data indicate that in chronic venous ulcers, where the normal coordination of the wound-healing process is lost, the induction of tenascin expression in the wound bed does not depend on keratinocyte migration and subsequent wound closure. Downregulation of tenascin protein to normal levels was not seen in our dynamic models, either in excisional wounds that were already covered by a multilayered neo-epidermis or in the explants in which keratinocytes had enclosed the dermal part of the cultured biopsies. This may be due to continued production of tenascin or to slow turnover of pre-existing tenascin.

A prerequisite for tenascin having a direct in vivo effect on keratinocytes, whether it concerns an influence on migration, proliferation, or other cellular activities, is that keratinocytes can directly interact with tenascin in the dermis. In normal intact skin, the basement mem- brane at the dermakpidermal junction precludes con- tact between keratinocytes and tenascin: the size of the hexameric tenascin molecules, which in humans is approximately 19 00 kD,34 prevents them from moving to the epidermis by diffusion. In this study we have shown that during the later stages of normal wound healing, tenascin is expressed diffusely in the wound bed and that at the same time no heparan sulphate is detectable beneath the neo-epidermis covering the wound bed. We speculate that in these conditions, keratinocytes in the neo-epidermis may interact directly with tenascin because no complete basement membrane is present. These histological data, however, cannot exclude the possibility that a provisional basement mem- brane, containing laminin and possibly other basement membrane component^,^^ hampers interaction between keratinocytes and tenascin. Further studies using immunoelectron microscopic analysis of the dermal- epidermal junction structures are mandatory to obtain more conclusive answers.

TENASCIN IN HEALING SKIN WOUNDS 35

Others have suggested previously that during wound healing, keratinocytes migrate over a surface rich in tenascin.' According to our data, tenascin is not a substrate for migrating keratinocytes: keratinocyte migration precedes tenascin expression both in excisional wounds and in skin explants.

Our hypothesis that tenascin upregulation correlates with keratinocyte proliferation was partially confirmed by the results of this study: at the borders of excisional wounds and chronic venous ulcers, tenascin was mark- edly upregulated in the dermis beneath epidermal hyper- proliferative areas. Because at these borders a basement membrane is present that separates proliferative kerati- nocytes from dermal tenascin, a mitogenic effect of tenascin in this is not likely. Hyperproliferative kerati- nocytes, however, may produce a factor that induces tenascin expression in the underlying dermis. Candidates are the growth factors interleukin-I (IL-l), tumour necrosis factor-a (TNFa), and transforming growth factor$ (TGFP), which in vitro stimulate tenascin pro- duction by cultured fibroblasts.36337 Other aspects of cytokine action in wound healing are dealt with at length in the review by Slavin in this issue of the journaL3* Proliferation of keratinocytes, however, appears not to be a prerequisite for tenascin upregulation in our in vitro model; in the explants, tenascin was clearly upregulated, although very few proliferating cells were detected. Likewise, inflammation cannot be mandatory for tenascin upregulation during explant culture, as the explants are excluded from influx of inflammatory cells from the blood. Further studies should be directed at identifying keratinocyte-derived and other possible factors that may regulate dermal tenascin expression during wound healing.

ACKNOWLEDGEMENT We wish to thank J. van den Born from the Nephrol-

ogy Department of the University Hospital Nijmegen for providing us with the monoclonal antiserum JM-403.

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

Kanzler MH, Gorsulowsky DC, Swanson NA. Basic mechanisms in the healing cutaneous wound. J Dermatol Surg Oncol 1986; 1 2 1156-1 164. Whitby DJ, Ferguson MW. The extracellular matrix of lip wounds in fetal, neonatal and adult mice. Development 1991; 1 1 2 651-668. Juhasz I , Murphy GF, Yan HC, Herlyn M, Albelda SM. Regulation of extracellular matrix proteins and integrin cell substratum adhesion receptors on epithelium during cutaneous human wound healing in vitro. Am J Pathol 1993; 143 1458-1469. Martin P, Hopkinson Woolley J, McCluskey J. Growth factors and cutaneous wound repair. Prog Growth Factor Res 1992; 4 25-44. Wokalek H. Cellular events in wound healing. CRC Crit Rev Biocomput- ibility; 4 209-246. Marks R. The healing and nonhealing of wounds and ulcers of the skin. In: Glynn LE, ed. Handbook of Inflammation. Volume 3. Tissue Repair and Regeneration. Amsterdam: Elsevier North Holland Biomedical Press, 1981; 31C342. Whitby DJ, Longaker MT, Harrison MR, Adzick NS, Ferguson MW. Rapid epithelialisation of fetal wounds is associated with the early depo- sition of tenascin. J Cell Sci 1991; 9 9 583-586. Mackie EJ, Halfter W, Liverani D. Induction of tenascin in healing wounds. J Cell Biol 1988; 107: 2757-2767. Betz P, Nerlch A, Tubel J, Penning R, Eisenmenger W. Localization of tenascin in human skin wounds-an immunohistochemical study. Inr J Legal Med 1993; 105 325-328.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Chuong CM, Chen HM. Enhanced expression of neural cell adhesion molecules and tenascin (cytotactin) during wound healing. Am J Pathol 1991; I38 427440. Murakami R, Yamaoka I, Sakakura T. Appearance of tenascin in healing skin of the mouse: possible involvement in seaming of wounded tissues. Int J Dev 5iol1989; 33: 439444. Lightner VA, Gumkowski F, Bigner DD, Erickson HP. Tenascin/ hexabrachion in human skin: biochemical identification and localization by light and electron microscopy. J Cell Bid 1989; 108: 2483-2493. Chiquet Ehrismann R, Mackie EJ, Pearson CA, Sakakura T. Tenascin: an extracellular matrix protein involved in tissue interactions during Fetal development and oncogenesis. Cell 1986; 47: 131-139. Crossin KL, Hoffman S, Grumet M, Thiery JP, Edelman GM. Site- restricted expression of cytotactin during development of the chicken embryo. J Cell Biol 1986; 102: 1917-1930. Thesleff I, Mackie E, Vainio S, Chiquet Ehrismann R. Changes in the distribution of tenascin during tooth development. Development 1987; 101: 289-296. Aufderheide E, Chiquet Ehrismann R, Ekblom P. Epithelial-mesenchymal interactions in the developing kidney lead to expression of tenascin in the mesenchyme. J Cell Biol 1987; 105 599608. Vainio S, Jalkanen M, Thesleff I . Syndecan and tenascin expression is induced by epithelial-mesenchymal interactions in embryonic tooth mesen- chyme. J Cell Biol 1989; 108: 1945-1953. Schalkwijk J, Van Vlijmen I, Oosterling B, et al. Tenascin expression in hyperproliferative skin diseases. 5 r J Dermatol 1991; 124: 13-20, Shikata N, Oyaizu T, Andachi H, Tsubura A. Tenascin expression in normal human adult skin and skin appendage tumours. Virchows Arch 1994; 424 51 1-516. Stamp GW. Tenascin distribution in basal cell carcinomas. J Pathol 1989; 159 225-229. Verstraeten AA, Mackie EJ, Hageman PC, et al. Tenascin expression in basal cell carcinoma. Br J Derrnatol 1992; 127: 571-574. Schalkwijk J, Steijlen PM, van Vlijmen Willems IM, Oosterling B, Mackie EJ, Verstraeten AA. Tenascin expression in human dermis is related to epidermal proliferation. Am J Pathol 1991; 139 1143-1 150. Chiquet Ehrismann R, Kalla P, Pearson CA, Beck K, Chiquet M. Tenascin interferes with fibronectin action. Cell 1988; 5 3 383-390. Spring J, Beck K, Chiquet Ehrismann R. Two contrary functions of tenascin: dissection of the active sites by recombinant tenascin fragments. Cell 1989; 5 9 325-334. Murphy Ullrich JE, Lightner VA, Aukhil I, Yan YZ, Erickson HP, Hook M. Focal adhesion integrity is downregulated by the alternatively spliced domain of human tenascin. J Cell 5iol 1991; 115 1127-1136. (Published erratum appears in J Cell Biol 1992; 116 833.) Prieto AL, Anderson Fisone C, Crossin KL. Characterization of multiple adhesive and counteradhesive domains in the extracellular matrix protein cytotactin. J Cell Biol 1992; 119 663-678. Friedlander DR, Hoffman S, Edelman GM. Functional mapping of cyto- tactin: proteolytic fragments active in cell-substrate adhesion. J Cell Biol 1988; 107: 2329-2340. Prieto AL, Edelman GM, Crossin KL. Multiple integrins mediate cell attachment to cytotactinltenascin. Proc Natl Acud Sci USA 1993; 90: 101 54-101 58. Rueee CR. Chiauet Ehrismann R. Alkan SS. Tenascin. an extracellular

I-

matrix protein, exerts immunomodulatory activities Proc Nut1 Acad Sci USA 1989, 86: 7437-7441 Hemesath TJ, Marton LS, Stefansson K. Inhibition of T cell activation by the extracellular matrix protein tenascin. J Immunol 1994; 152 5199-5207. Erickson HP. Tenascin-C, tenascin-R and tenascin-X: a family of talented proteins in search of functions. Curr Opin Cell Biol 1993; 5 869-876. Herrick SE, Sloan P, McGurk M, Freak L, McCollum CN, Ferguson MW. Sequential changes in histologic pattern and extracellular matrix deposition during the healing of chronic venous ulcers. Am J Puthol 1992; 141: 1085-1095. Van den Born J, van den Heuvel LP, Bakker MA, Veerkamp JH, Assmann KJ, Berden JH. A monoclonal antibody against GBM heparan sulfate induces an acute selective proteinura in rats. Kidney Int 1992; 41: 115-123. Taylor HC, Lightner VA, Beyer WF Jr, ef al. Biochemical and structural studies of tenascinfhexabrachion proteins. J Cell Biorhem 1989; 41: 71-90. Larjava H, Salo T, Haapasalmi K, Kramer RH, Heino J. Expression of integriiis and basement membrane components by wound keratinocytes. J CIin Invest 1993; 9 2 1425-1435. Tucker RP, Hammarback JA, Jenrath DA, Mackie EJ, Xu Y. Tenascin expression in the mouse: in situ localization and induction in v i m by bFGF. J Cell Sri 1993; 104: 69-77. Rettig WJ. Erickson HP, Albino AP, Garin Chesa P. Induction of human tenascin (neuronectin) by growth Factors and cytokines: cell type-specific signals and signalling pathways. J Cell Sri 1994; 107: 487497. Slavin J . The role of cytokines in wound healing. J Pathof 1996; 178 5-10.