Revascularization of skin transplanted into the brain: Source of the graft endothelium

MICROVASCULAR

RESEARCH

28, 113-124 (1984)

Revascularization of Skin Transplanted into the Brain: Source of the Graft Endothelium P. A. STEWART, L. G. CLEMENTS, AND M. J. WILEY Department of Anatomy, University of Toronto, Toronto, Ontario MSS IA8, Canada Received Muy 16, 1983 Transplantation of tissues into the brain is becoming a feasible therapeutic approach to some neurological diseases. The fate of the graft vasculature is not well understood. The purpose of the present study was to determine the source of the endothelium in the revascularized skin grafts transplanted into the brain. We hypothesized that if the skin endothelium were replaced by brain endothelium then we should observe in the grafts the following: (a) degenerating endothelium soon after grafting, (b) regenerating endothelium subsequently, (c) a time course for reestablishment of circulation that is consistent with the time required for vessel growth and invasion, and (d) doubling of the vascular basement membrane a few weeks later. We found only a few degenerating endothelial cells soon after transplantation and no evidence of regenerating endothelium or of double layers of basement membrane even after prolonged survival. The vessels within mature grafts had morphological characteristics typical of normal skin vessels. We concluded that when tissue fragments are transplanted to the brain, native graft vessels survive and anastomose with host vessels.

INTRODUCTION

The transplantation of tissues into the brain, although not a recent technique (Medawar, 1948; Glees, 1955) has been the object of renewed interest since it was shown that transplanted fragments of embryonic neural tissue survive and integrate functionally with the host brain (Bjorklund and Stenevi, 1971, 1979; Bjorklund et al., 1977). The transplantation of nonneuronal tissues into the brain has both experimental and therapeutic potential. Using similar techniques it may be possible to study interactions between central neurons and peripheral tissue. Furthermore, in hereditary metabolic defects, genetically normal tissue transplanted into the brain might serve as a source of normal enzyme that would not pass the blood-brain barrier if applied systematically. Fragments of skin have been successfully grafted into the brain (Raju and Grogan, 1977; Medawar, 1948), however, little attention has been paid to their revascularization. In the work reported here we have examined the process of revascularization of skin fragments grafted into the cerebral hemisphere. We specifically questioned whether the endothelial cells in the revascularized grafts originate in the brain or whether they are native skin endothelial cells that survive transplantation. We hypothesized that if the native skin vessels die and are replaced by ingrowing endothelium from the brain, then we should observe in the grafts the following: (a) large numbers of degenerating endothelial cells shortly 113 0026-286X84 $3.00 Copyright @ 19X4 by Academic Press, Inc. All rights of reproductmn in any form reserved. Printed in U.S.A.

114

STEWART, CLEMENTS, AND WILEY

after grafting, (b) regenerating cells subsequently, (c) a lag in the reestablishment of circulation consistent with the time required for vascular invasion, and (d) duplication of the vascular basement membranes 3-4 weeks after grafting. MATERIALS

AND METHODS

Fertile chicken eggs were obtained from Martindale Hatcheries, Caledonia, Ontario, and hatched in our own laboratory. Grafts were made using the young chicks within the first week after hatching. Autografts were used exclusively to avoid the problems of immunological rejection in this species (Clements and Stewart, unpublished observations). Surgical procedure (Fig. I). Pilot studies showed that transplanted skin grafts had a tendency to curl up such that the epidermis was always on the host side. Under these conditions blood vessels were unable to penetrate the grafts and the grafts died. The following technique was designed to ensure that the grafts were implanted “dermis side out.”

3.

FIG. 1. Method of implanting a skin fragment “dermis-out” into the brain. (1) The skin over the left hemisphere has been incised and the edges retracted. A sterile B-gauge needle is inserted through the cranium and dura and into the hemisphere. A core of brain tissue is removed with the needle. (2) The skin to be grafted is placed dermis-down over the opening in the cranium and the edges spread out. (3) The tips of fine forceps are placed on the graft over the opening and used to push the skin gently into the brain so that the dermis comes in contact with the brain tissue.

CEREBRAL SKINGRAFTS-REVASCULARIZATION

115

The surgery was carried out under local anesthesia. Approximately 0.05 ml of xylocaine (2%) was injected into the scalp overlying the left hemisphere and into the skin of the axillary region of l- to lo-day-old chicks. The axillary region was chosen as the donor site because it has few feather follicles. At least 20 min were allowed to pass to enable the xylocaine to diffuse through the cranium and into the dura. Brain tissue itself is insensitive to pain. The skin of the axilla was washed with 70% alcohol and a small fragment approximately 2-3 mm was removed and placed in sterile Hank’s saline (Hanks and Wallace, 1949). Down was removed from the scalp overlying the left hemisphere. The skin of the scalp was incised and retracted to expose the cranium. An opening was made in the cranium and dura using a sterile l&gauge needle. The graft was placed over the opening and the tips of fine forceps were used to push the graft into the hemisphere as illustrated in Fig. 1. The edges of the skin incisions were apposed and sealed with sterile skin closures and the birds were injected with approximately 15,000 units of penicillin. All instruments were sterilized with 70% alcohol prior to use. Circulation in the grafts. To determine when circulation was established in the grafts heparinized chick Ringer’s followed by filtered colloidal carbon was injected transcardially into the vascular systems of chicks that had undergone surgery l-4 days earlier. The grafts embedded in the surrounding brain were removed and fixed in 10% neutral buffered Formalin, cut into approximately lmm slices, and mounted flat in depression slides. The carbon-filled vascular channels were examined under low power to determine whether they extended into the graft. Histology. Grafts were harvested at various times from 3 hr to 6 weeks after transplantation. Tissues for study by light microscopy were fixed in Bouin’s fluid, routinely embedded in paraffin, sectioned, and stained with hematoxylin and eosin. For electron microscopy tissues were fixed either by immersion (3-hr to 4-day grafts) or by transcardiac perfusion (3- and 6-week grafts) with 2% glutaraldehyde plus 2% paraformaldehyde in 0.1 M cacodylate buffer as outlined previously (Stewart and Wiley, 1981). RESULTS Development of the Grafts. A total of 79 chicks were grafted. Of these, 47 (67%) living grafts and 12 dead or degenerating grafts were found when the chicks were sacrificed. In the remaining 20 birds, no graft could be found. By 1 day after grafting, some pyknotic nuclei were found in the areas most distant from the surrounding brain, e.g., at the tip of an infold (Fig. 2a). By the second day, focal necrosis was more advanced. The skin had a mosaic appearance with areas of living skin alternating with areas of necrotic skin (Fig. 2b). By 3 days the living epidermis had become thicker and mitotic figures could be seen at the edges where living and necrotic epidermis met (Fig. 2~). By 1 week after grafting, necrotic areas were no longer seen. The epidermis appeared to be continuous and accumulations of keratin and cellular debris occupied the central portions of the grafts (Fig. 2d). Although the grafts were implanted “dermis-out” (Fig. 3a), by 10 days some grafts had a distinct layer of epidermis on both internal and part of the external

116

STEWART,

CLEMENTS,

AND

WILEY

FIG. 2. Early changes in skin grafts transplanted to the cerebral hemisphere. (a) l-day graft. Tissue degeneration and pykontic nuclei can be seen in an infold (arrow). (X 400). (b) 2-day graft. Degeneration in focal necrotic areas (arrow) is more advanced. ( x 400). (c) 3-day graft. Hypertrophied area of epidermis (arrow) seems to be migrating to the right over a denuded area. (X 220). (d) 7day graft. The graft is entirely composed of healthy skin. The top of the photograph shows the central area of the graft containing sloughed keratin and cellular debris. ( x 220).

surfaces. Thickened areas of epidermis reminiscent of the regenerating epithelium seen in younger grafts, were found in the graft-brain interface (Fig. 3b). During the subsequent several weeks the external layer of dermis became more complete and thicker. In ares where keratin accumulated a slight inflammatory reaction FIG. 3. Late changes in skin grafts transplanted to the cerebral hemisphere. (a) 7-day graft. A distinct epidermal layer can be seen on the internal surface enclosing sloughed keratin and cellular debris. The dermal aspect of the graft is apposed to surrounding brain tissue. Most of the contiguous brain tissue separated from the graft during the histological procedure. ( x 75). (b) l&day graft. A layer of hypertrophied epidermis (arrow) appears to be migrating to the left along the brain (br) and the graft (gr) interface. k, Sloughed keratin. (x 190). (c) Mature graft. An epidermal layer covers both the internal surface and the brain surface of the graft. Keratin accumulation and some shrinkage are obvious in the graft-brain interface. An inflammatory reaction can be seen in the brain tissue adjacent to the graft. (X 75).

CEREBRAL

SKINGRAFTS-REVASCULARIZATION

117

118

STEWART,

CLEMENT&

AND WILEY

could be seen in the adjacent brain tissue (Fig. 3~). By 6 weeks after grafting 7 of 14 grafts examined were dead. The dead grafts were completely surrounded by a thick layer of epidermis and keratin. An inflammatory reaction could be seen in the contiguous brain tissue (Fig. 3~). Reestablishment of the Circulation As early as 2 days after grafting some carbon-filled vessels could be seen in the dermis of the graft (Fig. 4a). By 3 days carbon-filled vessels were more numerous (Fig. 4b), and by 4 days the density of the carbon-filled vessels in the graft was similar to that in normal dermis (Fig. 4~). Endothelial Ultrastructure Within 3 hr of transplantation, the endothelial cells of vessels in the grafts were swollen and the sizes of the vascular lumena reduced (Fig. 5). This generalized swelling was not evident in the vessels of grafts fixed at later stages, although on occasion, individual cells showed signs of edema. From 12 hr to 3 days after transplantation both healthy and degenerating vessels could be identified in the grafted samples. The endothelium of the healthy vessels resembled that of the vessels in control samples of dermis (Fig. 6). The cells had relatively smooth luminal and abluminal outlines and rested on well developed and continuous basement membranes. Adjacent cells were closely applied and tight junctions appeared to be present between some cells. No frank gaps were identified between cells. The cytoplasm of the cells was characterized by large numbers of pinocytotic vesicles, widely varying numbers of small mitochondria, and only occasional profiles of rough endoplasmic reticulum. The nuclei were large but contained relatively small amounts of heterochromatin. In degenerating vessels the endothelial cells displayed long villous luminal projections, fragmented basement membranes and frank gaps between cells. In addition, the nuclei contained larger amounts of condensed heterochromatin than those of healthy vessels and the cytoplasm was vesiculated with some accumulation of lipid. In grafted tissues fixed 3 or 6 weeks after transplantation, no signs of degenerating vessels were present. Furthermore, the ultrastructural characteristics of the vascular endothelia were comparable to control samples of dermis and to healthy vessels observed in grafts fixed at earlier stages. No signs of degenerating vessels were present (Fig. 7). DISCUSSION Successful grafting of skin to brain has been reported in other species by a number of investigators (Medawar, 1948; Nathaniel and Clemente, 1959; Raju and Grogan, 1977). In the experiments reported here 67% of avian skin autografts to cerebral hemisphere were viable when recovered and half of the grafts were FIG. 4. Vascularization of Grafts. (a) 2-day graft. Keratin and a blood clot occupy the central portion of the graft (dark amorphous area). Carbon-filled vessels have penetrated into the graft in some areas (arrowheads). (X 175). (b) 3-day graft. Carbon-filled vessels penetrate the full thickness of the graft in most areas. The graft has partly separated from the contiguous brain tissue during the histological procedure. (x 175). (c) 4-day graft. Large and small carbon-filled vessels fill the graft. (x 175).

CEREBRAL

SKINGRAFTS-REVASCULARIZATION

119

120

STEWART,

CLEMENTS,

AND

WILEY

FIG. 5. Electron micrograph showing edema in a dermal vessel 3 hr after grafting skin to brain. (x 13,000).

viable for up to 6 weeks. More prolonged studies are needed to evaluate the long-term survival of such grafts. Although some focal areas of the graft became necrotic during the first few days, these were soon replaced by healthy neighboring tissue (Fig. 2). The majority of the grafts remained viable for at least 1 month. After this time, however, the number of degenerating and dead grafts found at harvest increased. In progressively older grafts an epithelium appeared to migrate over the underside of the dermis from the cut edge of the graft epidermis and to eventually cover the entire graft-brain interface (Fig. 3). All of the dead and degenerating grafts were completely encysted. On the basis of these observations we concluded that the death of the older grafts was due to their progressive isolation from the host tissue. In grafted tissue in which there is no attempt to surgically connect the vessels, the origin of the endothelial cells in the vascularized tissue is uncertain and seems to depend on the tissue type, developmental age of the graft, and whether the tissue is normal or neoplastic. In transplanted tumors (Ausprunk et al., 1975; Ausprunk and Folkman, 1976; Folkman, 1971; Gimbrone et al., 1974) and in normal adult skeletal muscle grafts (Hansen-Smith et al., 1980; Vracko and Benditt, 1970) the vascular endothelium dies and new vessels grow into the graft from the host. When embryonic tissues are grafted, however, the existing vessels in the graft survive and are reperfused by joining with vessels from the host (Ausprunk et al., 1975; Ausprunk and Folkman, 1976). Both reuse of the existing vessels (Haller and Billingham, 1967) and invasion of host vessels into the graft

CEREBRAL

SKINGRAFTS-REVASCULARIZATION

121

FIG. 6. (a) Electron micrograph illustrating the normal appearance of dermal vessels. (x 12,100). (b) Electron micrograph of a healthy vessel in the dermis of the skin fixed 12 hr after grafting to brain. No differences could be consistently observed in the appearance of these vessels compared to those of ungrafted dermis. (X 7700).

(Converse and Ballantyne, 1962; Merwin and Algire, 1956; Williams, 1959; Zarem et al., 1967) have been reported in adult skin grafts. We have found that native graft vessels form the majority of the vascular channels in the reperfused grafts. This conclusion is based on the following evidence: (a) Few degenerating endothelial cells. (b) Lack of immature or regenerating vessels. Several features characterize

122

STEWART,

CLEMENTS,

AND

WILEY

FIG. 7. Typical appearance of a dermal vessel in skin fixed 3 weeks after grafting to brain. The ultrastructural features of the endothelial cells are comparable to those in ungrafted control tissues. (X 13,700).

immature or regenerating vascular endothelial cells (Ausprunk and Folkman, 1976; Cliff, 1963; Schoefl, 1963; Szalay and Pappas, 1970). These include irregular luminal and abluminal projections, frank gaps between adjacent cells, a discontinuous basement membrane, plentiful rough endoplasmic reticulum, numerous mitochondria, and few pinocytotic vesicles. In our grafts these features were not observed. (c) Lack of doubling of the basement membrane. Under conditions in which endothelial cells die and are subsequently replaced, new endothelial cells from the host grow into the grafted tissue preferentially along the original basement membrane-defined vascular channels and subsequently elaborate new basement membrane internal to the first one (Makitie, 1977; Hansen-Smith et al., 1980; Vracko and Benditt, 1970, 1972). As a result endothelium derived from the host can be identified in the graft by its association with a double basement membrane. We found no evidence of a double basement membrane in any of the grafts even after prolonged (12 weeks) survival. (d) Rapid reestablishment of circulation. In transplants in which host vessels must proliferate and invade the graft tissue, the time from surgery to full circulation in the graft is in the order of 4-5 days (Schoefl, 1963, 1964; Ausprunk and Folkman, 1976; Hansen-Smith et al., 1980). In the present study, continuity between host and graft circulations could be demonstrated as early as 2 days following transplantation. Although in the early grafts we did not observe gross endothelial cell death, we did see some degenerating endothelium, usually surrounded by other degenerating tissue. This is in keeping with the fact that some areas of skin, usually at the tip of an infold, did degenerate and were replaced from adjacent areas of skin (Fig. 2). In these areas one would expect vessel death and replacement and

CEREBRAL SKINGRAFTS-REVASCULARIZATION

123

it seems likely that the new vessels would originate from vessels in the adjacent regenerating skin. We considered the possibility that at least some growth of brain vessels into the contiguous dermis may have taken place. This possibility is difficult to assess since the brain was damaged at the site of implantation and showed a typical inflammatory response which obscured changes in the vessels. However, although some ingrowth may have taken place, the striking lack of evidence for endothelial death and replacement within the graft supports the conclusion that skin fragments transplanted into the brain are revascularized primarily by means of anastomoses between existing graft vessels and host vessels. ACKNOWLEDGMENTS The authors acknowledge with gratitude the technical assistance of Mrs. K. Hayakawa and Miss D. Randall. This work was supported by the Medical Research Council of Canada.

REFERENCES AUSPRUNK, D. H., AND FOLKMAN, J. (1976). Vascular injury in transplanted tissues. Fine structural changes in tumor, adult and embryonic blood vessels. Virchows Arch. B: 21, 31-44. AUSPRUNK, D. H., KNIGHTON, D. R., AND FOLKMAN, J. (1975). Vascularization of normal and neoplastic tissues grafted to the chick chorioallantois. Role of host and pre-existing graft blood vessels. Amer. J. Pathol. 79, 597-610. BJORKLUND, A., JOHANSSON,B., STENEVI, U., AND SVENDGAARD,N. A. (1977). Re-establishment of functional connections by regenerating central adrenergic and cholinergic axons. Nature (London) 253, 446-448. BJORKLUND, A., AND STENEVI, Il. (1971). Growth of central catecholamine neurones into smooth muscle grafts in the rat mesencephalon. Brain Res. 31, l-20. BJORKLUND, A., AND STENEVI, U. (1979). Reconstruction of the nigrostriatal dopamine pathway by intracerebral nigral transplants. Brain Res. 177, 555-560. CLIFF, W. J. (1963). Observations on healing tissues: a combined light and electron microscopic investigation. Proc. R. Sot. Ser. B 246, 305-325. CONVERSE, J. M., AND BALLANTYNE, D. L. (1962). Distribution of diphosphopyridine nucleotide diaphorase in rat skin autografts and homografts. Plasf. Reconsrr. Surg. 30, 415-425. FOLKMAN, J. (1971). Tumor angiogenesis: Therapeutic implications. New Engl. J. Med. 285, 1182-

1186. GIMBRONE, M. A., COTRAN, R. S., LEAPMAN, S. B., AND FOLKMAN, J. (1974). Tumor growth and neovascularization: An experimental model using the rabbit cornea. J. Natl. Cancer Inst. 52, 413427.

GLEES, P. (1955). Studies of cortical regeneration with special reference to cerebral implants. In “Regeneration of the Nervous System” (W. F. Windle, ed.), pp. 94-111. Thomas, Springfield, Ill. GROTH, C. G., COLLSTE, H., DREBORG, S., HAKANSSON, G., LUNDGREN, G., AND SVENNERHOLM, L. (1980). Attempt at enzyme replacement in Gaucher disease by renal transplantation. Birth Defects 16, 475-490. HALLER, J. A., AND BILLINGHAM, R. E. (1967). Studies on the origin of the vasculature in free skin grafts. Ann. Surg. 166, 896-901. HANKS, J. G., AND WALLACE, R. E. (1949). Relation of oxygen and temperature in the preservation of tissues by refrigeration. Proc. Sot. Exp. Biol. Med. 71, 196-200. HANSEN-SMITH, F. M., CARLSON, B. M., AND IRWIN, K. L. (1980). Revascularization of the freely grafted extensor digitorum longus muscle in the rat. Amer. J. Anal. 158, 65-82. MAKITIE, J. (1977). Skeletal muscle capillaries in intermittent claudication. Arch. Pathol. Lab. Med. 101, 500-503.

124

STEWART,

CLEMENTS,

AND

WILEY

P. B. (1948). Immunity to homologous grafted skin. iii. The fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. &it.

MEDAWAR,

.I. Expt. Pathol.

29, 58-69.

R. M., AND ALGIRE, G. H. (1956). The role of graft and host vessels in the vascularization of normal and neoplastic tissue. J. Nat/. Cancer Inst. 17, 23-33. NATHANIEL, E. J. H., AND CLEMENTE,C. D. (1959). Growth of nerve fibers into skin and muscle grafts in rat brains. Exp. Neural. 1, 65-81. RAJU, S., AND GROGAN, J. B. (1977). Immunological study of the brain as a privileged site. Transpl. Proc. 9, 1187-l 192. SCHOEFL, G. I. (1963). Studies on inflammation. iii. Growing capillaries: Their structure and permeability. Virchows Arch. A: Pathol. Anat. 337, 97-141. SCHOEFL, G. I. (1964). Electron microscopic observations on the regeneration of blood vessels after injury. Ann. N. Y. Acad. Sci. 116, 789-802. STEWART, P. A. AND WILEY, M. J. (1981). Developing nervous tissue induces formation of bloodbrain barrier characteristics in invading endothelial cells. Dev. Biol. 84, 183-192. SZALAY, J., AND PAPPAS, G. D. (1970). Fine structure of rat cornea1 vessels in advanced stages of wound healing. Invest. Ophthalmol. 9, 354-365. VRACKO, R., AND BENDITT, E. P. (1970). Capillary basal lamina thickening: Its relationship to endothelial cell death and replacement. J. Cell Biol. 47, 281-285. VRACKO, R., AND BENDITT, E. P. (1972). Basal lamina: The scaffold for orderly cell replacement. Observations on regeneration of injured skeletal muscle fibers and capillaries. J. Cell Biol. 55, 406-419. WILLIAMS, R. G. (1959). Experiments on the growth of blood vessels in thin tissue and in small autografts. Anat. Rec. 133, 465-486. ZAREM, H. A., ZWEIFACH, B. W., AND MCGEHEE,J. M. (1967). Development of microcirculation in full thickness autogenous skin grafts in mice. Amer. J. Physiol. 212, 1081-1085. MERWIN,