Use of nude athymic mice for the study of hypertrophic scars and keloidsVascular continuity between mouse and implants.код для вставкиСкачать
THE ANATOMICAL RECORD 225:189-196 (1989) Use of Nude (Athymic) Mice for the Study of Hypertrophic Scars and Keloids: Vascular Continuity Between Mouse and Implants C. WARD KISCHER, DON SHERIDAN, AND JANA PINDUR Department of Anatomy, University of Arizona College of Medicine, Tucson, Arizona 85724 ABSTRACT Hypertrophic scars and keloids appear to be unique to humans since animals are not known to form these lesions. Therefore, in an effort to develop an experimental model for their study, implants of these human lesions were made in nude (athymic) mice (ndnu) in suprascapular subcutaneous pockets. The implants were recovered from 2 to 246 days. By histological and fine structural parameters all implants remained viable and their morphological character was maintained. Selected mice were injected with barium to confirm by microangiography vascular flow between mouse and implant. Hoechst stain for DNA, used to distinguish mouse cells from human cells, confirmed vascular anastamosis between host and implant: barium-filled vessels in the interior of the implant demonstrated human endothelial cells. Peripheral vascularization of the implant with minimal ingrowth of mouse vessels occurs during the first 8 days. Anastamosis probably occurs sometime before 16 days postimplantation, or earlier, depending upon the availability of patent microvessels in the implanted tissue. The presence of the implant does not appear to prompt a continuing vascular growth into or throughout the implant. The time frame of 16 days postimplantation should be taken into account when developing schemata of experimental or therapeutic modalities. The hypertrophic scar and its closely related lesion, the keloid, are often the sequelae of deep surface injury. There are reports in the literature claiming that these lesions have been produced (Silverstein et al., 1976) or found (Marcenac, 1951) in animals. However, these claims have not been confirmed and the concensus is that no animal model exists for these lesions. Studies of hypertrophic scar and keloid implanted into subcutaneous pockets of the nude mouse1 demonstrated that they survive and retain their character after both short- and long-term implantation (Shetlar et al., 1985; Kischer et al., 1988). This permits experimental investigation of these lesions, which otherwise would be impossible from an ethical or legal point of view. Because the implants survive unchanged in their morphology up to 246 days (Kischer et al., 1988) one could assume vascular anastomosis occurs between host and implant. Yet, it remained to be demonstrated. Therefore, the objectives of the present study were 1) to demonstrate vascular continuity between host and implant, 2) to establish a time frame during which vascular continuity occurs, 3) to determine if anastomosis is a continuing event throughout the time of implantation, and 4) to determine if the presence of the im- ‘Nude mouse is the correct designation of this animal, which also is known genotypically as ndnu. The nude mouse is also always athymic (Krueger and Briggaman, 1982). 0 1989 ALAN R. LISS, INC. plant itself prompts a vascularization by the mouse into the implant. MATERIALS AND METHODS Thirteen samples of hypertrophic scars, six of keloids and six of Dupuytren’s contracture, were obtained as excess material from surgical procedures. These samples provided 194 pieces of tissue used as implants into 97 mice. Each piece was trimmed to a standardized size of 5 x 8 x 5 mm. In every case the donor tissues were taken from the nodular areas of the lesions. Nude mice were anesthetized with sodium pentabarbitol (40 mglkg body weight), and full-thickness incisions were made through the skin over each scapular area. The fine tips of surgical scissors were used to produce a subcutaneous pocket. A donor tissue was placed in each pocket and the skin was closed with 12 mm wound clips. The wounds were healed after 10 days and the clips were removed. Implants remained in the mice for varying lengths of time up to 246 days. Seventeen mice with 34 implants from hypertrophic scar, seven mice with 14 keloid implants, and two mice with four implants from Dupuytren’s contracture were selected for studies of vascular continuity. Twenty mice with 40 implants were studied from 2 to 20 days of implantation and six mice with 12 Received November 14, 1988;accepted February 21, 1989. 190 C.W. KISCHER ET AL. Fig. 1. A macrosectioned 16-day implant (I) with overlying mouse skin (S), injected with barium. Each section is approximately 1 mm thick. The bright areas are reinforced areas of barium due to vascular plexi. ~ 3 . 2 . Fig, 2. Macrosection of barium-injected 8-day implant with overly- ing mouse skin. Section shows minimal penetration of barium into implant. Other sections were negative. x 13. Fig. 3. Two macrosections of 8-day implant showing barium penetration. x 7.2. implants were studied from 40 to 106 days of implantation. For the injection procedure, a n injectate was prepared consisting of 33 g of barium sulphate suspended in 100 ml of water (Micropaque-Nicholas Labs. Ltd., Slough, England) and containing 4 g of gelatin. The mouse was prepared by injecting 0.1 cc of heparin I.P. After several minutes to allow circulation of the heparin, the mouse was anesthetized with ether, injected with a lethal dose of sodium pentabarbitol, and secured to a dissecting table. The thoracic cavity was opened and the aorta entered via the left ventricle with a 23- gauge plastic cannula attached to a 50 cc syringe. The tip of the cannula was passed into the aorta near the origin of the right brachiocephalic artery. The cannula was secured with ligatures around the aorta and the heart itself. Two volumes of 12 cc each of a dilute heparin solution (approximately 0.2 cc in 30 ml of physiological saline) was then injected into the circulatory system. The left jugular vein was nicked to allow for drainage (which also occurred through the intercostal veins) a s a n indication of perfusion. Three to 5 volumes (12 cc each) of the barium solution was then injected. Upon completion of the barium perfusion the whole 191 VASCULARIZATION O F IMPLANTS mouse was placed in the freezer for 30 minutes t o allow the barium-gelatin solution to solidify and then was placed in Karnovsky's fixative overnight. The suprascapular skin with attached implants was removed and placed in a waxed petri dish for observation and covered with Karnovsky's fixative. One implant, along with overlying skin, was sectioned with a hand-made, hand-held macro-tissue slicer (Sheridan et al., 1989), which assures sections of uniform thickness approximating 1 mm. These sections were laid out in serial order for microangiography on 2" x 2 Kodak highresolution plates. The sections and plate were wrapped with a film of clear plastic to hold them in place and then exposed by using a Faxitron microradiographic system with 18 kVP for 24 minutes. The developed x-ray plates were photographed on a Zeiss photoscope with a 1x objective with a halogen light source and a 2-minute 30-second exposure on Kodak Panatomic-X film. The other, or companion, implant with overlying skin was processed for histological evaluation. To determine if anastomoses occurred between host and implant, tissue sections from 21 implants were evaluated by Hoechst dye DNA staining (Grimwood et al., 1986). Three implants were recovered from 4 to 20 days. Eighteen of the 21 were recovered between 27 and 179 days. Four of the 18 were barium-injected implants. The Hoechst staining procedure distinguishes mouse cell nuclei from human cell nuclei. Hoechst dye 33258 is a benzimidazole compound which, when applied to cultured mouse cells, induces an enlargement of the pericentric area of metaphase chromosomes (Kim and Grzeschik, 1974). There is no similar effect on human chromosomes. Grimwood et al. (1986) demonstrated in tissue cultures that the same Hoechst dye caused a bright discrete or punctate staining of mouse fibroblast nuclei, whereas human fibroblast nuclei showed a diffuse staining pattern. A similar staining pattern was seen in frozen sections of basal cell carcinomas grown subcutaneously in nude mice. Thus, we analyzed cells in the periphery of the implant and within the interior. We especially looked at the type of endothelial cell in barium-filled vessels. We have used the Hoechst dye on Karnovsky-fixed paraffin tissue sections. All of the tissues were originally fixed in Karnovsky's fixative, dehydrated in graded alcohols, and embedded in paraffin. Tissue sections from injected and noninjected implants were cut from paraffin blocks a t 6-8 p,m. After deparaffinization, the sections were placed in undiluted Hoescht's stain (#33258, Flow Labs, Inc.) for 30 minutes. The slides were drained and then washed 3 x in dHzO. After air drying the slides were coverslipped with Permount. Identification of mouse nuclei is made by observing the specific fluorescing nuclear pattern in the form of punctate clumps after exposing the stained sections to U.V. light by using a Zeiss 01 filter set (excitation 365 nm, emission 480 nm). Human nuclei show the diffuse pattern. All exposures on film in the photomicroscope were made under identical conditions. Likewise processing of the film and printing were performed under identical conditions. Histological sections of one implant were evaluated by assigning arbitrary values for magnitude of vessels present and the extent of microvascular occlusion. These values were compared with those made from the TABLE 1. Implant tissues examined by barium injection for host-implantvascular continuity' Days of implantation K1 HS1 HS2 HS3 HS4 HS5 K2 HS6 DC1 DC2 2 - 4 - - _ 6 8 + - -+ 9 + 10 11 + + + 12 14 16 + + 20 40 + 62 + 69 70 75 106 + - + + + + + 'Total = 26 mice, 52 implants; + = confirmed barium inflow; 5 = minimal barium inflow; - = no barium inflow; HS = hypertrophic scar; K = keloid; DC = Dupuytren's contracture. histological sections of nonimplanted (zero-time) tissues from the same surgical specimen. This would tell us if implantation prompted a change in vascular availability for anastomosis. By evaluating implants of varying ages and comparing them with their nonimplanted (zero-time) samples, we could determine if anastomosis was a continuing process. RESULTS By examining serial sections completely through the implant we could distinguish barium-filled vessels within the implant from those in the periphery (Fig. 1). All implanted tissues retained their zero-time morphological character with no evidence of implant rejection as evidenced by lymphocyte or small cell infiltration. Gross examination of all implants showed that most are supplied within the first several days by a single branch of a subcutaneous artery which then arborizes throughout the periphery of the implant. In a few cases of the older implants a second branch was observed. Table 1 shows the samples of tissues used, the number of mice injected, the number of implants injected, the days postimplantation studied in each mouse, and the results for each injection. Of the four lesions implanted into 20 mice from which implants were examined from 2 to 20 days, one shows minimal ingrowth up to 16 days (Fig. 2). The other three lesions examined show more extensive barium inflow by 8, 11, and 16 days (Figs. 3-5, respectively). Each of the lesions examined a t 40-106 days demonstrates barium inflow, which varies in magnitude (Figs. 6,7). Stained sections from implants not x-rayed show that every implant demonstrates extensive vascularization in the periphery, extending into the implant from 100 to 300 pm, most of which is filled with the barium mixture (Fig. 8). Each of these companion implants was examined for barium within the vessels deeper into the implant. All but one sample confirm the presence of internalized 192 C.W. KISCHER ET AL. Fig. 4. Three macrosections of 11-day implant with barium penetration. x6.3. Fig. 5. Three macrosections of a 16-day implant with substantial barium penetration. x 7.2. Fig. 6. Macrosection of 70-day implant showing minimal barium penetration. x 12.2. Fig. 7. Two macrosections of barium injected implant of 75 days. Extensive barium penetration. X 15.6. 193 VASCULARIZATION O F IMPLANTS Fig. 8. Tissue section of implant injected with barium. Periphery of implant shows extensive vascular arborization with barium filling, extending into interior of implant. x 90. Fig. 9. H & E section of companion implant to the one used for microangiogram in Figure 2 showing minimal penetration of barium. Interior of implant contains very few microvessels. x 90. barium. In the one exception the microangiogram shows minimal barium inflow (Fig. 2). The companion implant demonstrates very few interior vessels (Fig. 9). In addition to the barium results we examined a minimum of eight sections of every implant (total = 776) by hematoxylin and eosin staining and Masson’s trichrome method for evidence of microvascular ingrowth. Six of those implants were serially sectioned and similarly examined. Figure 1Oa-d is representative of the evidence observed by routine tissue section examination. Hoechst staining of tissue sections in which barium was found in the peripheral vascularization demonstrates that these vessels are clearly of mouse origin (Fig. 1la,b). Punctate patterns of nuclear chromatin were evident in the peripheral vessels. However, those vessels within the interior of the implant where barium is found show the endothelial cells to be of human origin (Fig. 12a,b). Hoeschst-stained sections from all other implants demonstrate only human endothelial cells around internal vessels. In the case in which tissue sections from the implants were arbitarily graded as to the extent of vessels and magnitude of microvascular occlusion, it was found that neither the number of vessels nor the level of oc- clusion in any implants, regardless of length of time, changed from that in the nonimplanted parent tissue. DISCUSSION The combined barium injection, histological section study, and Hoechst staining demonstrate that vascular anastomosis occurs between host and implant and is maintained throughout extended implantation times. Among the implants in this study, vascular channeling into the implant occurred by the eighth day in three cases and by the 1l t h day in the other case. The degree of barium inflow varied from implant to implant and in early detection was confined to the outer portions of the implants. Older implants showed internalized barium deeper into the implant. The results of Hoechst dye staining are critical for supporting anastomosis. Barium inflow into the implant may be occurring via arterial side channels growing from the mouse without a venous return. Consequently, the microangiograms do not, of themselves, prove anastamosis. Only when we demonstrate human endothelial nuclei surrounding a barium-filled lumen do we know anastomosis truly occurred. The histology of the implants complemented the results with barium in that the degree of internalized barium did not appreciably differ from what was 194 C.W. KISCHER ET AL. Fig. 10. Set of four serial sections from implant of 111 days, noninjected. Continuity of peripheral vessel with a deeper vessel shown by arrows. Masson’s trichrome stain. x 200. VASCULARIZATION OF IMPLANTS 195 Fig. 11. Section of barium-injected implant of 6 days stained only with Hoechst stain. a: Transmitted light micrograph of periphery of implant with barium-filled vessels. The endothelial nuclei are identified by light areas (arrows). Barium (B). x 500. b Section observed under U.V. light only. Fluorescence shows endothelial nuclei with punctate pattern characteristic for mouse cells (arrows). Fig. 12. Section of implant of 75 days, stained only with Hoechst stain. a: Transmitted light micrograph of interior barium-filled vessel of implant. The endothelial nuclei are identified by the light areas (arrows). x 500. b Section viewed under U.V. light only. Fluorescence shows endothelial nuclei around barium-filled vessel with diffuse pattern (arrows). observed in the microangiograms. The magnitude of vascular anastomosis appeared to reflect the degree of patent capillaries in the implant. The presence of the implant does not appear to prompt a continuous or extensive neovascularization into the implant. Several implants showed very little barium inflow, and these also demonstrated very few patent microvessels. Vascular growth from the mouse appears extensive at first only at the periphery but then must rely on available patent microvessels within the donor tissue. This seems confirmed by the Hoechst stain data, which show older implants with interior barium-filled vessels with only human endothelial cells. Blood vessels in the implant which are available for anastomosis at the time of implantation will determine the extent of anastomoses and, henceforth, the extent of anast omoses does not change. This fact is important 196 C.W. KISCHER ET AL. for characterizing this potential model because it suggests that what is implanted is not complicated by invasive growth from the host. From the data presented it is suggested that when considering the use of hypertrophic scar or keloid implants as a potential model for the study of therapeutic applications, enough time should be given the implant/ host system to initiate internal anastomosis. We judge this to be at least 8 days and it could reasonably be about 16 days postimplantation. A final note: it should be mentioned which system appears to be anastomosing first, arterial or venous. In wound-healing systems, a search of the literature reveals no studies indicating which system regenerates first. Hunt et al. (19781, in a study of vascular outgrowth in healing wounds, showed that capillary budding is prompted by low oxygen tension and proceeds toward the gradient of high oxygen tension. Capillary growth, presumably, occurs preemptively from the arterial side. This is reasonable because these microvessels are filled with erythrocytes. Yet, the time when arterial and venous sides are both reestablished is not known. Apparently, the delay of a complete vascular routing within the implant does not affect the viability of the tissue. The fibroblasts of the hypertrophic scar and keloid are facultative anaerobes, anyway (Hunt and Van Winkle, 1979). If nutrients and oxygen are available only by tissue diffusion from the periphery, their condition would continue to be sustained. Hypoxia of hypertrophic scars has been demonstrated (Sloan et al., 1978). Hypoxia also is known to stimulate collagen synthesis (Chvapil et al., 1970; Chvapil, 1974). This would explain the excess of collagen and, consequently, the bulk of the hypertrophic scar. The hypoxia of the scar is explained by widespread microvascular occlusion demonstrated by Kischer et al. (1982).Thus, the long-term clinical course of this lesion may be explained. ACKNOWLEDGMENTS The authors wish t o express their sincere appreciation t o Andrzej Fryczklowski, M.D., and Bridget Gar- rity, M.D., for technical assistance with the injection technique and radiographic procedures, and to Mike Jaqua for technical assistance. This study was supported, in part, by NIH research grant 5R0134928. REFERENCES Chvapil, M. 1974 Pharmacology of fibrosis and tissue injury. Environ. Health Perspect., 9:283-294. Chvapil, M., J . Jurych, and E. Mirejorska 1970 Effects of long-term hypoxia on protein synthesis in granuloma and in some organs in rats. Proc. SOC.Exp. Biol. Med., 135:613-617. Grimwood, R.E., C.F. Ferris, L.D. Nielsen, J.C. Huff, and R.A.F. Clark 1986 Basal cell carcinomas grown in nude mice produce and deposit fibronectin in the extracellular matrix. J . Invest. Dermatol., 87:42-46. Hunt, T.K., and W. Van Winkle, J r . 1979 Normal repair. In: Fundamentals of Wound Mangement. T.K. Hunt and J.E. Dunphy, eds. Appleton-Century-Crofts, New York, pp. 2-62. Hunt, T.K., W.B. Conolly, S.B. Aronson, and P. Goldstein 1978 Anaerobic metabolism and wound healing: An hypothesis for the initiation and cessation of collagen synthesis in wounds. Am. J. Surg., 135:328-332. Kim, M.A., and K.H. Grzeschik 1974 A method for discriminating murine and human chromosomes in somatic cell hybrids. Exp. Cell Res., 88:406-410. Kischer, C.W., A.C. Theis, and M. Chvapil 1982 Perivascular myofibroblasts and microvascular occlusion in hypertrophic scars and keloids. Hum. Pathol., 13:819-824. Kischer, C.W., J. Pindur, M.R. Shetler, and C.L. Shetlar 1988 Implants of hypertrophic scars and keloids into the nude (athymic) mouse: Viability and morphology. J . Trauma, in press. Krueger, G.G., and R.A. Briggaman 1982 The nude mouse in the biology and pathology of skin. Exp. Clin. Res., 2:301-322. Marcenac, N. 1951 Cheloides cicatricelles du cheval. Rec. Med. Vet., 127:385-398. Shetlar, M.R., C.L. Shetlar, L. Hendricks, and C.W. Kischer 1985 The use of athymic nude mice for the study of human keloids. Proc. Sac. Exp. Biol. Med., 179:549-552. Sheridan, D., C.W. Kischer, and J. Pindur 1989 A cutting device to provide uniform serial macro sections of tissue. Lab. Med., in press. Silverstein, P., M.N. Goodwin, G.L. Raulston, and B.A. Pruitt 1976 Hypertrophic scar in the experimental animal. In: The Ultrastructure of Collagen. Its Relation to the Healing Wounds and to the Management of Hypertrophic Scar. J.J. Longacre, ed. Chas C. Thomas, Springfield, IL, pp. 213-236. Sloan, D.F., R.D. Brown, C.H. Wells, and J.G. Hilton 1978 Tissue gasses in human hypertrophic burn scars. Plast. Reconstr. Surg., 61:43 1-436.