- Email: [email protected]
In vitro and in vivo calcification of vascular bioprostheses
Biomaterials 19 (1998) 1651 — 1656
In vitro and in vivo calcification of vascular bioprostheses Jyotirmay Chanda*, Katsuyuki Kondoh, Keitaro Ijima, Makoto Matsukawa, Ryosei Kuribayashi Department of Cardiovascular Surgery, Akita University School of Medicine, Akita 010, Japan Received 18 July 1997; accepted 25 January 1998
Abstract Efficacy of different chemical treatments on calcification of vascular graft in vitro and in vivo was studied. Culture medium-filled rat aortas were separately treated in 0.2% glutaraldehyde and epoxy compound, and photooxidized in 0.01% methylene blue for a shorter period (group 1). Another group of rat aortas were separately treated in the same chemicals for a longer period (group 2). All fresh and treated aortas of both groups were cultured for 21 days in an organ culture medium and implanted (except for group 1) in weanling rats for five months. Histology and immunohistochemistry revealed that differently treated aortas of group 1 grow and calcify, and the smooth muscle cells between elastin fibers are the primary site of calcium deposition. In contrast, differently treated aortas of group 2 neither grew, nor did calcify in the medium except the epoxy compound cross-linked aorta of group 2 which did not grow but did calcify. Untreated aorta did not calcify. All fresh and differently treated aortic homografts calcified severely in rats. Our whole arterial segment-calcification system would be useful for analyzing the molecular and cellular mechanisms of both bioprosthetic and atherosclerotic calcification of vascular graft. New anticalcification technique is the only hope for better outcome of future vascular bioprostheses. ( 1998 Published by Elsevier Science Ltd. All rights reserved Keywords: In vitro calcification; Homograft; Photooxidation; Epoxy compound; Glutaraldehyde
1. Introduction Conduit failures from early degeneration and calcification in homograft conduits continue to limit the reconstructive congenital heart surgery. Recently, several techniques [1, 2] of cross-linking have been proposed to limit the calcification in bioprosthetic valves. In this study we have tried to find out whether these cross-linking processes are effective in prevention of calcification in homograft conduits. Unlike in leaflets or pericardium, calcification prevention in vascular bioprostheses is difficult to achieve. With in vivo calcification model factors, whether host or graft, generally important in calcification process of vascular bioprostheses cannot be easily distinguished. To exclude the host factor an in vitro model of calcification in a physiologic medium is necessary. To date well-established model of in vitro calcification in a whole segment of vascular bioprosthesis does not exist. Both in vivo and in vitro calcification
*Corresponding author. Present address: Bojpura, Sreedhar Tank Road, Jessore 7400, Bangladesh. Tel.: 0188 341111 10 3148; fax: 0188 36 2625; e-mail: [email protected]
model of vascular bioprostheses would allow more detailed investigation of the process resulting in calcification and development of further strategies for prevention of calcification of the prostheses in future. 2. Materials and methods Thoracic aortas were removed under sterile conditions from adult male Wistar rats (250—300 g) sacrificed with an overdose of ether, collected in tissue culture medium RPMI 1640 with HEPES buffer 25 mmol l~1 and Lglutamine (Gibco/BRL, Life Technologies, Inc., Paisley, UK), containing penicillin : streptomycin, 100 lg ml~1 (Gibco), gentamicin, 2.5 lg ml~1 (Schering-Plough, Osaka, Japan), amphotericin B, 5 lg ml~1 (Fungizone, Gibco) and transferred to the laboratory at room temperature. 2.1. Chemical treatment Upon arrival at laboratory, vascular clip was put at one end of the aorta followed by filling the aorta with culture medium (described above) and closing the other
0142-9612/98/$ — See front matter ( 1998 Published by Elsevier Science Ltd. All rights reserved. PII S 0 1 4 2 - 9 6 1 2 ( 9 8 ) 0 0 0 3 8 - 6
1652
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
end of the aorta with another vascular clip followed by chemical treatment: 2.2. Glutaraldehyde-treated rat aortas Culture medium-filled rat aortas were immersed in 0.2% glutaraldehyde (Glutaraldehyde EM 25%, TAAB Laboratories Equipment Ltd., Reading, UK) in phosphate buffer saline (PBS) (pH 7.4) at room temperature for 2 min followed by washing in a solution containing 1% glycine (Sigma chemical Co., St. Louis, MO) for 5 min (group 1). Another group of rat aortas were freely crosslinked in 0.2% glutaraldehyde for 4 weeks (group 2). 2.3. Epoxy-compound-treated rat aortas Culture-medium-filled rat aortas were kept in a solution of 4% glycerol polyglycidyl ether (Denacol' EX313, Nagase Chemicals Ltd., Osaka, Japan) and 2% ethylene glycol diglycidyl ether (Denacol' EX-810, Nagase) in PBS (pH 7.4) for 5 min at room temperature followed by washing in PBS (group 1). Another group of rat aortas were freely cross-linked in 4% glycerol polyglycidyl ether for 6 days followed by treating in 2% ethylene glycol diglycidyl ether for 7 days (group 2).
2.6. Cell viability assay To assess the effect of different chemicals on viability (live or dead) of cells, live/dead reduced biohazard viability cytotoxicity kit (L-7013) (Molecular Probes Inc., Eugene, OR, USA) was used. Working solution of the dye was prepared by pipetting 5 ll of Component A and 5 ll of Component B into a common 2.5 ml volume of Hank’s balanced salt solution (1 : 500 dilution of each, HBSS, Gibco). Segments of fresh and treated aortas were incubated in the diluted dye in complete darkness for 30 min at room temperature followed by washing in HBSS, fixing in freshly prepared 4% glutaraldehyde in HBSS for more than 15 min at room temperature, and finally washing in HBSS. After that, segments were submerged in OCT compound (Tissue Tek', Miles Inc., Elkhart, IN, USA) and quickly frozen in liquid nitrogen. The frozen samples were cut at 5 lm on a cryostat, mounted on glass slides, and observed under epifluorescent microscope (Olympus BH2, Tokyo, Japan). At excitation wavelength 485$11 nm, live (green fluorescent) and dead (red fluorescent) cells were viewed simultaneously. The fluorescence intensity of live cells was significantly lower than that of dead cells. 2.7. Implantation in rats
2.4. Photooxidized rat aortas Culture-medium-filled rat aortas were placed in a solution of photoactive dye (0.01% methylene blue in PBS, pH 7.4) (Methylene blue, Wako Chemical Co., Osaka, Japan). The solution was exposed to light of a 1000 W slide projector under continuous stirring for 5 min at 2°C followed by washing in PBS (group 1). Another group of rat aortas were photooxidized for 24 h (group 2). All fresh (untreated) and treated aortas of both groups were opened up longitudinally, cut into 5 mm lengths and cultured (n"30 of each group) for 21 days in the medium developed by Soyombo and colleagues [3]. Before culture, the viability of cells of the freshly isolated and treated aortas was checked with immunofluorescent staining on frozen sections described later. 2.5. In vitro culture protocol The culture medium consisted of: tissue culture medium RPMI 1640 with HEPES buffer 25 mmol l~1 and L-glutamine (Gibco), containing 30% fetal bovine serum (Gibco), penicillin:streptomycin, 100 lg ml~1 (Gibco), gentamicin, 2.5 lg ml~1 (Schering-Plough), amphotericin B, 5 lg ml~1 (Gibco), and maintained for 21 days at 37°C in a humidified atmosphere of 5% carbon dioxide in an air incubator. The tissue culture medium was replaced every 3 days. At the end of the culture period, aortic segments were washed in PBS fixed overnight in 10% formalin in PBS followed by processing and paraffin embedding.
Untreated (fresh) and differently treated rat aortas of group 2 (each 2 cm in length) were washed in copious amounts of normal saline and implanted subcutaneously in the back of both sides of 153 week-old male Wistar rats (30—50 g). Wounds were closed with 4-0 polypropylene. Samples were retrieved five months after implantation. Of two specimens from each rat, one was fixed overnight in 10% formalin in PBS followed by processing and paraffin embedding and another was processed for quantitative calcium estimation by atomic absorption spectroscopy [4]. 2.8. Histology and immunohistochemistry Serial sections of 2—3 lm were cut from each paraffin block and mounted on glass slides. To study the morphology of the specimens, hematoxylin and eosin (H & E), Masson’s trichrome, and elastin van Gieson stains were used. von Kossa stain was used to identify calcium phosphate. For immunohistochemistry, deparaffinized and rehydrated tissue sections were blocked with 0.1% hydrogen peroxide (H O ), incubated with 10% normal 2 2 swine serum (Gibco) for 20 min, and washed in PBS. Smooth muscle cells were detected with mouse anti alpha smooth muscle actin monoclonal antibody (clone 1 A4, 1 : 10 dilution, Immunon, Pittsburgh, PA, USA). To identify endothelial cells, anti-human von Willebrand factor serum (1 : 40 dilution, DAKO A/S, Glostrup, Denmark) was used. Bound monoclonal antibodies were
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
detected using horseradish peroxidase (HRP)-conjugated goat anti rabbit antibodies (IgG) (1 : 2500 dilution, MBL, Nagoya, Japan) or anti-mouse antibodies (IgG) (1 : 100 dilution, Tagoimmuno-chemicalsTM, Biosource International, Camarillo, CA, USA). Then, slides were exposed to 0.04% solution of 3,3@-diaminobenzinidine (Sigma) containing 0.03% hydrogen peroxide in 0.05 mol/Tris/ HCL buffer (pH 3.0) for color development, washed in running water for 20 min and counterstained with H & E. For detection of proliferative cell nuclear antigen (PCNA) in cultured aorta, incubation of the sections with mouse anti-PCNA/cyclin monoclonal antibody (clone PC10, 1 : 25 dilution, Novocastra Laboratories, Newcastle upon Tyne, UK) was carried out at 4°C overnight. After washes in PBS, biotinylated anti-mouse IgG (Dako) was applied at a dilution of 1 : 50 and incubated for 30 min at room temperature followed by incubation with strepavidin-peroxidase (1 : 500 dilution, Dako) at room temperature for 30 min. Slides were incubated with peroxidase-conjugated 3,3@-diaminobenzinidine (1 : 500 dilution, Dako) in Tris buffer (pH 7.6) for 5—7 min at room temperature and counterstained with H & E. Sec-
1653
tions were evaluated with side by side comparisons of the different stains on sequential tissue sections to allow correlation of the location of pathological changes. Statistical analyses were carried out with a two-tailed unpaired t-tests. Significant differences were found to exist at the P(0.01 level. 3. Results 3.1. Cell viability assay Live/Dead Reduced Biohazard Viability Cytotoxicity Kit demonstrated that cell viability of rat aorta was totally lost in differently treated aortas of group 2. However, short-term treatment did not affect the viability of endothelial cells of rat aortas of group 1. Cells other than endothelial also remained live in aortas of group 1. 3.2. Culture of rat aorta Untreated (fresh) and differently treated aortas of group 1 grew well in the medium. However, neointimal
Fig. 1. These photomicrographs show serial sections of 2 min glutaraldehyde cross-linked rat aorta (group 1) cultured for 21 days. Formation of neointima (arrow) (a), and migration of smooth muscle cells (arrow) to the neointima (b) can be seen. Deposition of calcium (black) in smooth muscle cells (between elastin fibers) can be noticed (c). Presence of PCNA (proliferating cell nuclear antigen)-positive cells (arrow) in different layers of the cultured aorta can be seen (d). (a) Elastin van Gieson stain, (b) anti-alpha actin antibody stain, (c) von Kossa stain, (d) anti-PCNA antibody stain. a, b, c, d, original magnification ]100.
1654
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
formation was more prominent in fresh aortas than that in treated aortas of group 1 after culture for 21 days (Fig. 1). PCNA-positive cells have increased in number (data not shown here), suggesting that both untreated (fresh) and differently treated aortas of group 1 grow in in vitro. Comparative study of adjacent sections of differently treated aortas of group 1 stained with either von Kossa or anti-alpha actin or elastin van Gieson stains revealed the deposition of calcium inside the smooth muscle cells between elastin fibers (Figs. 1 and 2). Calcified cells also stained PCNA-positive indicates that the cells were live and were capable of proliferation during calcification (Fig. 1). In contrast, differently treated aortas of group 2 neither grew, nor did calcify in the medium (Fig. 3) except the epoxy compound cross-linked aorta of group 2 which did not grow but did show calcific changes after culture. No calcification was observed in untreated (fresh) aorta.
Fig. 2. Histologic appearance of a photooxidized aorta of group 1 cultured in vitro. Notice calcific deposits (asterisk) in smooth muscle cells between elastic laminae. Arrow indicates the neointima. H & E stain. original magnification ]100.
Fig. 3. Morphologic feature of a glutaraldehyde cross-linked aorta of group 2 cultured in vitro. No neointima and no calcific deposit can be noticed. H & E stain. original magnification ]100.
Fig. 4. Microscopic appearance of untreated (fresh) aortic homograft implanted in rat for five months. Severe calcific changes can be noticed. Note that deposition of calcium (arrow) along elastin fiber can clearly be distinguished. von Kossa stain. original magnification ]100.
Fig. 5. This photomicrograph shows the calcific feature of epoxy compound cross-linked homologous aorta of group 2 implanted in rat for five months. Severe calcification of the graft can be seen. von Kossa stain. original magnification ]100.
Fig. 6. Calcium level in untreated (fresh) and differently treated aortic homograft of group 2 implanted in rat for five months. RA/Untr."Untreated (fresh) aorta, RA/GA"Glutaraldehyde-treated aorta of group 2, RA/MB"Methylene blue-mediated photooxidized aorta of group 2, RA/Epoxy"Epoxy compound-treated aorta of group 2. RA/Untr. vs. RA/GA: P"0.0002, RA/Untr. vs. RA/MB: P(0.0001, RA/Untr. vs. RA/Epoxy: P"0.0881.
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
3.3. Implantation in rats Despite the higher calcium level in differently treated than that in untreated (fresh) grafts, histology revealed that all fresh and differently treated aortic homografts calcified severely when implanted subcutaneously in weanling rats for 5 months (Figs. 4—6). On contrary to in vitro-grown grafts, elastin fibers are the primary site of calcification in fresh and differently treated aortic homografts implanted in rats (Fig. 4).
4. Discussion For the use of the in vitro calcification model, not only to simulate the calcification in vascular bioprostheses but also in atherosclerotic vascular organ, we were keen on finding out what would be the maximum period of treatment without affecting the viability of endothelial cells of rat aorta during a particular chemical treatment. The period of exposure of rat aortas of group 1 to different chemicals (see Section 2) was according to the protocol standardized for serving our purpose mentioned above. The chance of immunologic reaction can not be ignored in vivo study. This was the reason why differently treated (inadequate to suppress the antigenicity) aortas of group 1 were not implanted in rats. Carpentier [5] has introduced glutaraldehyde as a cross-linking agent of xenograft heart valve bioprostheses. Calcification is the main problem with glutaraldehyde-treated grafts. In order to inhibit calcification, alternative to glutaraldehyde, epoxy compounds are proposed as cross-linking agents of bioprosthetic heart valves [1]. Photooxidation of amino acids sensitized by methylene blue was described by Weil and associates [6]. To avoid glutaraldehyde-induced calcification, Mechanic [7] has used dye-mediated (methylene green) photooxidation technique for stabilization of heterologous tissue, and this tissue is used for fabrication of PhotofixTMa Pericardial Bioprosthesis (Carbomedics, Austin, Texas, USA) [2]. Methylene blue is occasionally applied to the adventitia of blood vessels during coronary artery bypass and other vascular procedures to assist in the orientation of the vessel. In our experiment, calcification did not occur in dead cells of aorta (group 2) cultured in vitro. Whereas, calcification did occur in chemical agent-induced injured cells of aorta (group 1) in vitro, suggesting that calcium influx into injured/dead cells is not a passive process and some how regulated at cellular level in vitro and at organic level in vivo. These facts would be helpful for understanding the underlying processes contributing to pathogenesis of calcification of a graft in general. For ischemic myocardiocytes [8—10] to calcify, reperfusion is necessary. Unlike our observation, Kim [11] has observed that glutaraldehyde-fixed cells calcify within a week when
1655
cultured in the medium. Glutaraldehyde fixation may be construed as a form of cell injury that results in calcification, however, in view of devitalization of the cells by fixation, increases in calcium and phosphorous as the mechanism of the prosthetic calcinosis are questionable [12]. Elastin-associated calcification was observed in allograft aortic wall [13] and glutaraldehyde-cross-linked heterologous vascular grafts [14] implanted subdermally in rats. The elastin fibers within the middle of the media were identified as the principal site of calcium accumulation, and the calcification of human aorta increases throughout aging [15]. Decrease in degree of cross-linkage in elastin was found in atherosclerotic aortas [16]. An open space of the macromolecular low-cross-linked elastin may provide a nucleation center for calcification of elastin [17]. In the media, the synthetic, cross-linked poly-pentapeptide of tropoelastin has been shown as the seeding matrix for calcium deposition [18]. However, the prevalence of discrepancy in primary sites of calcium deposition in a vascular tissue cultured in vitro or implanted in animal is difficult to explain. Tingfei and coworkers [19] have reported calcification in fresh, glutaraldehyde and epoxy compound cross-linked bovine pericardial tissue in vitro. No histology of in vitro-calcified tissue is available with this study. The amount of calcium (approx. 0.5 lg mg~1) [19] in the cultured tissue is hardly convincing in favor of calcification. However, epoxy compound cross-linked aorta (Fig. 3) of group 2 calcified in vitro in our study and this was quite demonstrateable histologically. Since histology was quite enough to indicate the calcification, the calcium level of cultured tissue was not quantified in our study. Why glutaraldehyde cross-linked or photooxidized aorta of group 2 does not but epoxy compound cross-linked aorta does calcify (despite the lack of growth) in vitro, remains illusive. The in vitro calcification model of bioprosthetic tissue proposed by others [20, 21] does not satisfy our requirements described earlier. Because the cells grow and proliferate, and endothelial layer remains intact (the viable and intact endothelium is not necessary to induce calcification), our whole arterial segment-calcification (not in isolated smooth muscle cells) system would be useful for analyzing the molecular and cellular mechanisms of both bioprosthetic and atherosclerotic calcification of vascular graft. In vivo calcification study shows that neither the epoxy compound cross-linking, nor the methylene-blueinduced photooxidation inhibits the calcification of homologous vascular graft implanted for five months. This fact again proves that calcification problem in vascular graft is difficult to solve. Untreated (fresh) rat aorta does not calcify in vitro, but does calcify in vivo, suggesting that conceivably, due to the immunogenic reactions fresh homograft aortas calcify in vivo. In contrast, the
1656
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
reason of calcification of treated tissue both in vivo and in vitro, might be due to the cross-linking process.
5. Conclusions Our whole arterial segment-calcification system would be useful for analyzing the molecular and cellular mechanisms of both bioprosthetic and atherosclerotic calcification of vascular graft. Because of calcification, homografts are no longer the reliable tissue to repair the congenital heart defects. Cross-linking in epoxy compound or photooxidation in methylene blue would not change the calcification feature of homologous vascular grafts. New anticalcification technique is the only hope for better outcome of future vascular bioprostheses.
Acknowledgements We gratefully acknowledge Professor Kentaro Yoshimura, Head, Department of Parasitology of Akita University School of Medicine, Akita, Japan for allowing us to use his biochemistry and cell culture laboratories for our ongoing work. We wish to express our sincere gratitude to Dr. Tatsuya Abe of Department of Parasitology of Akita University School of Medicine for his suggestions during the course of this work.
References [1] Sung HW, Shen SH, Tu R et al. Comparison of the cross-linking characteristics of porcine heart valves fixed with glutaraldehyde or epoxy compounds. ASAIO J 1993;39:M532—M536. [2] Moore MA, Bohachevsky IK, Cheung DT, Boyan BD, Chen WM, Bickers RR, McIlroy BK. Stabilization of pericardial tissue by dye-mediated photooxidation. J Biomed Mater Res 1994; 28:611—8. [3] Soyombo AA, Angelini GD, Bryan AJ, Newby AC. Neointima formation is promoted by surgical preparation and inhibited by cyclic nucleotides in human saphenous vein organ cultures. J Thorac Cardiovasc Surg 1995;109:2—12.
[4] Chanda J, Kuribayashi R, Abe T, Sekine S, Shibata S, Yamagishi I. Is dog a useful model for accelerated calcification study of cardiovascular bioprostheses? Artif Organs 1997;21:391—5. [5] Carpentier A, Lemaigre G, Robert L, Carpentier S, Dubost C. Biological factors affecting long-term results of valvular heterografts. J Thorac Cardiovasc Surg 1969;58:467—83. [6] Weil L, Gordon WG, Buchert AR. Photoo¨xidation of amino acids in the presence of methylene blue. Arch Biochem Biophys 1951;33:90—109. [7] Mechanic GL. Process for cross-linking collagenous material and resulting product. US Patent no. 5,147,514. 1992. [8] Burton KP, Hagler HK, Templeton GH, Willerson JT, Buja LM. Lanthanum probe studies of cellular pathophysiology induced by hypoxia in isolated cardiac muscle. J Clin Invest 1977; 60:1289—302. [9] Shen AC, Jennings RB. Kinetics of calcium accumulation in acute myocardial ischemic injury. Am J Pathol 1972;67:441—52. [10] Hagler HK, Lopez LE, Murphy ME, Grecio CA, Buja M. Quantitative X-ray microanalysis of mitochondrial calcification in damaged myocardium. Lab Invest 1981;45:241—7. [11] Kim KM. Cell death and calcification of canine fibroblasts in vitro. Cells Mater 1994;4:247—61. [12] Kim KM. Apoptosis and calcification. Scanning Microsc 1995; 9:1137—78. [13] Webb CL, Nguyen NM, Schoen FJ, Levy RJ. Calcification of allograft aortic wall in a rat subdermal model. Am J Pathol 1992;141:487—96. [14] Paule WJ, Strates BB, Nimni ME. Calcification of implanted vascular tissues associated with elastin in an experimental animal model. J Biomed Mater Res 1992;26:1169—77. [15] Elliot RJ, McGrath LT. Calcification of the human thoracic aorta during aging. Calcif Tissue Int 1994;54:268—73. [16] Yu SY. Cross-linking of elastin in human atherosclerotic aortas: 1. A preliminary study. Lab Invest 1971;25:121—5. [17] Partridge SM. Biosynthesis and nature of elastin structures. Fed Proc 1966;25:1023—9. [18] Urry DW, Long MM, Okamoto K. Cross-linked polypentapeptide of tropoelastin: an insoluble, serum calcifiable matrix. Biochemistry 1976;15:4089—94. [19] Tingfei X, Jiazhen M, Wenhua T, Shuhui L, Baoshu X. Prevention of tissue calcification on bioprosthetic heart valve by using epoxy compounds: a study of calcification test in vitro and in vivo. J Biomed Mater Res 1992;26:1241—51. [20] Tomazic BB, Siew C, Brown WE. A comparative study of bovine pericardium mineralization: a basic and practical approach. Cells Mater 1991;1:231—41. [21] Bernacca GM, Mackay TG, Wheatley DJ. In vitro calcification of bioprosthetic heart valves: report of a novel method and review of the biochemical factors involved. J Heart Valve Dis 1992; 1:115—30.
In vitro and in vivo calcification of vascular bioprostheses Jyotirmay Chanda*, Katsuyuki Kondoh, Keitaro Ijima, Makoto Matsukawa, Ryosei Kuribayashi Department of Cardiovascular Surgery, Akita University School of Medicine, Akita 010, Japan Received 18 July 1997; accepted 25 January 1998
Abstract Efficacy of different chemical treatments on calcification of vascular graft in vitro and in vivo was studied. Culture medium-filled rat aortas were separately treated in 0.2% glutaraldehyde and epoxy compound, and photooxidized in 0.01% methylene blue for a shorter period (group 1). Another group of rat aortas were separately treated in the same chemicals for a longer period (group 2). All fresh and treated aortas of both groups were cultured for 21 days in an organ culture medium and implanted (except for group 1) in weanling rats for five months. Histology and immunohistochemistry revealed that differently treated aortas of group 1 grow and calcify, and the smooth muscle cells between elastin fibers are the primary site of calcium deposition. In contrast, differently treated aortas of group 2 neither grew, nor did calcify in the medium except the epoxy compound cross-linked aorta of group 2 which did not grow but did calcify. Untreated aorta did not calcify. All fresh and differently treated aortic homografts calcified severely in rats. Our whole arterial segment-calcification system would be useful for analyzing the molecular and cellular mechanisms of both bioprosthetic and atherosclerotic calcification of vascular graft. New anticalcification technique is the only hope for better outcome of future vascular bioprostheses. ( 1998 Published by Elsevier Science Ltd. All rights reserved Keywords: In vitro calcification; Homograft; Photooxidation; Epoxy compound; Glutaraldehyde
1. Introduction Conduit failures from early degeneration and calcification in homograft conduits continue to limit the reconstructive congenital heart surgery. Recently, several techniques [1, 2] of cross-linking have been proposed to limit the calcification in bioprosthetic valves. In this study we have tried to find out whether these cross-linking processes are effective in prevention of calcification in homograft conduits. Unlike in leaflets or pericardium, calcification prevention in vascular bioprostheses is difficult to achieve. With in vivo calcification model factors, whether host or graft, generally important in calcification process of vascular bioprostheses cannot be easily distinguished. To exclude the host factor an in vitro model of calcification in a physiologic medium is necessary. To date well-established model of in vitro calcification in a whole segment of vascular bioprosthesis does not exist. Both in vivo and in vitro calcification
*Corresponding author. Present address: Bojpura, Sreedhar Tank Road, Jessore 7400, Bangladesh. Tel.: 0188 341111 10 3148; fax: 0188 36 2625; e-mail: [email protected]
model of vascular bioprostheses would allow more detailed investigation of the process resulting in calcification and development of further strategies for prevention of calcification of the prostheses in future. 2. Materials and methods Thoracic aortas were removed under sterile conditions from adult male Wistar rats (250—300 g) sacrificed with an overdose of ether, collected in tissue culture medium RPMI 1640 with HEPES buffer 25 mmol l~1 and Lglutamine (Gibco/BRL, Life Technologies, Inc., Paisley, UK), containing penicillin : streptomycin, 100 lg ml~1 (Gibco), gentamicin, 2.5 lg ml~1 (Schering-Plough, Osaka, Japan), amphotericin B, 5 lg ml~1 (Fungizone, Gibco) and transferred to the laboratory at room temperature. 2.1. Chemical treatment Upon arrival at laboratory, vascular clip was put at one end of the aorta followed by filling the aorta with culture medium (described above) and closing the other
0142-9612/98/$ — See front matter ( 1998 Published by Elsevier Science Ltd. All rights reserved. PII S 0 1 4 2 - 9 6 1 2 ( 9 8 ) 0 0 0 3 8 - 6
1652
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
end of the aorta with another vascular clip followed by chemical treatment: 2.2. Glutaraldehyde-treated rat aortas Culture medium-filled rat aortas were immersed in 0.2% glutaraldehyde (Glutaraldehyde EM 25%, TAAB Laboratories Equipment Ltd., Reading, UK) in phosphate buffer saline (PBS) (pH 7.4) at room temperature for 2 min followed by washing in a solution containing 1% glycine (Sigma chemical Co., St. Louis, MO) for 5 min (group 1). Another group of rat aortas were freely crosslinked in 0.2% glutaraldehyde for 4 weeks (group 2). 2.3. Epoxy-compound-treated rat aortas Culture-medium-filled rat aortas were kept in a solution of 4% glycerol polyglycidyl ether (Denacol' EX313, Nagase Chemicals Ltd., Osaka, Japan) and 2% ethylene glycol diglycidyl ether (Denacol' EX-810, Nagase) in PBS (pH 7.4) for 5 min at room temperature followed by washing in PBS (group 1). Another group of rat aortas were freely cross-linked in 4% glycerol polyglycidyl ether for 6 days followed by treating in 2% ethylene glycol diglycidyl ether for 7 days (group 2).
2.6. Cell viability assay To assess the effect of different chemicals on viability (live or dead) of cells, live/dead reduced biohazard viability cytotoxicity kit (L-7013) (Molecular Probes Inc., Eugene, OR, USA) was used. Working solution of the dye was prepared by pipetting 5 ll of Component A and 5 ll of Component B into a common 2.5 ml volume of Hank’s balanced salt solution (1 : 500 dilution of each, HBSS, Gibco). Segments of fresh and treated aortas were incubated in the diluted dye in complete darkness for 30 min at room temperature followed by washing in HBSS, fixing in freshly prepared 4% glutaraldehyde in HBSS for more than 15 min at room temperature, and finally washing in HBSS. After that, segments were submerged in OCT compound (Tissue Tek', Miles Inc., Elkhart, IN, USA) and quickly frozen in liquid nitrogen. The frozen samples were cut at 5 lm on a cryostat, mounted on glass slides, and observed under epifluorescent microscope (Olympus BH2, Tokyo, Japan). At excitation wavelength 485$11 nm, live (green fluorescent) and dead (red fluorescent) cells were viewed simultaneously. The fluorescence intensity of live cells was significantly lower than that of dead cells. 2.7. Implantation in rats
2.4. Photooxidized rat aortas Culture-medium-filled rat aortas were placed in a solution of photoactive dye (0.01% methylene blue in PBS, pH 7.4) (Methylene blue, Wako Chemical Co., Osaka, Japan). The solution was exposed to light of a 1000 W slide projector under continuous stirring for 5 min at 2°C followed by washing in PBS (group 1). Another group of rat aortas were photooxidized for 24 h (group 2). All fresh (untreated) and treated aortas of both groups were opened up longitudinally, cut into 5 mm lengths and cultured (n"30 of each group) for 21 days in the medium developed by Soyombo and colleagues [3]. Before culture, the viability of cells of the freshly isolated and treated aortas was checked with immunofluorescent staining on frozen sections described later. 2.5. In vitro culture protocol The culture medium consisted of: tissue culture medium RPMI 1640 with HEPES buffer 25 mmol l~1 and L-glutamine (Gibco), containing 30% fetal bovine serum (Gibco), penicillin:streptomycin, 100 lg ml~1 (Gibco), gentamicin, 2.5 lg ml~1 (Schering-Plough), amphotericin B, 5 lg ml~1 (Gibco), and maintained for 21 days at 37°C in a humidified atmosphere of 5% carbon dioxide in an air incubator. The tissue culture medium was replaced every 3 days. At the end of the culture period, aortic segments were washed in PBS fixed overnight in 10% formalin in PBS followed by processing and paraffin embedding.
Untreated (fresh) and differently treated rat aortas of group 2 (each 2 cm in length) were washed in copious amounts of normal saline and implanted subcutaneously in the back of both sides of 153 week-old male Wistar rats (30—50 g). Wounds were closed with 4-0 polypropylene. Samples were retrieved five months after implantation. Of two specimens from each rat, one was fixed overnight in 10% formalin in PBS followed by processing and paraffin embedding and another was processed for quantitative calcium estimation by atomic absorption spectroscopy [4]. 2.8. Histology and immunohistochemistry Serial sections of 2—3 lm were cut from each paraffin block and mounted on glass slides. To study the morphology of the specimens, hematoxylin and eosin (H & E), Masson’s trichrome, and elastin van Gieson stains were used. von Kossa stain was used to identify calcium phosphate. For immunohistochemistry, deparaffinized and rehydrated tissue sections were blocked with 0.1% hydrogen peroxide (H O ), incubated with 10% normal 2 2 swine serum (Gibco) for 20 min, and washed in PBS. Smooth muscle cells were detected with mouse anti alpha smooth muscle actin monoclonal antibody (clone 1 A4, 1 : 10 dilution, Immunon, Pittsburgh, PA, USA). To identify endothelial cells, anti-human von Willebrand factor serum (1 : 40 dilution, DAKO A/S, Glostrup, Denmark) was used. Bound monoclonal antibodies were
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
detected using horseradish peroxidase (HRP)-conjugated goat anti rabbit antibodies (IgG) (1 : 2500 dilution, MBL, Nagoya, Japan) or anti-mouse antibodies (IgG) (1 : 100 dilution, Tagoimmuno-chemicalsTM, Biosource International, Camarillo, CA, USA). Then, slides were exposed to 0.04% solution of 3,3@-diaminobenzinidine (Sigma) containing 0.03% hydrogen peroxide in 0.05 mol/Tris/ HCL buffer (pH 3.0) for color development, washed in running water for 20 min and counterstained with H & E. For detection of proliferative cell nuclear antigen (PCNA) in cultured aorta, incubation of the sections with mouse anti-PCNA/cyclin monoclonal antibody (clone PC10, 1 : 25 dilution, Novocastra Laboratories, Newcastle upon Tyne, UK) was carried out at 4°C overnight. After washes in PBS, biotinylated anti-mouse IgG (Dako) was applied at a dilution of 1 : 50 and incubated for 30 min at room temperature followed by incubation with strepavidin-peroxidase (1 : 500 dilution, Dako) at room temperature for 30 min. Slides were incubated with peroxidase-conjugated 3,3@-diaminobenzinidine (1 : 500 dilution, Dako) in Tris buffer (pH 7.6) for 5—7 min at room temperature and counterstained with H & E. Sec-
1653
tions were evaluated with side by side comparisons of the different stains on sequential tissue sections to allow correlation of the location of pathological changes. Statistical analyses were carried out with a two-tailed unpaired t-tests. Significant differences were found to exist at the P(0.01 level. 3. Results 3.1. Cell viability assay Live/Dead Reduced Biohazard Viability Cytotoxicity Kit demonstrated that cell viability of rat aorta was totally lost in differently treated aortas of group 2. However, short-term treatment did not affect the viability of endothelial cells of rat aortas of group 1. Cells other than endothelial also remained live in aortas of group 1. 3.2. Culture of rat aorta Untreated (fresh) and differently treated aortas of group 1 grew well in the medium. However, neointimal
Fig. 1. These photomicrographs show serial sections of 2 min glutaraldehyde cross-linked rat aorta (group 1) cultured for 21 days. Formation of neointima (arrow) (a), and migration of smooth muscle cells (arrow) to the neointima (b) can be seen. Deposition of calcium (black) in smooth muscle cells (between elastin fibers) can be noticed (c). Presence of PCNA (proliferating cell nuclear antigen)-positive cells (arrow) in different layers of the cultured aorta can be seen (d). (a) Elastin van Gieson stain, (b) anti-alpha actin antibody stain, (c) von Kossa stain, (d) anti-PCNA antibody stain. a, b, c, d, original magnification ]100.
1654
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
formation was more prominent in fresh aortas than that in treated aortas of group 1 after culture for 21 days (Fig. 1). PCNA-positive cells have increased in number (data not shown here), suggesting that both untreated (fresh) and differently treated aortas of group 1 grow in in vitro. Comparative study of adjacent sections of differently treated aortas of group 1 stained with either von Kossa or anti-alpha actin or elastin van Gieson stains revealed the deposition of calcium inside the smooth muscle cells between elastin fibers (Figs. 1 and 2). Calcified cells also stained PCNA-positive indicates that the cells were live and were capable of proliferation during calcification (Fig. 1). In contrast, differently treated aortas of group 2 neither grew, nor did calcify in the medium (Fig. 3) except the epoxy compound cross-linked aorta of group 2 which did not grow but did show calcific changes after culture. No calcification was observed in untreated (fresh) aorta.
Fig. 2. Histologic appearance of a photooxidized aorta of group 1 cultured in vitro. Notice calcific deposits (asterisk) in smooth muscle cells between elastic laminae. Arrow indicates the neointima. H & E stain. original magnification ]100.
Fig. 3. Morphologic feature of a glutaraldehyde cross-linked aorta of group 2 cultured in vitro. No neointima and no calcific deposit can be noticed. H & E stain. original magnification ]100.
Fig. 4. Microscopic appearance of untreated (fresh) aortic homograft implanted in rat for five months. Severe calcific changes can be noticed. Note that deposition of calcium (arrow) along elastin fiber can clearly be distinguished. von Kossa stain. original magnification ]100.
Fig. 5. This photomicrograph shows the calcific feature of epoxy compound cross-linked homologous aorta of group 2 implanted in rat for five months. Severe calcification of the graft can be seen. von Kossa stain. original magnification ]100.
Fig. 6. Calcium level in untreated (fresh) and differently treated aortic homograft of group 2 implanted in rat for five months. RA/Untr."Untreated (fresh) aorta, RA/GA"Glutaraldehyde-treated aorta of group 2, RA/MB"Methylene blue-mediated photooxidized aorta of group 2, RA/Epoxy"Epoxy compound-treated aorta of group 2. RA/Untr. vs. RA/GA: P"0.0002, RA/Untr. vs. RA/MB: P(0.0001, RA/Untr. vs. RA/Epoxy: P"0.0881.
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
3.3. Implantation in rats Despite the higher calcium level in differently treated than that in untreated (fresh) grafts, histology revealed that all fresh and differently treated aortic homografts calcified severely when implanted subcutaneously in weanling rats for 5 months (Figs. 4—6). On contrary to in vitro-grown grafts, elastin fibers are the primary site of calcification in fresh and differently treated aortic homografts implanted in rats (Fig. 4).
4. Discussion For the use of the in vitro calcification model, not only to simulate the calcification in vascular bioprostheses but also in atherosclerotic vascular organ, we were keen on finding out what would be the maximum period of treatment without affecting the viability of endothelial cells of rat aorta during a particular chemical treatment. The period of exposure of rat aortas of group 1 to different chemicals (see Section 2) was according to the protocol standardized for serving our purpose mentioned above. The chance of immunologic reaction can not be ignored in vivo study. This was the reason why differently treated (inadequate to suppress the antigenicity) aortas of group 1 were not implanted in rats. Carpentier [5] has introduced glutaraldehyde as a cross-linking agent of xenograft heart valve bioprostheses. Calcification is the main problem with glutaraldehyde-treated grafts. In order to inhibit calcification, alternative to glutaraldehyde, epoxy compounds are proposed as cross-linking agents of bioprosthetic heart valves [1]. Photooxidation of amino acids sensitized by methylene blue was described by Weil and associates [6]. To avoid glutaraldehyde-induced calcification, Mechanic [7] has used dye-mediated (methylene green) photooxidation technique for stabilization of heterologous tissue, and this tissue is used for fabrication of PhotofixTMa Pericardial Bioprosthesis (Carbomedics, Austin, Texas, USA) [2]. Methylene blue is occasionally applied to the adventitia of blood vessels during coronary artery bypass and other vascular procedures to assist in the orientation of the vessel. In our experiment, calcification did not occur in dead cells of aorta (group 2) cultured in vitro. Whereas, calcification did occur in chemical agent-induced injured cells of aorta (group 1) in vitro, suggesting that calcium influx into injured/dead cells is not a passive process and some how regulated at cellular level in vitro and at organic level in vivo. These facts would be helpful for understanding the underlying processes contributing to pathogenesis of calcification of a graft in general. For ischemic myocardiocytes [8—10] to calcify, reperfusion is necessary. Unlike our observation, Kim [11] has observed that glutaraldehyde-fixed cells calcify within a week when
1655
cultured in the medium. Glutaraldehyde fixation may be construed as a form of cell injury that results in calcification, however, in view of devitalization of the cells by fixation, increases in calcium and phosphorous as the mechanism of the prosthetic calcinosis are questionable [12]. Elastin-associated calcification was observed in allograft aortic wall [13] and glutaraldehyde-cross-linked heterologous vascular grafts [14] implanted subdermally in rats. The elastin fibers within the middle of the media were identified as the principal site of calcium accumulation, and the calcification of human aorta increases throughout aging [15]. Decrease in degree of cross-linkage in elastin was found in atherosclerotic aortas [16]. An open space of the macromolecular low-cross-linked elastin may provide a nucleation center for calcification of elastin [17]. In the media, the synthetic, cross-linked poly-pentapeptide of tropoelastin has been shown as the seeding matrix for calcium deposition [18]. However, the prevalence of discrepancy in primary sites of calcium deposition in a vascular tissue cultured in vitro or implanted in animal is difficult to explain. Tingfei and coworkers [19] have reported calcification in fresh, glutaraldehyde and epoxy compound cross-linked bovine pericardial tissue in vitro. No histology of in vitro-calcified tissue is available with this study. The amount of calcium (approx. 0.5 lg mg~1) [19] in the cultured tissue is hardly convincing in favor of calcification. However, epoxy compound cross-linked aorta (Fig. 3) of group 2 calcified in vitro in our study and this was quite demonstrateable histologically. Since histology was quite enough to indicate the calcification, the calcium level of cultured tissue was not quantified in our study. Why glutaraldehyde cross-linked or photooxidized aorta of group 2 does not but epoxy compound cross-linked aorta does calcify (despite the lack of growth) in vitro, remains illusive. The in vitro calcification model of bioprosthetic tissue proposed by others [20, 21] does not satisfy our requirements described earlier. Because the cells grow and proliferate, and endothelial layer remains intact (the viable and intact endothelium is not necessary to induce calcification), our whole arterial segment-calcification (not in isolated smooth muscle cells) system would be useful for analyzing the molecular and cellular mechanisms of both bioprosthetic and atherosclerotic calcification of vascular graft. In vivo calcification study shows that neither the epoxy compound cross-linking, nor the methylene-blueinduced photooxidation inhibits the calcification of homologous vascular graft implanted for five months. This fact again proves that calcification problem in vascular graft is difficult to solve. Untreated (fresh) rat aorta does not calcify in vitro, but does calcify in vivo, suggesting that conceivably, due to the immunogenic reactions fresh homograft aortas calcify in vivo. In contrast, the
1656
J. Chanda et al. / Biomaterials 19 (1998) 1651—1656
reason of calcification of treated tissue both in vivo and in vitro, might be due to the cross-linking process.
5. Conclusions Our whole arterial segment-calcification system would be useful for analyzing the molecular and cellular mechanisms of both bioprosthetic and atherosclerotic calcification of vascular graft. Because of calcification, homografts are no longer the reliable tissue to repair the congenital heart defects. Cross-linking in epoxy compound or photooxidation in methylene blue would not change the calcification feature of homologous vascular grafts. New anticalcification technique is the only hope for better outcome of future vascular bioprostheses.
Acknowledgements We gratefully acknowledge Professor Kentaro Yoshimura, Head, Department of Parasitology of Akita University School of Medicine, Akita, Japan for allowing us to use his biochemistry and cell culture laboratories for our ongoing work. We wish to express our sincere gratitude to Dr. Tatsuya Abe of Department of Parasitology of Akita University School of Medicine for his suggestions during the course of this work.
References [1] Sung HW, Shen SH, Tu R et al. Comparison of the cross-linking characteristics of porcine heart valves fixed with glutaraldehyde or epoxy compounds. ASAIO J 1993;39:M532—M536. [2] Moore MA, Bohachevsky IK, Cheung DT, Boyan BD, Chen WM, Bickers RR, McIlroy BK. Stabilization of pericardial tissue by dye-mediated photooxidation. J Biomed Mater Res 1994; 28:611—8. [3] Soyombo AA, Angelini GD, Bryan AJ, Newby AC. Neointima formation is promoted by surgical preparation and inhibited by cyclic nucleotides in human saphenous vein organ cultures. J Thorac Cardiovasc Surg 1995;109:2—12.
[4] Chanda J, Kuribayashi R, Abe T, Sekine S, Shibata S, Yamagishi I. Is dog a useful model for accelerated calcification study of cardiovascular bioprostheses? Artif Organs 1997;21:391—5. [5] Carpentier A, Lemaigre G, Robert L, Carpentier S, Dubost C. Biological factors affecting long-term results of valvular heterografts. J Thorac Cardiovasc Surg 1969;58:467—83. [6] Weil L, Gordon WG, Buchert AR. Photoo¨xidation of amino acids in the presence of methylene blue. Arch Biochem Biophys 1951;33:90—109. [7] Mechanic GL. Process for cross-linking collagenous material and resulting product. US Patent no. 5,147,514. 1992. [8] Burton KP, Hagler HK, Templeton GH, Willerson JT, Buja LM. Lanthanum probe studies of cellular pathophysiology induced by hypoxia in isolated cardiac muscle. J Clin Invest 1977; 60:1289—302. [9] Shen AC, Jennings RB. Kinetics of calcium accumulation in acute myocardial ischemic injury. Am J Pathol 1972;67:441—52. [10] Hagler HK, Lopez LE, Murphy ME, Grecio CA, Buja M. Quantitative X-ray microanalysis of mitochondrial calcification in damaged myocardium. Lab Invest 1981;45:241—7. [11] Kim KM. Cell death and calcification of canine fibroblasts in vitro. Cells Mater 1994;4:247—61. [12] Kim KM. Apoptosis and calcification. Scanning Microsc 1995; 9:1137—78. [13] Webb CL, Nguyen NM, Schoen FJ, Levy RJ. Calcification of allograft aortic wall in a rat subdermal model. Am J Pathol 1992;141:487—96. [14] Paule WJ, Strates BB, Nimni ME. Calcification of implanted vascular tissues associated with elastin in an experimental animal model. J Biomed Mater Res 1992;26:1169—77. [15] Elliot RJ, McGrath LT. Calcification of the human thoracic aorta during aging. Calcif Tissue Int 1994;54:268—73. [16] Yu SY. Cross-linking of elastin in human atherosclerotic aortas: 1. A preliminary study. Lab Invest 1971;25:121—5. [17] Partridge SM. Biosynthesis and nature of elastin structures. Fed Proc 1966;25:1023—9. [18] Urry DW, Long MM, Okamoto K. Cross-linked polypentapeptide of tropoelastin: an insoluble, serum calcifiable matrix. Biochemistry 1976;15:4089—94. [19] Tingfei X, Jiazhen M, Wenhua T, Shuhui L, Baoshu X. Prevention of tissue calcification on bioprosthetic heart valve by using epoxy compounds: a study of calcification test in vitro and in vivo. J Biomed Mater Res 1992;26:1241—51. [20] Tomazic BB, Siew C, Brown WE. A comparative study of bovine pericardium mineralization: a basic and practical approach. Cells Mater 1991;1:231—41. [21] Bernacca GM, Mackay TG, Wheatley DJ. In vitro calcification of bioprosthetic heart valves: report of a novel method and review of the biochemical factors involved. J Heart Valve Dis 1992; 1:115—30.