WO1996003093A1 - Bioprosthetic implants and method of making and using same - Google Patents

Abstract

The invention provides bioprosthetic implants from which the lipids have been removed by treatment with enzymes which degrade lipids, and methods of making and using the same. The enzymes include lipases and phospholipases. The removal of lipids from the bioprosthetic implants according to the present invention decreases the calcification of the bioprosthesis. The bioprosthetic implants preferably may be recellularized with fibroblasts or myofibroblasts before implanting the bioprosthesis in the body of a person.

Description

BIOPROSTHETIC IMPLANTS AMD METHOD OF MAKING

AND ys-tyg SAME

Field of the Invention

The present invention relates to bioprosthetic implants from which the lipids have been removed by treatment with enzymes which degrade lipids, and methods of making and using same. Treatment with the lipid degrading enzymes, lipase and phospholipase, decreases the calcification of the bioprosthesis upon implantion in vivo. The bioprosthetic implant may preferably be recellularized prior to implantation.

Background of the Invention

The replacement of diseased or defective body parts with synthetic and biological prostheses is well known in the art. A bioprosthesis is a prosthesis made of biological as opposed to synthetic material. In some cases, a bioprosthesis may be more effective than a synthetic prosthesis because it both physiologically and mechanically more closely resembles the body tissue which is to be replaced. This may be especially true in the case of a bioprosthesis serving both a structural and functional role within the recipient, such as a bioprosthesis implanted as a substitute blood vessel, heart valve, skin or other tissue. However, certain such bioprostheses may elicit an immune response in recipients due to the presence of antigenic components within the implant. Moreover, such a bioprosthesis may calcify over time and induce thrombogenesis.

Accordingly, techniques have been developed to reduce the antigenicity of bioprostheses, and to reduce their tendency to calcify in vivo. Along these lines, United States Patent Nos. 4,801,299 and 4,776,853 describe bioprostheses and methods of preparing same wherein the tissue excised from a body structure is processed to remove cellular and antigenic components such as cell membranes, nucleic acids, lipids and cytoplasmic components. Tissue processing according to these patents comprises extracting the tissue with some type of detergent, which tissue may require more extensive extractions with detergent, as well as with enzymes capable of degrading nucleic acids, in order to render the tissue suitable for its intended use. Similarly, United States Patent No. 3,318,774 discloses extraction with detergent and with an organic fat solvent. This latter patent is directed in particular to transplantation applications involving bone.

Despite the advances in the transplantation field which the disclosed methodologies represent, the current long term performance of bioprosthesis, and, in particular, bioprosthesis for the replacement of blood vessels and heart valves, clearly needs to be improved. An unacceptably high number of bioprosthesis fail following implantation in vivo (Levy et al., 1986). In fact, the current most widely used bioprosthesis offers recipients only about a 40% chance of survival after about fifteen years postimplantation (Bortolotti et al., 1985) . The failure of the implants such as prosthetic valves results from both calcification and tearing (Hammermeister et al., 1993). Ironically, replacement of an aortic valve with the bioprosthetic valve in the first place is typically necessitated by the calcification of the endogenous valve (Levy et al., 1986). However, the replacement valve tends to calcify at a more rapid pace than the replaced endogenous valve (Levy et al., 1986).

Both tearing and cuspal calcification occur in well defined patterns. Whereas tearing generally occurs at the free edge and at the base of the cusps where they attach to the supporting stent (Ishihara et al., 1981; Pomar et al, 1984; Grabenwoger et al., 1992), calcification usually initiates within the valve cusps and spreads outward through the cusp surface. This intrinsic calcification is believed to be associated with the occurrence of matrix vasicles (tarimura et al, 1983; Kim et al, 1976) , cell debris (Levy et al, 1976; Levy et al, 1977; Schon et al, 1993) and elastin (Paula et al, 1992) . Free edge tearing appears to occur due to highly concentrated flexural stresses (Vesely et al., 1988; Vesely et al., 1986; Krucinsko et al, 1993) and has been linked empirically to calcification (Ishihara et al., 1981; Pomar et al., 1984; Thubrikar et al., 1988). In contrast, the source of tearing at the base of the cusp is not as well defined. It is more probable than not, however, that such tearing is related to calcification, since significant calcification has been found at the base of the cusps (Ishihara et al., 1981; Pomar et al., 1984; Grabenwoger et al., 1992; Stein et al., 1985; Gallo et al. , 1987) . In this respect, it is known that lipids are associated with calcification and abound in atherosclerotic lesions (Knuth, 1984; Pasquinelli et al., 1989) . In humans, the calcium content of the heart valve increases concurrently with the lipid content as part of the aging process (Kim et al., 1976). Calcification studies performed in vitro using liposomes as models confirm that acidic phospholipids in particular mediate the nucleation and spread of calcification (Eanes, 1989; Boyar et al, 1989) . Thus, while it is desirable to ensure that no lipids are present in bioprosthetic implants, as disclosed herein, lipids remain within the bioprosthetic implants prepared by detergent extraction. Recently, European Patent Application 0 564 786 has attempted to address some of the shortcomings of currently available bioprostheses. While the disclosed methodology should result in reduction of some of the potentially damaging events that may occur in the processing of collagen-based tissues prior to transplantation (e.g., mechanical and biochemical events occurring during tissue procurement) and, in this sense, should reduce calcification resulting from structural damage to the bioprosthesis, the disclosed methodology is not specifically directed toward increased removal of lipids from the implant.

Similarly, an alternative approach to detergent extraction of tissue implants has been to directly remove components which elicit the production of host antibodies through the use of enzymes. For instance, U.S. Patent Nos. 4,098,571 and 4,233,360 disclose enzymatic treatment and implantation in a mammal of a collagen-based product, as does U.S.S.R. Patent No. 559,701. Moreover, U.S. Patent No. 2,900,644 and its foreign counterpart Great

Britain Patent 826,577 disclose ficin enzymatic treatment of collagen from mammals, which may then be fixed to aid in tissue handling, and implanted. The enzymes used in these approaches, however, are proteolytic in nature, and thus are not expected to reduce the amount of lipid remaining in the bioprosthetic implant. Accordingly, these enzymatic approaches, like the approach employed in the European Patent Application 0 564 786, do not appear to remedy the ostensible failure of the methodology of United States Patent Nos. 4,801,299 and 4,776,853 to remove substantially all the lipids from the resultant bioprosthetic implant.

Thus, there remains a need for an improved bioprosthesis, particularly a bioprosthesis in which substantially all of the lipid components have been removed from the implant. It is an object of the present invention to provide such a bioprosthetic implant, as well as methods of preparing and using such an implant. These and other objects and advantages of the present invention, as well as additional inventive features, will be apparent from the description of the invention provided herein.

Brief Summary of the Invention

The invention relates to a bioprosthetic implant which includes an extracellular matrix obtained by treatment of a collagen-based body-derived tissue with a 3093 PCΪ7US95/09590

lipid-degrading enzyme. The invention also relates to a method of implantation which comprises introducing the bioprosthetic implant into a recipient. Further, the invention relates to a method of preparing a bioprosthetic implant which comprises treating a collagen-based body-derived tissue with a lipid-degrading enzyme. The invention also relates to a method of recellularizing a bioprosthetic implant which comprises contacting the implant with cells, and maintaining the implant with the cells in tissue culture medium.

Brief Description of the Drawings

FIGURE 1 is a photograph of porcine heart valve cusps stained with Oil-Red-o after the tissue has been extracted with a series of buffered solutions containing salts, a detergent, and enzymes that degrade nucleic acids. These valve cusps have been treated as set forth in Example 3.

FIGURE 2 is a photograph of porcine heart valve cusps that were extracted with detergent and then subsequently treated with enzymes that degrade lipids, and stained with Oil-Red-0 as set forth in Example 3.

Detailed Description of the Invention

Diseased or defective body structures are now routinely replaced in therapeutic procedures. In certain cases, it is preferable to use healthy biological tissue rather than synthetic polymer as replacement tissue since the biological tissue is equivalent in its size, shape, mechanical properties and biochemistry to the tissue that is to be replaced. Once the components which elicit an antigenic reaction are removed, the advantageous properties of the biological tissue can be exploited. Furthermore, the biological material has the potential to be repopulated with host cells, improving its acceptance by the recipient. Accordingly, the present invention relates to bioprosthetic implants from which the lipids have been removed by treatment with enzymes which degrade lipids, and methods of making and using same. The removal of the lipids from the bioprosthetic implants decreases the calcification of the bioprosthesis, resulting in increased durability of the bioprosthesis in vivo. The present invention further provides bioprosthesis derived from a mammalian body-derived tissue from which lipids have been removed through the use of a lipid-degrading enzyme, comprising a matrix of collagen and elastin, and lacking substantially all cellular and extracellular lipids, nucleic acids, cellular membranes and cytoplas ic components. The bioprosthesis of the present invention may be used to treat tissue failure, including heart disease due to the failure of a heart valve. In the context of the present invention, "heart valve" includes the whole or part of a heart valve, heart valve cusp or heart valve leaflet. Accordingly, a bioprosthetic implant according to the present invention comprises extracellular matrix obtained by treatment of a collagen-based body-derived tissue with a lipid-degrading enzyme. In the context of the present invention, extracellular matrix comprises the intricate meshwork of interacting, extracellular acromolecules found in the extracellular space of most tissues. Preferably, the extracellular matrix of the present invention is intact, i.e., the structure of collagen and elastin within the matrix is analogous to that found n vivo. The extracellular matrix may or may not further comprise cells, cell membranes, nucleic acids, lipids and cytoplasmic components.

Body-derived tissue according to the present invention generally comprises any suitable collagen- containing tissue removed from the body and in the form of extracellular matrix, which may or may not further comprise elastin. For instance, the collagen-containing tissue used to derive the bioprostheses of the present invention may be selected from the group consisting of skin, blood vessels, heart valves, ligaments, tendons, bone, trachea, cartilage, dura mater, nerves and other such tissues. The collagen-containing tissue may be obtained from any source, and preferably, will be obtained from a mammalian host. Even more preferably, the tissue will be obtained from a human, an ungulate (e.g., a caprine, bovine or porcine species), a canine or a feline. Human tissue including cadaver tissue is available through tissue banks and hospitals. Other mammal tissue can be obtained through suppliers of laboratory mammals, or through the meat processing industries.

A lipid-degrading enzyme according to the present invention is any fat-splitting or lipolytic enzyme which cleaves a fatty acid residue from the glycerol residue in a neutral fat or a phospholipid. A preferred lipid- degrading enzyme is selected from the group consisting of Upases and phospholipases, and mixtures thereof. Even more preferred are the lipases triacylglycerol and diacylglycerol lipase, and the phospholipases phospholipase A,, A₂, B, C and D. Preferred triacyglycerol lipases according to the present invention are triacylglycerol lipase from porcine pancreas (Boehringer Mannheim, Laval, Quebec) and triacylglycerol lipase Type XIII from Pseudomonas species (Sigma Chemical Co. , St. Louis, MO) . The lipases and/or phospholipases may be employed alone, or in appropriate combination to comprise the lipid degrading enzymes. In a further preferred embodiment of the present invention, triacylglycerol lipase from porcine pancrea and triacylglycerol lipase Type XIII from Pseudomonas species will be used in combination. These preferred triacylglycerol lipases may also be used in conjunction with other lipases and/or phospholipases.

It will be recognized by the ordinary skilled artisan that various other components present within a particular solution may impact upon the activity of the lipid-degrading enzyme. Accordingly, a broad range of enzyme concentrations is contemplated for use in the context of the present invention, and the Ordinary skilled artisan is capable of adjusting the concentration in a particular application to achieve the desired effect (i.e., release of a free fatty acid). Each lipid- degrading enzyme may be employed at a concentration ranging from about 0.25 units/milliliter (ml) to about 50 units/ml. It is expected that the concentration of a particular lipid-degrading enzyme to be employed will differ according to its efficacy. For instance, whereas Pseudomonas triacylglycerol lipase Type XIII may preferably be employed at a concentration ranging from about 0.2 units/ml to about 10 units/ml, more preferably about 0.25 units/ml to about 5.0 units/ml, porcine pancreatic triacylglycerol lipase may preferably be employed at a concentration ranging from about 0.5 units/ml to about 70 units/ml, more preferably, 1.0 units/ml to about 50 units/ml.

In the context of the present invention, the lipid- degrading enzyme will preferably be employed in a buffered solution. Even more preferably, the solution will be buffered at a pH of about 7.4. Suitable buffers to employ in the lipase solution, as well as the other tissue processing solutions of the present invention, include Tris-HCl, and other appropriate buffers such as are known in the art. preferably, however, the buffer will not result in the inactivation of the lipid- degrading enzyme.

Processing of the collagen-based body-derived tissue in the solution comprised of the lipid-degrading enzyme can be carried out for any suitable length of time such that the amount of lipids remaining in the tissue is less than the amount of lipids which would be present were the tissue subjected to some other means of effecting lipid removal in the absence of processing using a lipid- degrading enzyme, such as, for example, detergent processing. Preferably such processing with use of the lipid-degrading enzyme will be carried out for up to about two hours, more preferably for up to about eight (8) hours, and even more preferably for up to about 24 hours.

The ordinary skilled artisan will recognize that there is a relationship between the amount of lipid- degrading enzyme to be employed, and the length of time allowed for tissue processing using the lipid-degrading enzyme, inasmuch as the amount of enzyme utilized can be decreased if the processing time is increased, and the processing time can likely be decreased if the amount of enzyme utilized is increased. This relationship is one which need be determined empirically, however, and will vary with different factors such as the particular tissue type being processed, etc. Moreover, at some point, a further increase in the processing time or amount of lipid-reducing enzyme employed will not result in a further decrease in the amount of lipid-reducing enzyme employed or processing time, respectively, based on such factors as the half-life of the lipid-degrading enzyme, the intrinsic ability of the lipid-degrading enzyme to hydrolyze lipid esters, etc. Accordingly, it is well within the means of the ordinary skilled artisan to optimize the precise manner in which tissue processing is to be carried out.

According to the present invention, processing with the lipid-degrading enzyme can be carried out at any suitable temperature. In some applications (e.g., wherein the tissue is not extracted with detergent or by any other means prior to processing with the lipid- degrading enzyme) it may be desirable to reduce possible proteolytic degradation of the tissue, and thus a low temperature, such as a temperature less than about 20°C may be preferred. Alternately, certain lipases may be more effective at an increased temperature, and in these cases, it may be desirable to carry out processing with the lipid-degrading enzyme at a temperature greater than about 20°C. In the context of the present invention, preferably processing with the lipid-degrading enzyme can be carried out at a temperature of between about 20°C and about 45°C, and even more preferably, at a temperature of about 37°C.

As prepared by the present invention, a bioprosthetic implant may be suitable for implantation without further processing to reduce the immune potential of the implant. However, it may be desirable to stabilize the implant to render the implant material less immune, which may provide for longer durability in applications where host interaction is not necessary or desired. Glutaraldehyde has been used to stabilize collagen-based bioprosthetic implants and has demonstrated clinical utility for at least twenty (20) years for tissue heart valve implants and pericardial augmentation materials. Glutaraldehyde is a bifunctional agent and stabilizes collagen-based materials by covalently binding to free amino groups (i.e., crosslinking) , preventing recognition by the recipient of the implant of the collagen protein in the implant as a foreign material. Any version of the commercial processes using glutaraldehyde or other acceptable crosslinking agents are suitable to prepare the implant of the present invention for implantation in cases where such further optional crosslinking is desired. Crosslinking agents and methods of using same are well known to those skilled in the art. A preferred process would mildly crosslink the material while maintaining native biomechanical properties Such a preferred process would still reduce the potential for the material to elicit an immune response. In this sense, the bioprosthesis either may not require fixation at a conventional concentration of glutaraldehyde or may require fixation at a weaker concentration of glutaraldehyde than conventional concentrations, preferably in a solution of glutaraldehyde that is substantially less than 0.5% glutaraldehyde.

The ideal bioprosthesis according to the present invention is biologically active tissue. The tissue prepared according to this invention preferably consists of a structurally sound collagen and elastin matrix which further may preferably be seeded with the recipient's fibroblasts or myofibroblasts and endothelium obtained from biopsy before the scheduled implantation surgery. The implant may be cultured in vitro before implantation as a viable graft. The implant preferably lacks substantially all lipids. Further, the implant preferably lacks other solubilized cellular and extracellular components, and may preferably lack fixatives (such as cross-linking agents),and therefore has minimal antigenicity. Furthermore, the implant preferably is not chemically modified or crosslinked. Therefore, its mechanical behavior should substantially be the same as the natural biological tissue. As disclosed herein, calcification of such an implant occurs at a rate which more closely approximates the natural physiological rate, giving the implant a lifespan which is much longer than conventional bioprostheses. Accordingly, the present invention provides an implant according to the present invention wherein all lipids are substantially removed. The invention also provides a method of preparing a bioprosthetic implant which comprises treating a collagen-based body-derived tissue with a lipid-degrading enzyme such that Figure 1 illustrates porcine heart valve cusps stained with Oil- Red-0 after the tissue has been extracted with a series of buffered solutions containing salts, a detergent, and enzymes that degrade nucleic acids, as in Example 3. These valve cusps have been treated in particular by the methods described in Kle ent et al. Red stained lipid clusters are clearly visible. In contrast, Figure 2 is a photograph of porcine heart valve cusps that were extracted in accordance with the methods described in Klement et al . , then subsequently were treated with enzymes that degrade lipids in accordance with this invention, and stained with Oil-Red-O. As can be seen from this figure, stainable lipid clusters are dramatically reduced following lipase treatment, indicating that significant amounts of the lipid clusters have been removed. Table 1 presents quantitative calcification levels, measured by atomic absorption analysis, of various bioprosthetic implants prepared as in Example 5 implanted in young growing rats with an in vivo time of three weeks. The glutaraldehyde fixed specimens were prepared using a typical commercial process with a glutaraldehyde concentration of 0.5M. The "Cell Extracted" implants were prepared using a modified process as identified in the prior art by Klement, et al . The various Lipase treated specimens (i.e., "Lipase A", "Lipase B", "Lipase C", and "Lipase D") were prepared using extracted specimens that were further treated with varying concentrations of lipid-degrading enzymes, as set forth in Example 5. The Fresh specimen was prepared by washing tissue samples received directly from the abattoir. These samples were not further processed. The data in Table 1 confirms that there is an additional reduction in calcification in specimens treated with lipid-degrading enzymes as compared with those specimens prepared using only the extraction processes identified in the prior art. This data suggests that with the addition of a lipid-degrading enzyme treatment step following detergent extraction, reduced calcification levels, and subsequently longer clinical useful life of bioprosthetic implants may be achieved. The invention also provides a method of preparing a bioprosthetic implant which comprises treating a collagen-based body-derived tissue with a lipid-degrading enzyme. Accordingly, the present invention provides a method of treating a collagen based tissue sample of an animal, preferably a mammal, to remove cellular and extracellular lipids before implanting the material in a body of a person. The method includes extracting the tissue with detergents and lipid-degrading enzymes, including lipases or phospholipases, and mixtures thereof. The collagen-containing tissue may include skin, blood vessels, heart valves, ligaments, tendons, bone, cartilage, dura mater, nerves and other such tissues.

The method of preparing a bioprosthetic implant which comprises a collagen-based body-derived tissue with a lipid-degrading enzyme may further preferably comprise extracting the implant with a solution comprising salt.

Even more preferably, this method may comprise extracting the implant with a solution comprising detergent. Moreover, preferably this method may further comprise extracting the implant with a solution comprising nuclease. These steps may be performed in any order, and optionally, any combination of steps may be combined with extraction with the lipid-degrading enzyme to obtain the bioprsothesis.

Accordingly, the method of the present invention may comprise one or more steps, including extracting the implant tissue sample with one or more buffered salt solutions, extracting the tissue sample with one or more detergents, treating the tissue sample with one or more enzymes which degrade lipids, and storing the tissue sample in a physiologically buffered solution. The method may additionally include the steps of isolating a tissue sample of biological material from a suitable donor, extracting the tissue sample with one or more buffered salt solutions to rupture the cells of the tissue sample, extracting the tissue sample with buffer solutions containing one or more detergents, treating the tissue sample with buffer solutions containing one or more enzymes, such as nucleases, which degrade nucleic acids, treating the tissue sample with buffer solutions containing one or more enzymes which degrade lipids, i.e., a lipase, phospholipase, or mixture thereof, reextracting the tissue sample with one or more buffered salt solutions, re-extracting the tissue sample with one or more detergents, and storing the tissue sample in a physiologically buffered solution. These steps may be performed in any order, and any one or more steps may be omitted except for use of the lipid-degrading enzyme. In a preferred method according to the present invention, tissue may be decellularized by extraction in a first solution comprised of a hypotonic buffer, protease inhibitors and antibiotics, which results in cell lysis. Subsequently, the tissue may be extracted with a solution comprised of a high concentration of salt, a non-ionic detergent such as Tris-HCl, protease inhibitors and antibiotics. In this step, soluble components of the extracellular matrix, as well as cytoplasmic components, are extracted. The tissue may then be extracted with urea, which swells the tissue, making it more amenable to extraction. Following rinsing of the tissue in an hypotonic salt solution, the tissue may be extracted with a suitable detergent, such as Chapso detergent. Antibiotics, antifungal agents and protease inhibitors may be included in the extraction mixture as necessary or desired. Nucleases such as purified and protease-free ribonuclease and deoxyribonuclease may optionally be used to remove nuclear material (i.e., DNA and RNA) from the tissue, followed by detergent extraction and rinsing of the tissue. The tissue may be extracted with lipid-degrading enzymes to remove cellular and extracellular lipids and phospholipids, and washed in a solution comprised of heparin. If desired, the tissue may be further extracted using a detergent such as Chapso, and rinsed in an hypotonic solution. Antibiotics and antifungal agents may be employed in processing solutions and tissue culture media to maintain sterility of the bioprosthetic implant at various stages of processing. Antibiotics may be selected from the group consisting of gentamicin, kanamycin, penicillin, streptomycin, neomycin. vancomycin and mixtures thereof. A preferred antibiotic according to the present invention is gentamicin. Antifungal agents may be selected from the group consisting of amphotericin B, polymyxin B, fungizyme and nystatin. Other suitable antibiotics and antifungal agents such as are known in the art may also be employed in the present invention.

Protease inhibitors may also be employed in the present invention to inhibit proteolytic enzymes embedded in the collagen-based matrix which can cause degradation of various components of the matrix. Preferred protease inhibitors according to the present invention include phenylmethylsulfonyl fluoride (PMSF) , N-ethylmaleimide (NEM) , ethylene glycol-bis (2-aminoethyl ether)- N,N,N' ,N^,-tetraacetic acid (EGTA) , leupeptin, aprotinin, pepstatin, and ethylenediaminetetraacetic acid (EDTA) . Also, according to the present invention, detergents may be used to remove cellular and antigenic components. It is further contemplated that such removal can be accomplished using other chemical treatments (e.g. , using proteolytic enzymes such as chymotrypsin) , and that various approaches may be employed in combination (e.g., as in a preferred method set forth herein, wherein tissue is processed by incubation in solutions comprised of detergent, salts and enzymes) .

Acceptable detergents for use in the present invention include those set forth in U.S. Patent No. 4,776,853, as well as polyoxyethylene (80) sorbitan nono- oleate (i.e., Tween 80), sodium deoxycholate, and 3-[3- cholamidopropyl)-dimethyla monio]-2-hydroxy-l- propansulfonate (Chapso; Boehringer Mannheim, Laval, Quebec) , DeoxyBigChap detergent, and Little Chap detergent. A preferred detergent according to this invention is Chapso detergent. Moreover, other detergents such as are known and routinely used by those skilled in the art may also be employed. In the context of the present invention, preferably all solutions are processed, such as by filtering the solutions through a 0.22 uM filter prior to use, to ensure sterility of the resultant bioprosthetic implant. The bioprosthetic implant may also be sterilized by other means known to those skilled in the art (e.g., by exposure to ionizing radiation) . Hypotonic solutions which may be employed for rinsing and the high salt buffer solution which may be used to extract cytoplasmic and extracellular components may preferably be buffered, even more preferably at a pH of about 7.4.

The present invention also provides a method of recellularizing a bioprosthetic implant obtained according to the present invention. This method comprises contacting an implant according to the present invention with cells, and maintaining the implant with the cells in suitable tissue medium.

In a preferred method of the present invention, appropriate cells,preferably of a non-m=neoplastic source of connective tissue, and particularly fibroblasts or myofibroblasts, may be used to recellularize the bioprosthetic implant prior to implantation in vivo. In other applications, for example with bioprostheses comprising heart valves or vascular conduits, the endothelial cell (which typically lines the inner surface of this type of tissue) may be employed for recellularization. Preferably the fibroblasts or myofibroblasts are isolated from the same individual that will be the recipient for the bioprosthetic implant. Optionally, the cells may be isolated from a mammal. The fibroblasts or myofibroblasts may be derived from organs or skin, as appropriate, which can be obtained by biopsy or other appropriate means such as are known to those skilled in the art. Alternately, the fibroblasts or myofibroblasts may be obtained from cadavers or fetal tissue. A preferred source of fibroblasts according to the present invention is gingiva or foreskin, and a preferred source of myofibroblasts is heart. Also preferred are cells taken from skin, gingiva or granulation tissue.

Fibroblasts or myofibroblasts can be isolated by desegregating an organ or tissue, which can be done mechanically (e.g. , through use of blenders, grinders, homogenizers, pressure cells, etc.), through use of enzymes or chelating agents (e.g., trypsin, chymotrypsin, dispase, etc.) which allow single cell dispersions to be obtained, or by any means or combination of means which are known to those skilled in the art. Upon reduction of the tissue to a single cell suspension, the fibroblasts may be isolated using techniques which are standard such as cell separation, selective destruction of unwanted cells, cloning of specific cell types, filtration, fluorescence-activated cell sorting, etc. Methods of disaggregation and cell sorting are the subject to numerous reviews (see, for example, Freshney, Culture of Animal Cells, A Manual of Basic Technique, 2nd Ed., (NY: A.R. Liss, Inc., 1987) 107-126) . In the context of the present invention, any commercially available medium can be used for the growth of cells on the bioprosthetic implant, such as RPMI 1640 and Dulbecco's Modified Minimal Medium. The medium may be supplemented as appropriate,for instance, with fetal calf serum, and may be further supplemented with cytokines, growth factors, and various interleukins as well as other growth-promoting agents, to maximize growth of cells in culture. Standard sterile tissue culture techniques should be employed, and the medium should be changed as appropriate.

The invention also discloses a method of implantion which comprises introducing an implant according to the present invention into a recipient. Preferably, the bioprosthetic implant has been recellularized prior to implantation. The invention also discloses a method of implanting in a body of a person a bioprosthesis in which substantially all of the cellular and extracellular lipids have been removed.

The present invention further provides a kit for treating a tissue sample of a mammal to remove substantially all cellular and extracellular lipids before implanting the sample in a body of a person. The kit includes one or more receptacles containing, buffer solutions, and enzymes which degrade lipids, including lipases or phospholipases. The kit may include a receptacle containing enzymes which degrade nucleic acids and a receptacle containing detergents.

The following examples further illustrate the present invention but, of course, should not be construed as in any way limiting its scope.

Example 1 - Materials and Methods Employed in Experiments

1. General Reagents. The following reagents were obtained from Boehringe Mannheim (Quebec, Canada) : Aprotinin, Leupeptin, Pepstatin, Ethylene dinitrilo tetra- acetic acid (EDTA), DNase I, RNase A, 3 - [ (3- cholamidopropyl) dimethyl ammonia] - 2-hydroxy - 1- propansulfonate (CHAPSO) , Tris Base, porcine pancreatic triacylglycerol Lipase; Phospholipase A₂ from porcine pancreas. The following reagents were obtained from Sigma Chemical Co. (St. Louis, MO) : Lipase: Type XIII from Pseudomonas species. The following reagents were obtained from Gibco (Ontario, Canada): BRI, Heparin, and Phenylmethyl Sulfonylfluoride (PMSF) . Urea was obtained from Fisher Scientific (Ontario, Canada) . Other reagents used are commercially available from commercial suppliers. (2) Morphometry to detect and quantitate lipids

Whole hearts were obtained from freshly slaughtered pigs at the abattoir. These pigs ranged in age from 3 to 6 months and weighed an average of 220 pounds. The hearts were transported to the lab, the aortic valve cusps cut out and kept in chilled saline solution at normal Ph during all handling.

The cusps were placed on the stage of a dissecting stereo microscope with the ventricularis side up. By gently grasping and carefully pulling on the ventricularis with forceps, the two layers could be separated along their natural cleavage plane, which appeared to be the midline of the spongiosa. Dissection progressed from the base, where the lipids were most abundant, through the body of the cusp towards the coaptation region. Since the spongiosa did not continue into the Nodulus Arantius and into the coaptation region, it was not possible to separate the layers and examine these areas for lipids. Saline solution was applied repeatedly to keep the tissue moist.

Once the ventricularis was peeled back, the lipid clusters could be easily visualized and their distribution could be quantified using a point counting technique. The dissected valve cusp was first placed on a piece of celluloid divided into 144 quadrants (12 x 12) and each quadrant was examined at high magnification for the presence of lipid clusters. The presence of lipids was quantified using a conventional stereological point counting method. An eyepiece with a 10 x 10 grid, sized to fit exactly into one of the 144 quadrants, was used to view the valve cusps. An estimate of the spatial density of lipids was obtained by counting the number of grid points in the eyepiece that overlay lipid clusters in each quadrant. The distribution of lipids in the cusp was visualized by expressing the lipid count in each quadrant as a "density". Since each quadrant had a total of 100 grid points, the values ranged from 0 to a maximum of 100. This density or darkness of each quadrant was therefore proportional to the number of occurrences of lipid clusters in that particular quadrant. A density of "zero" or white, meant that no lipids were found in a given quadrant. To determine an average distribution of lipids in a typical cusp, the density mapimages of all cusps were spatially averaged together. Spatial averaging was done by summing up the pixel values of all the images at a particular x,y pixel location and dividing by the number of images. This type of averaging, however, could not be done on the original density maps because the valve cusps had different sizes and shapes. Before averaging, all the density map images had to be warped to a standard cusp shape. This was done using commercially available software (Warplt, MIDIapolis Systems, Minneapolis, MN.) on a NeXT computer (Redwood City, California) . First, the cusp boundary, as viewed through the microscope, was drawn over the density map images. Then using the warping software, a mesh or grid was superimposed over the density map image and the nodes of the grid were aligned with recognizable features on the cusp boundary. This was defined as the "source image" with a "source mesh" overlaid. The target image was a simplified, idealized cusp shape, to which all the real cusp density plot images were warped. The same grid used for the source image was applied to the ideal cusp shape, the "target image", and the nodes of the "target mesh" were positioned to correspond to lie over the same landmarks as they did on the source image. Once the source and target meshes were defined, the source mesh was warped to align the source mesh nodes with the target mesh nodes. The underlying source image was therefore warped along with the mesh, so that its outline corresponded to the ideal cusp shape. Once all the images were warped to this standard shape, they could be averaged pixel by pixel, producing a mean density map of lipid distribution in each of the left, right and non coronary cusps. (3) Lipid analysis

The lipid clusters were excised from the aortic valve cusps, and the lipids were extracted and identified by thin layer chromatography (TLC), and gas/liquid chromatography (GLC) , as per Huff et al., 1993. In brief, 2 cusps from each extraction were placed in a 3:2 mixture of hexane/isopropanol, were dried under nitrogen, resuspended in 2:1 chloroform/methanol, and were separated by TLC with petroleum ether/diethyl ether/acetic acid (84:15:1). The free cholesterol, esterified cholesterol, and triglyceride spots were identified by exposure to iodine. The spots were then scraped from the TLC plate and extracted, and the mass for each constituent was determined by GLC. (4) Light Microscopy

Since lipids are soluble in standard histological processing solvents, all histology was done on fresh or frozen sections. These specimens were stained with Oil- Red-O, a known marker for the presence of lipid droplets (Kruth et al . , 1984; Mitchison et al . , 1985; Guyton et al . , 1989) . FIGURES 1 and 2 were taken through the dissecting microscope, showing the lipid clusters under white light illumination and following staining with Oil-Red-O.

Example 2 - Presence of Lipids in Untreated Tissue

Sample

We dissected the fibrosa and the ventricularies of porcine aortic heart valve cusps using a microdissection technique (Vesely et al . , 1992). Following such dissections, we discovered clusters of spherically shaped objects or droplets, roughly 50 micrometers in size. Staining with Oil-Red-0 indicated that these droplets were lipids (Knuth, 1984; Boyan et al . , 1989).

Since lipids are associated with calcified atherosclerotic plaques (Kruth et al . , 1984), and phospholipids are known to calcify readily (Boyan et al . , 1989) , we determined that the presence of lipids in aortic valves was significant in the development of prosthetic valve cusp pathology. While the presence of lipid droplets in aortic valve cusps had been reported previously (Grabenwoger et al . , 1992; Skold et al . , 1966; Ferrans et al . , 1978), a thorough characterization of their distribution and morphology had not been done. We therefore set out to investigate the distribution and the nature of aortic valve cusp lipids. Oil-Red-0 staining confirmed the presence of lipids within the spongiosa. Thin layer chromatography showed that the largest component of the lipid clusters was triglycerides and free cholesterol, while cholesterol ester existed only in small amounts. The lipid droplets were distributed within the porcine aortic valve cusps in a definite pattern. Lipid clusters occurred with high frequency at the base of the cusps, where the cusps attach to the aorta, and occurred less often along the attachment line leading up towards the commissures. Lipids generally did not extend into the belly of the valve cusp, and no lipids were observed in the coaptation region or near the free edge. The amount of lipid found in a given cusp was highly variable, ranging from zero occurrences up to 3% coverage of cusp area. The distribution of lipids in each of the three cusps was similar, with a greater occurrence in the left coronary cusp.

Lipids were found in 19 of 20 valves in the left coronary cusp, 15 of 20 in the non coronary cusp, and 11 of 20 in the right coronary cusp. More importantly, 20 out of 20 valves had lipids in at least two of the three cusps.

This demonstrated that lipid clusters occur with surprising abundance in the aortic valve cusps of juvenile pigs. A sampling of valves from three abattoirs in the Ontario region was positive in all cases, suggesting that the presence of lipids may be universal and not dependent upon a particular strain of pig or method of farming. While the reason for the presence of lipid droplets in aortic valve cusps is presently unclear, it may be related to the high protein, high fat diet that farm pigs are fed to encourage rapid growth.

Thubrikar has shown that calcification occurs at the base of the valve cusps (Thubrikar et al. , 1983), in the same areas that we found large amounts of lipids. Less calcification has generally been found in the belly of the cusps and towards the Nodulus Arantius, areas where we found no lipids. The correlation between the distribution of lipids in porcine aortic valve cusps, and clinically significant calcification of the prosthetic valves is particularly evident from images of x-rayed valves, (Cipriano et al . , 1984) . Thus, there are two categories of lipid deposits in implanted porcine aortic valves; endogenous and accumulated. Large amounts of accumulated lipids have been reported previously in heavily degenerated, explanted bioprostheses (Ferrans et al . , 1983). Routine findings of lipids in explanted valves, however, do not necessarily result from progressive lipid accumulation. Our work suggests that some of these lipids could be of porcine origin, implanted with the graft. If endogenous lipids in porcine aortic valves are related to calcification, then extraction of lipids prior to implantation, or simple screening and discarding would be useful. These results suggest that lipid extraction prior to implantation of xenograft valves will be useful in reducing the calcification of xenograft valves.

Example 3 - Presence of Lipids in Extracted Tissue

Sample

The tissue of porcine aortic valve cusps was extracted with a series of buffered solutions containing salts, a detergent, and enzymes that degrade nucleic acids. These valve cusps were treated using methods described in the prior art (ie, particularly as set forth in US Patent No. 4,776,853). Lipids present in valve cusps prepared according to the prior art were compared to the present example, and as set forth in Example 4. We dissected the fibrosa and the ventricularis of such extracted valve cusps using a icrodissection technique (Vesely et al . , 1992). Following such dissections, we discovered clusters of spherically shaped Objects or droplets, roughly 50 micrometers in size. Staining with Oil-Red-0 indicated that these droplets were lipids (Knuth, 1984; Boyan et al . , 1989).

We determined that there was little difference in the amount of lipids present in the valves dissected and stained with Oil-Red-0 in accordance with Example 2 and in the valves extracted, dissected and stained with Oil-Red-0 in accordance with Example 3. Thus, as described in Example 2, it would be helpful if endogenous lipids could be more effectively removed from tissue samples before implantation. Such a novel methodology is set out in Example 4.

Example 4 - Lipase Treated Tissue Sample

In the extraction procedure of this invention, biological material is removed from a suitable donor. In one example, heart valves were removed from slaughtered pigs and subjected to a multistep process designed to lyse the cells and extract the cellular constituents. Namely, the valves were placed in large beakers and were subjected to a series of processing steps, constantly agitating the solutions with mechanized stirrers.

The valves were first soaked in buffers (0.05M Tris- HC1, pH 7.4) containing 0.05M NaCl and enzyme inhibitors (lmM PMSF, lmM EDTA, 0. luM Aprotinin, luM Leupeptin, luM Pepstatin) for 24 hours at 4°C. The valves were then soaked in buffers (0.05M Tris-HCl, pH 7.4) containing 1.5M NaCl and enzyme inhibitors (lmM PMSF, lmM EDTA, O.luM Aprotinin, luM Leupeptin, and luM Pepstatin) for 24 hours at 4°C. The valves were rinsed in 0.05 M Tris buffer (pH 7.4) with enzyme inhibitors (lmM PMSF, lmM EDTA, O.luM Aprotinin, luM Leupeptin, luM Pepstatin) at 4°C numerous times. The valves were then soaked in buffer (0.05M Tris-HCl, pH 7.4) containing 8M Urea for 1 hour at 4°C. Then the valves were rinsed six (6) times in Tris buffer pH 7.4 with enzyme inhibitor (lmM PMSF, lmM EDTA, O.luM Aprotinin, luM Leupeptin, luM Pepstatin) for one hour at 4°C. The valves were soaked in buffer (0.05M Tris-Hcl, pH 7.4) containing 8 mM Chapso detergent and enzyme inhibitors (lmM PMSF, lmM EDTA, O.luM Aprotinin, luM Leupeptin, luM Pepstatin) for 24 hours at 4°C. Next, the valves were rinsed six times in Tris buffer pH 7.4 with enzyme inhibitors (lmM PMSF, lmM EDTA, O.luM Aprotinin, luM Leupeptin, luM Pepstatin) for one hour each rinse at 4°C. Gentamicin (lmL) was added as necessary. The valves were then soaked in 200 mL buffer (0.05 M Tris-HCl, pH 7.4) containing 4 mg DNase and 20 mg RNase for 24 hours at 37°C. The valves were exhaustively rinsed six times in 0.05M Tris buffer pH 7.4 with enzyme inhibitors (lmM PMSF, lmM EDTA, O.luM Aprotinin, luM Leupeptin, luM Pepstatin) for one hour with each rinse at 4°C. One mL of Gentamicin was added as necessary. The valves were soaked again in buffer (0.05M Tris-HCl, pH 7.4) containing 8mM Chapso detergent for 24 hours at 4°C. The valves were subjected to six washes in 0.05M Tris-HCl pH 7.4 for one hour each rinse at 4°C.

To remove substantially all cellular and extracellular lipids from the biological material, the valves were soaked in suitable buffers (0.05M Tris-HCl pH 7.4) containing varying concentrations of lipases and phospholipases for 24 hours at 37°C. Then the valves are soaked in buffer (0.05M Tris-Hcl) pH 7.4) containing 10 mg/ml heparin at 4°C. The valves are soaked in buffer ).05M Tris-HCl pH 7.4) containing 8mM Chapso detergent for 24 hours at 4°C. Then the valves were rinsed six times in 0.05M Tris buffer (pH 7.4) at 4°C.

The valves were then soaked in sterile culture medium (DMEM) with antibiotic (Gentamicin) with six (6) changes of antibiotic-containing medium over 1 hour and then stored in the same medium. The valves may then be implanted or cultured as described below. This process was found to remove substantially all of the cellular and extracellular lipids. In the method disclosed, the urea, DNase and RNase break up the cellular components. The detergent solubilizes and removes these components from the biological tissues through multiple extractions and rinses." The urea also solubilizes and removes various connective tissue components.

Variations may be made to the above process of extraction including using different detergents, different concentrations of lipases or phospholipases to extract lipids or phospholipids and performing the extraction steps at different pHs or at different temperatures. The process of extraction could be used on any biological material and the steps of extraction could be repeated or lengthened or several steps could be combined or one or more steps could be eliminated in an effort to remove substantially all of the lipids and phospholipids.

Example 5 - Comparative Analysis of Valve Cusps Obtained by Different Means

Fresh valve cusps were compared against those obtained according to the method of U.S. Patent No. 4,776,853, and those obtained according to the method set forth in Example 4.

For these experiments, four different conditions for use of the lipid-degrading enzymes were employed. Namely, porcine pancreatic triacylglycerol lipase and Pseudomonas triacylglycerol lipase Type XIII were used concurrently in the same lipid-degrading solution as follows:

Triacylglycerol Lipase (units/ml) Pseudomonas Type XIII Porcine Pancreatic

Condition

Lipase A 0.25 1.0

Lipase B 1.0 5.0

Lipase C 2.5 10.0

Lipase D 5.0 50.0

The amount of lipids present in the various valve tissues was determined by a combination of TLC and GLC as set forth in Example l. Whereas the amounts of free cholesterol, esterified cholesterol and triglyceride were greatest in the fresh tissue which was not subjected to any extraction protocol, these components were reduced in the tissue valves prepared according to U.S. Patent No. 4,776,853, and were further reduced in the tissue valves prepared according to the present invention. In the Lipase B and Lipase C condition in particular, levels of triglycerides were about two to three times less than levels observed for tissue valves prepared according to U.S. Patent No. 4,776,853. Similarly, whereas extraction according to U.S. Patent No. 4,776,853 resulted in levels of cholesterol ester which were only about seven times less than levels of cholesterol ester observed for fresh tissue, cholesterol ester was not detectable in tissue valves prepared using the Lipase B and Lipase C conditions.

These results confirm that whereas standard detergent extraction protocols result in lipid remaining in the resultant bioprosthesis, a bioprosthetic implant prepared according to the method of the present invention is substantially free of lipids. In particular, use of a lipid-degrading enzyme according to this invention is effective at reducing lipid levels below those remaining after conventional detergent extraction, e.g., below about 50 ug triglycerides per gram of wet tissue, and/or about 8 ug cholesterol ester per gram of wet tissue, especially with respect to pig heart valves.

Example 6 - Calcification of Tissue Valves Prepared by

Different Means and Implanted in vivo

The heart tissue valves prepared according to Example

5 were implanted into rats. Also, a commercially available glutaraldehyde fixed tissue valve was further employed for comparative purposes. After three weeks, the rats were sacrificed, and the tissue valves were recovered and examined for any presence of calcification. For these experiments, the amount of calcium in each tissue valve was determined by atomic absorption. Results of these experiments are set forth in Table 1.

Table 1

Glutaraldehyde Cell Lipase Lipase Lipase Lipase Fixed Fresh Extracted A B C D mg Ca/g mg Ca/g mg Ca/g mg Ca/g mg Ca/g mg Ca/g mg Ca/g tissue tissue tissue tissue tissue tissue tissue

Mean 79.6275 10.7498 4.1211 1.2075 - 0.9548 0.1759 0.2391

SD 51.7255 11.0368' 2.8818 2.2312 1.2803 0.2292 0.3953

Count 8 8 8 11 10 9 8

SEM 18.2877 3.9021 1.0189 0.6727 0.4049 0.0764 0.1398

As illustrated in Table 1, the glutaraldehyde fixed tissue valve exhibited substantially greater calcification than the fresh tissue, which was not processed in any fashion. In contrast, calcification was substantially reduced in the tissue valves prepared according to the present invention, and in the tissue valves prepared according to the Lipase C and Lipase D conditions in particular. Calcification in these tissues was further substantially reduced over the calcification observed for the tissue valve prepared according to U.S. Patent No. 4,776,853.

These results confirm that the method of the present invention results in a bioprosthetic implant which exhibits a reduced tendency to calcify as compared with an unprocessed tissue graft, a commercially available glutaraldehyde fixed implant, when implanted in vivo.

Example 7

Following the extraction process, the biological material may be cultured with the recipient's cells to partially repopulate the valve. The cells may be fibroblasts or myofibroblasts or both. The cells are

28

c r- ^■ explanted from a suitable, non-neoplastic connective tissue source taken from the recipient. Examples include cells taken from skin, gingiva, or granulation tissue.

In the case of gingiva, one or several small 2 X 2 mm. biopsies were taken from minimally or noninflamed marginal gingiva. The tissues were handled aseptically and washed 6 times by centrifugation (1000 rpm 10 min. at lOfiC) in tissue culture medium (typically Dulbecco's modified Eagle's medium - DMEM) supplemented with 50 g/ml gentamicin and fungizone. The tissue pieces were cut in a containment cabinet with a sterile scalpel into 1 X 1 mm pieces and transferred to a sterile tissue culture dish where they were covered with a sterile glass coverslip and incubated at 37ac in 5% C0₂ in culture medium containing 10% fetal calf serum, 50 μg/ l gentamicin.

Once fibroblasts or myofibroblasts reach confluency from the explanted gingival tissue, the material was removed, the cells washed with sterile balanced salt solution, two times, and the cells were treated with buffered trypsin solutions by standard techniques to passage the cells. Following several passages, selected dishes of cells were washed and fixed in situ with formalin, then immunohistochemically stained with antibodies to vimentin, cytokeratin, muscle specific actin, and human fibroblasts to verify that the cells were fibroblasts or myofibroblasts.

To repopulate the extracted mammalian biological material with the recipient's cells, the biological material was washed with tissue culture medium, then placed in a sterile culture dish or flask, and once adherent to the dish, covered with medium containing 5 X 10⁴ fibroblasts or myofibroblasts and incubated at 37sc with 5% C0₂ in culture medium supplemented with 10% fetal calf serum (FCS) , cytokines and antibiotics. The cells were cultured in the presence of cytokines, growth factors, transforming growth factors, various interleukins and other generally anabolic cytokines. Using porcine aortic valve cusps, fibroblasts or myofibroblasts were typically present on the surface and superficially within the extracted cusp connective tissue matrix by the tenth day of culture, as determined by histological studies.

After the biological material was repopulated with the fibroblasts or myofibroblasts of the recipient, the material was implanted in the body of the recipient using common surgical techniques. All the references cited herein, including patents, patent applications and publications are hereby incorporated in their entireties by reference.

It will be obvious to those skilled in the art that variations in the preferred methods may be used, including variations due to improvements in the art, and that it is intended that the invention be practiced otherwise than as specifically described herein to encompass these variations. Accordingly, this invention includes all modifications encompassed within the spirit and scope of the invention as defined by the following claims.

References

[1] Hufnagel CA. Basic concepts in the development of cardiovascular prostheses. Am.J.Surg. 137:285-300, 1979.

[2] Levy R.J., Shoen F.J., Golo b G. Bioprosthetic heart valve calcification: Clinical features, pathobiology, and prospects for prevention. CRC Critical Reviews in Biocompatibility. 2 (2) : 147-187, 1986.

[3] Bortolotti U. , Milano A., Mazzucco A., Valfre

C. , Talenti E. , Guerra F. , Thiene G., Gallucci V. Results of reoperation for primary tissue failure of porcine bioprostheses. Thorac.Cardiovasc. Surg. 90:564-569, 1985.

[4] Hammermeister K.E., Sethi G.K., Henderson W.G.,

Claims

Oprian C. , Kim T. , Rahimtoola S. A comparison of outcomes in men 11 years after heart-valve replacement with a mechanical valve or bioprostheses. N.Eng.J.Med. 328(18) :1289-96, 1993.[5] Ishihara T., Ferrans V.J., Boyce S.W., JonesM. , Roberts W.C. Structure and classification of cuspal tears and perforations in porcine bioprosthetic cardiac valves implanted in patients. Am.J.Cardiol. 48:665-678, 1981.[6] Pomar J.L., Bosch X., Chaitman B.R. , Pelletier C, Grodin CM. Late tears in leaflets of porcine bioprostheses in adults. Ann. horac.Surg. 37(l):78-83, 1984.[7] Grabenwoger M. , Grimm M. , Eybl E. , Kadletz M. , Havel M. , Kostler P., Plenk H. , Bock P., Wolner E. New aspects of the degeneration of bioprosthetic heart valves after long-term implantation. J.Thorac.Carciovasc.Surg. 104(1) :14-21, 1992.[8] Tani ura A., McGregor D.H., Anderson H.CMatrix vesicles in artherosclerotic calcification. Proc.Soc.Exp.Biol.Med. 172:173-177, 1983.[9] Kim K.M. , Valigorsky J.M., Mergner W.J. , Jones R.T. , Pendergrass R.F., Trump B.J. Aging changes in the human aortic valve in relation to dystrophic calcification. Human.Pathol. 7(l):47-60, 1976.[10] Levy R.J. , Schoen F.J. , Levy J.T. , FlowersW.B., Staelin S.T. Initiation of mineralization in bioprosthetic heart valves: Studies of alkaline phosphatase activity and its inhibition by A1C13 or FeC13 preincubation. J.Biomed. ater.Res. 25:905-935, 1991. [11] Schoen F.J., Collins J.J., Cohn L.H. Long term failure rate and morphologic correlations in porcine bioprosthetic heart valves. Am.J.Cardiol. 51:957-964, 1983.[12] Paule W.J., Bernick S., Strates B., Nimni M.E. Calcification of implanted vascular tissues associated with elastin in an experimental animal model. J.Biomed. ater.Res. 6:1169-1177, 1992.[13] Vesely I., Boughner D.R. , Song T. Tissue buckling as a mechanism of bioprosthetic valve failure. Ann.Thorac.Surg. 46:302-308, 1988.[14] Vesely I., Boughner D.R. A comparison of the bending behavior of fresh and glutaraldehyde treated porcine aortic valve leaflets. Circulation 74 (4) :II-339, 1986.[15] Thubrikar M.J. , Aouad J. , Nolan S.P. Patterns of calcific deposits in operatively exicised stenotic or purely regurgitant aortic valves and their relation to mechanical stress. Am.J.Cardiol. 54:304-308, 1986.[16] Krucinski S., Vesely I., Dokainish M.A.,Campbell G. Numerical Simulation of Leaflet Flexure in Bioprosthetic Valves Mounted on Rigid and Expansile Stents. J.Biomech. 26(8) :929-943 , 1993.[17] Stein P.D., Kemp S.R., Riddle J.M., Lee M.W. , Lewis J.W. , Magilligan D.J. Relation of calcification to torn leaflets of spontaneously degenerated porcine bioprosthetic valves. Ann.Thorac.Surg. 40(2) :175-180, 1985.[18] Gallo I., Nistal E. , Artinano E. , Fernandez D. , Cayon R. , Carrion M. , Garcia-Martinez V. The behavior of pericardial versus porcine valve xenografts in the growing sheep model. J.Thorac.Cardiovasc.Surg. 93:281-290, 1987.[19] Vesely I., Noseworthy R. Micromechanics of the fibrosa and ventricularis of aortic valve leaflets. J.Biomech 25(1) :101-113, 1992.[20] Kruth H.S. Localization of unesterified cholesterol in human atherosclerotic lesions. Am.J.Pathol. 114(2) :201-208, 1984.[21] Mitchison M.J. , Hothersall D.C, Brooks P.N. I De Burbure CY. The distribution of ceroid in human atherosclerosis. J.Pathol. 145:177-183, 1985.[22] Kruth H.S., Fry D.L. Histochemical detection and differentiation of free and esterified cholesterol in swine atherosclerosis using filipin. Exp.Molec.Pathol. 40:288-294, 1984.[23] Boyan B.D., Schwartz Z., Swain L.D., Khare A. Role of lipids in calcification of cartilage. The Anatomical Record 224:211-219, 1989.[24] Skold B.H., Getty R. , Ramsey F.K. Spontaneous atherosclerosis in the arterial system of aging swine. Am.J.Vet.Res. 27(116) :257-273, 1966.[25] Ferrans V.J., Spray T.L., Billingha M.E., Roberts W.C Structural changes in glutaraldehyde- treated porcine heterografts used as substitute cardiac valves. Am.J.Cardiol. 41:1159-1184, 1978.[26] Ferrans V.J. , McManus B. , Roberts W.C. Cholesteryl ester crystals in a porcine aortic valvular bioprosthesis implanted for eight years. Chest 83:698- 701, 1983. [27] Huff M.W., Sawyez C.G., Connelly P.W. , Maguire G.F., Little J.A. , Hegele R.A. B-VLDL in nepatic lipase deficiency induces apoE-mediated cholesterol ester accumulation in acrophageε. Arterioscler. Thromb. 13(9) :1282-1290, 1993.[28] Guyton J.R., Klemp K.F. The lipid-rich core region of human atherosclerotic fibrous plaques. Am.J.Pathol. 134 (3) :705-717, 1989.[29] Pasquinelli G., Preda P., Vici M. , Gargiulo M. , Stella A. , D'Addato M. , Laschi R. Electron microscopy of lipid deposits in human atherosclerosis. Scanning Microscopy 3 (4) : 1151-1159, 1989.[30] Eanes E.D. Biophysical aspects of lipid interaction with mineral: Liposoome model studies. The Anatomical Record 224:220-225, 1989.[31] Thubrikar M.J., Deck J.D., Aouad J. , Nolan S.P. Role of mechanical stress in calcification of aortic bioprosthetic valves. J.Thorac.Cardiovasc.Surg. 86:115-125, 1983.[32] Cipriano P.R., Billingha M.E., Miller D.C Calcification of aortic versus mitral porcine bioprosthetic heart valves: A radiographic study comparing amounts of calcific deposits in valves explanted from the same patient. Am.J.Cardiol. 54:1030- 1032, 1984.[33] Jorge-Herrero E.; Fernandez P., Gutierrez M. , Casillo-Olivares J.L. Study of the calcification of bovine pericardium: analysis of the implications of lipids and proteoglycans. Biomaterials 12:683-689, 1991. What is Claimed:

1. A bioprosthetic implant comprising an extracellular matrix obtained by treatment of a collagen- based body-derived tissue with a lipid-degrading enzyme.

2. The implant of claim 1 wherein said enzyme is selected from the group consisting of lipases and phospholipases, and mixtures thereof.

3. The implant of claim 1 wherein said tissue is selected from the group consisting of skin, blood vessels, heart valves, ligaments, tendons, bone, cartilage, dura mater and nerves.

4. The implant of claim 1 wherein said tissue is a heart valve.

5. A method of preparing a bioprosthetic implant which comprises treating a collagen-based body-derived tissue with a lipid-degrading enzyme.

6. The method of claim 5 wherein said enzyme is selected from the group consisting of lipases, phospholipases, and mixtures thereof.

7. The method of claim 5 wherein said tissue is selected from the group consisting of skin, blood vessels, heart valves, ligaments, tendons, bone, cartilage, dura mater and nerves.

8. The method of claim 5 wherein said tissue is a heart valve.

9. The method of claim 5 which further comprises extracting said implant with a solution comprising salt.

10. The method of claim 9 which further comprises extracting said implant with a solution comprising detergent.

11. The method of claim 10 which further comprises extracting said implant with a solution comprising nuclease.

12. A method of implantation which comprises introducing the implant of claim 1 into a recipient.

13. The method of claim 12 wherein said implant has been recellularized prior to implantation.

14. The method of claim 12 wherein said implant has been recellularized with fibroblasts prior to implantation.

15. A method of recellularizing a bioprosthetic implant which comprises contacting the implant of claim 1 with cells, and maintaining said implant with said cells in tissue culture medium.

16. The method of claim 15 wherein said cells are selected from the group consisting of fibroblasts and myofibroblasts.

17. The implant of claim 1 wherein said treatment is carried out such that all lipids are substantially removed.

18. The method of claim 5 wherein treating of said tissue is carried out such that all lipids are substantially removed.