Ann Thorac Surg 2001;71:S396-S400
© 2001 The Society of Thoracic Surgeons
Basic research
Effect of human immunoglobulins on the immunogenicity of porcine bioprostheses
Olivier Schussler, MD, PhDa,b,
Ming Shen, MD, PhDa,
Lin Shen, MDa,
Sophie M. Carpentier, PhDa,
Srini Kaveri, PhDb,
Alain Carpentier, MD, PhDa
a Laboratoire d’Etude des Greffes et Prothèses Cardiaques, Université de Paris VI, Paris, France
b INSERM U430, Hôpital Broussais, Paris, France
Address reprint requests to Dr Schussler, Laboratoire d’Etude des Greffes et Prothèses Cardiaques, UPRES 264, Hôpital Broussais, 96 rue Didot, 75014 Paris, France
e-mail: labo.legpc{at}brs.ap-hop-paris.fr
Presented at the VIII International Symposium on Cardiac Bioprostheses, Cancun, Mexico, Nov 3–5, 2000.
 |
Abstract
|
|---|
Background. Glutaraldehyde fixation (GT) is known to reduce immunologic reactions and tissue degeneration after implantation in humans. Sterilization after glutaraldehyde fixation (G-ST) improves the safety and reduces the tendency of GT valves to calcify. Intravenous immunoglobulins (IVIg) have been shown to reduce xenogeneic response against porcine tissue. We have investigated the effect of these fixation procedures combined with and without IVIg on the antigenicity of bioprostheses.
Methods. Lewis adult rats were implanted subcutaneously with a fresh, GT, or G-ST porcine heart valve pre- or posttreated with different amounts of IVIg. We followed by enzyme-linked immunosorbent assay and IgM and IgG titers against protein extracts from the porcine heart valves after implantation. Cellular reactivity was assessed in xenogeneic lymphoendothelial coculture experiments. Calcification content was also examined.
Results. Glutaraldehyde fixation partially decreased the humoral response against proteins of the implant but elicited a cellular xenogeneic response. Sterilization reduced these reactivities, but retained antigenicity. Intravenous immunoglobulin postincubated with GT valves before implantation reduced the antigenicity of the tissue to the same extent as G-ST valves, but had no effect on valvular tissue calcification.
Conclusions. Our studies demonstrate that IVIg or the sterilization procedure (ST) reduced the cellular response against glutaraldehyde-fixed valves (GT), whereas reduced calcification was observed only with ST.
|
Introduction
|
|---|
Fresh porcine tissues are highly immunogenic. Glutaraldehyde fixation (GT) has been introduced to reduce valve antigenicity and to improve its mechanical strength [1]. Such a procedure has allowed the use of xenogeneic tissue for valvular tissue replacement in humans. Bioprostheses have demonstrated excellent hemodynamics, low thrombogenicity, and acceptable long-term longevity in adults. However, degeneration of these implants mainly by tears or calcification after several years of implantation, especially in young adults and in children, has restricted their use. Immunologic reactions against porcine or bovine implants are often cited as one of the factors involved in bioprosthesis alteration in animal models [2] or in humans [3]. This has been demonstrated by immunohistologic analysis of the explants, which have shown the presence of cellular infiltration (mainly macrophages and some T lymphocytes) [4], the deposition of complement and the deposition of immunoglobulins early on after implantation that colocalize with collagen fiber alteration and calcifications [4]. At the same time, animal models have revealed a modification of cellular and humoral reactivity against porcine or bovine antigen after implantation of the respective tissues fixed in glutaraldehyde. Some early degeneration of bioprostheses has been attributed directly to an immunologic rejection in infants or adults. This finding was further confirmed by the longevity of bioprostheses in patients under corticoid therapy.
Intravenous immunoglobulin (IVIg) is a therapeutic preparation of polyclonal human IgG prepared from pools of plasma of several thousand healthy blood donors. In addition to its use as substitutive therapy for primary and secondary antibody deficiencies, IVIg has been shown to be beneficial in immunotherapy of patients with a variety of autoimmune diseases, systemic inflammatory disorders, and more recently in allotransplantation and xenotransplantation. In xenotransplantation, IVIg has been used to delay the hyperacute rejection of a pig heart implanted into a baboon [5]. The mechanisms of action of IVIg in xenotransplantation are complex. However, IVIg contains natural antibodies against xenogeneic tissues [5] and may modify the immune response against them.
In this study, we have investigated the effect of the glutaraldehyde fixation (GT) and sterilization after glutaraldehyde fixation (G-ST) [6] on the antigenicity of porcine heart valves implanted subcutaneously in rats, and the effect of IVIg on such reactivity by enzyme-linked immunosorbent assay and proliferative assays. We have further investigated the effect of these treatments on tissue calcification.
|
Material and methods
|
|---|
Intravenous immunoglobulin and preparation of F(ab')2 fragments
Intravenous immunoglobulin was provided by Baxter (Lessines, Belgium). F(ab')2 fragments were prepared from IVIg by pepsin digestion (2% w/w) as described elsewhere [7].
Tissue preparation
Porcine aortic valve leaflets were obtained at slaughter, washed twice in Dulbecco’s modified Eagle’s medium containing Penicillin/Streptomycin (PS) and fungizone, and preserved on ice until implantation within 4 hours. These valves were used either fresh (F) or placed in 0.625% glutaraldehyde (Merck, Darmstadt, Germany) in phosphate buffer pH 7.4 at room temperature for a minimum of 6 weeks to obtain glutaraldehyde fixed valve (GT). The sterilization procedure (ST) consisted of incubation (GT) samples in surfactant solution to obtain G-ST valves as described [6]. Intravenous immunoglobulins (5 to 40 mg/mL) or F(ab')2 fragments of IVIg were incubated with the different valves (F, GT, G-ST) for 2 hours at 37°C before implantation. In some cases, F valves were preincubated with IVIg before fixation with glutaraldehyde.
Subcutaneous tissue implantation and retrieval
Adult female Wistar rats (7 weeks old) were implanted subcutaneously for a period of 3 months with fresh, cryopreserved, or glutaraldehyde-fixed porcine aortic tissue [6] incubated or not with different amount of IVIg. Before tissue implantation, the rats were anesthetized with diethyl ether and four aortic cups were placed in subcutaneous pouches dissected in the dorsal wall. Only one treatment was investigated per animal and 6 to 12 rats were tested for each condition. Serum was collected at days 0 to 90 under anesthesia by retroorbital bleeding and stored at -80°C. Implants were removed at 1 and 3 months for calcium content measurement. The rats were sacrificed at 3 months by an overdose of diethyl ether and the spleens were collected.
Antibody reactivity against protein extracts from the porcine heart valves
An enzyme-linked immunosorbent assay [3] was designed to examine the reactivity of serum of the rats toward protein extracts from the porcine valves (PVE). The PVE was extracted from porcine valves or porcine endothelial cells as described previously [8].
Lymphoendothelial coculture
Xenogeneic lymphoendothelial coculture [9] was performed by mixing responding rat peripheral blood mononuclear cells (105 cells/well) with stimulating irradiated (20 Gy) porcine endothelial cells (PAEC) (3 x 104 to 3 x 105 cells/well) prepared and maintained as described previously [9] and pulsed after 6 days with [3]H-thymidine (1 mCi/well).
Calcium measurement was performed by atomic absorption spectrometer and expressed as percentage of dry tissue.
Statistical analyses were performed by the analysis of variance test and simple regression analysis.
|
Results
|
|---|
The animals implanted with F valves developed humoral responses against their implants. The titer of their IgM and IgG antibodies against PVE increased in their sera as in a classic humoral immunologic response against foreign antigens. During the same period of time, control animals implanted with Dacron and bred in the same conditions did not develop such modifications in their serum reactivity. These findings suggest the response was directed specifically against the implant (Fig 1).
View larger version (36K):
[in this window]
[in a new window]
|
Fig 1. Evolution of the titer of IgG and IgM against PVE in the serum of animals implanted with fresh or glutaraldehyde-treated porcine implants (n = 6 per group). (NS = not significant; **p = 0.001; ***p = 0.0001; DO = optical density at 405 nm; J = day.)
|
|
Glutaraldehyde fixation decreased the antigenicity of the xenogeneic tissue, because during the first 2 months of implantation the animals of the GT group produced only a low amount of IgM anti-PVE (-61%, n = 6, p = 0.03 analysis of variance test). However, control of tissue antigenicity was not total, because the secondary IgG response was unaffected. The sterilization procedure (G-ST) in fact drastically reduced the IgM response to the level of the control (-100%) and partially reduced the IgG response (-41%, n = 6, p = 0.03). However, even in this group, a 58% (p = 0.007) increase in the serum reactivity was observed when compared with the control (Fig 2A).
View larger version (26K):
[in this window]
[in a new window]
|
Fig 2. (A) Titer of IgG or IgM antibodies against PVE in the serum of rat implanted with fresh, glutaraldehyde (Glut), or sterilized (Glut-ST) porcine valves. (B) Evolution of this reactivity if the Glut or Glut-ST valves were incubated in the presence of intravenous immunoglobulins (IVIg, 20 mg/mL), F(ab')2 fragment (10 mg/mL), or human albumin (20 mg/ml) before implantation. These measurements were taken 3 months after implantation (n = 6 per group; Glut-IVIg, n = 12). (NS = not significant; *p = 0.01; **p = 0.001; ***p = 0.0001.)
|
|
The incubation of GT tissue with IVIg (20 mg/mL) reduced the antigenicity of the tissue since the IgM and IgG anti-PVE responses were decreased during the 2-month study period (-61% and -46%, respectively, p = 0.006, n = 12). The beneficial effect of the IVIg was observed for concentrations of IVIg ranging between 5 and 40 mg/mL, with a plateau at 20 mg/mL. This effect was specific, because a heterologous protein (human albumin) used at the same concentration did not yield similar results. The F(ab')2 portion of IVIg, which contains the antigen binding sites, is responsible for these effects and can mimic the effect of the whole molecule if used at the same concentration (Fig 2B). However, this effect was lost if the IVIg were used before and not after glutaraldehyde fixation.
All the animals implanted with porcine tissue modified their cellular reactivity against PAEC. This finding was demonstrated by a fivefold augmentation of the proliferative response of the splenocyte isolated from these animals against PAEC compared with the moderate proliferative response of the splenocyte of normal rat implanted with Dacron (50,000 and 9,000 counts per minute [CPM], respectively). The intensity of this proliferation was the same whether fresh valves had been fixed or not in glutaraldehyde before implantation (Fig 3A). However, the sterilization procedure or the incubation of GT valves with IVIg (20 mg/mL) significantly decreased the antigenicity of the porcine tissue (-70% for G-ST, n = 6, p = 0.004). The F(ab')2 fragments of IVIg (10 mg/mL) were even more efficient than the entire intact molecule in this control and reduced the antigenicity of GT valves to the same extent as the G-ST procedure did (-62%, n = 6) (Fig 3B).
View larger version (21K):
[in this window]
[in a new window]
|
Fig 3. (A) Proliferation of the splenocyte of rat implanted with fresh, glutaraldehyde-treated (Glut), or sterilized (Glut-ST) porcine tissue in the presence of increasing amounts of irradiated porcine endothelial cell. (B) Evolution of this response if the Glut or Glut-ST valves were incubated in the presence of intravenous immunoglobulins (IVIg, 20 mg/mL), F(ab')2 fragment (10 mg/mL), or human albumin (20 mg/ml) before implantation. These measurements were taken 3 months after implantation (n = 6 per group; IVIg, n = 12).
|
|
As reported earlier [6], the calcium content was lower in fresh than in GT-fixed xenogeneic tissue and the G-ST procedure decreases the tendency of GT tissue to calcify (p = 0.0001, n = 18). Although IVIg or F(ab')2 fragments decreased the antigenicity of (G) xenogeneic tissue, this decrease was not accompanied by a reduced amount of calcium content in these tissues. Moreover, we were unable to identify a positive correlation between the serum reactivity IgM or IgG of the animal before implantation or at the time of the explantation and the degree of calcification of the tissue or between calcium content and xenogeneic lymphoendothelial coculture at the time of explantation. However, IVIg or F(ab')2 fragment did not increase whereas albumin did increase the calcification of GT or G-ST tissues (Fig 4).
View larger version (27K):
[in this window]
[in a new window]
|
Fig 4. Calcification of fresh, cryopreserved (Cryo), glutaraldehyde (Glut), or sterilized (Glut-ST) porcine tissues implanted in adult rats for 3 months. Effect of the incubation of Glut or Glut-ST with intravenous immunoglobulins (IVIg, 20 mg/mL), F(ab')2 fragment (10 mg/mL), or human albumin (20 mg/ml) before implantation on the calcium content after 3 months of implantation (n = 18 per group). (ns = not significant; ***p = 0.0001.)
|
|
|
Comment
|
|---|
Previous reports have shown that porcine bioprostheses of bovine or porcine origin are immunogenic in human adults and infants. However, in animals some discrepancy exists between the origin of the tissue, because only glutaraldehyde-fixed bovine pericardium seems to elicit an immune response if this tissue is subcutaneously implanted in rats [2, 3]. The control of the antigenicity of porcine aortic valve by glutaraldehyde fixation has been shown by the lack of cellular infiltration of the implants (although severe degeneration of collagen fiber exists) and by the absence of the modification of humoral reactivity against porcine collagen in the implanted animals [3].
We have found that glutaraldehyde fixation (GT) only partially decreased the humoral IgM response against proteins of the implant but the implants retained antigenicity and elicited a cellular xenogeneic response comparable to the one observed in control animal implanted with fresh porcine tissue (F). It is known that the cellular immunologic response is involved in the switch from the IgM to the IgG class of antibodies elaborated against foreign antigens. The lack of control of the cellular xenogeneic response against GT was substantiated firstly by a similar humoral IgG response against PAEV in the group of animals implanted with F or GT valves and secondly by a similar proliferation index of the splenocyte isolated from both groups in the presence of irradiated porcine endothelial cells.
The difference we have observed as compared with previous reports [3] can be explained first by a difference in the process of tissue fixation. In our experiments, the valvular tissue was fixed in glutaraldehyde, whereas most of the commercially available porcine valve bioprostheses used in other studies are in fact treated in glutaraldehyde and subsequently "sterilized." Furthermore, we checked the humoral reactivity against a large panel of protein antigens that we extracted from fresh porcine valves, whereas other studies assessed this reactivity against the main component of the extracellular matrix of porcine implants: the collagen. We analyzed different isotypes of immunoglobulins and we found significant variation among the isotypes.
The sterilization procedure, which consists of posttreating GT tissue with ethanol, formol, and tween, had been used initially to decrease the risk of contamination of bioprostheses. Secondarily, the beneficial effect of this treatment on valvular calcifications has been reported [6]. Here, we have shown that in addition to that benefit, this treatment decreases the cellular response against xenogeneic proteins, which are still immunogenic after the GT procedure. We do not know yet which of the three components are mainly involved in this effect. As previously reported, detergents such as tween can release uncross-linked lipoproteins that are still immunogenic after the GT procedure or can modify the collagen structure and thus render the bioprostheses less immunogenic. However, it has been shown that the use of ethanol before GT decreases the tendency of the tissue to calcify by decreasing the amount of phospholipid that initiates the calcification process [10]. The use of ethanol after fixation by a similar mechanism could theoretically eradicate uncross-linked lipoproteins.
In this study, we have shown that it is possible to decrease the cellular xenogeneic response against porcine GT tissue by incubating the tissue with IVIg and that the F(ab')2 fragments of IVIg are mainly involved in this effect. It is known that the perfusion of IVIg can prevent the hyperacute xenograft rejection of a porcine heart transplanted into a primate [5]. However, the physiopathology of the rejection of a xenogeneic vascularized organ or a tissue is totally different. At least two mechanisms could explain the effect of IVIg on GT tissue: a competitive inhibition of the binding of preformed natural antibodies (XNA) against porcine valve tissue or the masking of xenogeneic epitope to natural killer cells (NK). XNA do exist in humans against GT tissue and can provoke in vitro and in vivo activation of the complement through the classic pathway. However, different studies have documented that the XNA of the IgM isotype but not XNA IgG provoke complement activation on xenogeneic tissue and that there is a competition for the binding of human XNA IgM and IgG on porcine tissue. The IVIg preparations also contain XNA IgG against porcine antigen [5] that can compete for the binding with XNA IgM of another origin. This would prevent the activation of the complement through the classic pathway and the generation of anaphylatoxin involved in the recruitment of inflammatory cells. More XNA fixed by their F(ab')2 portion can interact by their Fc portion with Fc receptor present on cells with antibody-dependent cell cytotoxicity activity. The use of F(ab')2 fragments of IVIg instead of the whole IVIg would prevent this interaction. The NK cells play an essential role in xenogeneic tissue rejection. There is evidence that human NK cell and human XNA IgG would recognize the same antigen on foreign porcine tissues. It is thus possible that incubation of GT tissue with human IgG would mask those antigen motifs to the immune system. The lack of effect of the IVIg when the glutaraldehyde fixed tissue has been sterilized suggests that the sterilization procedure somehow alters the antigen motifs normally recognized by IVIg or by rat XNA on GT tissue.
In this particular experiment, F valves did not calcify possibly because these highly immunogenic tissues are degraded by the inflammatory and immune response before calcification can occur. On the contrary, GT valves are tanned and are more resistant, therefore calcification can occur. In fact, we were not able to show a correlation between the serum titer against PVE in the animal before the implantation and the calcium content of the implants after 1 and 3 months. Although IVIg decreases the antigenicity of GT, this treatment has no impact on the propensity of the tissue to calcify. Moreover, interindividual variability for the level of antibodies against PVE was detected in the serum of animal in each treatment group. However, there was not a clear correlation between the titer of these antibodies and the level of calcification of the implants. Our results are in agreement with previous reports that have shown in immunodeficient rats no modification of the tendency of GT to calcify if compared with the wild type. However, nude rats still possess a normal NK cell response, which is important in the rejection of xenogeneic tissue. The direct involvement of glutaraldehyde in the tendency of the fixed tissue to calcify is still under discussion. However, other mechanisms such as excessive stress or metabolic or immune reactions, combined or not, may provoke by themselves or participate in the calcification process. Our experiments do not rule out the fact that an immunologic process could participate in the calcification of bioprostheses in humans. It is possible that in patients, early and late calcifications of bioprostheses do not have the same physiopathology; moreover, it is always difficult to reproduce in 3 months in animal models what happens to circulatory flow most of the time in 10 years in humans.
Together, our studies confirmed that GT bioprostheses retained immunogenic proteins that elicited a xenogeneic cellular response. Human IVIg or the sterilization procedure partially reduced this antigenicity.
|
References
|
|---|
-
Carpentier A., Lemaigre G., Robert L., Carpentier S., Dubost C. Biological factors affecting long-term results of valvular heterografts. J Thorac Cardiovasc Surg 1969;58:467-483.[Medline]
-
Gong G., Seifter E., Lyman W.D., Factor S.M., Blau S., Frater R.W. Bioprosthetic cardiac valve degeneration: role of inflammatory and immune reactions. J Heart Valve Dis 1993;2:684-693.[Medline]
-
Dahm M., Husmann M., Eckhard M., Prufer D., Groh E., Oelert H. Relevance of immunologic reactions for tissue failure of bioprosthetic heart valves. Ann Thorac Surg 1995;60:S348-S352.
-
Srivatsa S.S., Harrity P.J., Maercklein P.B., et al. Increased cellular expression of matrix proteins that regulate mineralization is associated with calcification of native human and porcine xenograft bioprosthetic heart valves. J Clin Invest 1997;99:996-1009.[Medline]
-
Magee J.C., Collins B.H., Harland R.C., et al. Immunoglobulin prevents complement-mediated hyperacute rejection in swine-to-primate xenotransplantation. J Clin Invest 1995;96:2404-2412.
-
Carpentier A., Nashef A., Carpentier S., Ahmed A., Goussef N. Techniques for prevention of calcification of valvular bioprostheses. Circulation 1984;70(Suppl I):165-168.[Abstract/Free Full Text]
-
Vassilev T.L., Kazatchkine M.D., Van Huyen J.P., et al. Inhibition of cell adhesion by antibodies to Arg-Gly-Asp (RGD) in normal immunoglobulin for therapeutic use (intravenous immunoglobulin, IVIg). Blood 1999;93:3624-3631.[Abstract/Free Full Text]
-
Mouthon L., Nobrega A., Nicolas N., et al. Invariance and restriction toward a limited set of self-antigens characterize neonatal IgM antibody repertoires and prevail in autoreac-tive repertoires of healthy adults. Proc Natl Acad Sci U S A 1995;92:3839-3843.[Abstract/Free Full Text]
-
Xu L., Oluwole S.F., Edwards N.M., et al. In vitro evaluation of neonatal human immunity against the pig. J Thorac Cardiovasc Surg 1996;111:920-929.[Abstract/Free Full Text]
-
Vyavahare N., Hirsch D., Lerner E., et al. Prevention of bioprosthetic heart valve calcification by ethanol preincubation. Efficacy and mechanisms. Circulation 1997;95:479-488.[Abstract/Free Full Text]
Top
Abstract
Introduction
