The ability to perform successful xenotransplantation would alleviate the disparity between the large number of human organs needed for transplants compared to the small number available (1). The pig has been the donor of choice for xenotransplants because of its similar organ physiology to the human. However, most transplantations of vascularized organs from pig to human are rejected within minutes or a few hours after implantation due to hyperacute rejection that occurs when naturally occurring anti-Galα(1,3) Gal antibodies of the host (human) interact with the Galα(1,3) Gal epitope found on endothelial cells that line the blood vessels of the donor (pig) organ. This results in complement activation and subsequent rejection of the organ (2–5). Delayed xenograft rejection can also present problems in xenograft transplants. It occurs 3 to 5 days after grafting and is manifested by anti-donor antibodies binding to the endothelium of the graft. This interaction activates endothelial cells and results in the production of proinflammatory cytokines and chemokines that attract and activate host macrophages and natural killer cells (6–8).
Extracellular matrix (ECM) biomaterials of xenogeneic origin, such as porcine small intestinal submucosa (SIS), are beginning to be used as acellular, resorbable bioscaffolds for tissue repair. Because these materials are avascular, hyperacute rejection mechanisms involving endothelial activation and intravascular thrombosis of the graft are avoided. When used as xenografts, ECMs appear to induce site-specific remodeling in the organ or tissue into which they are placed. ECM biomaterials have been used successfully in many species to repair cardiovasculature tissue (9), abdominal wall defects (10), urinary bladder defects (11), and ligament and tendon damage (12,13). Despite the xenogeneic nature of the graft material and the fact that no immunosuppression is used, there has never been any clinical or histological evidence of immediate or delayed rejection. It is intuitive that the graft recipient should mount an immune response to the foreign xenograft but whether such a response occurs and its nature is presently unknown.
Several studies have analyzed the role of Th1 and Th2 lymphocytes in cell mediated immune responses to xenografts (14,15). Th1 lymphocytes produce interleukin- (IL)2, interferon- (IFN) γ, and tumor necrosis factor- (TNF) β leading to macrophage activation, stimulation of complement-fixing antibody isotypes (IgG2a and IgG2b in mice) and differentiation of CD8+ cells to a cytotoxic phenotype (16,17). Activation of this pathway is associated with both allogeneic and xenogeneic transplant rejection (14,15,18). Th2 lymphocytes produce IL-4, IL-5, IL-6, and IL-10, cytokines that do not activate macrophages and lead to production of non-complement-fixing antibody isotypes (IgG1 in mice). Activation of the Th2 pathway is associated with transplant acceptance (19–21).
Our studies were designed to determine if ECM can in fact elicit an immune response and, if so, to determine the nature and kinetics of that response. The results suggest that implanted ECM elicits a vigorous immune response but this response is restricted to the Th2 pathway, which may allow acceptance and remodeling of the graft tissue.
MATERIALS AND METHODS
Female BALB/c and C57BL/6 mice 5–6 weeks of age were obtained from Charles River Laboratories (Raleigh, NC) through the National Cancer Institute (Bethesda, MD). C57BL/6 T cell KO mice were obtained from The Jackson Laboratory (Bar Harbor, ME). These mice have gene disruptions in T cell receptor (TCR) β and δ constant regions and therefore, they lack both αβ and γδ T cells. Animals were maintained in a specific pathogen-free environment with food and water provided ad libitum.
Representative extracellular matrix materials, porcine SIS, and urinary bladder submucosa (UBS) (Cook Biotech, Inc., West Lafayette, IN) were prepared as previously described (22). Briefly, porcine intestine was used to prepare small intestinal submucosa. The intestine was washed free of debris, everted, and subjected to mechanical delamination to remove the superficial layers of the mucosa. Then, it was reverted to its original orientation, split open longitudinally, rinsed in water to lyse any cells present and to remove cellular debris. The SIS was cut into 1-cm3 pieces, irradiated at 3000 rads for 15 min, and stored in sterile phosphate-buffered (PBS) at 4°C until used for implantation.
Tris extracts of ECM.
Soluble preparations of SIS and UBS were obtained by pulverization of frozen ECM under liquid nitrogen and suspension at 25% (w/v) in 50 mM Tris-HCl, pH 7.4, containing protease inhibitors. This was followed by vigorous stirring for 24 hr at 4°C. The mixture was centrifuged at 12,000×g for 30 min at 4°C and the supernatants were dialyzed extensively against deionized water. Protein concentrations were determined using a Bradford protein assay kit (Bio-Rad, Hercules, CA).
Mice were anesthetized with methoxyflurane and their abdomens clipped and prepared for sterile surgery. A 1-cm2 graft of SIS, UBS, BALB/c, or C57BL/6 mouse abdominal muscle (syngeneic control), or rat abdominal muscle (xenogeneic control) was placed in a ventral midline subcutaneous pocket and secured to the underside of the skin with 6–0 prolene sutures. Sham-treated animals were subjected to surgery without implantation of tissue. Groups of four animals each were killed at days 1, 4, 7, 10, and 28 days postsurgery. A sham-operated control animal was included at each time point. At the time of sacrifice, blood was drawn from anesthetized animals by retroorbital bleeding. Animals were then euthanized by halothane inhalation and the graft site, including implanted tissue, underlying abdominal wall, and overlying skin was excised and divided longitudinally. One portion was snap-frozen in liquid nitrogen and stored at −80°C for RNA analysis and the remainder of the tissue was fixed in 10% neutral buffered formalin for histopathological evaluation.
Paraffin-embedded tissue was sectioned to a thickness of 7 μm and stained with hematoxylin and eosin by standard methods. Pathology at the graft site was determined in a double-blind fashion.
Cytokine RNA analysis.
Total RNA was isolated from frozen implant tissue using Trizol reagent (Life Technologies, Rockville, MD) according to the manufacturer’s directions. Briefly, frozen tissue was homogenized with a mortar and pestle, then transferred to a tube containing 2 ml of Trizol reagent. The samples were centrifuged at 12,000×g and the supernatants were extracted with chloroform. The RNA was precipitated with isopropanol, washed in 75% ethanol and dissolved in diethylpyrocarbonate treated water. RNA was quantitated by spectrophotometric analysis at 260 nm and determined to be intact and free of DNA contamination by electrophoresis on 1% agarose gels. Cytokine expression at the graft site at various times post implantation was determined by reverse transcriptase polymerase chain reaction (RT-PCR) using primers for IFN-γ(23), a representative Th1 cytokine, and IL-4 (23), a representative Th2 cytokine. Primers for the housekeeping gene, hypoxanthine phosphoribosyl transferase (HPRT), were included in all assays to ensure equal loading of nucleic acid. PCR amplification of 2 μl cDNA prepared from total RNA was performed as previously described (23). The PCR products were separated on 2.5% agarose gels and visualized under UV light after staining with ethidium bromide. Real time PCR was performed as described (24)
ELISA for determination of ECM specific antibodies.
Serum samples were analyzed by ELISA for IgM, IgG1, IgG2a, IgG2b, and IgG3 ECM specific antibodies. Microtiter plates were coated overnight at 4°C with 10 μg/ml of a Tris-HCl extract of ECM. The plates were washed and blocked with PBS containing 5% fetal bovine serum and 0.3% Brij-35. Serial 2-fold dilutions of serum were added to the wells and incubated for 2 hr at room temperature (RT). The plates were washed and incubated with alkaline phosphatase conjugated goat anti-mouse isotype specific antibodies (Southern Biotechnology Associates, Birmingham, AL) for 1 hr at RT. Appropriate working dilutions and the isotype specificities of these reagents were tested using myeloma proteins of defined isotypes (Sigma, St. Louis, MO). The plates were again washed, p-nitrophenyl phosphatase substrate (Sigma) was added and absorbance was read at 405 nm on a microplate reader (Bio-Tek Instruments, Winooski, VT). Pooled serum from BALB/c mice immunized i.p. with 50 μg Tris extract of ECM in complete Freund’s adjuvant followed 28 days later by a booster injection of 50 μg Tris extract in incomplete Freund’s adjuvant was used as a positive control. Preimmune serum from implanted mice was included as a negative control.
Histopathology analysis of engrafted mice.
The histopathological response of BALB/c mice implanted with SIS, syngeneic tissue, or xenogeneic tissue was used as a marker for acceptance or rejection. Syngeneic and xenogeneic muscle tissue were used simply as proven models for visualization of acceptance and rejection, respectively. Implantation of rat tissue caused an acute inflammatory reaction consisting mostly of polymorphonuclear leukocytes (PMNs) beginning 1 day after implantation (Fig. 1). By day 10, the inflammation was chronic and hematoxylin and eosin staining revealed neutrophils at the graft site. Immunohistochemical analysis revealed an influx of T cells based on positive staining for CD3 and negative staining for B220 (data not shown). Additionally, there was focal degeneration of the implant and evidence of giant cell formation (Fig. 2). By day 28, there was dramatic worsening of the inflammatory process both within the implant and the surrounding interface between the implant and host tissue. The implant was completely necrotic and “walled off” from the host tissue. Additionally, tissue at this time exhibited granuloma formation at the interface between the implant and host tissue (Fig. 3). Taken together, these observations are indicative of rejection of the xenogeneic implant.
Implantation of syngeneic tissue resulted in an initial, acute inflammatory response both within the graft and at the interface with the host tissue (Fig. 1). By day 10, there was minimal chronic inflammation and the acute inflammation had mostly resolved (Fig. 2). By day 28, the implant was well organized and all tissue degenerative changes associated with inflammation had completely resolved (Fig. 3). These observations are consistent with graft acceptance.
Implantation of SIS resulted in acute inflammation consisting of PMNs mainly at the interface of the host tissue and the SIS implant (Fig. 1). By day 10, the inflammation was greatly reduced and consisted of mononuclear cells with a few scattered neutrophils (Fig. 2). The implant was surrounded by increased tissue organization and fibroblastic proliferation indicating that tissue remodeling was occurring. On day 28, the inflammatory process was completely resolved (Fig. 3). Together, these observations are indicative of acceptance. Thus, the host response to SIS was similar to that of syngeneic tissue with a transitory, acute inflammatory response, followed by resorption and remodeling by native tissue.
Cytokine mRNA analysis of SIS-grafted mice.
The results of the histopathological studies were consistent with failure of the mice to reject xenogeneic SIS. Rejection of xenografts is thought to be mediated by Th1 responses involving cytotoxic T cells induced by IFN-γ(14). Therefore, we measured expression of Th1 and Th2 cytokines at the graft site after implantation into BALB/c mice. The results showed that at the time of mononuclear cell infiltration (7 days after implantation) xenografts contained both IFN-γ and IL-4 mRNA (Fig. 4). Mice implanted with syngeneic tissue expressed IL-4 mRNA, as did SIS-implanted mice, which actually demonstrated augmented IL-4 mRNA expression compared to control animals (Fig. 4). The RT-PCR results were confirmed by real-time RT-PCR which showed that SIS-implanted mice had 100-fold fewer copies of IFN-γ mRNA at the graft site compared to mice implanted with xenogeneic tissue. Interestingly, mice implanted with SIS had 10-fold fewer copy numbers of IFN-γ compared to mice implanted with syngeneic tissue suggesting that SIS is suppressive for IFN-γ expression. The graft sites from all groups were also analyzed for IL-5, IL-6, and TGF-β. Very low levels of IL-6 were detected in some grafts although IL-5 and TGF-β were not detected (data not shown).
Anti-SIS antibody analysis of SIS-implanted mice.
Induction of SIS-specific antibodies of defined Th1 isotypes (IgG2a and IgG2b) and Th2 isotypes (IgG1) was next examined as a marker of Th subset activation. Sera from BALB/c mice immunized with a Tris extract of the SIS in complete Freund’s adjuvant followed by a boost with the same material in incomplete Freund’s adjuvant contained high levels of IgG1, IgG2a, and IgG2b SIS-specific antibodies (Fig. 5). This shows that mice can, under these conditions, respond to xenogeneic SIS with activation of both the Th1 and Th2 pathways. SIS implanted mice, however, expressed specific antibodies that were restricted to the IgG1 isotype. As expected, serum from normal mice did not contain SIS-reactive antibody, nor did sera from mice implanted with syngeneic or xenogeneic muscle tissue. No SIS-reactive IgM or IgG3 antibodies were detected in any of the animals (data not shown). Thus, although there was a vigorous immune response to the xenografted ECM after implantation, this response was restricted to the Th2 pathway. These results are in accordance with the pattern of cytokine expression noted at the graft site on day 7 (Fig. 4).
To confirm that the immune response to SIS was due to Th2 restriction rather than lack of sufficient antigen stimulation, antibody responses in BALB/c mice receiving two sequential SIS implants 28 days apart were measured. These animals showed a significant secondary SIS-specific antibody response but the response was still exclusively of the IgG1 isotype (Fig. 6). Cytokine profiles at the secondary graft site similarly showed no evidence for induction of Th1 cytokines (data not shown). Furthermore, to determine if the SIS graft was absorbing out IgG2a antibodies, antibody deposition on the SIS graft was analyzed. Immunohistochemical results showed no evidence for deposition of IgG1 or IgG2a antibodies on the graft (data not shown).
To determine if ECM derived from another source induced an immune response similar to that of SIS, UBS was implanted into BALB/c mice. As in the SIS immune response, specific antibodies were expressed by UBS implanted mice but were restricted to the IgG1 isotype (Fig. 7). Additionally, the results demonstrated that UBS-reactive serum cross-reacted with SIS (Fig. 7). Thus, the source of ECM did not alter the restricted Th2 immune response.
The immune response to SIS is T cell dependent.
To determine if the observed cytokine and antibody responses to SIS were T cell-dependent, C57BL/6 T cell KO mice were implanted with SIS and analyzed as above. Histologically, the graft sites showed a mild inflammatory response that resolved by day 10 postimplantation. By day 28 postimplantation, the graft was completely remodeled and indistinguisable from native tissue (Fig. 8). Cytokine analysis showed that T cell KO mice graft sites did not contain IL-4 mRNA (Fig. 9), demonstrating that T cells are the source of IL-4 mRNA observed in SIS implanted wild-type mice. Analysis of SIS-specific antibody levels showed the absence of anti-SIS antibodies in these mice (Fig. 10). Examination of C57BL/6 μMT KO mice, which have a targeted gene disruption in IgM that prevents B cell expression, demonstrated the absence of SIS-specific antibody responses in these mice but the presence of IL-4 mRNA (data not shown). Again, SIS was remodeled in mice lacking B cell expression. Together, these results indicate that T and B cells do indeed respond to SIS, but these cells are not necessary for SIS acceptance.
ECM grafts serve as bioscaffolds for site appropriate host tissue growth and remodeling and, in the process, are resorbed. In addition, ECM materials are acellular and avascular which makes unlikely immune-mediated destruction of tissue blood supply. The cells that populate the ECM graft are host-derived and thus nonimmunogenic. Nevertheless, it has been unclear why such xenogeneic material does not appear to induce an inflammatory response that leads to rejection.
Our study shows that there is in fact a vigorous host response to implanted ECM xenografts as evidenced by an early, acute inflammatory response at the graft site that consists mostly of PMNs. Cytokine and antibody isotype analysis demonstrated the presence of a Th2 response but the absence of a Th1 response. These results differ from typical cytokine and antibody responses to xenografts, which induce a Th1 cytokine response (14). Clearly, porcine ECM is not an inert material but is associated with the production of antiinflammatory cytokines and noncomplement-fixing antibodies, which is compatible with graft acceptance. The components within SIS that elicit the immune response are not defined, but preliminary western blot analysis has shown immunoreactive bands of approximately 28 and 50 kD (Turner LA, Allman AJ, Raeder RH, Metzger DW, unpublished observation). Interestingly, our previous studies using an SIS-specific mAb have shown the absence of SIS epitopes in remodeled tissue. Instead, host-derived and site-specific ECM is present where the graft was implanted (25).
It is possible that T and/or B cells are necessary for SIS remodeling. However, C57BL/6 T cell KO mice remodel SIS grafts in an environment devoid of IL-4 and SIS-specific antibodies. In fact, antibodies do not appear to be necessary for SIS acceptance because C57BL/6 μMT KO mice, which have a targeted gene disruption in IgM ablating B cell expression, do not produce SIS-specific antibody responses nor do they reject SIS grafts. However, when implanted with SIS, these mice do show the presence of IL-4 and the absence of IFN-γ (Allman AJ, Badylak SF, and Metzger DW, unpublished observations). Furthermore, graft sites in wild-type mice were devoid of IgG1 and/or IgG2a antibody deposition, also indicating that antibodies are not necessary for SIS tissue remodeling.
It is possible that an immunomodulatory activity within ECM is at least partially responsible for the observed Th2 dominant response. A recent report has suggested that unknown factors in ECM can inhibit expression of IFN-γ and TNF-α from naïve T cells stimulated by anti-CD3 antibody or staphylococcal enterotoxin (26). In parallel, real time PCR data suggest that SIS suppresses IFN-γ expression based on the fact that mice implanted with SIS have significantly fewer copy numbers of IFN-γ mRNA compared to mice implanted with a syngeneic control graft.
The Th2 dominant immune response to ECM may play a pivotal role in the favorable clinical response to xenogeneic biomaterials. Bach et al. (20) found that a Th2 response was associated with acceptance of xenogeneic rat heart transplants in hamsters and a Th1-mediated response was associated with rejection. A Th2 dominant response may prevent cell-mediated inflammation of the ECM graft through the suppressive action of IL-10 on Th1 cells. The secretion of pro-inflammatory cytokines by macrophages and eosinophils is also inhibited by IL-10 (27,28). It is noteworthy that increased levels of IL-10 mRNA were seen in all groups of ECM-implanted mice in our study. Thus, Th2-mediated anti-inflammatory activity may be an important factor in promoting ECM remodeling by modulating postsurgical inflammation. Additionally, SIS graft sites were screened for IL-5, IL-6, and TGF-β mRNA. Very low levels of IL-6 were detected and IL-5 and TGF-β were not detected.
Although there is evidence for an association between graft acceptance and a Th2 immune response, the relative roles of Th1 and Th2 cells in graft rejection remain controversial. For example, rejection of cardiac allografts has been reported to be exacerbated in mice treated with antibody to IL-12 to inhibit Th1 differentiation (29). In addition, humans receiving liver allografts have elevated levels of IL-10 and IL-4 locally at the graft rejection site but systemic levels of Th1 cytokines such as IL-2 and IFN-γ are elevated (30). Furthermore, there is evidence that graft rejection can still occur in mice with genetic defects in the IFN-γ gene (31). Experiments in progress are designed to investigate whether Th1 activation and/or Th2 inactivation leads to ECM rejection.
In summary, the mouse immune response to porcine ECM is restricted to the Th2 pathway both locally and systemically. An understanding of this immune response and of the potential immunomodulatory activity of the ECM itself may prove useful in future clinical applications of xenotransplantation in humans.
1. Auchincloss H Jr., Sachs DM. Xenogeneic transplantation. Annu Rev Immunol 1998; 16: 433.
2. Bach FH, Soares M, Lin Y, Ferran C. Barriers to xenotransplantation. Transplant Proc 1999; 31: 1819.
3. Sandrin MS, McKenzie IFC. Recent advances in xenotransplantation. Curr Opin Immunol 1999; 11: 527.
4. Good AH, Cooper DKC, Malcolm AJ, et al. Identification of carbohydrate structure that binds human antiporcine antibodies: Implications for discordant xenografting in humans. Transplant Proc 1992; 24: 559.
5. Vaughn HA, Loveland BE, Sandrin MS. Galα(1, 3)Gal is the major xenoepitope expressed on pig endothelial cells recognized by naturally occurring cytotoxic human antibodies. Transplantation 1994; 58: 879.
6. Soares MP, Lin Y, Sato K, Stuhlmeier KM, Bach FH. Accommodation. Immunol Today 1999; 20: 434.
7. Hancock WW. Delayed xenograft rejection. World J Surg 1997; 21: 917.
8. Bach FH, Winkler H, Ferran C, Hancock WW, Robson SC. Delayed xenograft rejection. Immunol Today 1996; 17: 379.
9. Roeder R, Wolfe J, Lianakis N, Hinson T, Geddes LA, Obermiller J. Compliance, elastic modulus, and burst pressure of small-intestine submucosa (SIS), small-diameter vascular grafts. J Biomed Mater Res 1999; 47: 65.
10. Clarke KM, Lantz GC, Salisbury SK, Badylak SF, Hiles MC, Voytik SL. Intestine submucosa and polypropylene mesh for abdominal wall repair in dogs. J Surg Res 1996; 60: 107.
11. Shalhav AL, Elbahnasy AM, Bercowsky E, et al. Laparoscopic replacement of urinary tract segments using biodegradable materials in a large-animal model. J Endourol 1999; 13: 241.
12. Aiken SW, Badylak SF, Toombs JP, et al. Small intestinal submucosa as an intra-articular ligamentous repair material: A pilot study in dogs. Vet Comp Orthopedics Traumatology 1994; 7: 124.
13. Hodde JP, Badylak SF, Shelbourne KD. The effect of range of motion on remodeling of small intestinal submucosa (SIS) when used as an achilles tendon repair in the rabbit. Tissue Eng 1997; 3: 27.
14. Zhai Y, Ghobrial RM, Busuttil RW, Kupiec-Weglinski JW. Th1 and Th2 cytokines in organ transplantation: paradigm lost? Crit Rev Immunol 1999; 19: 155.
15. Strom TB, Roy-Chaudhury P, Manfro R. The Th1/Th2 paradigm and the allograft response. Curr Opin Immunol 1996; 8: 688.
16. Abbas AK, Murphy KM, Sher A. Functional diversity of helper T lymphocytes. Nature 1996; 383: 787.
17. Matsumiya G, Shirakura R, Miyagawa S, Izutani H, Nakata S, Matsuda H. Assessment of T-cell subsets involved in antibody production and cell-mediated cytotoxicity in rat-to-mouse cardiac xenotransplantation. Transplant Proc 1994; 26: 1214.
18. Chen N, Gao Q, Field EH. Prevention of Th1 responses is critical for tolerance. Transplantation 1996; 61: 1076.
19. Piccotti JR, Chan SY, VanBuskirk AM, Eichwald EJ, Bishop DK. Are Th2 helper T lymphocytes beneficial, deleterious, or irrelevant in promoting allograft survival? Transplantation 1997; 63: 619.
20. Bach FH, Ferran C, Hechenleitner P, et al. Accommodation of vascularized xenografts: expression of “protective genes” by donor epithelial cells in a host Th2 cytokine environment. Nature Med 1997; 3: 196.
21. Chen N, Filed EH. Enhanced type 2 and diminished type 1 cytokines in neonatal tolerance. Transplantation 1995; 59: 933.
22. Voytik-Harbin SL, Brightman AO, Kraine M, Waisner B, Badylak SF. Identification of extractable growth factors from small intestinal submucosa. J Cell Biol 1997; 67: 478.
23. Arulanandam BP, Van Cleave VH, Metzger DW. IL-12 is a potent neonatal vaccine adjuvant. Eur J Immunol 1999; 29: 256.
24. Overbergh L, Valckx D, Waer M, Mathieu C. Quantification of murine cytokine mRNAs using real time quantitative reverse transcriptase PCR. Cytokine 1999; 11: 305.
25. Badylak SF, Kropp B, McPherson T, Liang H, Snyder PW. Small intestinal submucosa: a rapidly resorbed bioscaffold for augmentation cystoplasty in a dog model. Tissue Eng 1998; 4: 379.
26. Beilfuss BA, van Seventer GA. SIS inhibits generation of human T helper-1 cells by dendritic cells. Orlando, FL: Second SIS Symposium, 1998: 35.
27. Livesay SA, Herndon DN, Hollyoak MA, Atkinson YH, Nah A. Transplanted acellular allograft dermal matrix. Transplantation 1995; 60: 1.
28. Grunig G, Corry DB, Leach MW, Seymour BWP, Kurup VP, Rennick DM. Interleukin-10 is a natural suppressor of cytokine production and inflammation in a murine model of allergic bronchopulmonary aspergillosis. J Exp Med 1997; 185: 1089.
29. Piccotti JR, Chan SY, Goodman RF, Magram J, Eichwald EJ, Bishop DK. IL-12 antagonism induces T helper 2 responses, yet exacerbates cardiac allograft rejection. Evidence against a dominant protective role for T helper 2 cytokines in alloimmunity. J Immunol 1996; 157: 1951.
30. Cosenza CA, Shirwan H, Cramer DV, Sher L, Podesta L, Makowka L. Intragraft cytokine gene expression in human liver allografts. Liver Transpl Surg 1995; 1: 16.
31. Steiger JU, Nickerson PW, Hermle M, Thiel G, Heim MH. Interferon-gamma receptor signaling is not required in the effector phase of the alloimmune response. Transplantation 1998; 65: 1649.