Immune evasion by neocartilage-derived chondrocytes: Implications for biologic repair of joint articular cartilage

Stem Cell Research (2010) 4, 57–68

available at www.sciencedirect.com

www.elsevier.com/locate/scr

REGULAR ARTICLE

Immune evasion by neocartilage-derived chondrocytes: Implications for biologic repair of joint articular cartilage H.D. Adkisson a,⁎, C. Milliman a , X. Zhang b , K. Mauch b , R.T. Maziarz b , P.R. Streeter b a

ISTO Technologies, Inc., 1155 Olivette Executive Parkway, St. Louis, MO 63132, USA Oregon Stem Cell Center, OHSU Knight Cancer Institute, Division of Hematology & Medical Oncology, 3181 SW Sam Jackson Park Road, Oregon Health & Science University, Portland, OR 97239, USA

b

Received 8 April 2009; received in revised form 28 August 2009; accepted 18 September 2009 Abstract Degeneration of joint articular cartilage is a leading cause of disability worldwide, and is due in large part to the fact that adult articular cartilage is unable to undergo effective intrinsic repair. To overcome this barrier, we have developed a tissue engineering strategy which harnesses the superior anabolic activity of juvenile chondrocytes to produce a scaffold-independent, living neocartilage graft. Preclinical studies demonstrate that bioengineered neocartilage survives allogeneic and xenogeneic transplantation, suggesting the utility of universal donor-derived neocartilage for joint repair. However, the mechanism underlying neocartilage transplant tolerance remains poorly understood. We show here that neocartilage-derived chondrocytes are unable to stimulate allogeneic T cells in vitro, and they do not constitutively express cell surface molecules required for induction of T cell immune responses, including major histocompatibility complex (MHC) Class II antigens and costimulatory molecules B7-1 and B7-2. Additionally, chondrocytes suppress, in a contact-dependent manner, the proliferation of activated T cells, with suppression associated with chondrocyte expression of multiple negative regulators of immune responses, including B7 family members (B7-H1, B7-DC, B7-H2, B7-H3, and B7-H4), chondromodulin-I and indoleamine 2,3-dioxygenase. Thus, the survival of transplanted bioengineered neocartilage may depend on both passive and active mechanisms of immune evasion. © 2009 Elsevier B.V. All rights reserved.

Introduction Adult articular cartilage has a limited regenerative capacity. As a consequence, adults whose cartilage has been damaged by arthritis, trauma, or other insult often suffer from chronic pain and disability. Transplantation of laboratory-engineered cartilage, known as neocartilage, may be a therapeutic option for these individuals (Feder et al., 2004; Hunziker, 2002). Neocartilage is generated in vitro using ⁎ Corresponding author. Fax: +1 314 995 6025. E-mail address: [email protected] (H.D. Adkisson).

chondrocytes, the cellular components of articular cartilage, obtained from juvenile cadaveric donors (Adkisson et al., 2001). Cells from a single donor can be scaled in vitro to meet the transplant needs of multiple recipients. Previous studies have demonstrated that juvenile human chondrocytes are up to 100-fold more efficient than adult chondrocytes in producing bioengineered cartilage (Feder et al., 2004; Adkisson et al., 2001). With most tissues, transplantation across allogeneic barriers carries a significant risk of immune rejection, necessitating tissue matching and/or the use of immunosuppressive agents. Immune rejection of transplanted tissue

1873-5061/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.scr.2009.09.004

58 occurs when the recipient's T lymphocytes recognize grafted tissue as “nonself” (Bretscher and Cohn, 1970). T lymphocyte priming, a requisite step in the rejection process, requires T cell receptor (TCR) ligation and engagement of CD28, the receptor for the costimulatory molecules B7-1 (CD80) and B72 (CD86) (Bretscher , 1999). These stimulatory signals can be delivered to the graft recipient's T lymphocytes by direct contact with cell surface molecules on transplanted allogeneic cells or by presentation of allogeneic antigens by the recipient's antigen presenting cells (APC). Articular cartilage is viewed as an immune privileged tissue. Cartilage is rarely a target of autoimmune disease (Balanescu et al., 2002; Fox, 1997), and transplanted cartilage from unrelated donors does not elicit immune rejection responses (Chesterman and Smith, 1968; Elves and Zervas, 1974; Lance and Fisher, 1970). Further, fresh shell osteochondral allografts have been used successfully in the clinic for more than three decades without immunosuppressive agents to repair large osteochondral defects in young active patients (Maury et al., 2007; Williams et al., 2007). Cartilage structure and cellular composition may contribute to the absence of immune rejection responses and subsequent allograft survival. Structural features of cartilage, including its avascular nature and the extensive extracellular matrix that surrounds chondrocytes, may limit leukocyte access into this tissue (Elves and Zervas, 1974; Bacsich and Wyburn, 1947; Heyner, 1969). At the cellular level, cartilage is devoid of APC such as histiocytes, and chondrocytes may not effectively stimulate antigraft immune responses. Chondrocyte immune privilege is further supported by the survival and subsequent formation of repair cartilage following implantation of allogeneic chondrocytes delivered in the presence or absence of a carrier matrix in higher order animals, including chickens, rabbits, sheep, and horses (Chesterman and Smith, 1968, Hendrickson et al., 1994; Lu et al., 2005; Rahfoth et al., 1998; Robinson et al., 1990). Preclinical evaluations of bioengineered neocartilage transplanted into surgical defects created in the knee joint have demonstrated that this tissue, like joint-derived articular cartilage, is not rejected when transplanted into genetically unrelated recipients (Feder et al., 2004; Lu et al., 2005). This report elucidates the immunobiological properties of chondrocytes recovered from neocartilage. We demonstrate that human juvenile chondrocytes derived from bioengineered neocartilage lack cell surface markers required for induction of immune responses; that they do not induce alloantigenspecific proliferative immune responses in vitro; and that they actively suppress, in a contact-dependent manner, the proliferation of activated T cells. These findings support the concept that both passive and active mechanisms of immune evasion facilitate neocartilage allograft survival.

H.D. Adkisson et al.

Figure 1 Gross macroscopic and histological appearance of cartilage repair in the sheep knee following allogeneic neocartilage implantation. (a) A 6 mm circular defect created in the central aspect of the medial femoral condyle showing removal of native articular cartilage to the subchondral bone surface without bleeding (0.75 mm). (b) At 8 weeks after implantation, the defect was filled with repair tissue and the interface between repaired tissue and surrounding host articular cartilage was unremarkable (white dashed circle). (c) Safranin-O/fast green stained midsagittal section from an ovine allograft showing integrative repair at both the lateral and the subchondral interface. The cell density of implanted neocartilage (NC) is greater than that of adjacent host cartilage (HC), illustrating intense metachromatic staining for sulfated glycosaminoglycans. Arrow designates the graft/host junction. Original magnification× 100.

Results Chondrocyte immunogenicity Preclinical studies have demonstrated the survival and integration of bioengineered neocartilage into experimental

Figure 2 Neocartilage-derived juvenile chondrocytes do not stimulate allogeneic T cell proliferation. (a) After 7 days of culture, bone marrow-derived dendritic cells (BMDC) generated in vitro from the same donor tissue used to produce neocartilage promoted potent dosedependent proliferation of allogeneic responder PBL (thymidine incorporation; counts per minute). In contrast, neocartilage-derived juvenile chondrocytes did not promote PBL proliferation at any cell concentration evaluated. Responder PBL alone also exhibited baseline proliferation. (b) FACS analysis comparing expression of cell surface markers characteristic of antigen presenting cells on BMDC and neocartilage-derived chondrocytes obtained from 6-week-old male and 8-year-old female donors. BMDC showed positive staining for MHC Class II (HLA-DR), CD123, CD40, B7-1(CD80), B7-2(CD86), and the adhesion molecule CD11c. Resting juvenile chondrocytes were negative for these markers. Specific staining illustrated by filled histogram; isotype control is shown with open histogram. (c) Neocartilage-derived juvenile chondrocytes stimulated with 200 Units/mL of recombinant human IFN-gamma show increased staining for MHC Class I (HLA-ABC) and II (HLA-DR). Specific staining is shown with black histogram; isotype control is illustrated with gray histogram.

Immune evasion by neocartilage chondroctyes defects created in healthy joint surfaces (Feder et al., 2004; Lu et al., 2005). These studies, conducted in xenogeneic and allogeneic transplant models, have revealed no gross macroscopic or histological evidence of immune rejection. Ovine neocartilage has been transplanted into ovine reci-

59 pients, and human neocartilage has been transplanted into leporine, ovine, and caprine recipients. A representative outcome following allogeneic transplantation of neocartilage into a young-adult sheep is presented in Fig. 1. A full thickness defect in the most weight bearing region of the

60

H.D. Adkisson et al.

knee is shown prior to transplantation (Fig. 1a) and at 8 weeks following transplantation (Figs. 1b and c). Gross macroscopic and histological examination revealed repair of the defect, with integration of donor neocartilage and no observable damage to the opposing joint surface. In light of these findings, the immunogenicity of human neocartilage-derived juvenile chondrocytes was investigated using nondividing chondrocytes as stimulator cells and human peripheral blood leukocytes (PBL) as responder cells in mixed lymphocyte–chondrocyte reaction assays (MLCR). For this series of assays, varying numbers of irradiated (3000 rad) chondrocytes (102–104) were mixed with 2 × 105 allogeneic PBL. Chondrocytes used for this aspect of the study were derived from enzyme-dissociated neocartilage (Adkisson et al., 2001). Lymphocyte proliferation was quantitated by [3H]thymidine incorporation after 5–6 days of coculture. Neocartilage-derived chondrocytes from 13 of 14 juvenile donors did not stimulate allogeneic PBL proliferation (Fig. 2a and Table 1). By contrast, one chondrocyte preparation elicited a weak response above baseline (Table 1; chondrocytes from a 1.2 year old female donor yielded a response from one of two donor PBL specimens that was 1.9 times higher than the negative control). As a positive control, the stimulatory potential of irradiated (3000 rad) bone marrow dendritic cells (BMDC) generated in vitro from each of the chondrocyte donors was evaluated. These potent APC readily stimulated the proliferation of allogeneic PBL (Fig. 2a and Table 1). Thus, in contrast to the potent stimulatory capacity of BMDC, allogeneic juvenile chondrocytes derived from the same donor tissues did not stimulate T lymphocyte proliferation in vitro. The inability of neocartilage-derived juvenile chondrocytes to stimulate T cell proliferation in vitro may relate to

Table 1 Comparison of chondrocyte and bone marrow dendritic cell allostimulatory activity in the mixed lymphocyte chondrocyte reaction Donor age

Donor sex

Chondrocyte MLCR a

BMDC MLCR a

5 weeks 6 weeks 2 months 3 months 3 months 3 months 3 months 4 months 1.2 years 2.6 years 2.75 years 5 years 6 years 8 years

Female Male Male Male Female Male Male Female Female Male Male Female Male Male

– – – – – – – – +/– b – – – – –

++ NA ++ ++ ++ + +++ +++ ++ +++ +++ ++ NA NA

a – = b1.5-fold above background; + = 1.5-fold to b10-fold stimulation at 104 cells; ++ = 10-fold to b40-fold stimulation at 104 cells; +++ = N 40-fold stimulation at 104 cells. b Chondrocytes from this donor elicited a weak proliferative response from one of two PBL donors (1.9 fold above background). The response of PBL from the second PBL donor was b1.5-fold above background.

Figure 3 Analyses of major histocompatibility complex and costimulatory molecule expression in neocartilage-derived juvenile chondrocytes and adult articular chondrocytes. (a) RT-PCR analyses demonstrating that B7-1 (CD80) and B7-2 (CD86) mRNA were undetectable in juvenile chondrocytes. By contrast, B7-1 (CD80) and B7-2 (CD86) mRNA was detected in adult articular chondrocytes. Glyceraldehyde-3-phosphate dehydrogenase (GAP) was included to demonstrate equal loading of RNA. Note that the 44-year-old adult sample was underloaded (40%) due to limited sample availability. (b) FACS analysis comparing expression of MHC Class I and Class II molecules on adult resting and recombinant human IFN-gamma-activated chondrocytes. Chondrocyte stimulated with 200 Units/mL of recombinant human IFN-gamma show increased staining for MHC Class I (HLA-ABC) and II (HLA-DR). Specific staining is shown with black histogram; isotype control is illustrated with gray histogram.

the absence of molecules required for immune response induction (Bretscher and Cohn, 1970; Bretscher, 1999). To test this concept, we evaluated cell surface marker expression on neocartilage-derived juvenile chondrocytes. Flow cytometric analysis revealed that resting juvenile chondrocytes express neither MHC Class II (HLA-DR), which presents peptide epitopes to CD4+ T cells via the TCR, nor the costimulatory molecules B7-1 (CD80) and B7-2 (CD86) (Fig. 2b) which serve as ligands for CD28 binding. Further, serial passage of these cells (from P0 to P2, approx 6-8 population doublings) did not alter this expression profile (not shown). In contrast, immune stimulatory BMDC were shown by flow cytometric analysis to express MHC Class II, B7-1 (CD80), and B7-2 (CD86), CD11c, CD40, CD123 and the leukocyte lineage (LIN) markers CD3, CD14, CD16, CD19, CD20, or CD56 (Fig. 2b). Thus, neocartilage-derived juvenile chondrocytes

Immune evasion by neocartilage chondroctyes appear to lack cell surface expression of molecules required for T cell activation. Recent studies of the mechanisms associated with allograft rejection suggest that rejection may occur as a consequence of trauma associated with the transplant procedure. Those studies implicate inflammatory cytokines, such as IFN-gamma, in the activation of APC and subsequent immune rejection (Mellor and Munn, 2004). To model this process in vitro, neocartilage-derived juvenile chondrocytes were treated with IFN-gamma, which resulted in the upregulated cell surface expression of MHC Class I (HLAABC) molecules, and a small increase in MHC Class II (HLA-DR).

61 However, no increase in expression of the costimulatory molecules B7-1 (CD80) and B7-2 (CD86) was observed by flow cytometry (Fig. 2c). Further, IFN-gamma-activated chondrocytes, like their unstimulated counterparts, failed to elicit proliferation of allogeneic PBL in vitro (not shown). In summary, exposure of neocartilage-derived juvenile chondrocytes to IFN-gamma promoted expression of MHC Class I and II molecules, but this change in surface antigen expression did not yield chondrocytes with the ability to simulate alloantigen-specific T cell proliferation in the MLCR. To further address the question of chondrocyte immunogenicity, we evaluated the phenotype and function of adult

Figure 4 Articular chondrocytes are immunosuppressive and display contact-dependent immunosuppressive activity in vitro. (a) Neocartilage-derived juvenile chondrocytes inhibited the proliferation of alloantigen-activated PBL, when irradiated (1000 rad) allogeneic PBL or irradiated (3000 rad) allogeneic BMDC were used as stimulator cells. (b) Juvenile (neocartilage-derived) and adult articular cartilage actively inhibited in a dose-dependent manner the proliferation of polyclonally activated CD4+ T lymphocytes. Polyclonal activation was achieved by exposure of CD4+ T cells to antibodies directed against CD3 and CD28. (c) Transwell separation of chondrocytes and polyclonally activated PBL demonstrated that chondrocyte-mediated immunosuppressive activity requires cell– cell contact with T lymphocytes. Data are representative of three separate experiments using neocartilage-derived chondrocytes.

62 chondrocytes. In contrast to juvenile neocartilage-derived chondrocytes, adult chondrocytes were found by RT-PCR to express B7-2 and to a lesser extent B7-1 mRNA (Fig. 3a). Adult chondrocytes also express MHC Class I and low levels of MHC Class II; and when exposed to IFN-gamma these cells were found to up-regulate cell surface expression of MHC Class I (HLA-ABC) and MHC Class II (HLA-DR; Fig. 3b). Further, cytokine-activated adult chondrocytes, like their juvenile counterparts, did not induce allogeneic T cell proliferation in vitro (not shown). Thus, despite expression of cell surface molecules involved in T cell activation, adult chondrocytes failed to stimulate T cell proliferation in the MLCR. These findings suggested that chondrocytes actively inhibit or suppress lymphocyte proliferation in vitro.

Chondrocyte-mediated immune suppression The immune suppressive potential of chondrocytes was evaluated using multiple strategies. Initial studies assessed suppression of proliferative responses to alloantigen. For these experiments, responder PBL were stimulated using either irradiated (1000 rad) allogeneic PBL or irradiated (3000 rad) allogeneic BMDC (Fig. 4a). Stimulation with either of these sources of alloantigen resulted in proliferation of responder PBL. The addition of irradiated (3000 rad) neocartilage-derived juvenile chondrocytes to these cultures

H.D. Adkisson et al. resulted in the inhibition of T cell proliferation induced by either source of alloantigen (Fig. 4a). The studies described above revealed that the addition of chondrocytes to mixed lymphocyte cultures inhibited T cell proliferation. To determine if chondrocytes acted directly on responding T cells or indirectly via accessory cells such as B cells or monocytes, neocartilage-derived juvenile chondrocytes were coincubated with purified CD4+ T cells that were activated by cross-linking both the TCR/CD3 complex with an anti-CD3 specific antibody and the B7-1/B7-2 receptor (CD28) with an antibody directed against CD28. This combination of antibodies delivers a strong activation signal to CD4+ T cells, as is evidenced by robust T cell proliferation (Fig. 4b). Addition of irradiated neocartilage-derived juvenile chondrocytes to these cultures resulted in a dose-dependent decrease in T cell proliferation (Fig. 4b). Similarly, adult chondrocytes potently inhibited proliferation of activated T cells (Fig. 4b). Thus, chondrocytes suppress CD4+ T cell proliferative responses by direct action on the responding T cells. The immune suppressive property of neocartilage-derived juvenile chondrocytes was further assessed by physically separating these cells from activated CD4+ T cells using 0.2μm pore size transwell inserts. These inserts allow passage of soluble mediators between cells in the two chambers, but prevent direct cell–cell contact. Under these conditions, neocartilage-derived juvenile chondrocytes failed to

Figure 5 RT-PCR analysis of gene expression for immune inhibitory members of the B7 family of costimulatory molecules in chondrocytes from neocartilage-derived juvenile and adult articular cartilage. Constitutive expression of B7-DC, B7-H2, B7-H3, and B7-H4 mRNA, reported negative regulators of T cell function, was detected in all chondrocyte populations surveyed, whereas B7-H1 mRNA, a negative costimulatory molecule with inducible kinetics (below) showed variable constitutive expression in juvenile and adult chondrocytes. Chondromodulin-I mRNA encoding the cartilage-specific cell surface protein reported to display immunosuppressive activity to polyclonally activated T cells in vitro (Setoguchi et al., 2004) was readily identified in juvenile and adult chondrocytes. Note, however, that ChM-I gene expression was undetectable in the chondrosarcoma cell line used as a positive control for detection of each of the B7 costimulatory family members.

Immune evasion by neocartilage chondroctyes suppress the proliferation of CD4+ T cells activated with antibodies directed against CD3 and CD28, suggesting that chondrocyte-to-T cell contact is required for inhibition of CD4+ T cell proliferation (Fig. 4c).

Candidate mechanisms of chondrocyte-mediated immune suppression Expression of immune inhibitory B7 family members In addition to providing positive stimulatory signals, the B7 family of cell surface molecules contains several members that transmit inhibitory signals to T cells (Greenwald et al., 2005). B7-H1 and B7-DC block the early stages of T cell activation via the PD-1/PD-2 receptor (Freeman et al., 2000; Latchman et al., 2004); B7-H4 inhibits activated T cells through an unknown receptor (Sica et al., 2003); and B7-H3 also negatively modulates T cell responses (Suh et al., 2003). We analyzed by RT-PCR neocartilage-derived juvenile chondrocytes and native adult articular cartilage chondrocytes for expression of B7 family members that have been implicated as negative regulators of T cell function. B7-DC, B7-H2, B7-H3, and B7-H4 mRNA were constitutively expressed at high levels in most specimens (Fig. 5). By contrast, constitutive expression of B7-H1 mRNA was low in both juvenile (neocartilage) and adult chondrocytes, requiring greater than 40 cycles to identify message in resting cells. The human chondrosarcoma cell line (CH-1), which expresses mRNA for each of the B7 family members, was used as a positive control for costimulatory molecule expression (Fig. 5). Thus, these data revealed that neocartilage-derived juvenile chondrocytes express message for multiple inhibitory B7 family members. Expression of chondromodulin-I, a recently identified inhibitor of T cell proliferation Chondromodulin-I (ChM-I) is a 25-kDa cysteine-rich transmembrane protein originally purified from bovine epiphyseal cartilage that promotes chondrocyte growth (Hiraki et al., 1999; Inoue et al., 1997), while inhibiting endothelial cell growth (Hiraki et al., 1997). Although originally described as cartilage-specific, Northern blot analysis of murine tissue revealed expression of ChM-I in thymus, suggesting that the protein may impact T cell development or function. Indeed, recombinant human ChM-I has been found to be immunosuppressive both in vitro and in an animal model of antigeninduced arthritis (Setoguchi et al., 2004). RT-PCR analysis demonstrated high-level expression of ChM-I mRNA by neocartilage-derived juvenile chondrocytes and by chondrocytes from adult articular cartilage (Fig. 5). Thus, ChM-I expression by chondrocytes may directly disrupt T cell function. Expression of indoleamine 2,3-dioxygenase, a mediator of immune evasion Indoleamine 2,3-dioxygenase (IDO) is a heme-dependent oxygenase that is thought to play a role in the induction of immune tolerance (Mellor and Munn, 2004). A growing body of evidence suggests that IDO activity is modulated in vitro and in vivo by a variety of cytokines (Babcock and Carlin, 2000; MacKenzie et al., 1999) and by receptor-mediated signaling through B7 family members (Munn et al., 2004). In light of these data, we investigated whether neocartilagederived juvenile chondrocytes express IDO and whether IDO

63 expression plays a role in chondrocyte-mediated inhibition of T cell proliferative responses. We showed that IDO mRNA and protein synthesis by cultured neocartilage-derived juvenile chondrocytes is upregulated following exogenous treatment with proinflammatory cytokines, including interleukin-1 (IL-1), tumor necrosis factor (TNF) alpha, stem cell differentiation factor (SDF)-1 alpha, and interferon (IFN)gamma (Figs. 6a and b). Increased quantities of these cytokines are reported to be present in joint synovial fluid following trauma or as a consequence of degenerative autoimmune arthritis (Lettesjo et al., 1998; Schlaak et al., 1996). Importantly, inhibition of IDO with the selective inhibitor 1-methyl tryptophan partially inhibited neocartilage-derived juvenile chondrocyte immunosuppressive

Figure 6 Articular chondrocytes express indoleamine 2,3dioxygenase (IDO) mRNA and protein on cytokine activation. (a) RT-PCR analysis showing inducible expression of IDO mRNA after a 72 h treatment of neocartilage-derived juvenile chondrocytes with 200 Units/mL IFN-gamma. GAP mRNA served as a loading control. (b) Protein expression was demonstrated by Western blotting of lysates prepared 72 h after activation with the indicated proinflammatory cytokines. The PVDF membrane was probed for IDO protein using a monoclonal antibody (clone 10.1) purchased from Upstate (Lake Placid, NY), according to manufacturer's instructions. (c) Pharmacologic inhibition of IDO activity using 1-methyl tryptophan (Sigma-Aldrich) partially inhibited the immunosuppressive activity of chondrocytes that were added to polyclonally activated T lymphocytes at the intermediate concentration (3 × 104) but not at the highest concentration of chondrocytes analyzed.

64 activity, suggesting that this enzyme may also participate in chondrocyte-mediated suppression of T cell responses.

Discussion Articular cartilage is considered to be an immune privileged tissue, much like the anterior chamber of the eye. Allogeneic transplantation either of fresh articular cartilage or of chondrocytes harvested therefrom is reported to be well tolerated and is generally accepted to not elicit an immune response resulting in graft rejection, even in the absence of immune suppressive therapy (Chesterman and Smith, 1968; Elves and Zervas, 1974; Lance and Fisher, 1970; Maury et al., 2007; Williams et al., 2007; Hendrickson et al., 1994; Rahfoth et al., 1998; Robinson et al., 1990; Wakitani et al., 1989). Successful transplantation of bioengineered neocartilage into allogeneic and xenogeneic animal models suggests that this bioengineered tissue is also immune privileged ((Feder et al., 2004; Lu et al., 2005); and data shown here). It has been suggested that failure of the host immune system to reject transplanted articular cartilage is due, in part, to the unique architecture of this tissue. Articular cartilage is avascular, and as such, entry of host immune cells is restricted. Further, cartilage-associated chondrocytes are buried in an extensive muccopolysaccharide-containing extracellular matrix, a barrier which limits potential interaction between cells of the immune system and chondroctyes (Elves and Zervas, 1974; Bacsich and Wyburn, 1947; Heyner, 1969). Although these physical barriers may play a role in maintaining cartilage immune privilege, the data described here support the concept that chondrocytes play a direct role in preventing immune responses that may lead to the destruction of transplanted neocartilage allografts. When allogeneic T cells were cocultured with neocartilage-derived juvenile chondrocytes, the chondrocytes failed to stimulate alloantigen-specific T cell proliferation (Fig. 2). To evoke a proliferative response from CD4+ T cells, allogeneic stimulator cells must express unrelated MHC Class II molecules as well as the costimulatory molecules B7-1 (CD80) and/or B7-2 (CD86) (reviewed in (Greenwald et al., 2005)). Under normal conditions, juvenile chondrocytes, like most nonhematopoietic cells, do not express these proteins (Fig. 2c). Exposure to IFN-gamma, which can occur as a consequence of cellular activation in the transplant setting, induces surface expression of MHC Class I and II in chondrocytes, but not B7-1 (CD80) and B7-2 (CD86) (Fig. 2c). Thus, the nonimmunogenic properties of neocartilage-derived juvenile chondrocytes may be attributed to their failure to provide adequate T cell costimulation. Surprisingly, adult chondrocytes, which constitutively express B7-1 (CD80) and B7-2 (CD86) mRNAs, and show up-regulated expression of MHC Class I and II antigen on IFN-gamma treatment (Fig. 3b), also were unable to elicit T cell proliferation in the MLCR (not shown). This intriguing observation suggested that both juvenile and adult chondrocytes actively suppress immune responses. The immune suppressive action of neocartilage-derived juvenile chondrocytes was confirmed using alloantigen-activated PBL (Fig 4a). The inhibition of polyclonally activated CD4+ T cells by both juvenile and adult chondrocytes revealed that chondrocyte-mediated suppression of CD4+ T cell proliferation does not require accessory cells (Fig. 4b).

H.D. Adkisson et al. Further, transwell studies revealed that direct T cell contact with chondrocytes was required to attenuate T cell proliferation (Fig. 4c). That chondrocytes are capable of inhibiting T cell receptor-mediated T cell proliferation via direct cell–cell contact was an unexpected finding and suggested that these cells actively function to maintain balance between tolerance and autoimmunity within the joint. It is noteworthy that a human stem cell population, mesenchymal stem cells, is reported to actively suppress immune responses. Although no consensus mechanism has been described for immune suppression by these cells, several reports indicate that suppression is cell–cell contact independent (Tse et al., 2003; Klyushnenkova et al., 2005; Zhao et al., 2008). Thus, the immune inhibitory properties of chondrocytes and mesenchymal stem cells may function via distinct mechanisms. We propose that chondrocytes may inhibit T cell proliferation via three distinct mechanisms. First, T cell proliferation may be inhibited through inhibitory cell surface B7 costimulatory molecules. It was observed that mRNA for immune inhibitory costimulatory molecules B7-H1, B7-DC, B7-H2, B7H3, and B7-H4 are present in unstimulated and/or IFNgamma-activated chondrocytes (Fig. 5). B7-DC is one of two ligands that bind to the PD-1 receptor on T cells (reviewed in (Greenwald et al., 2005)). Constitutive expression of B7-DC by chondrocytes and the chondrosarcoma (CH-1) is intriguing, as B7-DC expression is predominantly restricted to macrophages and dendritic cells (Liang et al., 2003). Signaling through PD-1 inhibits T cell immune responses: B7-DC and the other PD-1 ligand, B7-H1, inhibit T cell proliferation in vitro (Freeman et al., 2000; Latchman et al., 2004; Carter et al., 2002; Dong et al., 2002) and disruption of the PD-1 pathway in vivo has been linked to both autoimmune disease (Nishimura et al., 1999) and increased severity of graft versus host disease (Blazar et al., 2003). B7-H3 was initially thought to be stimulatory because it promoted T cell proliferation in vitro (Chapoval et al., 2001); however, studies of B7-H3 knockout mice indicate that B7-H3 may inhibit the Th1 subclass of T cells, which have been implicated in transplant rejection (Suh et al., 2003). B7-H4 blocks proliferation and cytokine production by T cells in vitro and protects against autoimmune disease in vivo (Sica et al., 2003). Chondrocytes also express high levels of B7-H2, and although this molecule has not been associated with immune inhibition, it may aid in the acceptance of transplanted allografts or xenografts by promoting the development of Th2 rather than Th1 immune responses (Harada et al., 2003). Thus, chondrocytes express members of the B7 family that are immune suppressive and may promote allograft or xenograft survival. Second, we demonstrated that neocartilage-derived juvenile and adult articular chondrocytes show high level expression of ChM-I mRNA. ChM-I has pronounced immunosuppressive activity in an animal model of disease (Setoguchi et al., 2004). This cysteine-rich transmembrane protein, previously identified in cartilage and thymus (Hiraki et al., 1997), has been shown to block in vitro proliferation of polyclonally activated T cells (Setoguchi et al., 2004). Finally, we showed that the ability of neocartilagederived juvenile chondrocytes to suppress immune responses in vitro is sensitive to an inhibitor of IDO (Fig. 6). It is generally accepted that IDO prevents rejection of the fetus during pregnancy (Munn et al., 1998), and that human dendritic cells transduced with the IDO gene are capable of

Immune evasion by neocartilage chondroctyes preventing in vitro T cell proliferation via a kynurenine metabolite that is cytotoxic to B and NK cells (Terness et al., 2002). This mechanism may also contribute to neocartilage implant survival. Our finding that chondrocytes derived from bioengineered juvenile neocartilage lack the ability to evoke an allogeneic response in vitro is in complete agreement with that of Jobanputra et al. (Jobanputra et al., 1992) who demonstrated that freshly dissociated chondrocytes derived from human adult articular cartilage were unable to stimulate the proliferation of allogeneic PBL in the MLCR. More important, we show that cultured juvenile and adult chondrocytes derived from human articular cartilage display unique immunosuppressive properties requiring cell–cell contact, and we identify three possible mechanisms underlying the suppressive action of neocartilage chondrocytes. While instances of chondrocyte immunogenicity have been reported for rodent species (Gertzbein and Lance, 1976; Hyc et al., 1997; Moskalewski and Kawiak, 1966), the apparent disparity with the current human data may be due to the presence of contaminating nonchondrogenic cells, including endothelial cells from the epiphysis or adipocytes from the fat pad, in the rodent cell preparations used for study. Clean dissection of articular cartilage from its underlying bony tissue components is technically difficult due to the extremely thin nature of this tissue in mice and rats. In conclusion, allogeneic and xenogeneic transplantation of bioengineered neocartilage derived from juvenile chondrocytes appears to be successful because the tissue is protected from the host immune system by both passive and active mechanisms. Cartilage-associated chondrocytes are separated physically from immune cells (Elves and Zervas, 1974; Bacsich and Wyburn, 1947; Heyner, 1969), and chondrocytes derived from juvenile neocartilage do not express cell surface molecules that promote T cell responses. Further, chondrocytes are immunosuppressive, and inhibitory B7 molecules, ChM-I and/or IDO, may mediate immune suppression. Due to their unique immunological properties and an inherent enhanced ability to synthesize cartilage matrix, juvenile articular chondrocytes derived from allogeneic donor tissue may have therapeutic applications in cartilage repair.

Methods Human subjects Articular cartilage and bone marrow were obtained from deceased human donors within 96 h of death. Tissue was procured by Mid-America Transplant Services (St. Louis, MO), Cell Dynamics, LLC (Smyrna, GA), and National Disease Research Interchange (Philadelphia, PA), and informed consent for use in medical research was obtained. Of 28 samples analyzed, 19 were obtained from juvenile donor tissue (birth to 14 years of age; mean 2.3 years) and 9 were from adult donor tissue (21 to 54 years of age; mean 35.9 years). Donors were screened for communicable disease agents, and only donors with negative serology reports were used. Communicable disease test results were typically reported within 48– 72 h, and tissues were stored in physiologically buffered solution at 4 °C until test results were received.

65

Neocartilage production and dissociation Juvenile articular chondrocytes were isolated and neocartilage disks grown to Days 45–60 of culture as described by Adkisson et al. (Adkisson et al., 2001). Briefly, cadaveric donor articular cartilage was cleaned of adventitious tissue by sharp dissection, and chondrocytes were enzymatically released by brief treatment with pronase, followed by overnight digestion with collagenase and hyaluronidase. Cells were washed, counted, and subsequently seeded into tissue culture plastic at a density of 1 × 106 cells per/mL in HL-1, a chemically defined medium manufactured by Cambrex Biosciences, containing ascorbate and glutamine. In the absence of three-dimensional scaffolds, juvenile (in contrast to adult chondrocytes) were previously demonstrated to aggregate and produce an hyaline extracellular matrix (ECM) resembling that of juvenile articular cartilage (Feder et al., 2004; Lu et al., 2005). These cultures retain their native phenotype by assembling an ECM showing positive staining for collagen types II, IX, and XI, with no measurable type I collagen (Feder et al., 2004; Lu et al., 2005). For immunophenotyping experiments, two neocartilage disks prepared from each individual donor were digested for 12 h in HL-1 medium (Cambrex BioScience, Rockland, ME) containing 60 Units/mL CLS4 collagenase (Worthington, Lakewood, NJ) and 50 Units/mL hyaluronidase (type VIII, Sigma, St. Louis, MO). Dissociated chondrocytes were washed twice in RPMI 1640 medium containing 2% FBS and resuspended in HL-1 medium. Cells were counted and diluted to 1×106/mL in RPMI 1640 media containing 10% heat-inactivated human AB serum and 10 mM HEPES. Cells were stored on ice prior to use. Adult chondrocytes harvested from cadaveric joints were maintained in monolayer culture using the same conditions described above for juvenile chondrocytes. Under such conditions, adult chondrocytes fail to produce an ECM-rich neocartilage because of their reduced synthetic capacity relative to juvenile chondrocytes (Adkisson et al., 2001). Therefore, complete dissociation of chondrocyte monolayers prepared with adult chondrocytes required only 4–6 h, using the conditions described above.

Generation of bone marrow dendritic cells Bone marrow mononuclear cells served as a source of progenitor cells that were expanded and differentiated into dendritic cells in vitro. Bone marrow dendritic cells (BMDC) served as positive controls for flow cytometric analyses and MLCR assays. The medullary space of the femur, tibia, and fibula from each donated knee was flushed with Ca2+/Mg2+-free PBS to collect bone marrow cells. Mononuclear cells were recovered from these cell suspensions using Histopaque 1077 (Sigma), and cells were washed three times in Ca2+/Mg2+-free PBS and cryopreserved. BMDC were generated using a method adapted from Caux and Dubois (Caux and Dubois, 2001). Briefly, cryopreserved bone marrow mononuclear cells were rapidly thawed at 37 °C and incubated at a concentration of 3×106 cells/5 mL of X-Vivo 15 serum-free medium (Cambrex, Walkersville, MD) containing Flt-3 ligand (100 ng/mL), TNF-alpha (10 ng/mL), GM-SCF (100 ng/mL), IL-3 (10 ng/mL), IL-7 (10 ng/mL), SCF (10 ng/ mL), 20 mM HEPES, 2 mM L-glutamine, and 5% human AB serum. All cytokines were purchased from Peprotech Inc. (Rocky Hill, NJ). On Day 7 of culture, these cultures were split 1:4 and maintained as described for an additional 72 h. On Day 10 of

66 culture, IL-4 (20 ng/mL) was added to promote maturation of immature dendritic cells during the final 72 h of culture.

Mixed lymphocyte–chondrocyte reaction assays Immunogenicity of chondrocytes was assessed using mixed lymphocyte–chondrocyte reaction (MLCR) assays. For these assays, chondrocytes (102 to 104/well) were evaluated for their ability to stimulate proliferation of human peripheral blood lymphocytes (PBL; 2 × 105/well) from unrelated normal donors. BMDC prepared from the same donor tissue used to generate neocartilage grafts in vitro served as a positive control. Cells were cultured in RPMI 1640 medium containing 10% FBS, 15 mM HEPES, 2 mM L-glutamine, 1 mM MEM sodium pyruvate, MEM nonessential amino acids, penicillin-streptomycin, and 5 × 10− 5 M ß–mercaptoethanol. Stimulator cells were gammairradiated (3000 rad) to prevent proliferation. In vitro proliferation of allogeneic PBL was quantitated on Days 5–6 of coculture after an 18-h pulse with tritiated thymidine (Amersham, 1 μCi/ mL, Piscataway, NJ). Cells were lysed in water and released DNA was bound to glass fiber filters using an automated cell harvester (Tomtec, Hamden, CT). The filters were dried and counted in a Wallac MicroBeta Scintillation Counter (Perkin Elmer, Boston, MA). Results are expressed as mean counts per minute (CPM) for replicates of three to six ± SEM.

Immune suppression assays The immune suppressive activity of chondrocytes was assessed using polyclonally activated CD4+ T lymphocytes. Briefly, human peripheral blood mononuclear cells (PBMC) were recovered using Histopaque 1077, and CD4+ T lymphocytes were purified from this mononuclear cell fraction using magnetic-activated cell separation (MACS) columns (Miltenyi Biotec, Auburn, CA). To assess the immunosuppressive activity of chondrocytes, naïve T cells (1×105/well) were activated in 96-well plates by crosslinking the T cell receptor complex (TCR/CD3) with 0.01 μg/mL of plate-bound anti-CD3 and the costimulator receptor (CD28) with 5 μg/mL anti-CD28 added to culture media. Irradiated chondrocytes were added to activated lymphocytes. In some experiments, activated lymphocytes were separated from chondrocytes using Anapore Transwells (Nunc, 0.2-μm pore size). Cells were cultured for 72 h, and tritiated thymidine was added to each well for the final 18 h of incubation.

H.D. Adkisson et al. assessed using reverse-transcription polymerase chain reaction (RT-PCR). Total RNA was extracted from disrupted neocartilage disks using Qbiogene Fast RNA Pro Green Kit (Qbiogene Inc., Carlsbad, CA) followed by RNeasy Mini Kit (Qiagen Inc., Valencia, CA) column purification with DNase digestion (Qiagen). A single tube RT-PCR with 0.5 μg of RNA was performed in a 25-μL reaction mixture using RNA-specific primers and the EZ rTth RNA PCR kit (Perkin Elmer/Applied Biosystems). All experiments were performed three times on individual donors. Primers were selected using DNA Star software to ensure specificity, and primers were purchased from Sigma Genosys: B7-1 (sense, 5′-GCCATCAACACAACAGTTTCCCAA-3′; antisense, 5′-CAGGGCGTACACTTTCCCTTCTCAA-3′, length 313 bp), B7-2 (sense, 5′CTCTCTGGTGCTGCTCCTCTGAA-3′; antisense, 5′-CTGTGGGCTTTTTGTGATGGATGATAC-3′, length 298 bp), B7-H1 (sense, 5′CTACCCCAAGGCCGAAGTCA-3′; antisense, 5′-CCCAGAATTACCAAGTGAGTCC-3′, length 258 bp), B7-DC (sense, 5′-ACCCTGGAATGCAACTTTGACACT-3′; antisense, 5′-ACTTGGACTTGAGGTATGTGGAACGA-3′, length 167 bp), B7-H2 (sense, 5′-GACTTCTCCCTGCGCTTGTTCA-3′; antisense, 5′-GCAGGCTGTTGTCCGTCTTATTG-3′, length 256 bp), B7-H3 (sense, 5′-CGTGGGGCTGTCTGTCTGTCTCATT-3′; antisense, 5′-TCAGGCTATTTCTTGTCCATCATCTTCT-3′, length 196 bp), and B7-H4 (sense, 5′-CACTCATCATTGGCTTTGGTATTTCAG-3′; antisense, 5′-CGACAGCTCATCTTTGCCTTCTTTG-3′, length 203 bp). IFN-gamma-induced human chondrosarcoma RNA (Ch-1 cells were a gift from Linda Sandell, Washington University, St. Louis, MO) was used as a positive control for B7 family member gene expression.

RT-PCR characterization of chondromodulin-I and indoleamine 2,3-dioxygenase RT-PCR analysis of ChM-I and IDO gene expression was carried out using the following probes: ChM-I (sense, 5′AGAAAATTCTCTTATCTGGGTGGCTGTA-3′; antisense, 5′CTCTGGATTTCTTTTGGATAGGTTGGT-3′, length 135 bp), IDO (sense, 5′-AGGCAACCCCCAGCTATCAGAC-3′; antisense, 5′-TCAGGGAGACCAGAGCTTTCACAC-3′, length 314 bp). The housekeeping gene, glyceraldehyde-3-phosphate dehydrogenase (GAP) was used as the loading control for RNA; GAP (sense, 5′-GCAAATTCCATGGCACCGTCA-3′; antisense, 5′CAGGGGTGCTAAGCAGTTGG-3′, length 320 bp).

Acknowledgments

Flow cytometric analyses Chondrocytes and BMDC were evaluated for expression of cell surface markers commonly associated with antigen presenting cells using three-color flow cytometry. Cells were evaluated using antibodies directed against HLA-DR, CD11c, CD123, HLAABC, CD40, B7-1, and B7-2, and the leukocyte lineage (LIN) markers CD3, CD14, CD16, CD19, CD20, or CD56; and negative control antibodies. Data were analyzed using CellQuest software.

RT-PCR characterization of costimulatory molecule expression The expression of mRNA specific to the B7 family of costimulatory molecules, including B7-1 (CD80), B7-2 (CD86), B7-H1 (PD-L1), B7-DC (PD-L2), B7-H2 (ICOS), B7-H3, and B7-H4 was

The authors thank Markus Grompe and Brian Johnstone for their constructive comments in reviewing the manuscript, and Mid-America Transplant Services, National Disease Research Interchange and Cell Dynamics for their efforts in the procurement of donor tissue. Dr. Streeter has a financial interest as an owner of stock in ISTO Technologies, Inc., a company that may have a commercial interest in the results of this research and technology. This potential conflict of interest has been reviewed and managed by the OHSU Conflict of Interest in Research Committee.

References Adkisson, H.D., Gillis, M.P., Davis, E.C., Maloney, W., Hruska, K.A., 2001. In vitro generation of scaffold independent neocartilage. Clin. Orthop. Relat. Res. S280–S294.

Immune evasion by neocartilage chondroctyes Babcock, T.A., Carlin, J.M., 2000. Transcriptional activation of indoleamine dioxygenase by interleukin 1 and tumor necrosis factor alpha in interferon-treated epithelial cells. Cytokine 12, 588–594. Bacsich, P., Wyburn, G.M., 1947. The significance of the mucoprotein content on the survival of homografts of cartilage and cornea. Proc. R. Soc. Edinb. [Biol.] 62, 321–327. Balanescu, A., Radu, E., Nat, R., Regalia, T., Bojinca, V., Predescu, V., Predeteanu, D., 2002. Co-stimulatory and adhesion molecules of dendritic cells in rheumatoid arthritis. J. Cell. Mol. Med. 6, 415–425. Blazar, B.R., Carreno, B.M., Panoskaltsis-Mortari, A., Carter, L., Iwai, Y., Yagita, H., Nishimura, H., Taylor, P.A., 2003. Blockade of programmed death-1 engagement accelerates graft-versushost disease lethality by an IFN-gamma-dependent mechanism. J. Immunol. 171, 1272–1277. Bretscher, P.A., 1999. A two step, two-signal model for the primary activation of precursor helper T cells. Proc. Natl. Acad. Sci. USA 96, 185–190. Bretscher, P., Cohn, M., 1970. A theory of self-nonself discrimination. Science 169, 1042–1049. Carter, L., Fouser, L.A., Jussif, J., Fitz, L., Deng, B., Wood, C.R., Collins, M., Honjo, T., Freeman, G.J., Carreno, B.M., 2002. PD1:PD-L inhibitory pathway affects both CD4(+) and CD8(+) T cells and is overcome by IL-2. Eur. J. Immunol. 32, 634–643. Caux, C., Dubois, B., 2001. Dendritic cell protocols. In: Robinson, S.P., Stagg, A.J. (Eds.), Methods in Molecular Medicine. Humana Press, Totowa, NJ, pp. 257–273. Chapoval, A.I., Ni, J., Lau, J.S., Wilcox, R.A., Flies, D.B., Liu, D., Dong, H., Sica, G.L., Zhu, G., Tamada, K., Chen, L., 2001. B7-H3: a costimulatory molecule for T cell activation and IFN-gamma production. Nat. Immunol. 2, 269–274. Chesterman, P.J., Smith, A.U., 1968. Homotransplantation of articular cartilage and isolated chondrocytes. An experimental study in rabbits. J. Bone Joint Surg. Br. 50, 184–197. Dong, H., Strome, S.E., Salomao, D.R., Tamura, H., Hirano, F., Flies, D.B., Roche, P.C., Lu, J., Zhu, G., Tamada, K., Lennon, V.A., Celis, E., Chen, L., 2002. Tumor-associated B7-H1 promotes Tcell apoptosis: a potential mechanism of immune evasion. Nat. Med. 8, 793–800. Elves, M.W., Zervas, J., 1974. An investigation into the immunogenicity of various components of osteoarticular grafts. Br. J. Exp. Pathol. 55, 344–351. Feder, J., Adkisson, D., Kizer, N., Hruska, K.A., Cheung, R., Grodzinsky, B.S.A., Lu, Y., Bogdanske, J., Markel, M., 2004. The promise of chondral repair using neocartilage. In: Sandell, L.J., Grodzinsky, A.J. (Eds.), The Promise of Chondral Repair Using Neocartilage. American Academy of Orthopedic Surgeons, Rosemont, IL, pp. 219–226. Fox, D.A., 1997. The role of T cells in the immunopathogenesis of rheumatoid arthritis: new perspectives. Arthritis Rheum. 40, 598–609. Freeman, G.J., Long, A.J., Iwai, Y., Bourque, K., Chernova, T., Nishimura, H., Fitz, L.J., Malenkovich, N., Okazaki, T., Byrne, M.C., Horton, H.F., Fouser, L., Carter, L., Ling, V., Bowman, M.R., Carreno, B.M., Collins, M., Wood, C.R., Honjo, T., 2000. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J. Exp. Med. 192, 1027–1034. Gertzbein, S.D., Lance, E.M., 1976. The stimulation of lymphocytes by chondrocytes in mixed cultures. Clin. Exp. Immunol. 24, 102–109. Greenwald, R.J., Freeman, G.J., Sharpe, A.H., 2005. The B7 family revisited. Annu. Rev. Immunol. 23, 515–548. Harada, Y., Ohgai, D., Watanabe, R., Okano, K., Koiwai, O., Tanabe, K., Toma, H., Altman, A., Abe, R., 2003. A single amino acid alteration in cytoplasmic domain determines IL-2 promoter activation by ligation of CD28 but not inducible costimulator (ICOS). J. Exp. Med. 197, 257–262.

67 Hendrickson, D.A., Nixon, A.J., Grande, D.A., Todhunter, R.J., Minor, R.M., Erb, H., Lust, G., 1994. Chondrocyte-fibrin matrix transplants for resurfacing extensive articular cartilage defects. J. Orthop. Res. 12, 485–497. Heyner, S., 1969. The significance of the intercellular matrix in the survival of cartilage allografts. Transplantation 8, 666–677. Hiraki, Y., Inoue, H., Iyama, K., Kamizono, A., Ochiai, M., Shukunami, C., Iijima, S., Suzuki, F., Kondo, J., 1997. Identification of chondromodulin I as a novel endothelial cell growth inhibitor. Purification and its localization in the avascular zone of epiphyseal cartilage. J. Biol. Chem. 272, 32419–32426. Hiraki, Y., Mitsui, K., Endo, N., Takahashi, K., Hayami, T., Inoue, H., Shukunami, C., Tokunaga, K., Kono, T., Yamada, M., Takahashi, H.E., Kondo, J., 1999. Molecular cloning of human chondromodulin-I, a cartilage-derived growth modulating factor, and its expression in Chinese hamster ovary cells. Eur. J. Biochem. 260, 869–878. Hunziker, E.B., 2002. Articular cartilage repair: basic science and clinical progress. A review of the current status and prospects. Osteoarthr. Cartil. 10, 432–463. Hyc, A., Malejczyk, J., Osiecka, A., Moskalewski, S., 1997. Immunological response against allogeneic chondrocytes transplanted into joint surface defects in rats. Cell Trans. 6, 119–124. Inoue, H., Kondo, J., Koike, T., Shukunami, C., Hiraki, Y., 1997. Identification of an autocrine chondrocyte colony-stimulating factor: chondromodulin-I stimulates the colony formation of growth plate chondrocytes in agarose culture. Biochem. Biophys. Res. Commun. 241, 395–400. Jobanputra, P., Corrigall, V., Kingsley, G., Panayi, G., 1992. Cellular responses to human chondrocytes: absence of allogeneic responses in the presence of HLA-DR and ICAM-1. Clin. Exp. Immunol. 90, 336–344. Klyushnenkova, E., Mosca, J.D., Zernetkina, V., Majumdar, M.K., Beggs, K.J., Simonetti, D.W., Deans, R.J., McIntosh, K.R., 2005. T cell responses to allogeneic human mesenchymal stem cells: immunogeneicity, tolerance, and suppression. J. Biomed. Sci. 12, 47–57. Lance, E.M., Fisher, R.L., 1970. Transplantation of the rabbit's patella. J. Bone Joint Surg. Am. 52, 145–156. Latchman, Y.E., Liang, S.C., Wu, Y., Chernova, T., Sobel, R.A., Klemm, M., Kuchroo, V.K., Freeman, G.J., Sharpe, A.H., 2004. PD-L1-deficient mice show that PD-L1 on T cells, antigenpresenting cells, and host tissues negatively regulates T cells. Proc. Natl. Acad. Sci. USA 101, 10691–10696. Lettesjo, H., Nordstrom, E., Strom, H., Nilsson, B., Glinghammar, B., Dahlstedt, L., Moller, E., 1998. Synovial fluid cytokines in patients with rheumatoid arthritis or other arthritic lesions. Scand. J. Immunol. 48, 286–292. Liang, S.C., Latchman, Y.E., Buhlmann, J.E., Tomczak, M.F., Horwitz, B.H., Freeman, G.J., Sharpe, A.H., 2003. Regulation of PD-1, PD-L1, and PD-L2 expression during normal and autoimmune responses. Eur. J. Immunol. 33, 2706–2716. Lu, Y., Adkisson, H.D., Bogdanske, J., Kalscheur, V., Maloney, W., Cheung, R., Grodzinsky, A.J., Hruska, K.A., Markel, M.D., 2005. In vivo transplantation of neonatal ovine neocartilage allografts: determining the effectiveness of tissue transglutaminase. J. Knee Surg. 18, 31–42. MacKenzie, C.R., Gonzalez, R.G., Kniep, E., Roch, S., Daubener, W., 1999. Cytokine mediated regulation of interferon-gammainduced IDO activation. Adv. Exp. Med. Biol. 467, 533–539. Maury, A.C., Safir, O., Heras, F.L., Pritzker, K.P., Gross, A.E., 2007. Twenty-five-year chondrocyte viability in fresh osteochondral allograft. A case report. J. Bone Joint Surg. Am. 89, 159–165. Mellor, A.L., Munn, D.H., 2004. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat. Rev. Immunol. 4, 762–774. Moskalewski, S., Kawiak, J., 1966. Local cellular response evoked by cartilage formed after auto- and allogeneic transplantation by isolated chondrocytes. Transplantation 4, 572–581.

68 Munn, D.H., Zhou, M., Attwood, J.T., Bondarev, I., Conway, S.J., Marshall, B., Brown, C., Mellor, A.L., 1998. Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 281, 1191–1193. Munn, D.H., Sharma, M.D., Mellor, A.L., 2004. Ligation of B7-1/B7-2 by human CD4+ T cells triggers indoleamine 2,3-dioxygenase activity in dendritic cells. J. Immunol. 172, 4100–4110. Nishimura, H., Nose, M., Hiai, H., Minato, N., Honjo, T., 1999. Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor. Immunity 11, 141–151. Rahfoth, B., Weisser, J., Sternkopf, F., Aigner, T., von der Mark, K., Brauer, R., 1998. Transplantation of allograft chondrocytes embedded in agarose gel into cartilage defects of rabbits. Osteoarthr. Cartil. 6, 50–65. Robinson, D., Halperin, N., Nevo, Z., 1990. Regenerating hyaline cartilage in articular defects of old chickens using implants of embryonal chick chondrocytes embedded in a new natural delivery substance. Calcif. Tissue Int. 46, 246–253. Schlaak, J.F., Pfers, I., Meyer Zum Buschenfelde, K.H., MarkerHermann, E., 1996. Different cytokine profiles in the synovial fluid of patients with osteoarthritis, rheumatoid arthritis and seronegative spondylarthropathies. Clin. Exp. Rheumatol. 14, 155–162. Setoguchi, K., Misaki, Y., Kawahata, K., Shimada, K., Juji, T., Tanaka, S., Oda, H., Shukunami, C., Nishizaki, Y., Hiraki, Y., Yamamoto, K., 2004. Suppression of T cell responses by chondromodulin I, a cartilage-derived angiogenesis inhibitory factor: therapeutic potential in rheumatoid arthritis. Arthritis Rheum. 50, 828–839. Sica, G.L., Choi, I.H., Zhu, G., Tamada, K., Wang, S.D., Tamura, H., Chapoval, A.I., Flies, D.B., Bajorath, J., Chen, L., 2003. B7-H4, a

H.D. Adkisson et al. molecule of the B7 family, negatively regulates T cell immunity. Immunity 18, 849–861. Suh, W.K., Gajewska, B.U., Okada, H., Gronski, M.A., Bertram, E.M., Dawicki, W., Duncan, G.S., Bukczynski, J., Plyte, S., Elia, A., Wakeham, A., Itie, A., Chung, S., Da Costa, J., Arya, S., Horan, T., Campbell, P., Gaida, K., Ohashi, P.S., Watts, T.H., Yoshinaga, S.K., Bray, M.R., Jordana, M., Mak, T.W., 2003. The B7 family member B7-H3 preferentially down-regulates T helper type 1-mediated immune responses. Nat. Immunol. 4, 899–906. Terness, P., Bauer, T.M., Rose, L., Dufter, C., Watzlik, A., Simon, H., Opelz, G., 2002. Inhibition of allogeneic T cell proliferation by indoleamine 2,3-dioxygenase-expressing dendritic cells: mediation of suppression by tryptophan metabolites. J. Exp. Med. 196, 447–457. Tse, W.T., Pendleton, J.D., Beyer, W.M., Egalka, M.C., Guinan, E.C., 2003. Suppression of allogeneic T-cell proliferation by human marrow stromal cells: implications in transplantation. Transplantation 75, 389–397. Wakitani, S., Kimura, T., Hirooka, A., Ochi, T., Yoneda, M., Yasui, N., Owaki, H., Ono, K., 1989. Repair of rabbit articular surfaces with allograft chondrocytes embedded in collagen gel. J. Bone Joint Surg. Br. 71, 74–80. Williams, S.K., Amiel, D., Ball, S.T., Allen, R.T., Tontz Jr., W.L., Emmerson, B.C., Badlani, N.M., Emery, S.C., Haghighi, P., Bugbee, W.D., 2007. Analysis of cartilage tissue on a cellular level in fresh osteochondral allograft retrievals. Am. J. Sports Med. 35, 2022–2032. Zhao, Z.G., Li, W.M., Chen, Z.C., You, Y., Zou, P., 2008. Immunosuppressive properties of mesenchymal stem cells derived from bone marrow of patients with chronic myeloid leukemia. Immunol. Invest. 37, 726–739.