*Abbreviations: APC, antigen-presenting cell; BM, bone marrow; BN, brown Norway; HBSS, Hanks' balanced salt solution; hsp, heat shock protein; IL, interleukin; Lew, Lewis; mAb, monoclonal antibody; OLTx, orthotopic liver transplant; PDI, protein-disulfide isomerase; SDS, sodium dodecyl sulfate.
Although the pathogenesis of chronic rejection is not well understood, it has become apparent that the "response to injury" concept represents an important mechanism (reviewed in 1 (2)
This hypothesis has been tested with an rat cardiac allograft model of chronic rejection, developed recently by Murase et al. (3) 1 ) rats into Brown Norway (RT1n ) recipients, pretreated 100 days before with Lewis bone marrow (BM) cells and a brief course of tacrolimus, exhibit long-term graft survival without immunosuppression. All transplanted hearts in the BM group develop chronic rejection, as evidenced by a chronic obliterative arteriopathy and the endocardium and epicardium contain lymphocytic infiltrates resembling Quilty lesions. On the other hand, recipients pretreated with an orthotopic liver allograft (the OLTx group) do not develop chronic rejection of the subsequent cardiac allograft, and this appears related to a hematolymphoid chimerism involving donor dendritic cells, but the exact mechanisms have yet to be determined. In the BM group, a persistent immunological injury during the early posttransplant period is associated with the development of chronic rejection. An important mechanism involves the disruption of lymphatic flow in the allograft as a result of a continuous process of interstitial inflammation initiated by graft-infiltrating alloreactive lymphocytes (20)
As graft-infiltrating lymphocytes exhibit low levels of donor-specific alloproliferation (3) (2) (4) (5) (6)
The studies reported here deal with mammalian hsp60, which is found primarily in the mitochondrial compartment of eukaryotic cells and, its prokaryotic equivalent mycobacterial hsp65. Three members of the mammalian hsp70 family were studied: the stress-inducible hsp72, the constitutively expressed heat shock cognate 73 (hsc73) and the glucose-regulated protein 78 (grp78; also called binding protein, Bip; 7), a chaperone normally localized in the endoplasmic reticulum (8) (9,10)
The data provide evidence for the increased expression of certain stress proteins in cardiac allografts undergoing chronic rejection and the presence of graft-infiltrating autoreactive lymphocytes that operate under hsp-dependent mechanisms.
MATERIALS AND METHODS
Experimental model of chronic cardiac allograft rejection. Male Brown Norway (BN, RTn ) rats were pretreated with a Lewis (Lew, RT1 ) orthotopic liver allograft (OLTx) or an infusion of BM cells (2.5Ă—108 cells) according to previously described procedures (3) (20)
Immunohistochemical staining of cardiac allograft tissues. The indirect avidin-biotin method was used for the immunostaining studies on the expression of hsps in 100-day cardiac allograft tissues in the BM group (n=6) and OLTx group (n=5), whereas 100-day syngrafts (n=3) and normal Lew rat hearts (n=4) were used as controls. All tissues were snap-frozen in optimal temperature compound (Miles Laboratories Inc., Elkhart, IN) and after storage at -70°C, 4-µm cryostat sections were fixed in cold acetone at -20°C. Four hsp-specific murine mAbs were used for immunostaining (Table 1 ); all were obtained from StressGen Biotechnologies Corp (Victoria, British Columbia, Canada). SPA-806 (previously referenced to as LK-1) was generated against human hsp60 and reacts with human and rat hsp60 (11) (12) (13) (14) (4)
Table 1: hsp-specific monoclonal antibodies used for immunostaining and immunoblot analysis
All microscopic images of stained tissue sections were made with a Sony Power HAD videocamera (DXC-970 MD) using the Autocyte image management system in a Windows 95-based personal computer. Images were enhanced with an Adobe Photoshop 4.0 software program.
Isolation of graft-infiltrating lymphocytes. Graft-infiltrating cells were isolated from 100-day cardiac allografts according to previously described procedures (5) g at 4°C for 15 min. The cell pellet was resuspended in 7 ml of a gradient density solution consisting of 18 ml of Percoll (Sigma Chemical, St, Louis, MO), 2 ml of 10 × strength phosphate-buffered saline, and 10 ml of RPMI-1640 medium supplemented with 5% fetal calf serum, 24 mM Hepes buffer, 4 mM L-glutamine, and 60 µg/ml gentamicin. The cell suspension was then centrifuged at 10,000×g at 4°C for 20 min. After centrifugation, the interface between stromal cells and red blood cells was collected, washed three times, and counted.
Lymphocyte proliferation assays. Graft-infiltrating cells were tested for their proliferative responses to irradiated spleen cells as APCs in 3-or 5-day cultures in 96-well round bottom microtiter plates, as described previously (5) 4 responder cells were incubated with hsps (5 µg/ml) in the presence of 5×104 irradiated (3.0 Gy) syngeneic spleen cells in 200 µl of tissue culture medium consisting of RPMI-1640 supplemented with 10% fetal calf serum, 24 mM Hepes buffer, 4 mM L-glutamine, 60 µg/ml gentamicin, 50 µM 2-mercaptoethanol, and 1 mM sodium pyruvate. The following recombinant hsp preparations were used: Mycobacterium leprae hsp65 (Mlep65), Bacillus Calmette-Guerin hsp65 (BCG65), Mycobacterium tuberculosis hsp71 (Mtub hsp71), and M. tuberculosis hsp10 (Mtub 10). They were supplied by Priv. Doz. Dr. M Singh, WHO Recombinant Protein Bank (Braunschweig, Germany). Human hsp70 was obtained from StressGen. Recombinant mouse grp78 was kindly provided by Dr. M.-J. Gething (University of Texas Southwestern Medical Center, Dallas, TX). Small quantities of interleukin-2 (IL-2; 0.2 U/well) were added to augment hsp-induced proliferation. During the final 18-20 hr of incubation, each culture was pulsed with 1 µCi of tritiated thymidine (specific activity: 20 mCi/mmol; New England Nuclear Products, Boston, MA) and harvested with a multiple-sample harvester (Skatron, Sterling, VA).
Generation of cardiac allograft-derived hsp71-dependent T-cell clones. hsp71-dependent T-cell lines were generated by culturing allograft lymphocytes with Mtub hsp71 plus irradiated syngeneic spleen cells and IL-2 as described previously (6)
To learn about its mode of action of hsp71 in the responses of hsp71-dependent T-cell clones, we treated Mtub hsp71 under various conditions as described previously (6) (15,16) (6) (17) (6)
Immunoblotting of cardiac stromal tissues. After the isolation of graft-infiltrating cells, the graft stromal tissue remaining on the mesh was retrieved into a 1-ml Eppendorf tube filled with HBSS at 4°C. After centrifugation at 11,500 rpm for 30 min, the tissue was stored at -70°C. The tissues were homogenized and lysed in 2 ml of lysis buffer consisting of 1% Triton X-100, 1% deoxycholate, 1% sodium dodecyl sulfate (SDS), 0.15 M sodium chloride, 0.02 M sodium phosphate pH 7.2, 0.01 M Tris buffer, pH 7.5, 2 mM ethylenediaminetetraacetic acid, 1 mM phenylmethylsulfonyl fluoride, 50 mM sodium fluoride, 30 mM sodium pyrophosphate, 1 mg/ml leupeptin, and 1 mg/ml pepstatin. The nuclei were excluded by centrifuging at 7,500 rpm for 10 min, and the total protein was extracted. Protein concentration was measured by the bicinchoninic acid protein assay (Sigma Chemical).
Protein samples (100 mg in 10 ml) were mixed with 10 ml of 2Ă—SDS denaturing loading buffer and loaded on a 5-8% Tris-glycine SDS-polyacrylamide stacking gel. The electrophoresis was done at 40 mA for 4 hr. Rainbow marker proteins (Amersham) were used as indicators of molecular weight. The quantity and quality of the protein samples were checked by Coomassie Brilliant Blue staining. After electrophoresis, the gel was equilibrated in transfer buffer for 20 min, and the protein was transferred to a nitrocellulose membrane (Schleicher & Schuell) by electroblotting at 15 mV overnight. After treatment with 5% Blotto (5% nonfat dried milk in 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Tween 20) for 1 hr, the membranes were incubated with hsp-specific mouse mAb for 2-4 hr. The membranes were washed three times with 1% Blotto and then incubated with horseradish peroxidase-linked goat anti-mouse Ig antibody for at 1 hr. After washing five times, the membranes were incubated in ECL solution (Amersham) for 1-2 min. and the antibody-bound protein was detected by exposing the membrane to X-Omat films (Kodak).
RESULTS
Immunohistochemical analysis of stress protein expression in cardiac allograft tissues. Previous histopathological studies have shown that this chronic rejection model has two major types of lesions involving inflammatory cells (3) (20)
Immunostaining on hsp expression was done with 100-day heart allografts in the BM and OLTx groups and 100-day syngrafts and normal hearts as controls. Three members of the hsp70 family were studied: the endoplasmic reticular chaperone grp78 (or BiP), the heat stress-inducible hsp72, and the constitutively expressed hsc73.
Figure 1 shows immunostaining results with SPA-827, a monoclonal antibody generated against the synthetic peptide KSEKDEL from rat grp78 conjugated to KLH (14) (4) Fig. 1, A-C ). The grp78seq staining pattern was granular and involved cells with dendritic morphology. Previous studies have shown similar clusters of OX62-positive dendritic cells amid lymphoid aggregates in the adventitia of arteries with intimal thickening (20) Fig. 1, D and E ), which, as previous studies have shown, consist of primarily CD4+ T cells and CD8+ T cells (20) (3) Fig. 1F ). Similarly, grp78seq expression was low in 100-day syngrafts and normal cardiac tissues (data not shown).
Figure 1: Immunostaining of cardiac tissues with monoclonal antibodies SPA-827 specific for grp78seq (panels A-F), SPA-820 specific for hsp72+hsc73 (panels G-K), and SPA-806 specific for hsp60 (panels L-O). Images were made with the following magnifications: Ă—100 (B, F, G, H, K, L, and O), Ă—200 (A, D, and J), and Ă—400 (C, E, I, M, and N).
Staining with the mAb SPA-810 specific for only hsp72 (the heat stress-inducible form of hsp70) yielded no significant expression in any heart allograft or syngraft tissue examined in this study (data not shown). This finding suggests that hsp72 is not involved with the stress response associated with chronic rejection. On the other hand, SPA-820, a more broadly reactive mAb specific for both constitutive hsc73 and inducible hsp72 (13) Fig. 1G ). On the other hand, the arterial walls themselves showed intense staining of hsc73, especially in the areas of intraluminal myofibroblast proliferation. No or a few SPA-820-positive cells were found in the Quilty lesions (Fig. 1H ), although often enough the underlying myocytes showed increased hsc73 expression as compared to myocytes in areas where there was no cellular infiltration. The intercalated disks of cardiac myocytes showed readily detectable expression of hsc73 in the BM group (Fig. 1I ) and the OLTx group (Fig. 1K ) and in normal hearts as well (data not shown). Increased hsc73 staining often affected myocytes in areas of lymphoid cellular infiltration (Fig. 1, I and J ). This increased expression of hsc73 seemed to reflect a stress response of myocytes exposed to infiltrating lymphoid cells. Some of these infiltrates contained hsc73-positive cells (Fig. 1J ), but it was not clear whether this was a result of an uptake of hsc73 released by stressed myocytes or whether these lymphoid cells had increased hsc73 expression as a result of the inflammatory environment. It should be noted that the expression of hsc73 was very low in areas of fibrosis, particularly in the BM group of transplanted hearts (Fig. 1G ).
The immunostaining with the hsp60-specific mAb SPA-806 was more pronounced for myocytes in heart allografts in the BM group (Fig. 1, L and M ) than in the OLTx group (Fig. 1O ) and normal hearts. The most intense staining involved myocytes near Quilty lesions (Fig. 1L ) and in other areas with cellular infiltration (Fig. 1M ) as compared to myocytes in areas where there was no cellular infiltration. The intercalated disks were not visible with SPA-806. The arterial lesions showed little hsp60 expression. No or a few SPA-806-positive graft-infiltrating cells were seen in the arterial walls (Fig. 1N ) or conjunction with the Quilty lesions (Fig. 1L ).
In summary, these immunostaining results demonstrate different patterns in hsp expression in cardiac allografts in the BM and OLTx groups. More immunostaining data with colored images are presented on the Immunobiology page of the web site of Transplantation Pathology Internet Services (http://tpis.upmc.edu). Most salient findings are the clusters of grp78seq-positive graft-infiltrating cells and the strong hsc73 expression in obliterating arteries. During chronic rejection, hsp60 and hsc73 expression is also higher in allograft myocytes especially in areas with lymphoid cell infiltration. This increased expression of these hsps probably reflects a stress response of injured myocytes. No staining was seen with antibodies against the inducible hsp72 and negative controls such as anti-HLA-A2.
Immunoblot analysis of stress protein expression in stromal tissues of cardiac allografts. Immunoblots were made with three or four different stromal tissue lysates of cardiac tissues that had remained after extraction of graft-infiltrating cells. Figure 2 shows representative data with 100-day cardiac allografts in the BM and OLTx groups in comparison with 100-day syngeneic grafts. The same four mAbs listed in Table 1 were used to determine the expression of hsp60, hsp72, hsc73, and grp78. All allograft extracts revealed two strong 50-55-kDa bands with secondary goat anti-mouse Ig antibody (Fig. 2A ). These dual bands were not seen with syngrafts or normal hearts, and they seem to represent rat immunoglobulin heavy chains that cross-react with this anti-mouse Ig antibody and three other antisera from different suppliers (data not shown). As the bands are found equally in the BM group with chronic rejection and the OLTx group without chronic rejection, their significance in allograft immunity remains uncertain. Previous immunoblot studies on an acute rejection model did not reveal such bands with the secondary anti-Ig antibody (4)
Figure 2: Immunoblots of cardiac graft stromal tissues with hsp-specific mAb. In each panel, lanes B and L represent 100-day cardiac allografts in the BM and OLTx groups, respectively, and lane S represents a 100-day syngeneic graft. These results were obtained with the same group of stromal preparations tested with no antibody (A; Ig control), SPA-806 (B; hsp60), SPA-810 (C; hsp72) SPA-820 (D; hsp72+hsc73), and SPA-827 (E; grp78). All allografts revealed a 50-55-kDa doublet (indicated with HC) with the Ig control and hsp-specific mAbs. The molecular size markers are indicated on the left.
The presence of this 50-55-kDa doublet might reflect the deposition of specific antibodies, perhaps as a result of a humoral immune response to cardiac graft antigens in long-term transplant survivors. It should be noted that the immunostaining of allograft tissues showed no overt immunoglobulin deposits.
The 50-55-kDa double band was seen in each immunoblot with the hsp-specific mouse mAb. SPA-806 gave a major band of about 60 kDa (corresponding to hsp60), and its intensity was slightly higher with the syngrafts than the allografts in the BM and OLTx groups (Fig. 2B ). Two extra bands of about 55 and 35 kDa were seen primarily for the BM group, and they were less pronounced for the OLTx group. These bands might represent hsp60 fragments or different molecules that cross-react with SPA-806. This mAb also detects a 40-kDa band in stromal tissues from rat cardiac allografts undergoing acute rejection (4)
Figure 2C shows the immunoblots with SPA-810 specific for the inducible hsp72. This mAb gave a ∼70-kDa band with stromal extracts of 100-day syngrafts, but this band was not seen with 100-day allografts in the BM or OLTx groups. Previous studies have shown that stromal extracts from acutely rejecting cardiac allografts show a ∼70-kDa band with SPA-810 and an additional 30-kDa band that might represent a breakdown product of hsp72 (4) Fig. 1 ). These findings indicate that hsp72 does not play a major role in the stress response associated with chronic rejection.
Immunoblots of 100-day syngraft stromal extracts with SPA-820 showed two bands in the 70-kDa range; one corresponded with hsc73, and the second weaker band had a molecular weight similar to the inducible hsp72 (Fig. 2D ). The hsc73 band was equally strong in the BM group of allografts but less intense in the OLTx group. The second ∼70-kDa band was stronger in both groups of allografts. As no significant bands were seen with SPA-810 (Fig. 1C ), it seems that this ∼70-kDa band is probably not the inducible hsp72, but rather another hsp70 protein. Most notable was the presence of several lower molecular weight bands in 100-day allograft stroma. Although no significant differences were noted between these extra band patterns in the BM and the OLTx groups, it seems likely that many bands could represent hsc73 fragments resulting from molecular degradation. These findings might be comparable to the increased hsc73 staining of injured myocytes in areas of lymphoid cellular infiltration (Figs. 1, I and J ). Previous immunoblot studies on an acute rat cardiac allograft model have also shown several extra bands with SPA-860 (4)
The grp78seq-specific mAb SPA-827 gave a major single band (presumably grp78) with 100-day syngrafts but not bands with 100-day allografts (Fig. 2E ). SPA-827 cross-reacts with grp94 and an undefined 50-kDa KDEL-containing protein. This mAb reacts readily with both corresponding bands in acutely rejecting allografts and, to a lesser extent, in syngrafts during the first few days after transplantation (4)
In summary, these immunoblots provide information about hsp levels in homogenized stromal tissues, which had been previously subjected to mechanical extraction procedures to remove graft-infiltrating cells. Allograft lysates contained lower molecular weight proteins that cross-react with the hsp60-specific and hsp70-specific mAb. They might represent hsp fragments that had been generated in vivo or that were the result of the extraction procedure. No fragments were detected for grp78 or hsp72.
Proliferative responses of allograft-infiltrating cells. Graft-infiltrating cells were isolated from 100-day cardiac allografts for in vitro proliferation assays with irradiated spleen cells and hsps. In the BM group, the average yield was 4.5Ă—105 graft-infiltrating cells per allograft. The cell recovery in the OLTx group was insufficient (less than 1Ă—104 cells) for performing the proliferation assays. Figure 3 summarizes the proliferative responses of three cell suspensions from allografts in the BM group. A small amount of exogenous IL-2 (0.2 U/ml) was necessary to achieve significant proliferation in these 3-day assays with only 1Ă—104 graft-infiltrating cells per well. In the presence of syngeneic BN spleen cells, mycobacterial hsp65 but not hsp10 and hsp71 induced proliferation under these conditions.
Figure 3: Proliferative responses of graft-infiltrating lymphoid cells isolated from chronically rejecting 100-day cardiac allografts to syngeneic (self-APCs) and allogeneic irradiated spleen cells (donor APCs) and different mycobacterial hsp preparations. These 3-day assays were done in the absence (dark bars) or the presence of 0.2 U/ml exogenous IL-2 (lighter bars). The bars represent mean cpm values calculated from data with four different preparations of allograft-infiltrating cells. All standard errors were below 20% of the mean values.
Analogous to previous observations (3) Fig. 3 ). Incubation with hsp65 led to increased proliferation, whereas only a marginal effect was seen for hsp71.
These data suggested that, in this chronic rejection model, the hsp reactivity of 100-day cardiac graft-infiltrating cells was primarily towards hsp65 rather than hsp71. They are somewhat in contrast with the findings on an acute rejection model, which showed a stronger hsp71-dependent reactivity of graft-infiltrating cells in rat cardiac allografts (5)
Hsp71-dependent autoreactivity of allograft-derived lymphocyte clones. Table 2 shows the proliferative responses of four T-cell clones propagated from cardiac allografts undergoing chronic rejection. These clones reacted with syngeneic BN APCs but only in the presence of Mtub hsp71. No effect was seen with Mlep hsp65 nor BCG hsp65 or Mtub hsp10 (data not shown) and these clones showed no donor-specific alloreactivity towards donor Lew APCs even in the presence of Mtub hsp71. Attempts to generate such hsp71-dependent, autoreactive T-cell clones have, thus far, been unsuccessful for the OLTx group of graft-infiltrating cells.
Table 2: Hsp effect on four T-cell clones cultured from 100-day heart allografts undergoing chronic rejection
The hsp71 dependency of these chronically rejecting allograft-derived T cells is not limited to mycobacterial hsp71 because murine grp78 has a similar effect (Table 3 ). On the other hand, human hsp70 is ineffective. In spite of their divergent origin, mycobacterial hsp71 and mammalian grp78 have a high degree of amino acid sequence homology and similar peptide binding characteristics by ATP-dependent mechanisms (10,18,19)
Table 3: Mtub hsp71- and murine grp78-dependent reactivity of a T-cell clone 2.1 cultured from a cardiac allograft undergoing chronic rejection
Hsp71-dependent clones were first described with an acute rejection model of ACI into Lew cardiac allografts (5,6)
To assess the mode of action of Mtub hsp71 on the hsp-dependent T-cell responses to autologous APCs, we have studied the effect of modified Mtub hsp71 with the T-cell clones derived from chronically rejecting allografts. These procedures were similar to those described in a report on hsp71-dependent T-cell clones propagated from acutely rejecting cardiac allografts (6) Table 3 ). Furthermore, treatment with trypsin or polymyxin B, which changes the structural integrity of hsp71 (6)
DISCUSSION
Previous studies by Murase et al. have shown that chronic rejection in this cardiac allograft model reflects a loss of donor-specific hematolymphoid chimerism and an accumulation of graft-infiltrating CD45-positive, ED1-positive recipient cells (3) (20)
The findings in the present study support the additional concept that chronic rejection is associated with a stress response in the allograft as indicated by the increased tissue expression of stress proteins and that graft-infiltrating lymphocytes might be activated by mechanisms involving hsps. Although these mechanisms of lymphocyte activation are not understood, one can postulate two pathways, one involving hsp65 and the other including members of the hsp70 family, especially grp78.
Immunostaining of allograft tissues shows increased expression of hsp60 and hsp70 in myocytes especially in areas of lymphoid infiltration. These findings are consistent with the immunoblot data suggesting the presence of hsp60 and hsp70 fragments in homogenized stroma from allografts but not syngrafts.
Graft-infiltrating cells from chronically rejecting heart allografts show low in vitro responsiveness to allogeneic and syngeneic APCs, but significant proliferation occurs if mycobacterial hsp65 has been added to the culture. This hsp65 effect has also been observed with lymphocytes isolated from acutely rejecting cardiac allografts (5) (4)
Many studies on experimental and clinical models of autoimmune diseases such as arthritis and diabetes have demonstrated immune responses to hsp65 (22) (23)
On the other hand, hsp71-dependent, autoreactive T cell clones were readily cultured from infiltrating cells isolated from allografts with chronic rejection. These clones respond to self-APCs, which must first be exposed to mycobacterial hsp71. Treatment of Mtub hsp71 with proteolytic enzymes, polymyxin B, which alters the electrophoretic migration of hsp71 (6) (19) (10,18,19)
The grp78 dependency of graft-derived autoreactive T cells becomes more interesting in view of the immunostaining data showing clusters of grp78seq-positive cells in the lymphocytic infiltrates of the arteriopathy and Quilty lesions. In the cellular infiltrates of chronically rejecting allografts, grp78seq-positive cells resemble morphologically macrophages and dendritic cells. The granular staining patterns are quite intense, and they seem to suggest a more widespread intracellular distribution rather than the endoplasmic compartment alone. This immunostaining with the grp78seq-specific SPA-827 cannot be considered as conclusive evidence for the grp78 expression in this subset of graft-infiltrating cells. SPA-827 cross-reacts with grp94 and an undefined 50-kDa protein; our studies show different immunostaining patterns between SPA-827 and a grp94-specific mAb SPA-850 (Duquesnoy et al., manuscript in preparation). We have also found that the SPA-827-immunostaining of graft-infiltrating cells is not related to a high expression of the KDEL sequence-containing protein-disulfide isomerase (PDI), which has a molecular mass of about 58 kDa (Duquesnoy et al., manuscript in preparation).
Several explanation can be forwarded for the presence of grp78seq-positive cells in the arteriopathy lesions. During rejection, the allograft may signal the recruitment of grp78seq-positive circulating accessory cells originating in the peripheral lymphoid organs. Increased numbers of grp78seq-positive cells have been identified in the marginal zones of the T-cell follicles of the spleen (manuscript in preparation; see also web site at http://tpis.upmc.edu). Alternatively, the intragraft inflammatory microenvironment itself may also promote the production of grp78 and other KDEL-containing proteins in graft-infiltrating macrophages and dendritic cells. Several proinflammatory cytokines stimulate the synthesis of grp78 and other hsps (24-28) (29)
How do grp78seq-positive cells function? As they cluster in certain areas of graft-infiltrating lymphocytes, they might represent a distinct population of APCs that interact with a subset of lymphocytes, perhaps the grp78-dependent autoreactive T cells that have been cultured from the cellular infiltrates. This postulate would place grp78 as a mediator of intragraft autoimmune processes involved in the pathogenesis of chronic rejection. Through its peptide-binding properties, grp78 might participate in the transport and processing of autologous peptides presented by APCs. Recent studies have demonstrated that stress proteins such as grp94 (also called gp96) and PDI participate, through their peptide-binding properties, in alternative pathways of antigen processing and presentation (21,30,31)
Finally, we have also observed a higher expression of hsc73 (but not the heat stress-inducible hsp72) in myocytes nearby cellular infiltrates of the allograft. Immunoblot studies with anti-hsc73 antibodies have shown lower molecular weight bands that might represent hsc73 fragments. One might speculate that such fragmentation of hsc73 leads to the generation of immunogenic peptides that elicit an autoimmune response.
In conclusion, this report presents the first evidence that the pathogenesis of chronic rejection involves a stress response and the activation of autoreactive lymphocytes that operate under mechanisms involving hsps.
REFERENCES
1. Demetris A, Murase N, Lee R, et al. Analysis of chronic rejection and obliterative arteriopathy: possible contributions of donor antigen presenting cells and lymphatic disruption. Am J Pathol 1997; 150: 563.
2. Duquesnoy R, Liu K, Moliterno R, Attfield D. Do heat shock proteins play a role in transplant immunity? Transplant Rev 1996; 10: 175.
3. Murase N, Starzl TE, Tanabe M, et al. Variable chimerism, graft-versus-host disease, and tolerance after different kinds of cell and whole organ transplantation from Lewis to brown Norway rats. Transplantation 1995; 60: 158.
4. Qian J, Moliterno R, Donovan-Peluso M, et al. Expression of heat shock proteins and lymphocyte reactivity in rat cardiac allografts undergoing cellular rejection. Transplant Immunol 1995; 3: 114.
5. Moliterno R, Valdivia L, Pan F, Duquesnoy RJ. Heat shock protein reactivity of lymphocytes isolated from heterotopic rat cardiac allografts. Transplantation 1995; 59: 598.
6. Liu K, Moliterno R, Fu X-F, Duquesnoy R. Identification of two types of autoreactive T lymphocyte clones cultured from rat cardiac allograft cells incubated with recombinant mycobacterial heat shock protein 71. Transplant Immunol 1996; 5: 57.
7. Haas I, BiP (GRP78), an essential hsp70 resident protein in the endoplasmic reticulum. Experientia 1994; 50: 1012.
8. Morimoto RI, Tissieres A, Georgopoulos C. The biology of heat shock proteins and molecular chaperones. Plainview, NY: Cold Spring Harbor Laboratory Press, 1994.
9. Gething JR, Sambrook JF. Protein folding in the cell. Nature 1992; 355: 33.
10. Roman E, Moreno C, Young D. Mapping of Hsp70-binding sites on protein antigens. Eur J Biochem 1994; 222: 65.
11. Boog CJP, Graeff-Meeder ERD, Lucassen MA, et al. Two monoclonal antibodies generated against human hsp60 show reactivity with synovial membranes of patients with juvenile chronic arthritis. J Exp Med 1992; 175: 1805.
12. Welch WJ, Suhan JP. Cellular and biochemical events in mammalian cells during and after recovery from physiological stress. J Cell Biol 1986; 103: 2035.
13. Mizzen LA, Riabol KT, Welch WJ. Heat shock is lethal to fibroblasts microinjected with antibodies against hsp70. Science 1988; 242: 433.
14. Vaux D, Tooze J, Fuller S. Identification by anti-idiotypic antibodies of an intracellular membrane protein that recognizes a mammalian endoplasmic reticulum retention signal. Nature 1990; 345: 495.
15. Flaherty KM, DeLuca-Flaherty C, McKay DB. Three-dimensional structural of the ATPase fragment of a 70-heat-shock cognant protein. Nature 1990; 346: 623.
16. Rippmann F, Taylor WR, Rothbard JB, Green NM. A hypothetical model for the peptide binding domain of hsp70 based on the peptide binding domain of HLA. EMBO J 1991; 10: 1053.
17. Issekutz A. Removal of gram negative bacteria from solution by affinity chromatography. J Immunol Methods 1983; 61: 275.
18. Mehlert A, Young D. Biochemical and antigenic characterization of the
Mycobacterium tuberculosis 71 kDa antigen, a member of the 70 kD heat-shock protein family. Mol Microbiol 1989; 3: 125.
19. Gething M-J, Blond-Elguindi S, Mori K, Sambrook JF. Structure, function, and regulation of the endoplasmic reticulum chaperone BiP. In: Morimoto RI, Tissieres A, Georgopoulos C, eds. The biology of heat shock proteins and molecular chaperones. Plainview, NY: Cold Spring Harbor Laboratory Press, 1994: 111.
20. Demetris A, Murase N, Ye Q, et al. An analysis of chronic rejection and obliterative arteriopathy: possible contributions of donor antigen presenting cells and lymphatic disruption. Am J Pathol 1997; 150: 563.
21. Lammert E, Stevanovic S, Brunner J, Rammensee HG, Schild H. Protein disulfide isomerase is the dominant acceptor for peptides translocated into the endoplasmic reticulum. Eur J Immunol 1997; 27: 1685.
22. van Eden W, Young DB. Stress proteins in medicine. New York: Marcel Dekker. Inc., 1996.
23. Wick G, Schett G, Amberger A, Kleindienst R, Xu Q. Is atherosclerosis an immunologically mediated disease? Immunol Today 1995; 27: 27.
24. Koller M, Hensler T, Konig B, Prevost G, Alouf J, Konig W. Induction of heat-shock proteins by bacterial toxins, lipid mediators and cytokines in human leukocytes. Int J Med Microbiol Virol Parasitol Infect Dis 1993; 278: 365.
25. Souza S, Antel J, Freedman M. Cytokine induction of heat shock protein expression in human oligodendrocytes: an interleukin-1-mediated mechanism. J Neuroimmunol 1994; 50: 17.
26. Satoh J, Kim SU. Cytokines and growth factors induce HSP27 phosphorylation in human astrocytes. J Neuropathol Exp Neurol 1995; 54: 504.
27. Hall TJ. Role of hsp70 in cytokine production. Experientia 1994; 50: 1048.
28. Anderson SL, Shen T, Lou J, et al. The endoplasmic reticular heat shock protein gp96 is transcriptionally upregulated in interferon-treated cells. J Exp Med 1994; 180: 1565.
29. Hightower LE, Guidon PT. Selective release from cultured mammalian cells of heat-shock (stress) proteins that resemble glia-axon transfer proteins. J Cell Physiol 1989; 138: 257.
30. Lammert E, Arnold D, Nijenhuis M, et al. The endoplasmic reticulum-resident stress protein gp96 binds peptides translocated by TAP. Eur J Immunol 1997; 27: 923.
31. Arnold D, Wahl C, Faath S, Rammensee HG, Schild H. Influences of transporter associated with antigen processing (TAP) on the repertoire of peptides associated with the endoplasmic reticulum-resident stress protein gp96. J Exp Med 1997; 186: 461.
