Phosphorylated S6 Ribosomal Protein: A Novel Biomarker of Antibody-Mediated Rejection in Heart Allografts

Transcription

1 American Journal of Transplantation 2006; 6: Blackwell Munksgaard C 2006 The Authors Journal compilation C 2006 The American Society of Transplantation and the American Society of Transplant Surgeons doi: /j x Phosphorylated S6 Ribosomal Protein: A Novel Biomarker of Antibody-Mediated Rejection in Heart Allografts E. J. Lepin a,b, Q. Zhang a,b, X. Zhang a,b, P. T. Jindra a,b, L. S. Hong c, P. Ayele a,b,c, M. V. P. Peralta c, D. W. Gjertson a,b, J. A. Kobashigawa d, W. D. Wallace c, M. C. Fishbein c and E. F. Reed a,b, a UCLA Immunogenetics Center, b Department of Pathology and Laboratory Medicine, David Geffen School of Medicine UCLA, Los Angeles, California, USA c Department of Pathology and Laboratory Medicine, Room A7 149, David Geffen School of Medicine UCLA, Los Angeles, California, USA d Division of Clinical Faculty Medicine, UCLA Heart Transplant Program, David Geffen School of Medicine UCLA, Los Angeles, California, USA Corresponding author: E. F. Reed, ereed@mednet.ucla.edu We tested the hypothesis that phosphorylation of S6 ribosomal protein (S6RP), a downstream target of the PI3K/Akt/mTOR pathway, is a biomarker of antibodymediated rejection (AMR) in heart allografts. Primary cultures of human aortic and microvascular endothelial cells (EC) were treated with anti-hla class I and class II antibodies (Ab) and cell lysates were studied for phosphorylation of S6 ribosmal protein at Serine 235/236 (p-s6rp). Treatment of cultured EC with anti-class I and class II Ab stimulated S6RP phosphorylation. Immunohistochemical techniques were used to detect the level of p-s6rp in endomyocardial biopsies (n = 131) from 46 heart transplant recipients and the results were correlated with histopathological diagnosis of rejection, C4d staining, production of posttransplant anti-hla Ab and clinical outcome. Increased phosphorylation of S6RP in endomyocardial biopsies was significantly associated with the diagnosis of AMR (p < ). No significant association between acute cellular rejection (ACR) and p-s6rp was observed. C4d staining was positively associated with both AMR and p-s6rp. Posttransplant anti-hla class II Ab production was also significantly associated with a positive p- S6RP status in cardiac biopsies. These results indicate that p-s6rp is a useful biomarker for the diagnosis of AMR. Key words: Alloantibodies, endothelial cells, heart/ lung transplantation, histocompatibility antigens, rejection Received 29 September 2005, revised 09 March 2006 and accepted 20 March 2006 Introduction Antibody-mediated rejection (AMR) is emerging as a leading cause of cardiac allograft rejection and graft loss (1 4). The ability to accurately diagnose AMR has been markedly improved due to advances in the ability to detect circulating alloantibodies and histologic and immunohistochemical abnormalities on endomyocardial biopsy; including complement split products. However, diagnosis of AMR in endomyocardial biopsy specimens is more challenging. Histologic findings may be subtle, and immunofluorescence and/or immunohistochemical studies are not performed routinely at most centers. Standardized criteria for the diagnosis of AMR have only recently been agreed upon. Therefore, a combination of clinical, histologic and immunopathologic findings, and demonstration of circulating donor-specific antibodies (DSA), in the absence of cellular rejection are recommended to diagnose acute AMR in heart allografts (5). The discovery of new molecular biomarkers of AMR is needed to improve diagnosis of cardiac allograft rejection as well as to select and monitor treatment strategies. The frequency of antibody-mediated cardiac transplant rejection is estimated to be 10 20% and is accompanied by the production of DSA. AMR is associated with a significantly worse survival and predisposes patients to cardiac allograft vasculopathy (CAV) and graft loss (1,3). CAV manifests itself as a diffuse, concentric intimal proliferative arteriosclerosis. Patients developing AMR following cardiac transplantation progress to CAV earlier and at an increased frequency when compared to control patients. A common feature among patients with CAV is the development of posttransplant anti-hla antibody (Ab) (3,6). Although the detection of circulating DSA is linked to AMR and CAV, the physiological and pathological effect of their binding to the transplanted organ has only recently been explored. Research studies from our group and others have shown that ligation of class I and class II molecules on the surface of endothelial cells (EC) by anti-hla Ab promotes diverse biological functions including cellular proliferation 1560

2 p-s6rp is a Marker of Antibody-Mediated Rejection and survival in a model relevant to the development of transplantation-associated vasculopathies (7 13). Engagement of HLA class I molecules by anti-hla Ab stimulates tyrosine phosphorylation of intracellular proteins including Src, focal adhesion kinase (FAK) and paxillin and enhances proliferative responses to basic fibroblast growth factor (7 11,14). Ligation of class I molecules also triggers a prosurvival signaling cascade resulting in phosphorylation of PI3 kinase (PI3K) and Akt and up-regulation of the antiapoptotic proteins Bcl-2 and Bcl-xL in EC (9,11,12). Similarly, ligation of HLA class II molecules on vascular EC has been shown to induce phosphorylation of protein kinase C and activation of Akt (13). It is well established that Akt signaling regulates cell survival and proliferation through activation of downstream targets including S6 kinase and S6RP (15). With the availability of phosphorylation-specific Ab that can detect activated signaling molecules in cultured cells and paraffin-embedded biopsy tissue, we set out to determine the significance of this signaling pathway in AMR. We report that ligation of HLA class I molecules on cultured human aortic EC (HAEC) and cardiac microvascular EC (MVEC) stimulated an increase in phosphorylation of S6RP at Serine 235/236 in a PI3K/Akt-dependent manner. Antibody ligation of class II molecules also stimulated increased phosphorylation of p-s6rp in MVEC and HAEC. Furthermore, increased capillary staining for p-s6rp in cardiac biopsies was highly associated with AMR. Elucidation of the signal transduction pathways involved in AMR has the potential to improve diagnosis of cardiac allograft rejection and to guide the development of new treatment strategies. Materials and Methods Western blotting Western blotting was performed as previously described (9,14). The following reagents were used in Western blotting experiments. W6/32 a murine mab directed against a monomorphic determinant on HLA class I molecules was produced from hybridoma HB-95 supplied by the American Type Culture Collection (Manassas, VA, USA). F002 6C6G1, a murine IgG2a mab directed against a monomorphic determinant on HLA class II molecules was provided as a gift from Dr. Jarhow Lee, One Lambda Inc. (Canoga Park, CA, USA). Antiphospho-S6 ribosmal protein Ab (Serine 235/236 )was purchased from Cell Signaling Technology (Beverly, MA, USA). Cytochalasin D and wortmannin were purchased from Sigma-Aldrich (St. Louis, MO, USA). Latrunculin A was obtained from Molecular Probes (Eugene, OR, USA). The Src kinase inhibitor PP2, inactive analog PP3 and LY were obtained from Calbiochem (La Jolla, CA, USA). H 2 O 2 was obtained from Sigma-Aldrich. Cells HAEC from single donors were obtained from Cambrex Research (Walkersville, MD, USA) and maintained in Medium 199, supplemented with 20% FCS, 50 ng/ml ECGS and 64 ng/ml heparin at 37 C in a humidified incubator (5% CO 2, 95% air). Human MVEC (Cambrex Research) were cultured in EBM-2 medium supplemented with 5% FCS and EGM-2 MV SingleQuots (Cambrex Research). Cells from passages 3 8 were used. EC were grown to a confluency of 80% and incubated for 12 h in Medium 199 with 2% serum before use. To induce HLA class II expression on HAEC, cells were treated for 4 days with 1000 U/mL INF-c (R&D Systems, Minneapolis, MN, USA). HLA class II antigens were up-regulated on MVEC by incubating the cells for 4 days with 500 U/mL INF-c. Flow cytometry analysis was performed to confirm class II molecule expression by indirect immunofluorescence using mab F002 6C6G1 (One Lambda). Greater than 80% of EC expressed HLA-class II antigens following 4 days of treatment with INF-c. Patient study population The patient population consisted of 46 adult heart allograft recipients transplanted between November 1996 and December Patients were asked to participate in this Internal Review Board-approved study at the time of routine outpatient clinical visits. The study group comprised 15 patients diagnosed with AMR, 11 patients with acute cellular rejection (ACR) (Grade 3a, 3b, or 4) five patients with both AMR and ACR and 15 patients without biopsy-proven evidence of rejection. A total of 131 endomyocardial biopsies were stained for p-s6rp. Clinical data were obtained by chart review and the clinical database. Data gathered included patient and donor demographics and posttransplant outcome variables including diagnosis of rejection and CAV. Diagnosis of rejection The diagnosis of AMR was made as described in the Revision of the 1990 working formulation for the standardization of nomenclature in the diagnosis of heart rejection by Stewart et al. (1,4). Biopsies were considered positive for AMR if they showed certain histologic and immunohistochemical features. Examples of AMR and ACR are presented in Figure 1. Pathologically, AMR is distinguished from normal myocardium (Figure 1A) or ACR (Figure 1B) by a combination of histological and immunohistochemical features primarily involving myocardial capillaries (1). In AMR, the capillaries appear congested, and show evidence of endothelial injury with swelling of cytoplasm and nuclear enlargement. There is prominent linear accumulation of intravascular macrophages sometimes associated with interstitial edema and patchy interstitial hemorrhage (Figure 1C). Immunofluorescence studies, if performed, showed immunoglobulin (IgG, IgM and/or IgA) plus complement deposition (C3, C1q and/or C4d) in capillaries. If frozen tissue for immunofluorescence was not available, immunoperoxidase studies were performed to document macrophages in capillaries (Figure 1D) in combination with a vascular marker, CD31 or CD34 (Figure 1E). C4d deposition in capillaries (Figure 1F) is often also present. C4d staining of myocardial capillaries showed a strong correlation with the presence of AMR; however, occasional cases of AMR did not show significant capillary staining for C4d and rare histologically normal biopsies showed C4d staining of capillaries. The diagnosis of AMR was made in the absence of knowledge of DSA status. The diagnosis of ACR was made using International Society for Heart and Lung Transplantation criteria (16). Coronary angiography All patients underwent coronary angiography annually following heart transplantation. CAV was defined as any angiographic coronary lesion with greater than or equal to 30% luminal stenosis or occlusive CAV confirmed at autopsy. Immunosuppressive therapy During the study period of November 1996 and January 2003, immunosuppression changed. Prior to April 1999, routine immunosuppressive therapy consisted of cyclosporine, corticosteroids and azathioprine or mycophenolate mofetil. After April 1999, routine therapy included tacrolimus, corticosteroids and mycophenolate mofetil. Patients with little or no rejection were weaned off from corticosteroids after 6 months. Moderate cellular rejections (ISHLT grade 3A) were treated with oral prednisone bolus (50 mg po bid for 3 days with taper to 10 mg po bid over 2 weeks). More severe cellular rejections (ISHLT 3B, 4) were treated with OKT3 monoclonal Ab. American Journal of Transplantation 2006; 6:

3 Lepin et al. Figure 1: H & E and immunoperoxidase staining of endomyocardial biopsies from heart transplant patients. H & E staining of normal myocardium (A); cardiac allograft with ACR showing perivascular and interstitial mononuclear cells (B); AMR in cardiac allograft showing capillaries with intraluminal macrophage accumulation and EC swelling (C); staining of intravascular macrophages with CD68 immunoperoxidase stain (D); EC staining by CD31 immunoperoxidase stain (E); and diffuse staining of myocardial capillaries in cardiac allograft with AMR with C4d immunoperoxidase stain (F). A and B original magnification is 200. C, D, E and F original magnification is 400. Asymptomatic patients with AMR with a 10 15% fall in echocardiographic left ventricular ejection fraction received high-dose oral corticosteroids as previously mentioned. In addition, those patients on azathioprine were switched to mycophenolate mofetil. Symptomatic AMR patients (i.e. shortness of breath, lightheadedness) were treated with intravenous immunoglobulin in a split dose of 2g/kg (maximum 140 g) over 2 days in association with high-dose intravenous corticosteroids (500 mg methylprednisolone daily for 3 days). For those with hemodynamic compromise (requiring inotropic support), part or all of the following combination therapy was administered: intravenous methylprednisolone (500 mg daily for 3 days); cyclophosphamide (1 3 mg/kg i.v. daily); plasmapheresis (daily for 5 days), OKT3 (5 mg daily for 14 days given immediately after each plasmapheresis); and intravenous herparinization for 7 days. Immunohistochemical analysis of phosphorylated S6 ribosomal protein and grading. Immunohistochemical staining was carried out as previously described (17). Endomyocardial biopsy sections were deparaffinized in xylene and rehydrated in graded alcohols. Antigen recovery was performed by placing tissue sections in a steamer with 10 mm sodium citrate buffer (ph 6.0) for 25 min. Endogenous peroxidase activity was inhibited by incubation in 3% hydrogen peroxidase in methanol for 15 min. Sections were then blocked for 30 min at room temperature with 10% normal goat serum (NGS) diluted in PBS. Phospho-S6 ribosomal protein (Serine 235/236 ) Ab (Cell Signaling Technology) was diluted 1/50 in 3% NGS and 100 ll was added to each section. The primary Ab was incubated overnight at 4 C. The secondary Ab, a biotinylated goat anti-rabbit IgG (Vector) diluted 1:200 in 3% NGS was incubated for 40 min at room temperature. After three washes in PBS, sections were incubated for 30 min with horseradish peroxidase avidin D (HRP, Vector) diluted 1:1000 with PBS. After three 5-min washes of PBS, the sections were developed with DAB kit (Vector). Slides were counterstained with dilute hematoxylin, rinsed with ammonia and then with tap water. Sections were dehydrated with graded ethanol, cleared in xylene and then coverslipped. Cardiac biopsies were scored by two blinded observers. Positive EC staining for S6RP was scored on a scale of 0 4 (0 = no staining; 1 = rare staining; 2 = focal staining; 3 = multifocal staining; 4 = diffuse staining). A score of 3 or greater was considered positive. Myocyte and leukocyte p-s6rp staining was scored on a scale from 0 to2(0= no staining; 1 = rare staining; 2 = multifocal staining). A score of 2 was considered positive. Evaluation of anti-hla antibodies The most recent serum obtained prior to transplantation was studied for the presence of anti-hla antibodies. This time period was selected for evaluation since the decision to transplant sensitized heart transplant candidates at UCLA is based on antibody and crossmatch tests performed on the serum obtained immediately prior to transplantation. In the study group, this represented sera collected within a 3-month period prior to transplantation. Sera were tested for the presence of anti-hla class I or class II IgG Ab and the positive sera subsequently tested for anti-hla Ab specificity using Luminex reagents purchased from Tepnel LifeCodes (Stamford, CT, USA) according 1562 American Journal of Transplantation 2006; 6:

4 p-s6rp is a Marker of Antibody-Mediated Rejection Figure 2: Ligation of HLA class I and class II molecules on EC results in increased phosphorylation of S6RP. EC were treated with different concentrations of anti-class I mab W6/32 or anti-class II mab F002 6C6G1 (0.01, 0.1 and 1 lg/ml) for 1, 10, 30 or 60 min or untreated (UN). For each experimental condition, total cell lysates were resolved by SDS-PAGE and immunoblotted with an Ab specific to p-s6rp and compared to the b-actin or total S6RP equal loading control. Quantification was performed by scanning densitometry. Histogram values are normalized to equal loading results and expressed as the percentage of the maximal increase in serine phosphorylation. The results are representative of three independent experiments. (A) HAEC treated with anti-class I Ab; (B) MVEC treated with anti-class I Ab; (C) HAEC treated with IFN-c followed by stimulation with anti-class II Ab; (D) MVEC treated with IFN-c followed by stimulation with anti-class II Ab; (E) HAEC treated with 300 mm H 2 O 2 ; (F) MVEC treated with 300 mm H 2 O 2. American Journal of Transplantation 2006; 6:

5 Lepin et al. to the manufacturer s protocol. Particle fluorescence was assessed by Luminex100 IS (Luminex, Austin, TX, USA). Class I and Class II single antigen beads were used to identify specificities in high panel reactive antibody patients (PRA > 75%), using LABScreen Single Antigen HLA Class I and II Ab Detection Test (One Lambda Inc.). Single-antigen particle fluorescence was measured on a Becton Dickenson FACScan. Statistical analysis Patient characteristics, their posttransplant statuses and outcomes among rejectors and nonrejectors were summarized using means ± standard deviations for continuous variables and counts (percents) for categorical variables. Associations between groups were tested via Wilcoxon s rank-sum test and Fisher s exact test for continuous and categorical variables, respectively. Actuarial estimates for CAV-free survival were calculated by the Kaplan-Meier method, and the log-rank test was used to test for equality between the groups. All p-values were two-sided and, when comparing groups, a significance level of 0.05 was used. Results Ab ligation of HLA class I and class II molecules stimulates S6RP phosphorylation in EC To determine the effect of anti-class I Ab on phosphorylation of S6RP, normal-cultured HAEC or MVEC were treated with varying concentrations of anti-hla class I Ab ranging from 0.01 to 1 lg/ml for different time points and immunoblotted with a phospho-specific anti-s6rp Ab. Class I ligation on the surface of HAEC resulted in a time- and dose-dependent increase in phosphorylation of S6RP at Serine 235/236 that peaked at 10 min, and declined thereafter (Figure 2A). Maximal phosphorylation of S6RP was achieved using 0.1 lg/ml of mab W6/32. Densitometric scanning revealed a 4.4-fold increase in phosphorylation of S6RP following treatment with 0.1 lg/ml W6/32 when compared with untreated EC. Similar results were obtained when MVEC were treated with anti-class I Ab and studied for S6RP phosphorylation (Figure 2B). Class I ligation resulted in a maximal stimulation of p-s6rp at 1 min following treatment with 0.1 lg/ml of mab W6/32. Similar studies were conducted to determine the capacity of anti-hla class II Ab to stimulate phosphorylation of S6RP. To induce HLA class II expression, HAEC and MVEC were cultured for 4 days in INF-c followed by stimulation with varying concentrations of anti-hla class II Ab ranging from 0.01 to 1 lg/ml for different time points. Treatment of class II-positive HAEC or MVEC with anti-class II Ab stimulated a time-dependent increase in phosphorylation of S6RP with maximal activation of min after ligation (Figure 2C, D). Densitometric scanning revealed a 2-fold increase in phosphorylation of S6RP in HAEC and an 8-fold increase in p-s6rp in MVEC following treatment with 0.1 lg/ml mab F002 6C6G1. The treatment of HAEC or MVEC with H 2 O 2, at concentrations (300 mm) known to induce oxidative stress and activation of the S6 kinase pathway (18), resulted in increased S6RP phosphorylation (Figure 2E, F). Figure 3: HLA class I-mediated phosphorylation of S6RP is dependent on Src, PI3K and actin cytoskeleton integrity. Serumstarved HAEC were pretreated for 30 min in the presence (+) or absence ( ) of 12.5 lm PP2, 12.5 lm PP3, 1.25 lm cytochalasin D, 0.5 lm latrunculin A, 0.1 lm wortmannin or 25 lm LY before stimulation for 10 min with 1 lg/ml W6/32 (+) or untreated ( ). For each experimental condition, total cell lysates were resolved by SDS-PAGE and immunoblotted with an Ab specific to p-s6rp and compared to the equal loading control. The results are representative of three independent experiments. p-s6rp is a downstream target of the Src/FAK/PI3K HLA class I signaling pathway To establish whether Src tyrosine kinase activity is required for HLA-mediated phosphorylation of S6RP, EC were preincubated in PP2, a selective inhibitor of Src, followed by stimulation with mab W6/32. As shown in Figure 3, pretreatment of EC with PP2, but not the inactive analogue PP3, blocked class I-induced S6RP phosphorylation at Serine 235/236. The role of FAK in class I stimulation of S6RP was investigated by using two distinct pharmacological agents known to disrupt the actin cytoskeleton and prevent FAK phosphorylation. Treatment of EC with either cytochalasin D or Latrunculin A abrogated phosphorylation of S6RP in response to class I ligation (Figure 3). We next examined the role of PI3K in class I-mediated phosphorylation of S6RP. Quiescent EC were pretreated with wortmannin or LY294002, two PI3K inhibitors with different mechanisms of action, and stimulated with anti-hla class I Ab. Treatment with either Ly or wortmannin significantly inhibited HLA class I-induced phosphorylation of p-s6rp. These results support the involvement of Src, FAK and PI3K in class I-mediated phosphorylation of S6RP. Immunostaining of p-s6rp in endomyocardial biopsies and correlation with acute rejection The use of phospho-specific Ab to analyze PI3K/Akt signaling in vivo has been successfully used in cancer research (17). Using a similar protocol, we investigated the staining pattern of an anti-p-s6rp Ab in 131 endomyocardial biopsies from 46 adult cardiac allograft recipients. The characteristics of the patient study group are presented in Table 1. Fifteen patients were diagnosed with AMR, 11 were diagnosed with ACR, five patients were diagnosed with both AMR and ACR and 15 patients had no evidence of rejection. Nine of 20 patients demonstrated AMR in two or more endomyocardial biopsy samples. Six of the patients with ACR were males and five were females. Seven of the 1564 American Journal of Transplantation 2006; 6:

6 Table 1: Patient characteristics Rejector p-s6rp is a Marker of Antibody-Mediated Rejection Nonrejector ACR AMR AMR and ACR Missing (n = 11) (n = 15) (n = 5) (n = 15) p-value 1 (n = 46) Age, years (mean ± SD) 54 ± ± ± ± years > Gender (males/females) 6/5 7/8 2/3 11/ HLA-ABDR mismatches (mean ± SD) 5.0 ± ± ± ± (9%) 0mm Retransplants (primary:regrafts) 10:1 15:0 5:0 13: Pretransplant sensitization Pre-PRA T (mean ± SD) (%) 0.2 ± ± ± ± (15%) Pre-PRA B (mean ± SD) (%) 5.0 ± ± ± ± (15%) Pre-XM [(%+)] Cytotoxicity 0:9 (0%) 0:11 (0%) 0:5 (0%) 0:13 (0%) 8 (17%) Flow 0:10 (0%) 3:9 (25%) 0:4 (0%) 3:11(21%) (13%) 1 p-values denote statistical significances in tests between rejectors (all forms combined into a single group) versus nonrejectors. Wilcoxon rank-sum test was used for continuous variables and Fisher s exact test was used for discrete variables. patients with AMR were males and eight were females compared to 11 males and four females in the nonrejection study group. The average age of patients with rejection (combined ACR + AMR) was lower (51 years) compared to the nonrejectors (60 years) (p = 0.04). There was also a higher degree of HLA-A, -B, -DR mismatching in the rejection study group (combined ACR + AMR) compared to nonrejectors (4.8 vs. 3.9, p = 0.05). One of 11 recipients with ACR and two of 15 in the nonrejection group were recipients of secondary allografts. The effect of presensitization on rejection was studied using the most recent serum obtained prior to transplantation. The average T- and B-panel reactive Ab was 0.2% and 5.0% in ACR, 3.3% and 16.2% in AMR, 0% and 3% in patients with both ACR and AMR and 0.7% and 1.9% in nonrejectors, respectively. None of the patients exhibited a pretransplant-positive cytotoxicity crossmatch with the donor. Also, a positive T- and/or B-cell flow cytometry crossmatch was not associated with rejection (p = 0.65). Specifically, three of 12 patients with just AMR and three of 14 patients without rejection displayed a positive pretransplant T- and/or B-cell flow cytometry crossmatch. None of the patients diagnosed with ACR exhibited pretransplant-positive T- and/or B-cell flow crossmatches. Multivariate analysis showed no association between diagnosis of rejection and gender, number of transplants and pretransplant sensitization between rejectors and nonrejectors (Table 1). A total of 131 paraffin-fixed endomyocardial biopsies from 46 adult cardiac allograft recipients were stained for phosphorylated S6RP. Figure 4 shows the staining pattern of p-s6rp in representative cardiac biopsy samples. Biopsies with rare or focal EC staining were considered negative for p-s6rp (Figure 4A C). Multifocal and diffuse EC staining were considered positive for p-s6rp (Figure 4D, E). Background staining of the intercalated discs for p- S6RP was consistently seen in all biopsies (Figure 4A E). Strong staining of necrotic myocytes and necrotic fat cells was also observed in biopsies with myocardial damage (Figure 4F, H). Positive staining of interstitial and intravascular leukocytes for p-s6rp was also seen in some biopsies (Figure 4G). The association between p-s6rp staining of capillary ECs and graft outcome was analyzed (Table 2). A strong association between AMR and expression of p-s6rp in capillary EC was observed (p < ). Nineteen of the 20 patients with AMR showed multifocal or diffuse capillary staining for p-s6rp. No significant association was observed between ACR (ISHLT Grades 3 or 4) and p-s6rp staining of capillary EC in endomyocardial biopsies. Positive p-s6rp staining of myocytes was observed in 11 of 20 biopsies with AMR and eight of 16 biopsies with ACR (Table 2, p = NS). Inflammatory infiltrates were scored in 43 of the patients in the study group. Ten of the 20 patients (50%) diagnosed with AMR showed p-s6rp staining of leukocyte infiltrates (p = 0.05). There was no association between p-s6rp-positive infiltrates and ACR (p = 0.32). The results presented above demonstrate a positive association between p-s6rp staining and diagnosis of AMR American Journal of Transplantation 2006; 6:

7 Lepin et al. Figure 4: p-s6rp immunoperoxidase staining of endomyocardial biopsies from heart transplant patients. p-s6rp staining grades are shown: grade 0 (A), grade 1 (B), grade 2 (C), grade 3 (D) and grade 4 (E). Staining of injured myocytes at biopsy site (F), intravascular leukocytes (G) and myocytes with coagulative necrosis secondary to ischemic injury (H). The original magnification is 200 (G 400). using all available biopsy data for each patient. In Table 3, we have more precisely defined rejection status by analyzing the concurrent presence of p-s6rp and rejection type. Eleven of the 13 patients diagnosed with AMR exclusively were p-s6rp-positive compared to five of 24 without evidence of rejection (p = 0.001). No association was found between the diagnosis of ACR and p-s6rp status. The relationship between C4d, p-s6rp and AR was also examined (Table 4). First-time biopsies from 31 patients that were jointly assessed for rejection and p-s6rp status were analyzed for capillary deposition of C4d by immunostaining of paraffin-embedded tissues. Four of 5 patients diagnosed with AMR were C4d-positive compared to four of 26 patients without AMR (p = 0.01, Table 4). There was also an association between C4d positivity and p-s6rp staining. Seven of 14 patients that were positive for p-s6rp were C4d-positive compared to one of 17 patients without p-s6rp (p = 0.01). The temporal relationship between p-s6rp EC staining and graft rejection was next examined (Figure 5). For this analysis we tested the next two to three available biopsy samples obtained after an initial biopsy that were jointly 1566 American Journal of Transplantation 2006; 6:

8 p-s6rp is a Marker of Antibody-Mediated Rejection Table 2: Association of p-s6rp staining with acute rejection ACR+ AMR+ AMR (Grade 3+) ACR S6RP EC+, > S6RP EC p-value < S6RP myocytes+, > S6RP myocytes p-value S6RP leukocytes+, > S6RP leukocytes p-value A score of >2 is considered positive for p-s6rp. Table 3: Association between p-s6rp and acute rejection at the time of first biopsy AMR+ACR AMR ACR+ AMR ACR S6RP EC+, > S6RP EC p-value Table 4: Association of p-s6rp or rejection with C4d staining Initial biopsy (n = 31) 1 AMR+ AMR S6RP+ S6RP C4d C4d p-value Thirty-one patients had all the three measures (C4d, AMR and p-s6rp) reported for initial biopsy. positive for S6, AMR or ACR. There were 12 patients with solely AMR and four patients with solely ACR, diagnosed at the time of their initial biopsy. No ACR patient exhibited p-s6rp beyond their initial biopsy whereas six AMR patients (54%) (ID04, ID05, ID06, ID08, ID11 and ID12) demonstrated persistent p-s6rp (p = 0.09). The median duration of persistent p-s6rp in these six patients with AMR was 133 days (range days). Sustained phosphorylation of S6RP was observed primarily in patients with multiple episodes of AMR. Thus, six of the 11 patients with AMR experienced two or more episodes of AMR and four of these patients had evidence of persistent p-s6rp. Five of these six patients with AMR and persistent p-s6rp also demonstrated circulating anti-hla Ab (ID04, ID05, ID06, ID08 and ID11). With the exception of patient 11, all generated DSA to class I and/or class II donor antigens. None of these patients with ACR produced posttransplant anti-hla Ab. We performed Kaplan-Meier estimates of CAV-free survival rates for heart transplant recipients who survived beyond 1 year posttransplant (n = 40). In seven patients with AMR and persistent p-s6rp, 58% were estimated to develop CAV within 3 years after transplantation compared to the remaining 33 patients among whom only 29% were expected to develop CAV. This difference was not significant by log rank statistics and requires further study on a larger sample number to determine the clinical relevance of this finding. Association of p-s6rp staining with circulating anti-hla Ab Sera samples were obtained at the time of cardiac biopsy and evaluated for the presence of anti-hla class I and class II Ab. As shown in Table 5, 16 of the 26 patients demonstrating multifocal or diffuse capillary staining of p-s6rp produced posttransplant Ab to class I and/or class II Ab. Ten of these patients showed Ab to class I antigens and 16 showed Ab to class II antigens. Ab production to class II antigens was positively associated with p-s6rp-positive staining (p = 0.02). There was no significant association between Ab to class I alloantigens and p-s6rp staining. Analysis of the association between DSA production and EC staining of p-s6rp confirmed a strong association between generation of DSA to class II antigens and p-s6rp staining (p = 0.01). There was no association between EC staining of p-s6rp with DSA to HLA class I antigens (p = 0.47). Discussion Morphologic classification of transplant rejection has limited sensitivity and reproducibility. Therefore, a new direction in the diagnosis of transplant rejection is the switch from morphological classification to an approach based on molecular biomarkers. In light of the development of new immunosuppressive agents inhibiting signaling pathways involved in transplant rejection, this goal is even more important for both diagnosis and selection of appropriate treatment strategies. To approach this objective we used phosphorylation-specific Ab profiling to elucidate the effects of anti-hla Ab-induced signal transduction in EC in vivo and in vitro models. Our data show that cross-linking of MHC class I and class II molecules by anti-hla Ab on the surface of cultured HAEC and MVEC results in increased phosphorylation of S6RP, a protein that promotes growth and proliferation of cells. Furthermore, we found a strong association between the diagnosis of AMR and the presence of phosphorylated S6RP in the capillary endothelium of the graft (p < ). These results demonstrate that p-s6rp is a useful marker to diagnose and classify AR episodes with a humoral component that will likely require specific immunotherapy. Studies from our group and others have shown that treatment of EC with anti-class I or class II Ab triggers a prosurvival signaling cascade resulting in phosphorylation of PI3K and Akt (9 14). Akt is known to activate mtor, which in turn regulates protein synthesis, and cell proliferation through the phosphorylation and activation of S6 kinase and S6RP. Consequently, we examined whether Ab ligation of class I and/or class II molecules induces phosphorylation of the downstream target S6RP. Our in vitro American Journal of Transplantation 2006; 6:

9 Lepin et al. Figure 5: Temporal relationship between p-s6rp EC staining, anti-hla Ab production and rejection. Upper panel, 10 AMR+/ACR cases; lower panel, 4 ACR+/AMR cases. Score of >2 is positive for p-s6rp staining. Symbols: Class I and II Ab; Class I Ab; Class II Ab; Negative Ab testing; DSA; Positive biopsy for AMR; Negative biopsy for AMR; Positive biopsy for ACR; Negative biopsy for ACR. Table 5: Association between p-s6rp and anti-hla antibody production Association of EC staining and HLA antibody production (both class I/II) Ab+ (ever) Ab Ab I+ (ever) Ab I Ab II+ Ab II S6RP EC+ (ever) S6RP EC p-value Association of EC staining and donor-specific antibody production (both class I/II) DSA+ (ever) DSA DSAI+ DSAI DSAII+ DSAII S6RP EC+ (ever) S6RP EC p-value American Journal of Transplantation 2006; 6:

10 p-s6rp is a Marker of Antibody-Mediated Rejection studies revealed that S6RP becomes immediately phosphorylated on serine residues in response to class I and class II ligation. Preincubation of EC in PP2-abrogated class I-mediated phosphorylation of p-s6rp providing evidence that activation of S6RP is dependent upon Src. Signal transduction induced by ligation of class I molecules on the surface of EC stimulates phosphorylation of Src and two elements of the focal adhesion signaling complex, FAK and paxillin. Src binds to the autophosphorylation site of FAK at Tyr 397 and facilitates Src-dependent phosphorylation of tyrosine residues Tyr 576 and Tyr 577 as well as Tyr 397 the autophosphorylation site (19). Phosphorylation of FAK at Tyr 397 promotes the assembly of a signaling complex with PI3K and provides a mechanism for class I activation of the PI3K/Akt signaling pathway and downstream targets S6 kinase and S6RP (15,20). Our data also showed that treatment of EC with either cytochalasin D or latrunculin A, at concentrations known to disrupt the actin cytoskeleton, inhibited class I-mediated increases in phosphorylation of S6RP indicating that an intact cytoskeleton is required for class I-induced phosphorylation of S6RP. In addition, cross-linking of MHC class I molecules in the presence of two distinct pharmacological inhibitors of PI3K, wortmannin and LY294002, resulted in a strong inhibition of phosphorylated S6RP levels. Recent studies by Le Bas- Bernardet et al. showed that similar to class I signaling, ligation of HLA class II molecules on vascular EC with anti-hla Ab induces activation of the PI3K/Akt pathway (13). Taken together, these results provide evidence that class I and class II-induced activation of S6RP is mediated by the PI3K/Akt pathway and consistent with a model in which anti-hla Ab-mediated clustering of class I and class II molecules stimulates protein phosphorylation and activation of downstream targets including PI3K, Akt, mtor, S6 kinase and S6RP causing EC activation and proliferation. Consistent with this interpretation, phosphorylation of ribosomal S6 protein and p70s6k has been previously shown to be critical steps in EC proliferation induced by growth factors (21 23). Furthermore, recent studies have implicated S6 kinase and S6RP in different cardiac pathologies (24). For example, cardiac xenografts undergoing acute vascular rejection showed increased p70s6k activity (25). Ultrastructural analysis of xenograft EC revealed a dramatic expansion of the rough endoplasmic reticulum typical of activated EC suggesting an increase in protein translation (25). Phosphorylation of a number of specific proteins of the translational apparatus, including 40 S ribosomal protein and p70s6k kinase is an early obligatory step in the mitogenic response by initiating and increasing the rate of protein synthesis (26). Many recent studies have focused on the heterogeneity of EC and differences in the responsiveness of EC from different organs as well as different locations within a given vasculature (27,28). Emerging data demonstrate that EC have specialized properties and molecular signaling that appear to be dependent upon the anatomical location. Based on these observations, we sought to determine if there were differences in the responsiveness of large vessel EC and microvascular EC to anti-hla Ab ligation. Our results show that S6RP phosphorylation was enhanced in both HAEC and MVEC in response to ligation of class I and class II molecules with anti-hla Ab. This response was observed at all doses of Ab studied and suggested to us that p-s6rp may be a potential molecular marker of AMR in vivo. Therefore, based on our in vitro findings we further evaluated p-s6rp staining in cardiac allograft biopsies with and without evidence of AMR and ACR. We found that 19 of 20 patients with AMR showed multifocal or diffuse capillary staining for p-s6rp. Furthermore, sustained capillary EC staining of p-s6rp was found in cardiac transplant recipients with multiple episodes of AMR indicating the potential of this biomarker to identify patients with chronic Ab-mediated rejection. Additional, although significantly weaker, links were found between p-s6rp and ACR. In the case of ACR, the time course of S6RP phosphorylation was short lived. Our data also suggest that persistent capillary staining of p-s6rp may identify a subgroup of patients with increased risk for the development of CAV and requires further validation in a larger study group. Finally, phosphorylated S6RP may be useful for stratifying patients for targeted kinase inhibitor therapy such as rapamycin. The characterization of p-s6rp in cardiac biopsies was associated with several unexpected findings and caveats. In addition to EC staining, leukocytes within the interstitium and macrophages within the vessels of biopsies diagnosed with rejection demonstrated expression of p- S6RP. Previous studies demonstrated increased phosphorylation of S6 kinase and S6RP in activated T cells following stimulation with anti-cd3 or INF-c (29,30). In addition, the Akt/mTOR signaling pathway has been shown to be induced in macrophages by phosphatidic acid, an essential regulator of inflammatory response (31). Therefore, expression of p-s6rp by graft infiltrating cells may be a characteristic of activated immune cells engaged in the rejection process. There was also S6RP staining in biopsies with myocardial damage including injured myocytes at the biopsy site, myocytes with coagulative necrosis secondary to ischemic injury and necrotic fat cells. Background staining of myocyte-intercalated discs was observed in all biopsies studied. We speculate that increased staining of p-s6rp in injured myocytes may be linked to protein synthesis. For example, the background staining of p-s6rp in intercalated discs could be explained by the specific distribution of ribosomes in this cell type (32). Ribosomes have been shown to be colocalized with desmin, an important constituent of the intercalated discs (32). Generally, colocalization of ribosomes and myosin mrna in the proximity of myofibrils suggests that in this cell type, translation of myosin mrna appears to be located close to the area where the protein will be used. Since it has been shown that phosphorylation of S6RP is linked to the translation of specific mrna, it is possible that strong phosphorylation of S6RP in regions of intercalated discs corresponds to the overexpression of protein(s) forming at the intercalated American Journal of Transplantation 2006; 6:

11 Lepin et al. disc. Previous studies have reported increased expression of ribosomes in necrotic tissues (32) and this could explain why we also observed strong p-s6rp staining in injured myocytes. Increased expression of ribosomes in general and p-s6rp in particular could be involved in the cellular response to injury. We did not see a correlation between increased p-s6rp staining of myocytes and diagnosis of rejection. Therefore, the diagnostic relevance of this finding is not clear. In light of these findings, several pitfalls of p-s6rp staining should be considered. Only capillary EC staining appears to be a reliable adjunct marker for diagnosis of AMR in cardiac allografts. Background p- S6RP staining of intercalated discs in nonrejecting hearts is always found and should not be considered specific for AMR. Although p-s6rp staining of myocytes occurs during AMR, this staining pattern is also observed in biopsies with evidence of ischemic injury or sampling of a previous biopsy site. Finally, graft infiltrating leukocytes can be positive for p-s6rp and although cases with infiltrates should be considered suspicious for rejection, classification for AMR would require additional histological and immunohistochemical evidence. Our results and previous studies indicate that Ab ligation of class I and class II antigens can activate the PI3K/Akt pathway (7 13). Therefore, we monitored the study group for the generation of posttransplant anti-hla class I and class II Ab and analyzed the relationship between Ab production and p-s6rp immunostaining. Although the majority of patients with positive staining for p-s6rp produced Ab, the correlation was not perfect. Seventy-five percent of patients demonstrating positive capillary staining of p-s6rp generated anti-hla Ab. Thus, the presence of p-s6rp in the absence of circulating alloab and vice versa was found in a number of cases. Furthermore, Ab production to class II antigens was positively associated with p-s6rp-positive staining; however, no significant association was found between p-s6rp and Ab to class I alloantigens. These findings may be explained by a number of factors. It is possible that in the absence of HLA Ab, AMR was mediated by non-hla Ab. Non-HLA Ab reactive with EC proteins including major histocompatibility class I chain-related A and vimentin have been shown to be associated with AMR and CAV (33). Blocking or adsorption of anti-hla class I Ab by donor antigens either on the graft or in the circulation may also explain our inability to detect circulating anti-class I Ab (34 36). Alternatively, the titer of circulating class I Ab may be below the threshold level of detection in our assays, whereas, accumulation of Ab in the graft is sufficient for signal transduction in vivo. Additional studies in a larger study population are required to determine if one or more of these scenarios explain the discrepancy. Furthermore, characterization of the combined influence of pre- and posttransplant sensitization and p-s6rp on the development of AMR and CAV await analysis in a larger study group. In summary, this study emphasizes the potential value of using a proteomic approach to identify signal transduction pathways involved in transplant rejection. Phosphorylated S6RP appears to be a useful adjunct molecular marker to diagnose AMR that will likely require specific and intensive immunotherapy. Acknowledgments This research was supported by National Institutes of Health grant RO1AI42819 to E.F.R and American Heart Association Grant-In-Aid Y to E.F.R. The authors thank Dr. Jarhow Lee, One Lambda Inc., for providing monoclonal antibodies for this work. References 1. Michaels PJ, Espejo ML, Kobashigawa J et al. Humoral rejection in cardiac transplantation: Risk factors, hemodynamic consequences and relationship to transplant coronary artery disease. J Heart Lung Transplant 2003; 22: Hammond EH, Yowell RL, Nunoda S et al. Vascular (humoral) rejection in heart transplantation: Pathologic observations and clinical implications. J Heart Transplant 1989; 8: Reed EF, Demetris AJ, Hammond E et al. Acute antibody mediated rejection of cardiac transplants. J Heart Lung Transplant 2006; 25: Stewart S, Winters GL, C. FM, Tazelaar HD. Revision of the 1990 working formulation for the standardization of nomenclature in the diagnosis of heart rejection. J Heart Lung Transplant 2005; 24: Rose AG, Cooper DK. A histopathologic grading system of hyperacute (humoral, antibody-mediated) cardiac xenograft and allograft rejection. J Heart Lung Transplant 1996; 15: Reinsmoen NL, Nelson K, Zeevi A. Anti-HLA antibody analysis and crossmatching in heart and lung transplantation. Transpl Immunol 2004; 13: Bian H, Reed EF. Alloantibody-mediated class I signal transduction in endothelial cells and smooth muscle cells: Enhancement by IFN-gamma and TNF-alpha. J Immunol 1999; 163: Bian H, Harris PE, Reed EF. Ligation of HLA class I molecules on smooth muscle cells with anti-hla antibodies induces tyrosine phosphorylation, fibroblast growth factor receptor expression and cell proliferation. Int Immunol 1998; 10: Jin YP, Fishbein MC, Said JW et al. Anti-HLA class I antibodymediated activation of the PI3K/Akt signaling pathway and induction of Bcl-2 and Bcl-xL expression in endothelial cells. Hum Immunol 2004; 65: Jin YP, Jindra PT, Gong KW, Lepin EJ, Reed EF. Anti-HLA class I antibodies activate endothelial cells and promote chronic rejection. Transplantation 2005; 79(3 Suppl): S19 S Narayanan K, Jaramillo A, Phelan DL, Mohanakumar T. Preexposure to sub-saturating concentrations of HLA class I antibodies confers resistance to endothelial cells against antibody complement-mediated lysis by regulating Bad through the phosphatidylinositol 3-kinase/Akt pathway. Eur J Immunol 2004; 34: Salama AD, Delikouras A, Pusey CD et al. Transplant accommodation in highly sensitized patients: A potential role for Bcl-xL and alloantibody. Am J Transplant 2001; 1: Le Bas-Bernardet S, Coupel S, Chauveau A, Soulillou JP, Charreau B. Vascular endothelial cells evade apoptosis triggered by human leukocyte antigen-dr ligation mediated by allospecific antibodies. Transplantation 2004; 78: American Journal of Transplantation 2006; 6:

12 p-s6rp is a Marker of Antibody-Mediated Rejection 14. Jin YP, Singh RP, Du ZY, Rajasekaran AK, Rozengurt E, Reed EF. Ligation of HLA class I molecules on endothelial cells induces phosphorylation of Src, paxillin, and focal adhesion kinase in an actin-dependent manner. J Immunol 2002; 168: Hay N, Sonenberg N. Upstream and downstream of mtor. Genes Dev 2004; 18: Billingham ME, Cary NR, Hammond ME et al. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: Heart Rejection Study Group. The International Society for Heart Transplantation. J Heart Transplant 1990; 9: Choe G, Horvath S, Cloughesy TF et al. Analysis of the phosphatidylinositol 3 -kinase signaling pathway in glioblastoma patients in vivo. Cancer Res 2003; 63: Bae GU, Seo DW, Kwon HK et al. Hydrogen peroxide activates p70(s6k) signaling pathway. J Biol Chem 1999; 274: Turner CE. Paxillin and focal adhesion signalling. Nat Cell Biol 2000; 2: E231 E Thomas G. The S6 kinase signaling pathway in the control of development and growth. Biol Res 2002; 35: Beitz JG, Kim IS, Calabresi P, Frackelton AR, Jr. Human microvascular endothelial cells express receptors for platelet-derived growth factor. Proc Natl Acad Sci U S A 1991; 88: Kanda S, Hodgkin MN, Woodfield RJ, Wakelam MJ, Thomas G, Claesson-Welsh L. Phosphatidylinositol 3 -kinase-independent p70 S6 kinase activation by fibroblast growth factor receptor-1 is important for proliferation but not differentiation of endothelial cells. J Biol Chem 1997; 272: Vinals F, Chambard JC, Pouyssegur J. p70 S6 kinase-mediated protein synthesis is a critical step for vascular endothelial cell proliferation. J Biol Chem 1999; 274: Hiraoka E, Kawashima S, Takahashi T, Rikitake Y, Hirase T, Yokoyama M. PI 3-kinase-Akt-p70 S6 kinase in hypertrophic responses to leukemia inhibitory factor in cardiac myocytes. Kobe J Med Sci 2003; 49: Holzknecht ZE, Kuypers KL, Plummer TB et al. Apoptosis and cellular activation in the pathogenesis of acute vascular rejection. Circ Res 2002; 91: Flotow H, Thomas G. Substrate recognition determinants of the mitogen-activated 70K S6 kinase from rat liver. J Biol Chem 1992; 267: Hendrickx J, Doggen K, Weinberg EO, Van Tongelen P, Fransen P, De Keulenaer GW. Molecular diversity of cardiac endothelial cells in vitro and in vivo. Physiol Genomics 2004; 19: Moldobaeva A, Wagner EM. Difference in proangiogenic potential of systemic and pulmonary endothelium: Role of CXCR2. Am J Physiol Lung Cell Mol Physiol 2005; 288: L1117 L Lekmine F, Sassano A, Uddin S et al. Interferon-gamma engages the p70 S6 kinase to regulate phosphorylation of the 40S S6 ribosomal protein. Exp Cell Res 2004; 295: Cham CM, Gajewski TF. Glucose availability regulates IFN-gamma production and p70s6 kinase activation in CD8+ effector T cells. J Immunol 2005; 174: Lim HK, Choi YA, Park W et al. Phosphatidic acid regulates systemic inflammatory responses by modulating the Akt-mammalian target of rapamycin-p70 S6 kinase 1 pathway. J Biol Chem 2003; 278: Larsen TH, Hesketh JE, Rotevatn S, Greve G, Saetersdal T. Ribosome distribution in normal and infarcted rat hearts. Histochem J 1994; 26: Rose ML. De novo production of antibodies after heart or lung transplantation should be regarded as an early warning system. J Heart Lung Transplant 2004; 23: Lee PC, Terasaki PI, Takemoto SK et al. All chronic rejection failures of kidney transplants were preceded by the development of HLA antibodies. Transplantation 2002; 74: Zhang Q, Liang LW, Gjertson DW et al. Development of posttransplant antidonor HLA antibodies is associated with acute humoral rejection and early graft dysfunction. Transplantation 2005; 79: Suciu-Foca N, Reed E, Marboe C et al. The role of anti-hla antibodies in heart transplantation. Transplantation 1991; 51: American Journal of Transplantation 2006; 6: