Anti-HLA Abs have been shown to contribute to the process of transplant vasculopathy by binding to HLA class I molecules expressed by the endothelial and smooth muscle cells of the graft and transducing intracellular signals that elicit cell proliferation. The aim of this study was to determine the role of mammalian target of rapamycin (mTOR) in HLA class I-induced endothelial cell proliferation and to explore in depth the relationship between mTOR complexes and their downstream targets following ligation of HLA class I molecules by anti-HLA Abs. We used small interfering RNA technology to abrogate mTOR, rapamycin-insensitive companion of mTOR (rictor), or regulatory associated protein of mTOR (raptor) to study the function of these gene products to activate proteins involved in MHC class I-induced cell proliferation and survival. Knockdown of mTOR inhibited class I-mediated phosphorylation of proteins downstream of mTOR complex 1 and mTOR complex 2. Furthermore, knockdown of mTOR, rictor, or raptor blocked HLA class I-induced endothelial cell proliferation. Long-term pretreatment with the mTOR inhibitor rapamycin significantly blocked both mTOR-raptor and mTOR-rictor complex formation. Interestingly, rapamycin also blocked class I-induced Akt phosphorylation at Ser473 and Bcl-2 expression. These results support the role of anti-HLA Abs in the process of transplant vasculopathy and suggest that exposure of the graft endothelium to anti-HLA Abs may promote proliferation through the mTOR pathway.
Chronic allograft rejection is the major limitation to solid organ transplantation (1). Chronic rejection manifests itself as accelerated transplant arteriosclerosis, a progressive vasculo-occlusive disease, resulting in ischemic injury and deterioration of organ function. The histologic appearance of transplant arteriosclerosis shows increased proliferation and hyperplasia of vascular smooth muscle and endothelial cells (EC).3 There are several lines of evidence supporting a role for anti-donor HLA Abs in both acute and chronic allograft rejection. Characteristics features of acute Ab-mediated rejection in renal and heart transplantation include Ig and C4d deposition in the vessels of the graft and the presence of anti-donor HLA Abs in the recipient’s circulation (2, 3). Furthermore, patients developing anti-donor HLA Abs following transplantation are at increased risk of transplant arteriosclerosis and graft loss (4, 5). Moreover, passive transfer of sera containing anti-donor HLA Abs has been shown to accelerate the development of transplant arteriosclerosis in experimental models of transplantation (6, 7, 8). Based on these findings, we hypothesize that anti-HLA Abs contribute to the development of acute and chronic Ab-mediated rejection by binding to the class I molecules on the endothelium of the graft and transducing intracellular signals that stimulate cell activation and proliferation.
Anti-HLA Abs have been shown to transduce signals that stimulate the phosphorylation of multiple proteins in endothelial cells that in turn mediate functional changes in the cell (9, 10). One of the earliest events is the activation of RhoA and stress fiber formation. Rho plays an important role in inducing the assembly of contractile actin and stress fibers needed for regulating cell migration, survival, and proliferation (11). Inhibition of RhoA blocks downstream activation of the PI3K/Akt pathway, emphasizing its importance in cytoskeletal rearrangement and cell proliferation (12). Engagement of class I molecules by anti-HLA Abs also stimulates phosphorylation of Src, focal adhesion kinase (FAK), and paxillin and assembly of focal adhesions and activation of the PI3K/Akt pathway (13, 14). The phosphorylation of PI3K and Akt leads to up-regulation of anti-apoptotic Bcl-2 and Bcl-xL protein expression in EC (15). Ligation of MHC class I molecules phosphorylates S6 ribosomal protein (S6RP), a downstream target of mammalian target of rapamycin (mTOR) complex 1 (mTORC1), suggesting a role for this pathway in cell proliferation and protein synthesis. In vivo evidence that anti-class I Abs initiate a cell proliferation signaling cascade was provided by immunostaining of cardiac transplant biopsies with evidence of AMR. Biopsies from heart allografts diagnosed with AMR showed increased expression of phosphorylated S6RP suggesting that class I signaling may elicit cell proliferation via the mTOR/S6RP pathway (15, 16).
mTOR is a serine-threonine kinase that plays a central role in the regulation of cell proliferation and protein synthesis through the activation of p70 ribosomal S6 kinase (S6K) and eukaryotic initiation factor 4E-binding protein 1 (4E-BP1) (17, 18). mTOR forms two distinct molecular complexes, allowing for greater functional diversity. mTORC1 containing mTOR, regulatory associated protein of TOR (raptor), and GβL activates S6K and 4E-BP1 (17, 19). Activation of these targets leads to increased ribosomal biosynthesis and translation of critical mRNAs of proteins required for G1 to S phase transition and proliferation. The recently discovered mTOR complex 2 (mTORC2), containing mTOR, rapamycin-insensitive companion of TOR (rictor), and GβL, has been show to phosphorylate Akt at Ser473 and is involved in cytoskeleton rearrangements through the activation of Rho GTPases (20, 21). The insensitivity of rictor to rapamycin has recently been debated due to reports that long-term exposure to rapamycin blocks mTORC2 formation (22). Additional adaptor proteins that associate with mTORC2 have been discovered. Stress-activated protein kinase-tenracting protein 1 (Sin1) acts to maintain mTORC2 integrity and facilitates Akt phosphorylation at Ser473 (23, 24). Additionally, protein observed with rictor was reported to be a rictor-binding subunit that associates with mTORC2 (25). The formation of mTOR signaling complexes and activation of downstream targets including S6K and 4E-BP1 following MHC class I ligation remain to be determined.
The studies described in this report were aimed at elucidating the role of mTOR in MHC class I-induced cell proliferation and cell survival. Using siRNA to knock down mTOR in EC, we showed that mTOR plays a critical role in anti-HLA Ab-induced protein synthesis via activation of proteins downstream of mTORC1 and mTORC2. Knockdown of mTOR inhibited class I-induced proliferative responses, demonstrating a role of mTOR in regulating class I-mediated cell protein synthesis and proliferation. Using siRNA knockdown of rictor or raptor, we identified mTORC2 as an upstream regulator of MHC class I-induced Akt at Ser473 phosphorylation. Long-term exposure of EC to rapamycin blocked MHC class I-induced phosphorylation of Akt at Ser473 and expression of prosurvival protein Bcl-2. These data suggest that mTOR can control MHC class I-induced cell proliferation and survival signals in EC. Lastly, we identified Src and FAK as upstream regulators of MHC class I-induced mTOR signaling.
Materials and Methods
Abs and chemicals
Mirus TransIT-TKO transfection reagent was purchased from Mirus Bio. The hybridoma (HB-95) W6/32, recognizing a monomorphic epitope on HLA class I, was purchased from the American Type Culture Collection and purified by protein A-agarose affinity chromatography. The mouse IgG mAb isotype control, mAb against vinculin (V9131), and protein A-agarose were purchased from Sigma-Aldrich. Rabbit polyclonal Abs against phospho-mTOR (Ser2448), phospho-S6K (Thr389), phospho-S6K (Thr421/Ser424), phospho-4E-BP1 (Thr37/46), phospho-Akt (Ser473), phospho-Akt (Thr308), phospho-S6 ribosomal protein (Ser235/236), mTOR, Akt, ERK, S6K, S6RP, and raptor were purchased from Cell Signaling Technology. Rabbit polyclonal Ab against phospho-FAK (Tyr576) was obtained from Invitrogen. Goat polyclonal Abs against mTOR and rabbit polyclonal FAK, protein A/G PLUS-agarose were purchased from Santa Cruz Biotechnology. Rabbit polyclonal anti-rictor was purchased from Novus Biologicals. Rapamycin was purchased from Sigma-Aldrich. The Src kinase inhibitor PP2 and was obtained from Calbiochem. CellTrace CFSE Cell Proliferation kit was purchased from Invitrogen.
Primary human aortic endothelial cells (HAEC) were isolated from the aortic rings of explanted donor hearts, as described previously (26). Cells were cultured in M199 complete medium containing M199, sodium pyruvate (1 mM) (Irvine Scientific), penicillin and streptomycin at 100 U/ml and 100 μg/ml, respectively (both from Invitrogen), 20% (v/v) FBS (HyClone), heparin (90 μg/ml) (Sigma-Aldrich), and endothelial cell growth supplement (20 μg/ml) (Fisher Scientific). Cells from passages 3–8 were used at a confluence of 70–80%. Before use in experiments, cells were grown for 16 h in medium containing 0.2% FBS.
Human aortic endothelial cells were grown in 150-mm dishes. They were rinsed once with ice-cold PBS and lysed in 800 μl of ice-cold lysis buffer (40 mM HEPES (pH 7.5), 120 mM NaCl, 1 mM EDTA, 10 mM β-glycerophosphate, 50 mM NaF, 1.5 mM Na3VO4, 0.3% CHAPS, 10 μg/ml aprotinin, and 10 μg/ml leupeptin). Samples were place on a rotator at 4°C for 20 min. After lysis, cell debris was spun down at 14,000 rpm for 10 min, 4 μg of anti-mTOR Ab was added to the cleared supernatant, and samples were placed on the rotator overnight. Protein A/G beads (40 μl) were added for 4 h. Immunoprecipitates were washed four times in wash buffer (10 mM HEPES (pH 7.5), 50 mM β-glycerophosphate, 5 mM NaCl), 30 μl of 2x SDS loading buffer was added, and samples were then boiled 5 min and loaded onto a 6% SDS-PAGE gel.
Western blot analysis
Serum-starved cultures of HAEC were stimulated with Abs, lysed in Triton X-100 lysis buffer, run on a SDS-PAGE gel, and proteins were transferred overnight on Immobilon-P membranes (Millipore). Membranes were blocked using 5% BSA in TBST for 1 h and incubated for an additional 2–15 h at room temperature with the phosphospecific Abs. Primary Abs to immunoreactive bands were visualized using HRP-conjugated anti-rabbit, anti-mouse, or anti-goat Abs.
siRNA design and transfection
We designed an mTOR siRNA duplex corresponding to bases 5309–5327 from the open reading frame of human FRAP1 mRNA 5′-CCA AAG UGC UGC AGU ACU AUU-3′. The RNA sequence used as a negative control (GL-2) for siRNA activity was 5′-CGU ACG CGG AAU ACU UCG A-dT.dT-3′. Human rictor 5′-ACU UGU GAA GAA UCG UAU C-dT.dT-3′ and human raptor 5′-GGA CAA CGG CCA CAA GUA C.dT.dT-3′ were purchased from Dharmacon. The FAK siRNA sequence and transfection procedure was previously described (14). In preliminary experiments, we optimized conditions for the efficient transfection of HAEC using siGLO Lamin A/C siRNA (Dharmacon). HAEC were cultured from 40 to 60% confluence maintained in M199 complete medium and transfected with siRNA as we previously described (14). Briefly, cells were placed in 800 μl of M199 medium along with 200 μl of transfection solution (Opti-MEM medium), Mirus transfection reagent, and siRNA (100 nM, or as noted). After 6 h of transfection, 1 ml of M199 containing twice the amount of FBS, endothelial cell growth supplement, and heparin were added, and experiments were conducted 48 h after transfection.
CFSE cell proliferation assay
HAEC were seeded into 35-mm dishes. Cells were transfected with 100 nM of mTOR, rictor, raptor, or control siRNA for 24 h. The cells were starved overnight in M199 containing 5% FBS. Cells were labeled with 2 μM of CFSE at 37°C for 15 min and then washed twice in warm PBS and stimulated with anti-MHC class I Ab or control mouse IgG for 72 h. Cells were detached using 0.125% trypsin/0.05% EDTA, washed, and analyzed by flow cytometry. Analysis was preformed on 10,000 cells/sample using a FACSCalibur flow cytometer (BD Biosciences). Data were analyzed using CellQuest Pro (BD Biosciences) and Modfit LT software (Verity Software House). Cell proliferation was calculated using the Proliferation Wizard Model (Verity Software House). The proliferation index is the sum of the cells in all generations divided by the computed number of original parent cells present at the start of the experiment.
siRNA inhibits mTOR expression in EC
To determine the effect of mTOR siRNA on mTOR expression, cells were transfected with increasing concentrations of mTOR siRNA, and mTOR protein expression was determined by Western blot (Fig. 1 A). mTOR expression was markedly inhibited at all concentrations tested, with maximal blockade observed at 100 nM of siRNA. In contrast, transfection with the nontargeting siRNA had no effect on mTOR expression.
To further characterize the efficiency of mTOR siRNA transfection, we performed immunofluoresence analysis of EC transfected with mTOR siRNA. Cells were transfected with increasing concentrations of a stable fluorescent form of lamin A/C that allows a direct correlation of fluorescence uptake with silencing activity. Analysis of intracelluar fluorescence revealed 60, 90, and 100% of cells silenced following transfection with 25, 50, and 100 nM of lamin A/C siRNA, respectively (Fig. 1 B). These data show that the efficiency of mTOR siRNA transfection is optimal at a concentration of 100 nM of siRNA.
To exclude the possibility of nonspecific effects of mTOR siRNA on MHC class I-induced endothelial cell activation, we examined the expression of a variety of proteins involved in the MHC class I and mTOR pathway in the same cell lysates of EC transfected with mTOR siRNA. Transfection with control or mTOR siRNA did not substantially change the expression of rictor, raptor, S6K, Akt, ERK, or Vinculin (Fig. 1 A). Collectively, these data indicate that mTOR siRNA efficiently and specifically inhibited protein expression of mTOR in EC.
siRNA knockdown of mTOR blocks MHC class I-induced cell proliferation
Ligation of HLA class I molecules triggers cell proliferation by up-regulating fibroblast growth factor receptor (FGFR) expression and increasing responsiveness to basic fibroblast growth factor (bFGF) (27, 28). Because mTOR has been reported to be a critical regulator of cell proliferation (29), we determined its role in class I-mediated EC proliferation. For this, EC were transfected with targeting or nontargeting mTOR siRNA, labeled with CFSE, and then stimulated with anti-HLA class I Abs or untreated. Seventy-two hours later, cells were harvested and analyzed by flow cytometry for decreased intracellular fluorescence as an indicator of cell proliferation. As a way to measure the increase in cell numbers over time, the proliferation index was calculated as the ratio of cells in all generations divided the number of original parent cells at the start of the experiment. Ligation of MHC class I by W6/32 stimulated increased cell proliferation in EC transfected with nontargeting control siRNA compared with untreated controls (Fig. 1, C and E). Conversely, cells transfected with mTOR siRNA failed to stimulate MHC class I-induced proliferation compared with control siRNA-transfected cells (Fig. 1, C and D). The proliferation index of cells transfected with control siRNA and stimulated with anti-class I Ab was 35 vs 25 in cells transfected with mTOR siRNA, a value comparable to untreated EC (23) (Fig. 1 C–E). These data support a role for mTOR in HLA class I Ab-mediated cell proliferation.
Ab ligation of HLA class I molecules stimulates the mTOR pathway in EC
To determine the effect of anti-class I Abs on phosphorylation of proteins in the mTOR signaling pathway, primary HAEC were treated with anti-HLA class I Abs for different time points and immunoblotted with phospho-specific Abs to S6K, S6RP, and Akt. Class I ligation resulted in a time dependent increase in phosphorylation of S6K at sites Thr389 and Thr421/Ser424, which peaked at 10 min and diminished after 30 min. The tempo of class I-mediated phosphorylation of Akt at site Ser473 differed from S6K in that it reached its highest level at 30 min and declined thereafter (Fig. 2,A). Assessment of signaling events downstream of S6K showed that class I ligation led to a time-dependent increase in the phosphorylation of S6RP Ser235/236 that peaked at 10–30 min and decreased by 60 min (Fig. 2 B). These data show that ligation of class I molecules on EC activates proteins involved in the mTOR signaling pathway.
Class I ligation induces formation of mTORC1 and mTORC2
mTOR exists as two distinct complexes within cells, mTORC1 and mTORC2, and these complexes play distinct roles in cell proliferation and survival (17, 19, 20, 30). To determine whether ligation of class I molecules with Abs stimulate the association of mTORC1 (mTOR with raptor) or mTORC2 (mTOR with rictor), quiescent EC were treated with anti-HLA class I mAbs for different periods of time, and the extracts were immunoprecipitated with anti-mTOR Abs. The immunoprecipitates were subsequently analyzed by immunoblotting with anti-rictor, anti-raptor, and anti-mTOR Abs. Anti-class I Abs stimulated mTOR-rictor and mTOR-raptor complex formation as early as 5 min, which peaked at 10 min and declined after 30 min (Fig. 2 C). mTOR complexed with raptor and mTOR complexed with rictor were recovered during a 5–30-min interval following addition of anti-class I Abs. mTOR complex formation was maximal at 10 min following class I ligation. These results demonstrate that binding of anti-HLA Abs to class I Ags on EC transduces signals resulting in the formation of mTORC1 and mTORC2.
siRNA knockdown of mTOR blocks MHC class I induction of the mTOR signaling pathway
We used mTOR siRNA in cultured EC to determine the role of the TOR pathway in class I-mediated phosphorylation of the targets of mTOR that have been implicated in cell proliferation. EC were transfected with control or mTOR siRNA, stimulated with anti-class I Abs for various time points, and the lysates were immunoblotted for proteins involved in the mTOR signaling pathway. As shown in Fig. 3,A, knockdown of mTOR markedly reduced class I-induced phosphoryation of S6K at sites Thr389 and Thr421/Ser424. Knockdown of mTOR also blocked class I-mediated phosphorylation of S6RP at Ser235/236. We next looked at 4E-BP1, a component of the translation machinery, downstream of mTOR. Knockdown of mTOR strikingly inhibited class I-induced phosphoryation of 4E-BP1 at Thr37/46 (Fig. 3 B) compared with EC transfected with nontargeting siRNA.
siRNA knockdown of rictor blocks Akt phosphorylation at site Ser473
To explore the role of mTOR in class I-mediated cell survival signaling we determined the effect of mTOR knockdown on class I-induced Akt phosphorylation. Treatment of EC transfected with nontargeting siRNA with anti-class I Abs stimulated a time-dependent increase in the phosphorylation of Akt at both sites Ser473 and Thr308 (Fig. 3,C). Transfection of EC with mTOR siRNA impaired class I-mediated Akt phosphorylation at site Ser473. In contrast, Akt site Thr308, which is phosphorylated by PDK1, was not affected (Fig. 3 C). Transfection with mTOR siRNA completely inhibited mTOR protein expression. These data demonstrate that knockdown of mTOR expression blocks class I-induced activation of proteins in the mTOR pathway and establishes mTOR as the upstream kinase to class I-induced Akt at Ser473 phosphorylation.
Emerging data indicate that the adaptor protein rictor complexes with mTOR and functions as an upstream kinase of Akt at Ser473 to regulate the cell survival machinery (21). Furthermore, previous studies have shown that knockdown of rictor protein does not affect mTORC1 or its downstream target S6K (20). To characterize the effects of mTORC2 on MHC class I-induced phosphorylation of Akt at Ser473, we transfected EC with a targeting or nontargeting rictor siRNA and compared the amount of phosphorylated Akt Ser473 following class I stimulation. EC transfected with the rictor siRNA blocked class I-induced Akt phosphorylation at site Ser473 (Fig. 4,A). In contrast, rictor knockdown had no effect on class I-mediated activation of S6K at Thr389, the downstream target of mTORC1 (Fig. 4,A). These data demonstrate that class I-induced phosphorylation of Akt at Ser473 requires mTORC2 assembly. To further demonstrate that mTORC1 is not essential for class I-induced phosphorylation of Akt at Ser473, EC were transfected with raptor or control siRNA for 48 h and then stimulated with anti-HLA class I Abs at various time points. EC transfected with raptor siRNA inhibited class I-induced S6K at Thr389 phosphorylation, whereas class I-induced Akt at Ser473 phosphorylation was not affected (Fig. 4 B). These results establish mTOR as the upstream kinase to MHC class I-induced Akt phosphorylation at Ser473 through mTORC2.
The effect of raptor and rictor siRNA knockdown on MHC class I-induced cell proliferation
To determine the effect of mTOR complex components raptor and rictor on MHC class I-induced cell proliferation, EC were transfected with raptor, rictor, or nontargeting siRNA, labeled with CFSE, and then stimulated with anti-HLA class I Abs. Seventy-two hours later, cells were harvested and analyzed by flow cytometry for decreased intracellular fluorescence as an indicator of cell proliferation. Ligation of MHC class I by W6/32 stimulated increased cell proliferation in EC transfected with nontargeting control siRNA compared with control siRNA alone (Fig. 4, C and D). siRNA knockdown of raptor completely inhibited class I-induced cell proliferation compared with EC transfected with control siRNA (Fig. 4, D and F). Transfection with rictor siRNA also inhibited MHC class I-induced cell proliferation by 60% compared with EC transfected with control siRNA (Fig. 4, C and E). These data provide evidence showing mTOR-raptor and mTOR-rictor are in involved in MHC class I-induced cell proliferation.
Rapamycin inhibits class I-induced phosphorylation of Akt at site Ser473 and expression of Bcl-2
Rapamycin is a U.S. Food and Drug Administration-approved immunosuppressive agent that blocks mTOR activity by inhibiting complex formation between mTOR, raptor, and GβL (mTORC1) (17, 31). Although mTORC2 was previously described to be insensitive to rapamycin, recent studies suggest that prolonged exposure to this agent prevents the assembly of mTORC2 and Akt at Ser473 phosphorylation, inhibits activation of Akt, and induces EC apoptosis (22, 32). To determine the effect of rapamycin on MHC class I-induced survival signaling in EC, we measured the levels of Akt phosphorylation at Ser473 as well as Bcl-2 expression. EC were stimulated with various doses of rapamycin for 24 h in the presence or absence of anti-class I Abs. Pretreatment of EC with rapamycin strongly inhibited class I Ab-induced phosphorylation of Akt at Ser473 within 24 h of addition to EC (Fig. 5,A). Exposure of EC to concentrations of rapamycin ranging from 30 to 100 nM produced similar effects on Akt at Ser473 phosphorylation at 24 h. Furthermore, pretreatment with 30 nM of rapamycin for as little as 2 h was sufficient to impair class I-mediated phosphorylation of Akt at Ser473 (Fig. 5,B). To determine whether rapamycin alters the levels of mTORC2, EC were stimulated for 10 and 30 min with anti-class I Abs in the presence or absence of 30 nM of Rapamycin, and we compared the amounts of rictor and raptor bound to mTOR at 24 h. Exposure to rapamycin completely prevented MHC class I-induced mTORC1 and mTORC2 complex formation (Fig. 5,C). Akt mediates prosurvival and anti-apoptosis by upregulating Bcl-2 (15, 33). Because we observed that exposure of EC to rapamycin blocked class I-mediated phosphorylation of Akt, we reasoned that rapamycin may also block class I-induced up-regulation of the anti-apoptotic protein Bcl-2. As shown in Fig. 5 D, treatment of EC with anti-class I Abs stimulated a statistically significant increase in Bcl-2 protein expression (p = 0.002) and this effect was inhibited by pretreatment of EC with 30 nM of rapamycin. These results raise the possibility that exposure to rapamycin may inhibit MHC class I-mediated survival signaling in EC.
mTOR and S6K are downstream targets of Src and FAK in the MHC class I signaling pathway
Our previous work identified Src as an upstream signaling tyrosine kinase following MHC class I ligation (13). Src-mediated phosphorylation FAK regulates adhesion dynamics, survival signaling, and growth (34, 35). To establish whether Src tyrosine kinase activity is required for HLA-mediated activation of the mTOR pathway, EC were pretreated with PP2, a selective inhibitor of Src, followed by stimulation with anti-MHC class I Abs. As shown in Fig. 6 A, pretreatment of EC with PP2 considerably blocked class I-induced phosphorylation of mTOR at Ser2448 and S6K at sites Thr389 and Thr421/Ser424. This result supports the involvement of Src as an upstream kinase to MHC class I-mediated phosphorylation of mTOR and S6K.
Src kinases were shown to associate with FAK in response to class I ligation and promote maximal FAK activation (14). Additionally, with the recruitment of Src into a FAK-Src signaling complex, Src facilitates phosphorylation of many FAK-associated proteins, including PI3K, Akt, and paxillin (13, 15). FAK plays a critical role in cell migration, cytoskeletal rearrangements, and cell survival (34, 35, 36). To determine whether FAK is an upstream regulator of class I-induced mTOR activation, we blocked FAK protein expression (14) and explored the mTOR pathway in the absence of FAK. The phosphorylation of mTOR at Ser2448 and S6RP at Ser235/236 were both markedly inhibited following FAK knockdown (Fig. 6,B). To confirm FAK is upstream of mTOR, we stimulated EC with anti-MHC class I Abs for 10 min in the presence of control or mTOR siRNA, and the cell lysates were immunoblotted for the phosphorylation of FAK at Tyr576. As shown in Fig. 6 C, knockdown of mTOR did not affect class I-induced phosphoryation of FAK at site Tyr576. These results demonstrate that FAK is an essential upstream regulator required for the mTOR signaling pathway.
The major finding of this study is that Abs to HLA class I molecules activate the mTOR pathway leading to cell proliferation. Importantly, we found that class I-mediated activation of Akt at Ser473 is regulated by mTORC2, whereas class I-induced cell proliferation is regulated by mTORC1. These data suggest that Abs may contribute to the process of transplant vasculopathy through Ab-mediated endothelial cell proliferation via an mTOR-dependent mechanism.
Using siRNA-mediated protein depletion of mTOR, we explored the contribution of the mTOR pathway to anti-HLA class I-induced EC proliferation. We found that transfection with mTOR siRNAs into primary aortic human endothelial cells efficiently knocked down the expression of mTOR protein without affecting the expression of other proteins involved in the MHC class I signaling pathway and provided a reliable experimental model to explore the relationship of mTOR with other upstream and downstream targets. We observed that inhibition of mTOR protein expression by siRNA blocked class I-mediated EC proliferation. Furthermore, treatment of EC with raptor siRNA or rictor siRNA also resulted in a decreased capacity to stimulate cell proliferation following ligation of class I molecules. Down-regulation of raptor completely abrogated class I-induced cell proliferation. Rictor silencing partially, although significantly, reduced class I-induced EC proliferation by 60%. Rictor knockdown may inhibit cell proliferation by interrupting cytoskeletal rearrangement, an important component of cell proliferation (30). This finding indicates that cooperating mechanisms between signals from both mTOR complexes promote class I-mediated proliferation of endothelial cells. MHC class I ligation lead to the phosphorylation of several downstream targets of mTORC1 signaling, including S6K at Thr389, S6RP at Ser235/236, and 4E-BP1 at Thr37/46 and the mTORC2 signaling target Akt at Ser473. These effects were consistent with previous findings showing that mTORC1 and mTORC2 complexes preceded activation of S6K at Thr389 and Akt at Ser473, respectively, in a model of hypoxia-induced endothelial cell proliferation (37). Furthermore, we show for the first time that MHC class I ligation led to an increase in both mTORC1 and mTORC2 formation following 10 min of stimulation. Fluctuations in mTOR complex formation have not been observed in cell lines treated with growth factors (17, 25). We conducted our experiments using primary cultured EC expressing endogenous levels of all signaling components, whereas previous work was done using a common cancer cell line HEK293. It is conceivable that MHC class I ligation activates mTOR via different pathways than those used by tyrosine kinase receptors in fibroblasts.
Knockdown of raptor expression inhibited the phosphorylation and activity of S6K at Thr389 but not Akt at Ser473. In contrast, knockdown of rictor completely blocked class I-induced Akt at Ser473 phosphorylation, yet failed to alter class I-induced phosphorylation of S6K at Thr389. These results support the contention that mTORC2 is the upstream kinase of MHC class I-induced Akt at Ser473 phosphorylation and are in agreement with recent studies showing that mTORC2 is the kinase responsible for phosphorylating Akt on Ser473 (21, 24, 38, 39).
We have previously shown that ligation of MHC class I molecules stimulates EC proliferation by increasing the cell-surface expression of FGFR and responsiveness to bFGF (27, 40). Activation of the mTOR pathway and FGFR up-regulation by MHC class I ligation appear to be separate events based on recent work showing that FGFR translocation is a FAK-independent process (14). Additionally, class I-induced phosphorylation of mTOR at Ser2448 and Akt at Ser473 occur in the presence of FGFR blocking Abs. However, inhibition of FAK protein expression by FAK siRNA reduced class I-mediated proliferation in the presence and absence of bFGF. This suggests that although class I-mediated up-regulation of FGFR expression was not inhibited by siRNA knockown of FAK, FAK is critically involved in class I-mediated EC proliferation downstream to bFGF ligand binding to FGFR (14). Whether there is a direct association between FGFR and FAK remains to be explored.
Our previous work and the current study together indicate that FAK is an upstream regulator of the mTOR pathway. Knockdown of FAK using siRNA blocked mTOR at Ser2448 and S6RP at Ser235/236 phosphorylation, whereas knockdown of mTOR did not affect class I-induced FAK at Tyr576 phosphorylation. Furthermore, FAK knockdown appreciably inhibited class I-mediated phosphorylation of Akt at Ser473 (14). Based on our data showing that mTORC2 is the kinase for class I-mediated phoshorylation of Akt at Ser473, it is tempting to speculate that class I-mediated alterations in the cytoskeleton may involve mTORC2. This hypothesis is consistent with previous studies showing knockdown of mTORC2, but not of mTORC1, prevents paxillin phosphorylation at Tyr118 and stress fiber formation, demonstrating that mTORC2 signals to the actin cytoskeleton (30). Jacinto et al. showed that serum-stimulated NIH3T3 fibroblast cells transfected with mTOR, GβL, or rictor siRNA had decreased cell spreading and presence of polymerized actin compared with raptor siRNA or controls, and they suggested that mTOR signals through Rac1 and Rho GTPases. This mechanism may occur through a protein guanine exchange factor, P-Rex1, which associates with mTORC2 to activate Rac, leading to increases in cell migration (41). Cross-linking of MHC class I molecules has also been shown to lead to the activation of Rho GTPases, assembly of focal adhesions, and stress fiber formation (11, 12, 42, 43). We have recently shown that transfection with FAK siRNA decreased class I-mediated phosphorylation of Src at Tyr418 and paxillin at Tyr118 and blocked paxillin redistribution to focal adhesions in EC (14). siRNA knockdown of FAK also completely disrupted class I-mediated stress fiber formation, implicating FAK as a crucial regulator of the actin cytoskeleton (14). In contrast, other studies suggest that components of mTORC2 affect the actin cytoskeleton by inhibiting stress fiber formation. Sarbassov et al. demonstrated that knockdown of either mTOR or rictor leads to actin stress fiber formation and migration of paxillin patches from the cell periphery within cellular extensions to the ends of thick actin fibers (20). The newly discovered adaptor protein Sin1 acts as an essential component of mTORC2 but does not associate with mTORC1. Sin1 is required for insulin-induced mTORC2 phosphorylation of Akt at Ser473, and its depletion increases F-actin stress fiber formation (24). These results imply that mTORC2 may function to suppress FAK-induced actin stress fiber formation. These conflicting cytoskeletal effects involving mTORC2 may result from the association of mTOR with distinct adaptor proteins that modify downstream functions. Future work is needed to sort out how mTORC2 adaptor proteins function with respect to each other while identifying additional adaptor proteins that might regulate this process.
Rapamycin is an immunosuppressive drug that targets mTOR and is used to prevent rejection of transplanted organs. Rapamycin has been recently shown to be effective at reducing the development of cardiac allograft vasculopathy in a randomized clinical trial, suggesting an anti-proliferative effect resulting in decreased intimal thickening (44). Rapamycin works by binding to FKBP12 to generate a drug-receptor complex that inhibits the kinase activity of mTORC1. Although mTORC2 is rapamycin insensitive, several studies have shown that prolonged exposure to rapamycin may also inhibit mTORC2 function by preventing assembly of mTORC2 complexes in the cell (22). Our data suggests that short-term exposure to rapamycin does not affect mTORC2 or Akt at Ser473 phosphorylation. However, long-term exposure to rapamycin (>2 h) disrupts anti-MHC class I-induced mTORC2 formation and phosphorylation of Akt at Ser473. These data are consistent with the results of other studies showing that rapamycin strongly inhibits Akt phosphorylation at Ser473 (22). Activated Akt can promote cell survival and graft accommodation by upregulating the anti-apoptotic protein Bcl-2 (15, 33, 45, 46) or by maintaining nutrient transporters and receptors through an mTOR-dependent mechanism (47). Based on our previous work showing that anti-MHC class I Abs stimulate Bcl-2 expression through the PI3K/Akt pathway (15), we reasoned that rapamycin may prevent class I-mediated up-regulation of Bcl-2. As suspected, we found that long-term exposure to rapamycin resulted in a decrease in MHC class I-induced Bcl-2 protein expression. This suggests that although rapamycin may be beneficial in preventing cell proliferation and intimal thickening, it may avert class I-mediated up-regulation of cell survival proteins, making the graft more susceptible to the apoptotic effects of immune effectors such as CTL (33). Our results are consistent with recent studies by Dormond et al. that showed that mTOR inhibition by rapamycin suppresses Akt-inducible prosurvival signals. Treatment of EC with rapamycin increased the number of EC undergoing apoptosis after serum withdrawal and after stimulation with vascular endothelial growth factor (32). In light of these findings it will be important in the future to develop drugs that will selectively target mTORC1 while preserving the integrity of mTORC2.
Our studies were designed to study the effects of anti-HLA class I Abs on the mTOR signaling pathway and proliferation using an siRNA knockdown cell culture model that allowed us to isolate the specific effects of anti-HLA Abs on EC without immune cells or complement. Therefore, the interpretation of these findings will require further validation in a suitable in vivo transplant model. We recently developed a mouse vascularized heterotopic cardiac allograft model in which B6.RAG1 knockout hosts receive a fully MHC-incompatible BALB/c heart transplant and are passively transfused with anti-donor MHC class I Abs (48). Cardiac allografts of mice treated with anti-MHC class I Abs showed characteristic features of Ab-mediated rejection and prominent phosphorylation of Akt at Ser473 and S6K at Thr421/Ser424. These data provide the first glimpse of the interrelationships between these signaling molecules in vivo and confirms the results of the signaling pathway derived from our in vitro experiments.
Our data are consistent with a model in which ligation of MHC class I molecules by anti-MHC class I Abs stimulates Src and FAK phosphorylation and formation of mTORC1 and mTORC2 (Fig. 7). mTOR complexed with rictor phosphorylates Akt at Ser473 and promotes cell survival by up-regulating anti-apoptotic proteins. mTOR complexed with raptor phosphorylates S6K at sites Thr389 and Thr421/Ser424, as well as 4E-BP1 at sites Thr37/46. Activation of S6K in turn phosphorylates S6RP at Ser235/236 and together with 4E-BP1 increases protein translation and cell proliferation. Long-term treatment with rapamycin inhibits mTORC1 formation, which may prove to be beneficial in preventing class I-mediated cell proliferation and transplant vasculopathy. Preventing mTORC2 formation by exposure to rapamycin has the potential to adversely affect endothelial cell survival responses. Future studies are planned to assess mTOR signaling in the endothelium of allografts from recipients treated with rapamycin to determine the effect of this drug on cell survival and proliferation signaling pathways in vivo. Elucidation of the signaling pathways activated following MHC class I ligation should also help to identify more specific molecular targets to inhibit Ab-mediated chronic allograft rejection.
The authors have no financial conflicts of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by National Institute of Allergy and Infectious Diseases Grant R01 AI 42819 and by the American Heart Association Grant-in-Aid 9750894A.
Abbreviations used in this paper: EC, endothelial cell(s); bFGF, basic fibroblast growth factor; 4E-BP1, initiation factor 4E-binding protein 1; FAK, focal adhesion kinase; FGFR, fibroblast growth factor receptor; HAEC, human aortic endothelial cells; mTOR, mammalian target of rapamcyin; mTORC1, mTOR complex 1; mTORC2, mTOR complex 2; rictor, rapamycin-insensitive companion of TOR; raptor, regulatory associated protein of TOR; S6K, p70 ribosomal S6 kinase; S6RP, S6 ribosomal protein; siRNA, small interfering RNA; Sin1, stress-activated protein kinase-interacting protein 1.