MAPK-independent impairment of T-cell responses by the multikinase inhibitor sorafenib

Sorafenib, originally developed as CRAF inhibitor but soon recognized as a multikinase inhibitor, is currently widely tested for the treatment of different cancers either alone or in combination therapy. However, the clinical success, particularly in immunogenic tumors such as melanoma, was less than anticipated. Because T-cell activation is tightly regulated by a multitude of kinases, we scrutinized effects of sorafenib on immune responses. To this end, comprehensive in vitro studies revealed that the presence of sorafenib concentrations comparable with observed plasma levels in patients strongly impairs the activation of T cells. Notably, even established tumor-specific immune responses are influenced by sorafenib. Indeed, ELISPOT data of peripheral blood lymphocytes obtained from melanoma patients vaccinated against survivin show markedly diminished survivin-specific immune responses in the presence of sorafenib. Surprisingly, inhibition of T-cell activation was not associated with reduced extracellular signal-regulated kinase phosphorylation. In fact, on T-cell receptor stimulation phospho-extracellular signal-regulated kinase and phospho-mitogen-activated protein kinase kinase levels were found to be elevated in the presence of sorafenib, showing the complexity of signal transduction events following T-cell receptor stimulation. In conclusion, our data show that T-cell function is sensitive toward the multikinase inhibitor sorafenib in a mitogen-activated protein kinase-independent fashion. This observation has important implications for the use of sorafenib as therapy for immunogenic cancers.

Anti-cancer therapies targeting the tumor stroma

For anti-tumor therapy different strategies have been employed, e.g., radiotherapy, chemotherapy, or immunotherapy. Notably, these approaches do not only address the tumor cells themselves, but also the tumor stroma cells, e.g., endothelial cells, fibroblasts, and macrophages. This is of advantage, since these cells actively contribute to the proliferative and invasive behavior of the tumor cells via secretion of growth factors, angiogenic factors, cytokines, and proteolytic enzymes. In addition, tumor stroma cells take part in immune evasion mechanisms of cancer. Thus, approaches targeting the tumor stroma attract increasing attention as anti-cancer therapy. Several molecules including growth factors (e.g., VEGF, CTGF), growth factor receptors (CD105, VEGFRs), adhesion molecules (alphavbeta3 integrin), and enzymes (CAIX, FAPalpha, MMPs, PSMA, uPA) are induced or upregulated in the tumor microenvironment which are otherwise characterized by a restricted expression pattern in differentiated tissues. Consequently, these molecules can be targeted by inhibitors as well as by active and passive immunotherapy to treat cancer. Here we discuss the results of these approaches tested in preclinical models and clinical trials.

Tumor stroma-associated antigens for anti-cancer immunotherapy

Immunotherapy has been widely investigated for its potential use in cancer therapy and it becomes more and more apparent that the selection of target antigens is essential for its efficacy. Indeed, limited clinical efficacy is partly due to immune evasion mechanisms of neoplastic cells, e.g. downregulation of expression or presentation of the respective antigens. Consequently, antigens contributing to tumor cell survival seem to be more suitable therapeutic targets. However, even such antigens may be subject to immune evasion due to impaired processing and cell surface expression. Since development and progression of tumors is not only dependent on cancer cells themselves but also on the active contribution of the stromal cells, e.g. by secreting growth supporting factors, enzymes degrading the extracellular matrix or angiogenic factors, the tumor stroma may also serve as a target for immune intervention. To this end several antigens have been identified which are induced or upregulated on the tumor stroma. Tumor stroma-associated antigens are characterized by an otherwise restricted expression pattern, particularly with respect to differentiated tissues, and they have been successfully targeted by passive and active immunotherapy in preclinical models. Moreover, some of these strategies have already been translated into clinical trials.

The HLA-A2 restricted T cell epitope HCV core 35-44 stabilizes HLA-E expression and inhibits cytolysis mediated by natural killer cells

Impaired activity of natural killer cells has been proposed as a mechanism contributing to viral persistence in hepatitis C virus (HCV) infection. Natural cytotoxicity is regulated by interactions of HLA-E with inhibitory CD94/NKG2A receptors on natural killer (NK) cells. Here, we studied whether HCV core encodes peptides that bind to HLA-E and inhibit natural cytotoxicity. We analyzed 30 HCV core-derived peptides. Peptide-induced stabilization of HLA-E expression was measured flow cytometrically after incubating HLA-E-transfected cells with peptides. NK cell function was studied with a (51)chromium-release-assay. Intrahepatic HLA-E expression was analyzed by an indirect immunoperoxidase technique and flow cytometry of isolated cells using a HLA-E-specific antibody. We identified peptide aa35-44, a well-characterized HLA-A2 restricted T cell epitope, as a peptide stabilizing HLA-E expression and thereby inhibiting NK cell-mediated lysis. Blocking experiments confirmed that this inhibitory effect of peptide aa35-44 on natural cytotoxicity was mediated via interactions between CD94/NKG2A receptors and enhanced HLA-E expression. In line with these in vitro data we found enhanced intrahepatic HLA-E expression on antigen-presenting cells in HCV-infected patients. Our data indicate the existence of T cell epitopes that can be recognized by HLA-A2 and HLA-E. This dual recognition may contribute to viral persistence in hepatitis C.

Truncated HLA-G isoforms are retained in the endoplasmic reticulum and insufficiently provide HLA-E ligands

The preferential expression of the non-polymorphic human leukocyte antigen G (HLA-G) on invading extravillous cytotrophoblast cells that are, with the exception of HLA-C and -E, HLA class I negative led to the hypothesis that HLA-G plays a major role in controlling the effector functions of the large granular leukocytes (LGL), a specialized natural killer (NK) cell population present in large numbers in the decidua. Transcription of the HLA-G gene is characterized by extensive alternative splicing producing at least seven potentially membrane bound or secreted isoforms. Except for HLA-G1 and its soluble variant (HLA-G1s), there is still dispute as to whether any of the other isoforms displays a major immunological function. Here we describe that the membrane-bound isoforms HLA-G2, -G3, and G4 as well as the soluble variant of HLA-G2 (HLA-G2s) do not egress the endoplasmic reticulum as determined by Endo H sensitivity assays. Moreover these isoforms seem not to have a major immunological function with respect to NK cell inhibition by providing a ligand for HLA-E, which would allow the interaction of this molecule with the inhibitory CD94/NKG2A NK cell receptor.

HLA-G modulates immune responses by diverse receptor interactions

HLA-G regulates immune responses as it binds different receptors expressed on natural killer (NK) cells, T cells and myeloid cells. HLA-G1 can inhibit NK- and T-cell-mediated lysis of target cells by its interaction with the inhibitory receptors ILT2 and ILT4. Engaging KIR2DL4 triggers different reactions depending on the activation state of the effector cells. The indirect recognition of HLA-G as peptide presented by HLA-E and recognized by the CD94/NKG2 receptor family might further power the battle between the immune system and tumor cells. Secreted HLA-G5 can also bind CD8 and induces Fas/Fas ligand-mediated apoptosis in activated CD8+ lymphocytes.

HCMV glycoprotein US6 mediated inhibition of TAP does not affect HLA-E dependent protection of K-562 cells from NK cell lysis

Human cytomegalovirus has evolved multiple strategies to interfere with immune recognition by the host. A variety of mechanisms affect antigen presentation by major histocompatibility complex class I molecules resulting in a reduced class I cell-surface expression. This downregulation is expected to trigger natural killer (NK) cytotoxicity, requiring counteraction by the virus to establish long-term infection. Here we describe that the human cytomegalovirus gpUS6 protein, which has been demonstrated to downregulate the expression of human leukocyte antigen (HLA) class I and the presentation of cytotoxic T lymphocyte epitopes by blocking transporter associated with antigen presentation (TAP function), does not affect the ability of HLA-E to inhibit NK cell mediated lysis of K-562 cells by interaction with CD94/NKG2A expressed on NK cells. Cell surface expression and function of HLA-E is not altered although gpUS6 inhibits TAP-dependent peptide transport by 95%. Moreover, HLA-E molecules presenting HLA class I signal sequence-derived peptides are functionally detectable on transfected TAP-deficient RMA-S cells.

The non-classical MHC molecule HLA-G protects human muscle cells from immune-mediated lysis: implications for myoblast transplantation and gene therapy

HLA-G is a non-classical MHC class I molecule with highly limited tissue distribution which has been attributed chiefly immune-regulatory functions. We previously have reported that HLA-G is expressed in inflamed muscle in vivo and by cultured myoblasts in vitro. Here, we used the in vitro models of human myoblasts or TE671 muscle rhabdomyosarcoma cells to characterize the functional role of HLA-G for muscle immune cell interactions. Gene transfer of the two major isoforms of HLA-G (transmembranous HLA-G1 and soluble HLA-G5) into TE671 rendered these cells resistant to alloreactive lysis by direct inhibition of natural killer (NK) cells, and CD4 and CD8 T cells. Further, HLA-G reduced alloproliferation, interfered with effective priming of antigen-specific cytotoxic T cells and reduced antigen-specific alloreactive lysis. HLA-G pre-induced on cultured myoblasts inhibited lysis by alloreactive peripheral blood mononuclear cells. This protection was reversed by a neutralizing HLA-G antibody. Interestingly, a few HLA-G-positive cells within a population of HLA-G-negative muscle target cells conveyed significant inhibitory effects on alloreactive lysis. Our results reveal further insights into the immunobiology of muscle and suggest that ectopic expression of HLA-G may promote the survival of transplanted myoblasts in the future treatment of hereditary muscle diseases. Further, HLA-G could represent a novel self-derived anti-inflammatory principle applicable in strategies against inflammatory aggression.

Binding analysis of HLA-G specific antibodies to hematopoietic cells isolated from leukemia patients

Expression of HLA-G on the surface of malignant hematopoietic cells isolated from leukemia patients was analyzed by flow cytometry using monoclonal antibodies (mAbs) recognizing both, intact HLA-G complex (87G, 01G and MEM-G9) as well as HLA-G free heavy chain (4H84, MEM-G/1 and MEM-G/2). Prerequisite of HLA-G detection by mAbs specific to free heavy chain was mild acid treatment, which dissociates intact HLA-G complex. All mAbs, with the exception of 4H84 mAb, did not indicate the presence of HLA-G antigen in leukemia cells. Positive staining with 4H84 mAb was detected in acid-treated cells isolated from 16 out of 30 patients. Intensity of staining increased after IFN-g pre-incubation in most cases. Immunoblot analyses and RT-PCR, however, failed to detect HLA-G antigen or HLA-G transcripts in cells that bind 4H84 mAb after acid-treatment. The binding of 4H84 mAb can be explained by the acid-induced cross-reactivity of this HLA-G specific mAb with classical HLA class I molecules [15]. The results described here further demonstrate that the HLA-G molecule is not expressed in freshly isolated human leukemia cells.

A functional role of HLA-G expression in human gliomas: an alternative strategy of immune escape

HLA-G is a nonclassical MHC molecule with highly limited tissue distribution that has been attributed chiefly immune regulatory functions. Glioblastoma is paradigmatic for the capability of human cancers to paralyze the immune system. To delineate the potential role of HLA-G in glioblastoma immunobiology, expression patterns and functional relevance of this MHC class Ib molecule were investigated in glioma cells and brain tissues. HLA-G mRNA expression was detected in six of 12 glioma cell lines in the absence of IFN-gamma and in 10 of 12 cell lines in the presence of IFN-gamma. HLA-G protein was detected in four of 12 cell lines in the absence of IFN-gamma and in eight of 12 cell lines in the presence of IFN-gamma. Immunohistochemical analysis of human brain tumors revealed expression of HLA-G in four of five tissue samples. Functional studies on the role of HLA-G in glioma cells were conducted with alloreactive PBMCs, NK cells, and T cell subpopulations. Expression of membrane-bound HLA-G1 and soluble HLA-G5 inhibited alloreactive and Ag-specific immune responses. Gene transfer of HLA-G1 or HLA-G5 into HLA-G-negative glioma cells (U87MG) rendered cells highly resistant to direct alloreactive lysis, inhibited the alloproliferative response, and prevented efficient priming of cytotoxic T cells. The inhibitory effects of HLA-G were directed against CD8 and CD4 T cells, but appeared to be NK cell independent. Interestingly, few HLA-G-positive cells within a population of HLA-G-negative tumor cells exerted significant immune inhibitory effects. We conclude that the aberrant expression of HLA-G may contribute to immune escape in human glioblastoma.

Th1- and Th2-like cytokine production by first trimester decidual large granular lymphocytes is influenced by HLA-G and HLA-E

During normal early pregnancy, a particular immune environment in the decidua and the expression of non-classical HLA-G and HLA-E molecules on the invading trophoblast are assumed to be essential for the tolerance of the fetus. To assess whether HLA-G and HLA-E influence the cytokine production of their putative target cells [large granular lymphocytes (LGL)], we analysed the concentrations of tumour necrosis factor (TNF-alpha), interferon (IFN)-gamma, interleukin (IL)-10, IL-13 and granulocyte-macrophage colony stimulating factor (GM-CSF) in supernatants of isolated first trimester LGL co-cultured with HLA-G or HLA-E transfected K-562 leukaemia cells lacking the classical HLA class I and II molecules. In comparison with that observed with untransfected K-562 cells, co-culture of LGL with HLA-G-expressing cells significantly reduced the concentration of all cytokines investigated (TNF-alpha, IL-10 and GM-CSF, P < 0.01; IFN-gamma and IL-13, P < 0.05). In contrast, co-culture of LGL with HLA-E-expressing cells significantly (P < 0.01) decreased only IL-10 production, although a strong tendency towards reduced IL-13 levels was also observed. In the co-culture system presented, membrane-bound HLA-G and, to a lesser extent, HLA-E expression affected cytokine release by decidual LGL in a manner not consistent with the Th1/Th2 paradigm. In conclusion, our data are indicative of a general immune-suppressive effect of HLA-G on LGL activity.

[Primary stability of press-fit acetabulum cups using a new acetabulum reamer]

The aim of the present study was to assess the initial stability of uncemented press-fit acetabular components using a newly developed reamer designed to optimize the surgical preparation of the acetabulum. Ten synthetic human pelves were used to investigate the stability of 20 uncemented press-fit acetabular components, each of which was tested in a servohydraulic testing machine for 6 cycles under an axial load of 2.4 kN. The results of the micrometric measurements revealed satisfactory stability for a reaming depth of 2 mm, and a press-fit of 2 mm. Micromotion was less than 200 microns in all the anatomical sections of the acetabulum (ischium 63 microns, pubis 150 microns, ilium 85 microns). A press-fit of 4 mm and the smaller reaming depth of 1 mm were associated with a substantial decrease in mechanical stability.

[New instruments for preparation of the prosthesis socket and primary stability of the acetabular Press-fit cups]

The aim of the present study was to develop surgical instruments necessary to achieve a precisely reamed surface and stable initial fixation. The instruments used to prepare the socket were a gauge-drill guide, a liner for the spigot hole and two spigotted reamers of different design and indentation intended to achieve a precisely reamed surface while preserving subchondral bone. For each reamer we implanted in a synthetic hip model 10 uncemented cups with 2 mm press-fit and loaded at 2.4 kN in the physiological axis (Mod. 8501, Instron, Canton, MA, USA). The micromotion between implant and bone socket was measured using an inductive micrometric measuring system (MultiNCDT-500, Micro-Epsilon, Ortenburg, Germany) and compared with that seen after using conventional instruments. The use of the new reamer of elliptical design significantly reduced the standard deviations of the measured values (p < 0.01 at the ischium and pubis) and also reduced maximum movement (p < 0.01 at the ischium); at the same time, all the components showed overall limited movement (< 150 microns at the ischium, pubis and ilium) under maximum loading (2.4 kN). Manufacturing tolerances, the quality and wear of the instruments, acetabular bone stock and surgical technique all impact on the degree of press-fit obtainable at surgery. The results of our study show that press-fit and initial stability can be optimized by using adequate instruments to prepare the socket.

[Effect of socket preparation on primary stability of an elliptical acetabular component]

The aim of this investigation was to examine, with the aid of a valid micrometric measuring system, the initial stability of an elliptical press-fit acetabular component with regard to the size of the reamer used to prepare the socket, and the depth of reaming. All 10 implanted cups showed marked movement (> 500 microns) at the first loading after implantation. Only the implant with a press-fit of 1 mm and a reaming depth of 2 mm showed small elastic micromovement (< 130 microns) at all three measuring points during cyclic loading. The results indicate that press fit and stability need to be optimized using adequate surgical instrumentation and techniques, and that it is necessary to develop new methods to optimize the cementless anchorage of acetabular components.

[Factors affecting the primary stability of acetabular components. An in vitro study]

The aim of the present study was to investigate the micromotion under physiological loading of various acetabular components with and without screws, to determine the most suitable anchoring, cup design and optimal surface structure. Six acetabular components with varying cup geometry and surface configuration were implanted with a 2 mm press-fit into polyethylene pelves. In the first set of trials, the uncemented cups with two peripheral screws were tested under an axial load of 240 kg (2,354 N). The screws were then removed and the cups, held in place only by press fit, were tested again. None of the uncemented cups achieved the high initial stability of the cemented reference cups. It was not possible to determine an optimal cup design. In this study, titanium plasma-spray-coated cups achieved the best results. Stability is determined only in part by the configuration of the prosthesis. Of equal importance is the quality of the preparation of the cup bed. The use of screws cannot be unreservedly recommended. With a good press fit, the use of screws enhances stability only minimally.