Resistance to apoptosis and escape from host immune surveillance - two related (?) survival mechanisms of malignant glioma cells

DSpace Repositorium (Manakin basiert)


Dateien:

Zitierfähiger Link (URI): http://nbn-resolving.de/urn:nbn:de:bsz:21-opus-14177
http://hdl.handle.net/10900/44547
Dokumentart: Dissertation
Erscheinungsdatum: 2004
Sprache: Englisch
Fakultät: 4 Medizinische Fakultät
Fachbereich: Sonstige
Gutachter: Weller, Michael
Tag der mündl. Prüfung: 2004-10-14
DDC-Klassifikation: 610 - Medizin, Gesundheit
Schlagworte: Gliom , Tumorimmunologie , Apoptosis
Freie Schlagwörter:
resistance to apoptosis , immune escape
Lizenz: http://tobias-lib.uni-tuebingen.de/doku/lic_mit_pod.php?la=de http://tobias-lib.uni-tuebingen.de/doku/lic_mit_pod.php?la=en
Gedruckte Kopie bestellen: Print-on-Demand
Zur Langanzeige

Inhaltszusammenfassung:

Gliomzellen zeichnen sich durch Resistenz gegenüber apoptotischen Stimuli aus. Die Überwindung dieser intrinsischen Apoptose-Resistenz ist daher ein wichtiges Ziel der Gliomforschung. Therapeutische Induktion von Apoptose in Tumorzellen führt in vivo zur Bereitstellung von antigenem Tumorzellmaterial. Dieses kann von dendritischen Zellen aufgenommen und präsentiert werden, sodass eine anti-Tumor-Immunantwort ermöglicht wird. Allerdings verfügen Gliomzellen über eine Reihe immuninhibitorisch wirksamer Mechanismen, die eine Tumorabstoßung verhindern. Zudem stellt die Apoptose-Resistenz von Gliomzellen auch für einen immuntherapeutischen Ansatz ein Hindernis dar, da immun-vermittelte anti-Tumor-Effekte ebenfalls auf der (durch NK und T Zellen vermittelten) Induktion von Apoptose in Krebszellen beruhen. Darüber hinaus können Immunzellen bei Kontakt mit Gliomzellen ebenfalls apoptotisch werden. Daher befasst sich diese Arbeit mit den zellbiologischen Prinzipien des programmierten Zelltods (Apoptose) in Gliomzellen sowie mit den wechselseitigen Beziehungen zwischen Immuneffektorzellen und Gliomzellen, ebenfalls unter besonderer Berücksichtigung der Apoptose. Ein interessantes Agens zur Induktion von Apoptose in Tumorzellen ist das experimentelle Pharmakon CP-31398, das von der Firma Pfizer in den USA mit dem Ziel der Wiederherstellung funktioneller Aktivität in mutanten p53-Proteinen entwickelt wurde. Zur Charakterisierung dieser Substanz wurde unter Zuhilfenahme verschiedene Gentransfer-Techniken ein p53-abhängiger und ein p53-unabhängiger Zelltodweg charakterisiert, die beide durch CP-31398 aktiviert werden. Somit erwies sich CP-31398 als interessante, für den klinischen Gebrauch auf Grund unspezifischer Effekte jedoch nicht hinreichend weit entwickelte Substanz. Die erzielten Ergebnisse könnten aber für die Entwicklung verbesserter p53-modulierender anti-Tumor-Agenzien von Bedeutung sein. Im Rahmen dieser Untersuchungen stellte sich heraus, dass klassische Reporter-Assays für transkriptionelle p53-Aktivität nicht ausschließlich die Aktivität von p53 reflektieren, sondern ein umfassendes Bild der Gesamt-Aktivität von p53, p63, p73 und möglicherweise weiterer, noch nicht bekannter Mitglieder dieser Proteinfamilie liefern. Aufgrund dieses Befundes müssen zahlreiche bisher mit derartigen Reporter-Assays publizierte Daten neu interpretiert werden. In Ergänzung zu CP-31398, das auf die Aktivierung des intrinsischen apoptotischen Mechanismus abzielt (und in Gliomzellen Caspasen-unabhängigen Zelltod induziert), wurde auch der extrinsische oder rezeptor-vermittelte Apoptosemechanismus untersucht. Die bekannten Todesrezeptoren gehören zur Familie der Tumor-Nekrose-Faktor (TNF)-Rezeptoren, die auch bei Immunreaktionen eine wichtige Rolle spielen. Da sich das Expressionsmuster dieser Rezeptoren auf Tumorzellen von demjenigen auf nicht-transformierten Zellen unterscheidet, bieten Todesrezeptoren potentielle therapeutische Angriffspunkte. Systemische Aktivierung des CD95/Fas-Rezeptors führt aber in Mäusen zu Leberversagen und auch die gegenwärtig verfügbaren Präparationen von Apo2L/TRAIL sind klinisch nicht einsetzbar oder ineffektiv. CD40L vermag nach dem bisherigen Kenntnisstand nur in transformierten Zellen Apoptose zu induzieren, was auf eine veränderte Signaltransduktion in Tumorzellen hindeutet. Zudem ist CD40L bereits in immuntherapeutischen Studien klinisch erprobt. Daher wurde im Kontext dieser Arbeit erstmals die Expression und funktionelle Bedeutung von CD40 auf Gliomen untersucht. Dabei wurde CD40-Expression auf allen 12 untersuchten Gliomzelllinien sowie in vivo nachgewiesen. Dennoch wurden keine zytotoxischen Effekte von CD40L auf Gliomzelllinien beobachtet. Da bei Tumoren eine Herunterregulation der CD40-Expression im Verlauf von Langzeit-Kulturen beschrieben ist, wurde eine CD40-überexprimierende Gliomzelllinie generiert. In dieser induziert CD40L Caspasen-abhängigen Zelltod, der durch Hemmung der Proteinbiosynthese potenziert wird. Gleichzeitig kommt es zu einem verstärkten proteasomalen Abbau von CD95/Fas, wodurch die Zellen ihre Sensitivität gegenüber CD95/Fas-vermittelter Apoptose weitgehend verlieren. TNF-alpha-abhängige Apoptose wird hingegen durch subtoxische Konzentrationen von CD40L verstärkt, was durch eine neu entdeckte intrazelluläre Wechselwirkung zwischen CD40 und TNF R1 (p55) erklärt werden kann. Schließlich wurde gezeigt, dass die endogene CD40-Expression von Gliomzellen dazu benützt werden kann, Apoptose mittels eines bispezifischen CD40xCD95 Antikörpers zu induzieren. Somit wurden in diesem Teil der Arbeit komplexe Wechselwirkungen zwischen CD40 und verschiedenen anderen TNF-Rezeptoren nachgewiesen. Ein weiteres Molekül der TNF-Rezeptor-Familie ist CD27, dessen Ligand CD70 im Rahmen einer Array-Analyse als radio-induzierbares Gen in U87MG-Gliomzellen identifiziert wurde. Dieser Befund war überraschend, weil CD70-Expression bisher lediglich auf T- und B-Zellen nachgewiesen worden war. Während CD70/CD27-Interaktionen dort immunstimulatorisch wirkt (T, B und NK Zell-Aktivierung), zeigten umfangreiche immunologische Versuche, dass die Expression von CD70 auf Gliomen CD27-vermittelte Apoptose in Immuneffektorzellen zu induzieren vermag und somit möglicherweise zur Immuninhibition bei Gehirntumorpatienten beiträgt. Als potentieller Schutzmechanismus wurde die Abspaltung von löslichem CD27 von Immunzellen beobachtet. Ein bereits bekannter Mediator der Immuninhibition in der Plazenta und bei Tumoren ist HLA-G, dessen funktionelle Relevanz auf Gliomen erstmals untersucht wurde. Dabei gelang der Nachweis der Expression in vitro und in vivo sowie die Charakterisierung einer Reihe T Zell-vermittelter immunsuppressiver Effekte: So wird sowohl die durch allogene Gliomzellen induzierte Proliferation als auch die direkte allogene und die antigen-spezifische Lyse nach Stimulation mit bestrahlten Tumorzellen durch lösliches und membranständiges HLA-G inhibiert. Interessanterweise genügt auch in Abwesenheit von löslichem HLA-G ein relativ geringer Anteil HLA-G positiver Tumorzellen, um einen Schutz der benachbarten HLA-G negativen Gliomzellen zu vermitteln. Neben diesen hier erstmals bei Gliomen beschriebenen Mediatoren der tumor-induzierten Immunsuppression wurden auch noch Untersuchungen zu TGF-beta durchgeführt, das bereits 1987 von Adriano Fontana als „glioblastoma derived T cell suppressive factor“ identifiziert wurde. Insbesondere wurden die Effekte von TGF-beta auf NKG2D und die zugehörigen Liganden untersucht. NKG2D ist ein aktivatorischer Rezeptor, der auf NK, CD8+ alpha beta und gamma delta T Zellen exprimiert wird. TGF-beta supprimiert die NKG2D-Expression auf NK und CD8+ T Zellen. Klinisch korreliert dies mit einer verringerten NKG2D-Expression bei Gliompatienten. Die NKG2D Liganden MICA/B (MHC class I-chain related molecule A/B) und ULBP1-4 (UL16-binding proteins 1-4) werden selektiv von infizierten und von transformierten Zellen exprimiert und können die immunvermittelte Lyse dieser Zellen induzieren. Anhand von TGF-beta1,2 siRNA-Zellen, in denen die Expression von TGF-beta1 und 2 durch plasmidbasierte RNA-Interferenz inhibiert ist, konnte gezeigt werden, dass TGF-beta nicht nur die Expression des NKG2D-Rezeptors, sondern auch die mRNA und Protein-Expression der NKG2D-Liganden MICA, ULBP2 und ULBP4 inhibiert, während sich kein Effekt auf MICB, ULBP1 und ULBP3 zeigte. Zudem fördert TGF-beta die Freisetzung der Matrix Metalloproteinasen MMP-2 und MMP-9 (nicht von MT1-MMP), die wiederum die Abspaltung von MICA von der Zelloberfläche vermitteln, sodass es auch noch einen posttranslationalen Effekt von TGF-beta auf NKG2D-vermittelte anti-Tumor Immunreaktionen gibt. Zudem führt der „knock-down“ von TGF-beta zum Verlust des migratorischen und invasiven Potentials der Tumorzellen, vermutlich ebenfalls aufgrund der Herunterregulation der Matrix Metalloproteinasen. Die geringere Motilität und die erhöhte Immunogenität der „TGF-beta knock-down Zellen“ führt zu einem Phänotyp, der in Nacktmäusen nicht mehr tumorigen ist. Diese Daten zeigen bisher unbekannte Wechselwirkungen zwischen TGF-beta und dem NKG2D Signalweg und belegen somit die Bedeutung von TGF-beta als einem zentralen Molekül der Immunsuppression bei Gliomen. Anti-TGF-beta-Strategien könnten daher therapeutische Bedeutung erlangen. In diesem Zusammenhang wurde die Wirksamkeit von SD-208, einem neuen TGF-betaRI Kinase Inhibitor, auf Wachstum und Immunogenität von murinen SMA-560 und humanen LN-308 Gliomzellen untersucht. SD-208 blockiert die autokrine und die parakrine TGF-beta Signaltransduktion in Gliom- und Immunzellen, wie über die Phosphorylierung von Smad2 und die Induktion TGF-beta abhängiger Reportergene nachgewiesen wurde. Zudem wird die TGF-beta-abhängige Migration und Invasion von Gliomzellen gehemmt. Die allogene Lyse von Gliomzellen durch restimulierte humane Lymphozyten wird durch SD-208 erhöht, zudem sezernieren mit Gliomzellen kokultivierte T Zellen in Gegenwart von SD-208 mehr IFN-beta und TNF-beta und weniger IL-10. Zudem kann SD-208 die durch Zugabe von Gliomzellüberstand bewirkte Inhibition polyklonaler NK Zellen aufheben. Die orale Bioverfügbarkeit von SD-208 wurde durch einen phospho-Smad2-ELISA aus Milz und Gehirn TGF-beta behandelter Mäuse nachgewiesen. Systemische Behandlung von VM-Dk Mäusen mit SD-208 verlängert das mediane Überleben der Mäuse nach intrakranieller Inokulation mit syngegen SMA-560 Gliomzellen von 18.6 auf 25.1 Tage. Dieser therapeutische Effekt liesse sich durch geeignete Kombinationen mit adjuvanter Chemo- oder Immuntherapie möglicherweise noch verbessern. Somit stellen TGF-betaRI Kinase Inhibitors wie SD-208 vielversprechende neue Therapeutika für die Behandlung von malignen Gliomen und anderen TGF-beta-abhängigen Erkrankungen dar. Somit könnten einzelne der im Rahmen dieser Dissertation zusammengefassten Projekte Grundlagen für neue, molekular definierte Therapieansätze bei Gliomen liefern. Bis zu einer klinischen Anwendung der hier vorgestellten Konzepte müssen die immun-inhibitorischen Eigenschaften von Gliomzellen noch weiter erforscht und insbesondere klinisch einsetzbare Agenzien für eine kombinierte Apoptose-Immuntherapie entwickelt werden.

Abstract:

Our improved understanding of cancer biology has helped to establish criteria that appear to be prerequisites for tumour formation by transformed cells. Cancer cells must (i) be able to grow autonomously, (ii) develop insensitivity to negative growth regulation, (iii) display unlimited replicative potential, (iv) develop the capacity for angiogenesis, (v) develop competence for invasive growth and metastasis and (vi) evade intrinsic apoptotic signals. A further emerging precondition for the malignant phenotype in vivo is the capacity of a malignant cell to evade the extrinsic tumour suppressor functions of the immune system. Glioma cells display all of these “hallmarks of cancer”. Their intrinsic resistance to apoptosis and their immunosuppressive properties (chapter 1) have been analyzed in our laboratory and were also the main topics explored in my Ph.D. work. p53 is a potent activator of the intrinsic apoptotic pathway. It represents an ideal target for anti-cancer drug design, since it is mutated in more than half of human tumours. Most of the remaining tumours, although carrying wild-type p53, have defects in the p53-mediated apoptotic pathway. Unlike most other tumour suppressor genes, mutant p53 is overexpressed in tumour cells. Therefore it can function as a tumour antigen. However, the approach that was investigated here (chapter 2) aimed at restoring a tumour-suppressor function in mutant p53 variants. CP-31398 is a prototype small molecule that stabilizes the active conformation of p53 and promotes p53 activity in cancer cell lines with mutant or wild-type p53. CP-31398 was found to induce p53-like activity in all of 11 glioma cell lines harbouring wild-type or mutant p53, but not in p53 null LN-308 cells. However, upon prolonged exposure to CP-31398, all glioma cell lines, irrespective of their p53 status, undergo caspase-independent cell death displaying some characteristics of apoptosis. Posttranscriptional repression of p53 by an intracellularly transcribed siRNA as well as analysis of p53 transgenes in p53-deleted glioma cells allowed to delineate two pathways of CP-31398-induced cell death: an early, p53-dependent pathway that requires new p53 protein synthesis and a late, p53-independent pathway characterized by calcium release and epiphenomenal free radical formation. These observations point out some of the liabilities of CP-31398 as a prototype p53-based therapeutic and define a rationale for the further refinement of small molecules which specifically target the p53 pathway, but lack the p53-independent effects. At the current stage of development, CP-31398 has promising effects on mutant and wild-type p53 in cancer cell lines, but its strong p53-independent side effects are likely to prevent its use in the clinic. Surprisingly, posttranscriptional repression of p53 by an intracellularly transcribed siRNA in a p53 mt cell line resulted in the induction of a low, but detectable p53-like reporter gene activity. This observation prompted us to investigate the specificity of Stratagene´s p53 PathDetect reporter plasmid (chapter 3), a widely used luciferase-coupled reporter gene for the detection of the transcriptional activity of p53. Both p63 and p73 were found to activate the promoter sequence contained in the p53-luc vector. This shows that data obtained using this system (or any similar one) should be interpreted with caution because they may not actually reflect p53 activity, but rather provide an integrated signal derived from different members of the p53 family. However, this does not seriously challenge the reporter gene data obtained with CP-31398, since a net effect of CP-31398 on p53, p63 and p73 is a physiologically perfectly sensible measure of CP-31398-dependent p53-related activity. In fact, treatment of glioma cells with CP-31398 also increases the cellular level of p73, while p63 is absent from most glioma cell lines (data not shown). Further, the potency of CP-31398 as an inducer of apoptosis may even be enhanced by a cooperative effect on further members of the p53 protein family. Whereas in glioma cells CP-31398 activates an intrinsic, caspase-independent cell death pathway, the extrinsic or death receptor mediated apoptotic pathway may also potentially be exploited for an anti-cancer therapy, since the distribution of death receptors found on tumour cells apparently differs from that found on non-transformed cells. Further, the signalling originating from these receptors may be different in tumour cells. This has been described for CD40, one of several members of the TNFR family that lack a death domain and yet can induce apoptosis. Physiologically, CD40 transduces antiapoptotic signalling in B and T cells whereas CD40-mediated apoptosis appears to be restricted to transformed cells. Therefore adverse effects like the massive hepatotoxicity of CD95/FasL are unlikely to occur with CD40. Instead, ligation of CD40 can induce antitumour immunity in vivo in mice. CD40 is expression was detected in all of 12 human glioma cell lines (chapter 4), even though in most cell lines the expression seems to be mainly intracellular. In contrast to various non-glial carcinoma cell lines, CD40-expressing glioma cells resist cytotoxic effects of CD40L. Resistance to CD40L-induced cell death has been described as the result of CD40 downregulation during long-term culture. CD40 gene transfer into LN-18 glioma cells induces (or restores?) susceptibility to cell death by soluble CD40L. Similar to CD95L- or Apo2L/TRAIL-induced apoptosis, CD40-dependent cell death is potentiated by CHX and involves receptor clustering and caspase 8 and 3 processing. Surprisingly, CD40-transfected LN-18 cells acquire resistance to CD95L, resulting from the down-regulation of CD95 expression via a posttranslational, proteasome inhibitor-sensitive mechanism. In contrast, subtoxic concentrations of CD40L strongly sensitize these cells for TNF-alpha-induced apoptosis via a mechanism involving intracellular crosstalk between CD40 and TNF-R1 (p55). This suggests complex patterns of modulation of death receptor-mediated glioma cell apoptosis by CD40/CD40L interactions. Further, bispecific CD40xCD95, but not CD20xCD95, antibodies kill glioma cells, disclosing the property of endogenous CD40 to facilitate death signalling. However, the data obtained on CD40 in glioma cells do not strongly support a therapeutical application of CD40L as an inducer of apoptosis. On the other hand, CD40L had no stimulatory or antiapoptotic effects on glioma cells and could therefore be used as an adjuvant in a potential anti-glioma immunotherapy. Another TNFR-related cell surface receptor implicated in T, B and NK cell activation as well as in apoptosis is CD27. The ligand for CD27, CD70 was identified as a radio-inducible gene in U87MG glioma cells (chapter 5). A screening of a panel of human glioma cell lines revealed that 11 of 12 cell lines expressed CD70 mRNA and protein. CD70 protein was also detected by immunocytochemistry in 5 of 12 glioblastomas and 3 of 4 anaplastic astrocytomas in vivo. CD27 expression was not detected in any glioma cell line, and there was no evidence for autocrine or backward signalling of the CD70 system in human glioma cells. Unexpectedly, CD70 expressed on glioma cells did not increase the immunogenicity of glioma cells in vitro. In contrast, CD70-positive glioma cells induced apoptosis in PBMC in a CD70-dependent manner. Neutralization of CD70 expressed on glioma cells prevented apoptosis and enhanced the release of TNF-alpha in cocultures of glioma cells and PBMC. The effects of CD70-expressing glioma cells on PBMC were mimicked by agonistic CD27 antibodies. Conversely, the shedding of CD27 by PBMC was identified as a possible escape mechanism from glioma cell-induced CD70-dependent apoptosis. Thus, induction of B cell and T cell apoptosis via interactions of CD70 expressed on glioma cells and CD27 expressed on B and T cells may be a novel way for the immune escape of malignant gliomas. Our in vitro results are in contrast to in vivo experiments describing CD27-mediated tumour rejection. Further, a study using CD27 knockouts supports a role for CD27 in costimulation. However, chronic stimulation via CD70 as occurs in CD70 transgenic mice results in B cell depletion and lethal T cell immunodeficiency. Further, the immune system of glioma patients does not reject CD70-expressing tumours in vivo. Instead, the high proportion of CD70-positive gliomas found in this and a related study suggests that there may rather be a selection pressure operating in favour of CD70-expressing glioma cells. These apparently contradictory data suggest that CD27 may be functioning in vivo together with other receptors that determine the outcome of CD27 stimulation. Thus CD27 may signal either activation or apoptosis, depending on the context. Given the capability of glioma cells to paralyse anti-tumour immune reactions, it is well conceivable that the stimulatory pathways originating from CD27 are inhibited in glioma patients whereas the pathway leading to T cell apoptosis may be enforced. One newly discovered mechanism by which glioma cells apparently inhibit anti-tumour T cell responses is expression and secretion of HLA-G molecules (chapter 6). HLA-G is a nonclassical MHC molecule with highly limited tissue distribution that has been implicated in foetal semi-allograft tolerance during pregnancy. 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 cell lines and brain tissue. HLA-G protein was detected in 4 of 12 cell lines in the absence, and in 8 out of 12 cell lines in the presence of IFN-gamma. Immunohistochemical analysis of human brain tumours revealed expression of HLA-G in 4 of 5 tissue samples. Functional studies showed that expression of membrane-bound HLA-G1 and soluble HLA-G5 inhibited alloreactive and antigen-specific immune responses. However, HLA-G dependent T cell apoptosis was not detected. Gene transfer of HLA-G1 or HLA-G5 into HLA-G-negative glioma cells rendered these 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 tumour cells exerted significant immune inhibitory effects. These data suggest that aberrant expression of HLA-G may contribute to immune escape in human glioblastoma. A long-established determinant of glioblastoma-induced immunosuppression is the cytokine TGF-beta. However, TGF-beta does not only compromise anti-tumour immune responses and induce T cell apoptosis, but also stimulates angiogenesis and invasion. Therefore, TGF-beta has become a major target for the experimental treatment of human malignant gliomas. The effects of TGF-beta on NKG2D-mediated anti-tumour immunity were investigated (chapter 8). NKG2D is an activating immunoreceptor expressed by NK, CD8+ alpha beta and gamma delta T cells. Human NKG2DL comprise MHC class I-chain related molecules A (MICA) and MICB and UL16-binding proteins (ULBP) 1, 2, 3, and 4. These ligands are expressed by infected and transformed cells and transmit danger signals to immune cells, leading to the lysis of NKG2DL-expressing cells. An effect of TGF-beta on the expression of the activating immunoreceptor NKG2D in CD8+ T and NK cells has been shown by us and others in vitro. Accordingly, a downregulation of NKG2D is also observed in glioblastoma patients in vivo. Moreover, TGF-beta inhibits the transcription of the NKG2D ligands MICA, ULBP2 and ULBP4 and reduces their protein level at the cell surface. In contrast, MICB, ULBP1 and ULBP3 are unaffected by TGF-beta. Further, TGF-beta promotes the release of matrix metalloproteinases (MMP) by glioma cells which in turn mediate the shedding of MICA from the surface of glioma cells. siRNA-mediated “knock-down“ of TGF-beta synthesis in LN-229 glioma cells suppresses MMP expression and strongly enhances the immunogenicity of glioma cells by enhancing MICA, ULBP2 and ULBP4 expression and by preventing NKG2D down-regulation on immune cells treated with glioma cell SN. Further, the TGF-beta knock-down leads to a decreased migratory and invasive tumour phenotype. Altogether, the knock-down of TGF-beta results in a glioma cell phenotype which is less motile, more sensitive to immune cell lysis, and non-tumorigenic in nude mice. These data define downregulation of NKG2D expression on immune effector cells and of NKG2DL expression on glioma cells as a further mechanism by which TGF-beta may promote immune escape. This confirms the importance of TGF-beta as a principle inhibitor of anti-glioma immune responses. Strategies to counteract the immunosuppressive activities of TGF-beta may be crucial to improve the current cancer immunotherapeutic approaches. In this context, a novel TGF-beta RI kinase inhibitor, SD-208, was characterized for its effects on the growth and immunogenicity of murine SMA-560 and human LN-308 glioma cells in vitro and the growth of, and immune response to, intracranial SMA-560 gliomas in syngeneic VM/Dk mice in vivo (chapter 7). SD-208 blocks autocrine and paracrine TGF-beta signalling in glioma cells as detected by phosphorylation of Smad2 or TGF-beta reporter assays and strongly inhibits constitutive and TGF-beta-evoked migration and invasion. PBL or purified T cells, cocultured with TGF-beta-releasing LN-308 glioma cells in the presence of SD-208, exhibit enhanced lytic activity against LN-308 targets. The release of IFN-gamma and TNF-alpha by these immune effector cells is enhanced by SD-208 whereas the release of IL-10 is reduced. SD-208 restores the lytic activity of polyclonal NK cells against glioma cells in the presence of recombinant TGF-beta or TGF-beta-containing glioma cell SN. The oral bioavailability of SD-208 was verified by demonstrating the inhibition of TGF-beta-induced Smad phosphorylation in spleen and brain. Systemic SD-208 treatment initiated three days after the implantation of SMA-560 cells into the brains of syngeneic VM/Dk mice prolongs their median survival from 18.6 to 25.1 days. Importantly, SD-208 could be administered together with adjuvant chemotherapy or an immunotherapeutic vaccine. Such a combined treatment might further improve the effect achieved by SD-208 alone. Thus, TGF-beta RI kinase inhibitors such as SD-208 are promising novel agents for the treatment of human malignant glioma and other conditions associated with pathological TGF-beta activity. In my Ph.D. research I have thus explored several novel approaches aiming at the selective induction of apoptosis in glioma cells and at an improved understanding of the immune-paralysing capacities of glioma cells. While the connections between these various aspects of glioma biology may not be apparent at first sight, there are good reasons for taking a combined look at apoptosis and tumour immunology: Tumour immunosurveillance is mediated by immune effector cells that selectively induce apoptosis in cancer cells. Activated T and NK attack their targets by expression death ligands and by secretion cytotoxic granules and thus seek to activate both the death receptor and the intracellular apoptotic pathway. An apoptosis-resistant target cell is therefore less susceptible to immune-mediated lysis. Sensitization of tumour cells for apoptotic stimuli could therefore enhance the efficacy of ongoing anti-tumour immune responses. Further, the tumour may strike back and induce apoptosis in tumour infiltrating lymphocytes by various mechanisms, including the expression of CD95/FasL, HLA-G, TGF-beta and, under specific circumstances, CD70 (see section 5). Prevention of tumour cell-induced apoptosis of immune effector cells could also support endogenous anti-tumour responses. Finally, therapeutic induction of apoptosis in tumour cells is thought to provide a source of antigens that can be taken up and presented by professional APC, resulting in an anti-tumour immune response. This, of course, implies that the low rates of spontaneously occurring apoptosis in vivo are simply insufficient to trigger an anti-glioma immune response. Further, the mounting of a productive anti-tumour immune response may be prevented by immune-inhibitory signals sent out by the tumour itself. Therefore, an apoptosis-based reduction of large tumour masses may require the simultaneous relief of tumour-dependent immunosuppression in order to be followed by a productive immune response that can finally clear residual glioma cells dispersed in the brain. Whether the emerging concepts outlined in this Ph.D. thesis have the potential to improve current glioma treatments will, of course, have to be explored in further models. Thus, the present work finally confirms George Bernard Shaw’s notion: “Science... never solves a problem without creating ten more.”

Das Dokument erscheint in: