Tumour-associated vasculature provides an accessible target for systemic gene therapy using targeted adenoviruses. The aim of this thesis is to develop strategies for targeting adenovirus infection to tumour-associated endothelium. Adenovirus expressing luciferase (Adluc) was coated with an amino-reactive polymer based on poly [N-(2-hydroxypropyl) methacrylamide] [pHPMA] to ablate normal infection pathways¬. This was a pre-requisite to redirecting virus tropism to infect endothelial cells via specific receptors. Direct attachment to the pHPMA-adenovirus (pcAdluc) of ligands including vascular endothelial growth factor (VEGF165) and a monoclonal antibody (RAFL) recognising VEGF receptor 2 (VEGFR-2) retargeted infectivity to VEGFR-2-positive endothelial cells and not to receptor-negative cells. Specificity of transduction in vitro was shown by competition with excess antibody. In vivo however, the VEGF165-retargeted virus failed to transduce tumour-associated endothelia following systemic administration. Similarly, direct linkage of a monoclonal antibody against E-Selectin (MHES) demonstrated E-Selectin-specific transduction of tumour necrosis factor-α (TNF-α)-activated endothelial cells, although overall levels of infection were not increased compared to unmodified Adluc. A two-component targeting system using protein A or protein G as ‘bridging’ agents was developed to ensure the required orientation of targeting antibodies. Using this system MHES mediated greater transduction of TNF-α-activated endothelial cells than Adluc. Conjugation using protein A also gave non-specific effects which were not seen with protein G. Whereas the unmodified Adluc virus failed to transduce TNF-α-activated endothelium in an umbilical vein model ex vivo, the MHES-protein G-pHPMA-adenovirus (MHES-StrepGpcAdluc) mediated good transduction. Similarly, StrepGpcAdluc retargeted with a chimeric P-Selectin Glycoprotein Ligand-1 (PSGL-1)-Fc fusion protein, showed good circulation kinetics and significant uptake into HepG2 xenografts following intravenous administration. Histological studies suggested selective targeting to tumour-associated endothelial cells. Overall these findings support the assertion that tumour-associated vasculature is an accessible target for systemic gene delivery, and the use of protein G as bridging agent facilitates rapid screening of Fc-bearing ligands for retargeting pcAd infection to tumour-associated endothelium.

Radiation-induced damage to the genomic DNA of cells may lead to errors in transcription and replication and, if not repaired correctly, these may result in mutations, genomic instability and cell death. Laser microbeams have generally been used by many research groups to investigate the real-time dynamics of protein recruitment in response to DNA insults in mammalian cells; however, such irradiations induce a plethora of DNA damage (including UV base damage, base damage, SSBs and DSBs and complex damage). A novel experimental setup has been designed capable of following the dynamics of protein recruitment in response to DNA insults in mammalian cells shortly following submicrosecond- pulsed electron irradiation of living mammalian cells, not possible using conventional irradiation techniques. This arrangement was developed based on a 6 MeV electron pulse linear accelerator, to deliver sparsely ionising radiation, coupled to an automated, time-lapse inverted epifluorescence microscope imaging system. An integrated robotic system contained within a physiological environment of 37°C and 5% CO₂ was used to transfer remotely and repetitively custom-designed cell dishes containing the mammalian cells between irradiation and imaging locations. Following the development of the linear accelerator and associated imaging devices, preliminary ‘proof-of-principle’ investigations were carried out using living HT1080 mammalian cells containing YFP-tagged 53BP1, an established biomarker of DSB, to follow the recruitment and loss of 53BP1 to sites of radiation-induced DNA damage in real-time. This novel experimental setup has allowed for the first time observations of the appearance and disappearance of radiation-induced foci in the same cell population at very early times. These single-foci studies have provided evidence for the formation of not only promptly formed DSBs but also late appearing DNA damage signalled by 53BP1. These data highlight different classes of DSBs formed in response radiation damage. Additionally, the role of cell cycle on the repair kinetics was undertaken using HT1080- 53BP1-YFP cells which also express Geminin-mCherry under appropriate selection. Geminin is increasingly expressed from early S-phase onwards, but is degraded following mitosis. Geminin-associated fluorescence can be used as a marker of progression through the cell cycle. 53BP1 repair kinetics were characterised in response to radiation damage in combination with ATM and PARP inhibitors. These studies provided supporting evidence for the existence of different classes of DSBs, potentially assigned to radiation-induced replication breaks and DSBs formed by enzymatic conversion of clustered damage. These preliminary ‘proof-of-principle’ findings using DNA damage repair as an example, emphasize the use of this novel technology to explore the dynamics of numerous other biochemical processes in living cells in real-time with the knowledge of being able to quantify the range of damage induced by IR coupled with accurate dosimetry. The knowledge obtained may be used to identify potential biological targets coupled with drug discovery for translation into adjuncts for radiotherapy.

Hypoxia (low physiological O₂ levels) is a characteristic of solid tumours. It not only alters the chemical microenvironment of a tumour but initiates a number of mechanisms which enable cells to cope and thrive under these conditions, resulting in therapy-resistant and aggressive tumours. The replication stress induced by severe hypoxia activates a DNA damage response which involves the kinases ATR and Chk1. Moreover, periods of hypoxia are often followed by reoxygenation, which induces DNA damage. Chk1 inhibitors have been used to potentiate chemotherapy with cytotoxic agents and have recently been proposed as single agents in tumours with high levels of replication stress. However, inhibition of Chk1 also affects normal DNA replication, cell cycle progression and DNA repair. The herein presented study chose known inhibitors of Chk1 and, with methods of synthetic organic chemistry, modified them into agents to selectively target hypoxic cells. Three different Chk1 inhibitors were selected and bioreductive analogues synthesised which were evaluated in chemical, biochemical and cellular assays. We found a convenient route to access a precursor of the bioreductive 2-nitroimidazole group and established a three-step protocol for the testing of bioreductive drugs. This protocol allows us to determine whether a bioreductive drug undergoes reduction and prodrug activation. In addition, bioreductive Chk1 inhibitors were shown to induce DNA damage and cellular toxicity in a hypoxia-selective fashion. While reduction of the prodrugs occurred in all three cases, fragmentation was always the rate-limiting step. We propose that the use of bioreductive Chk1 inhibitors is a promising strategy to target the most therapy-resistant tumour fraction while sparing normal tissue.

Zusammenfassung Einleitung: Die Kraniotomie mit umfassender Tumorresektion bleibt Therapie der Wahl zur Behandlung von Hirntumoren. Bei der geforderten Radikalität des therapeutischen Vorgehens kommt der präoperativen Lokalisationsdiagnostik eloquenter Hirnareale eine besondere Bedeutung zu. In der vorliegenden Arbeit wird ein Verfahren vorgestellt, daß den präzisen Übertrag der Ergebnisse funktioneller MRT-Studien in die Therapie von Hirntumoren ermöglicht. Desweiteren wird das klinische Potential der fMRT untersucht. In einem hierarchischem System erfolgt die Beurteilung der klinischen Wirksamkeit der Methode zur präoperativen Lokalisationsdiagnostik eloquenter Hirnareale bei Patienten mit Hirntumoren. Methode: Bei 40 Patienten mit supratentoriellen Hirntumoren wurden insgesamt 144 präoperative funktionelle MRT-Studien durchgeführt. Die Bewertung der klinischen Wirksamkeit erfolgte in einem hierarchischem Modell unter Betrachtung der aufgeführten Dimensionen: 1. Ebene: Technische Wirksamkeit 2. Ebene: Wirksamkeit in Bezug auf die diagnostische Genauigkeit 3. Ebene: Wirksamkeit in Bezug auf das diagnostische Denken 4. Ebene: Therapeutische Wirksamkeit 5. Ebene: Wirksamkeit in Bezug auf das Patient-Outcome 6. Ebene: Wirksamkeit in Bezug auf die Gesellschaft Die Ergebnisse der funktionellen MRT-Untersuchungen wurden in ein neurochirurgisches Navigationssystem eingebracht. Intraoperativ besteht für den Operateur die Möglichkeit sich die Lagebeziehung zu den gekennzeichneten Arealen in das Okular des Operationsmikroskop einzuspielen. Ergebnisse: Das geschilderte Verfahren ermöglicht die navigierte Operation von Hirntumoren unter besonderer Berücksichtigung eloquenter- das heißt funktionstragender Hirnareale. Die beschriebene Methode zur Integration der fMRT weist ein hohe Praktikabilität auf. Wie diese Arbeit zeigen konnte, erbringt die fMRT als Methode auch bei Patienten mit Hirntumoren für die klinische Anwendung ausreichend zuverlässige Ergebnisse. Schlussfolgerung: 1. Mit dem geschilderten Verfahren gelingt die zuverlässige Integration von fMRT-Daten in die Therapie von Hirntumoren. 2. Die fMRT ist für den klinischen Einsatz zur präoperativen Lokalisation eloquenter Hirntumoren mit Einschränkungen geeignet. / Abstract Introduction: Craniotomy and maximal tumour resection remains the major therapy in patients with brain tumours. Preoperatively it is of great importance to identify eloquent brain areas. In this study we develop a method which allows the precise integration of functional MR-data into the therapy of brain tumours. Additionally we investigated the diagnostic potential of functional MRI in a clinical setting. We were assessing the effectiveness of functional MRI in patients with brain tumours in a hierarchical system. Methods: We performed 122 preoperative, functional MRI studies in 40 patients with supratentorial brain tumours. We evaluated the effectiveness in a hierarchic model. 1. level: technical effectiveness 2. level: diagnostic effectiveness 3. level: diagnostic effectiveness 4. level: therapeutical effectiveness 5. level: patient-outcome 6. level: society The acquired parametric maps were integrated into a neurosurgical navigation system. Intraoperatively the neurosurgeon can have the localisation of functional brain areas displayed within the optical system of the microscope. Conclusion: 1. The described approach allows image guided neurosurgery paying special attention to eloquent brain areas. The method is very practicable and reliable. 2. As we could demonstrate in our work, the functional MRI is sufficiently robust for clinical application

Bei der Replikation monotoner Sequenzen tritt theoretisch ein Vorgang auf, den wir als „Basenkonkurrenz“ bezeichnen: Da sich an jeder Replikations-Stelle mehrere Basenbausteine bewerben, aber immer nur einer benötigt wird, bewerben sich die übrig gebliebenen Bausteine an den jeweils nächsten Replikations - Positionen und erlangen wegen der fortwährenden Beschleunigung durch elektrostatische Anziehung immer größere kinetische Energien. Das führt dazu, dass an einer bestimmten Stelle der replizierenden monotonen Sequenz der eine Partner der Wasserstoffbrückenbindung ein hohes Energieniveau erreicht. Es wird berechnet, dass sich dadurch kurzzeitig eine sehr hohe Bindungsenergie zwischen den beiden Partnern der Wasserstoffbrückenbindung einstellt, wodurch der in dieser kurzen Zeitspanne wirkende DNA-Reparaturmechanismus unterdrückt wird. Die Auswirkungen der hohen Basenkonkurrenz – Energien werden berechnet (hohe Bindungsenergien der Wasserstoffbrückenbindungen, Tunnelvorgänge, irreparable Mutationen). Die Folgen dieser Erscheinung sind Tumorbildung, Alterung, Veränderung der DNA – Struktur, Beeinflussung der Evolution, worauf im Einzelnen eingegangen wird. Es zeigt sich, dass die negativen Auswirkungen der Basenkonkurrenz vorwiegend bei zu niedriger Viskosität des Zellplasmas auftreten.

High interstitial fluid pressure (IFP) in tumours is recognized as a barrier to drug delivery resulting in reduced efficacy. High IFP impedes the normal process of convective fluid flow (CFF) from blood vessels into the interstitium. The aim of this study was to develop an in vitro model that could be used to measure CFF and to study its effects on drug delivery. The model consists of a transwell cell culture insert which supports the growth of multicell layers (MCL) on collagen coated membranes. A graduated tube is inserted into the transwell and a pressure gradient is applied across the membrane by raising the volume of medium in the tube above that of the bottom chamber. CFF is determined by measuring the weight of medium in the bottom chamber as a function of time. CFF was inversely proportional to MCL thickness and 41.1±3.6µm thick MCL has completely stopped CFF. Using a physiologically relevant hydrostatic pressure of 28mmHg, a CFF of 21µL/min was recorded using a DLD-1 MCL that was 12.21±3.2µm thick. Under these conditions, the rates of penetration of doxorubicin, imatinib and gefitinib were respectively 42, 26 and 13 folds greater than when no CFF exists. Reversing the CFF so that it opposed the drug diffusion gradient significantly impairs drug penetration. In conclusion, a novel in vitro model for assessing the impact of CFF on drug delivery has been developed. This model could be used to evaluate strategies designed to increase drug delivery to solid tumours by modifying the CFF.

No description available.

Introduction and aims. South Africa is the epicenter of the HIV/AIDS pandemic. Hypertensive disorders of pregnancy (15.7%) are the second cause of maternal deaths of which pre-eclampsia represents 83%. Normal pregnancy requires a balance between pro- and anti-angiogenic factors to necessitate effective vasculogenesis, angiogenesis and placental development, however, pre-eclampsia is characterised by an excess anti-angiogenic state. The hypoxic placenta releases excess anti-angiogenic factors into the maternal circulation causing endothelial dysfunction. However, there is no data to verify if HIV infection affects pre-eclampsia, or if the angiogenic imbalance is affected. Contradictory data exists on the association between HIV infection and pre-eclampsia. In an attempt to assess the role of HIV infection in pre-eclampsia, this study examined the immunolocalisation of sFlt-1, sEng, PlGF and VEGF in placentae of HIV negative and positive normotensive and pre-eclamptic pregnancies at term using immunohistochemistry (IHC) and immunoelectron microscopy (IEM). Additionally, we estimated the placental expression of sFlt-1, sEng, PlGF and VEGF to verify if the HIV negative differed from the HIV positive cohorts. We further evaluated the maternal serum to determine if variations existed in the circulating levels of these factors in HIV negative and positive normotensive and pre-eclamptic pregnancies. Methods. Following institutional ethical approval and informed consent, placental biopsies and maternal serum were collected post-delivery. For IHC and IEM, 130 and 25 placentae were evaluated, respectively. Following conventional immunohistochemical processing, 5μm sections were immunostained & immunoexpression of the various antibodies were evaluated with the Zeiss Axioscope A1 interfaced with an AxioVision Image analysis software package (version 4.8.3) in combination with the auto-measurement module (Carl Zeiss, Germany). Post-conventional immunoelectron processing, ultra-thin sections were immunolabelled. Sections were post-fixed, contrast enhanced with uranyl acetate and Reynolds lead citrate and viewed on a Jeol 1011 Transmission Electron Microscope. Additionally, the placental expressions of these factors were assessed using RT-PCR. In an attempt to confirm if maternal circulating levels of these factors differed, we quantitatively evaluated these factors in serum from HIV negative normotensives, HIV negative pre-eclamptics, HIV positive normotensives, and HIV positive pre-eclamptics using ELISA techniques. Results and Discussion. The expression of sFlt-1, sEng, PlGF and VEGF was confirmed using immunohistochemistry, RT-PCR and ELISAs. Irrespective of the HIV status, sFlt-1 and sEng was elevated with the concomitant reduction in PlGF in pre-eclamptic compared to normotensive pregnancies. The levels of VEGF were however undetectable across all study groups. It is plausible that this lack of effect of HIV status on the factors under study may be attributed to the treatment regimen as HAART is known to restore the immune response of HIV positive preeclamptic women. However, a concise anti-retroviral treatment history in our study was unavailable. Additionally, this study is novel in that it ultrastructurally immunolocalises sFlt-1, sEng, PlGF and VEGF within the placenta. This immunoelectron localisation data corresponds to our immunohistochemical data. Our study further demonstrates strong immunoreactivity of both placental sFlt-1 and sEng in pre-eclampsia with concurrent elevations in the maternal circulation. A qualitative increase in the occurrence of syncytial knots in the pre-eclamptics compared to the normotensive pregnancies was noted. These observations support the detachment of antixxx angiogenic rich microparticles from syncytial knots in the pre-eclamptics compared to the normotensive pregnancies was noted. These observations support the detachment of antiangiogenic rich microparticles from syncytial knots and their subsequent deportation and elevation in the maternal circulation. Moreover, their consequent antagonistic effects on VEGF, PlGF and TGF-β, disrupts the vascular endothelial maintenance. The strong immunoreactivity of sFlt-1, sEng, PlGF and VEGF was observed in villous endothelial cells. Moreover, a strong sFlt-1 and sEng but a weak PlGF and VEGF immunoreactivity was noted in syncytio- and cytotrophoblasts. This immunoexpression within trophoblasts is suggestive of their autocrine mode of action on normal trophoblast functions including invasion, differentiation and production. It is plausible that the angiogenic imbalance observed in our study, will impact on placental function, by modifying trophoblast activity thereby contributing to abnormal placentation. Conclusion. Our study supports the hypothesis that pre-eclampsia is characterized by an imbalance between pro- and anti-angiogenic factors. Whether the pregnancy is complicated by immune insufficiencies or not, does not affect the role of the anti-angiogenic factors in pre-eclampsia development. Nevertheless, the neutralising effect of HIV infection on the immune system may be insufficient in the development of pre-eclampsia. To our knowledge, the quantification of serum pro-/anti-angiogenic factors in HIV-associated pre-eclampsia is novel. In conclusion, our data reinforces the hypothesis that increased concentrations of sFlt-1 and sEng are involved in the pathogenesis of pre-eclampsia and indicates their possible use as discriminatory factors between diseases. / Thesis (Ph.D.)-University of KwaZulu-Natal, Durban, 2013.

In der vorliegenden Arbeit wurde die Wirkung von Valproinsäure (VPA) auf Keimzelltumoren des Hodens in vivo und in vitro untersucht. Keimzelltumoren des Hodens (TGCT) sind die häufigsten soliden malignen Tumoren des Mannes zwischen dem 15. und 34. Lebensjahr. Nach therapeutischen und histogenetischen Kriterien werden die TGCT in Seminome und Nicht-Seminome unterteilt. VPA gilt seit 2001 als weiterer Wirkstoff der Gruppe der Histondeacetylaseinhibitoren die über die Wirkung auf die Chromatinstruktur und epi-gentische Modifikation unterschiedliche Effekte in den Zellen erzielen können. VPA führt in unterschiedlichen malignen Tumoren zu einer Proliferationshemmung, Apoptoseinduktion und kann den Differenzierungsgrad in Tumorzellen beeinflussen. Im In-vivo-Mausmodell konnte gezeigt werden, dass VPA eine antitumoröse Potenz auch auf TGCT besitzt. Es zeigte sich in vitro eine Proliferationshemmung und Apoptoseinduktion sowie eine Differen-zierungsinduktion unter VPA-Behandlung. Konkordant zu anderen Tumor konnte eine verstärkte Histonacetylierung unter VPA gezeigt werden. In einer Mikroarray-expressionsanalyse zeigten sich für die Zelllinie TCam-2 als Modell eines seminomatösen Keimzelltumors eine differentielle Expression von 1810 Genen unter VPA-Behandlung und eine differentielle Expression von 1061 Genen für die Zelllinie NTERA-2 als Modell eines embryonalen Karzinoms (Nicht-Seminom). Hierunter fanden sich eine Reihe von differentiell exprimierten Kandidatengenen, deren Regulation Einfluss auf die o.g. Proliferations-hemmung und Apoptoseinduktion haben können. In beiden Zelllinien wurde das Stammzell-genmuster durch Behandlung mit VPA verändert und vermehrt Differenzierungsmarker exprimiert. Insbesondere supprimierte VPA die Expression von NANOG, OCT3/4 in beiden Zelllinien und in der Nicht-seminomatösen Zelllinie NTERA-2 zusätzlich SOX2 als Schlüsselgene zur Erhaltung der Pluripotenz. Beides konnte mittels Mikroarray und q-RT-PCR gezeigt werden.

In der vorliegenden Arbeit wurde die Wirkung von Valproinsäure (VPA) auf Keimzelltumoren des Hodens in vivo und in vitro untersucht. Keimzelltumoren des Hodens (TGCT) sind die häufigsten soliden malignen Tumoren des Mannes zwischen dem 15. und 34. Lebensjahr. Nach therapeutischen und histogenetischen Kriterien werden die TGCT in Seminome und Nicht-Seminome unterteilt. VPA gilt seit 2001 als weiterer Wirkstoff der Gruppe der Histondeacetylaseinhibitoren die über die Wirkung auf die Chromatinstruktur und epi-gentische Modifikation unterschiedliche Effekte in den Zellen erzielen können. VPA führt in unterschiedlichen malignen Tumoren zu einer Proliferationshemmung, Apoptoseinduktion und kann den Differenzierungsgrad in Tumorzellen beeinflussen. Im In-vivo-Mausmodell konnte gezeigt werden, dass VPA eine antitumoröse Potenz auch auf TGCT besitzt. Es zeigte sich in vitro eine Proliferationshemmung und Apoptoseinduktion sowie eine Differen-zierungsinduktion unter VPA-Behandlung. Konkordant zu anderen Tumor konnte eine verstärkte Histonacetylierung unter VPA gezeigt werden. In einer Mikroarray-expressionsanalyse zeigten sich für die Zelllinie TCam-2 als Modell eines seminomatösen Keimzelltumors eine differentielle Expression von 1810 Genen unter VPA-Behandlung und eine differentielle Expression von 1061 Genen für die Zelllinie NTERA-2 als Modell eines embryonalen Karzinoms (Nicht-Seminom). Hierunter fanden sich eine Reihe von differentiell exprimierten Kandidatengenen, deren Regulation Einfluss auf die o.g. Proliferations-hemmung und Apoptoseinduktion haben können. In beiden Zelllinien wurde das Stammzell-genmuster durch Behandlung mit VPA verändert und vermehrt Differenzierungsmarker exprimiert. Insbesondere supprimierte VPA die Expression von NANOG, OCT3/4 in beiden Zelllinien und in der Nicht-seminomatösen Zelllinie NTERA-2 zusätzlich SOX2 als Schlüsselgene zur Erhaltung der Pluripotenz. Beides konnte mittels Mikroarray und q-RT-PCR gezeigt werden.