Alzheimer's disease (AD) is a late onset neurodegenerative disorder, affecting tens of thousands of people in New Zealand, as well as their families and caregivers. AD is characterised by extra cellular β-amyloid (βA) deposition, Tau-containing neurofibrillary tangles (NFTs) and progressive cortical atrophy. Abnormal protein accumulation is also a common feature of other late onset neurodegenerative diseases, including the heritable polyglutamine (polyQ) disorders such as Huntington disease (HD) and the spinocerebellar ataxias (SCAs). One of this family of disorders, SCA17, is caused by an expansion of a polymorphic polyQ repeat in TATA binding protein (TBP), an essential transcription factor. Surprisingly, the wild type TBP repeat length ranges from 25-42, and in Caucasian populations the most common allele is 38, a size large enough to cause HD if within the huntingtin protein. Wild type length TBP accumulates in the disease structures of HD and in at least some of the SCAs. The work described in this thesis investigates the hypothesis that the TATA binding protein (TBP) contributes to AD. This thesis describes the discovery that TBP accumulates in AD brain, localising to NFTs. Also, a proportion of the detectable TBP present is insoluble; a signature of the polyQ diseases. TBP positive structures are shown to be present differentially between patients and its amount and distribution is not directly proportional to that of Tau or β-amyloid positive structures. To investigate if the polyQ repeat length in TBP is associated with AD, alleles were genotyped from two large case/control cohorts. In one cohort, but not the other, older patients carried longer TBP polyQ repeats than controls, and the effect was enhanced when subjects homozygous for the ApoE4 allele were removed from the dataset. Evidence is presented that transcriptional activation by HMGBI, a chromatin associated protein that binds to the polyQ tract within TBP, is altered when co-expressed with TBP carrying 90 glutamines. Taken together this data provides evidence for the hypothesis that the accumulation or misfolding of this polyQ containing protein may be a contributing factor in Alzheimer's disease.

Clinical trial simulation (CTS) and modelling are techniques which play an important role in clinical pharmacology and drug development. The goal of this thesis is to apply techniques in CTS and modelling for studying disease progression and the pharmacokinetics and pharmacodynamics of levodopa in Parkinson's disease. Three components of levodopa response can be identified in patients with Parkinson's disease; a transient short-duration response, a sustained long-duration response to exogenous levodopa, and diurnally varying endogenous levodopa synthesis. A pharmacokinetic-pharmacodynamic model was developed to describe the individual components of levodopa response. Twenty-one previously untreated (de novo) patients with Parkinson's disease and 12 patients who had received levodopa orally for 9.7 ± 4.0 years were investigated. Twenty de novo patients were examined at 6, 12, 24 and 48 months. No systematic changes in levodopa pharmacokinetics were found in the first 4 years of long-term levodopa treatment. A lowered baseline motor response and an increase in efficacy with time result in an exaggerated short-duration response which might be key contributors to the development of motor fluctuations with long-term levodopa treatment. The capability of CTS to generate typical responses was evaluated. If progression rate of bradykinesia is influenced by age then it should be possible to detect this effect using reasonable clinical trial designs. The ELLDOPA trial aims to determine if levodopa slows or accelerates the rate of disease progression of Parkinson's disease. The power of the ELLDOPA trial design was evaluated using a clinical pharmacology model. The simulation results suggested that the short washout period for levodopa symptomatic effect could lead to a false conclusion about the nature of the levodopa treatment effect. This thesis has shown that pharmacodynamic or disease progression factors rather than pharmacokinetic changes are responsible for the development of motor fluctuations in Parkinson's disease. It has quantified the unexpected development of increased efficacy of levodopa over time. Clinical trials designed to evaluate disease progression without recognising washout of the long-duration response of levodopa can lead to misinterpretation of trial outcomes. However, model based approaches can disentangle these confounding factors and dissect the underlying effects of drug action on disease progression.

A number of studies have demonstrated directed migration of neural progenitor cells to sites of brain injury and disease. However, a detailed examination of when a cell is “born” in relation to injury induction and the migratory response of that cell has not previously been determined. This study therefore examined the temporal correlation between progenitor cell proliferation (“birth”) and neuroblast migratory response into the damaged striatum following quinolinic acid (QA) lesioning of the adult rat striatum. Retroviral labeling of subventricular zone (SVZ)–derived progenitor cells demonstrated that cell loss in the QA-lesioned striatum increased progenitor cell migration through the rostral migratory stream for up to 30 days. In addition, a population of dividing cells originating from the SVZ generated doublecortin positive neuroblasts that migrated into the damaged striatum in response to cell loss invoked by the QA lesion. The majority of doublecortin positive neuroblasts present in the damaged striatum were generated from progenitor cells dividing within two days prior to, or on the day of QA lesioning. In contrast, cells dividing two or more days following QA lesioning, migrated into the striatum and exhibited a glial phenotype. These results demonstrate that directed migration of SVZ-derived cells and neuroblast differentiation in response to QA lesioning of the striatum is acute and transient. We subsequently demonstrated a role for the chemokines MCP-1, MIP-1α and GRO-α in directing adult SVZ-derived progenitor cell migration following striatal cell death. MCP-1, MIP-1α and GRO-α were significantly upregulated in the striatum 2-3 days following QA-induced lesioning, correlating with maximum SVZ-derived progenitor cell recruitment into the lesioned striatum. We established that SVZ-derived progenitor cells express receptors for each chemokine, and demonstrated MCP-1, MIP-1α and GRO-α to be potent chemoattractants for SVZ-derived progenitor cells in vitro. Immunofluorescence revealed MCP-1, MIP-1α and GRO-α are predominantly expressed in the striatum by NG2-positive cells that appear to infiltrate from the bloodstream 6 hours following QA lesioning. These results indicate that upregulation of MCP-1, MIP-1α, GRO-α following striatal cell death leads to chemoattraction of SVZ-derived progenitor cells into the damaged striatum and raises a potential role for blood-derived cells in directing the recruitment of SVZ progenitor cells following brain injury. / Whole document restricted, but available by request, use the feedback form to request access.

The main aim of this study was to develop nanoparticulate drug delivery systems for chlorambucil (CHL) and asulacrine (ASL). CHL is a DNA alkylating agent. In an attempt to alter the pharmacokinetics (short half-life, rapid clearance and high volume of distribution) and improve the anticancer activity, CHL was incorporated into a lipid nanosphere (LN) formulation. The LN were composed of soybean oil as the internal oil phase, egg lecithin as the primary emulsifier, cholesterol as the phospholipid layer stabilizer, and water as the external phase. PEG-modified (long circulating) and DOTAP-modified (cationic) LN were prepared to evade the reticulo-endothelial system and enhance cellular delivery, respectively. LN prepared by ultra-sonicater and microfluidizer had an average particle size below 200 nm, with a CHL encapsulation efficiency of over 97%. Zeta potential of the LN ranged from -30.6 to +41.5 mV. The LN exhibited good physical stability over six months at 4°C and 25°C. In vitro evaluation on SKOV-3 cells showed that the cytotoxicity and the pro-apoptotic activity of CHL were significantly enhanced when given in the LN formulation compared to the CHL solution (P<0.05). An altered pharmacokinetics with increased plasma AUC and elimination half-life, and reduced clearance were observed after intravenous administration of CHL incorporated LN compared to a CHL solution (P<0.01). A marked reduction in the tissue distribution of CHL was also noted when it was given in LN. In addition, CHL incorporated in LN showed greater anticancer activity compared with the CHL solution in colon-38 tumour-bearing mice (P<0.01). These results suggest that LN could be an effective parenteral carrier for CHL delivery. ASL is an inhibitor of topoisomerase II. In an attempt to improve its efficacy by altering the pharmacokinetic profile, ASL was formulated as a nanocrystal suspension (NS). The NS was consisting of ASL nanoparticles produced by high pressure homogenization and stabilized by stabilizers. The lyophilized NS exhibited good stability over three months at 4°C and 25°C. The dissolution and solubility of ASL were enhanced in NS form compared to un-milled ASL (P<0.05). Initial peak plasma concentration and AUC were remarkably reduced in the plasma after intravenous administration of NS compared to the ASL solution (P< 0.01). This suggests reduced systemic exposure to the drug.

In intrauterine tissues, pro-inflammatory cytokines and prostaglandins (PGs) have been identified as key mediators in the maintenance of pregnancy and parturition. The rise in PGD2 detected in the amniotic fluid before labour prompted the research presented in this thesis which describes the effects of 15-deoxy Δ12,14 -prostaglandin J2 (15d-PGJ2), a non-enzymatic metabolite of PGD2, on amnion-derived WISH and JEG3 choriocarcinoma cells as models of the amnion and chorion trophoblasts, respectively.15d-PGJ2 induced apoptosis in both cell lines in a concentration-dependent fashion (2.5-10 µM). Apoptosis was characterised by condensation of chromatin (visualised after Hoechst 33342 staining), appearance of nucleosomal DNA fragmentation upon electrophoresis and flow cytometry analysis, and activation of caspase-3. Apoptotic cell death was inhibited in the presence of serum (0.5% w/v) and albumin, not serumderived growth factors (insulin growth factor (IGF)-1, IGF-2 or epidermal growth factor (EGF), was determined as the key survival factor. Since 15d-PGJ2 is reported to activate peroxisome proliferator activated receptor (PPAR)-γ, the activities of PPARs were assessed using JEG3 cells transfected with a PPAR-response element reporter construct (pTK-PPREx3-luc). The PPAR-γ-specific ligand, rosiglitazone, induced PPRE mediated activity in a concentration-dependent manner, while the PPAR-γ-specific irreversible inhibitor, GW9662, fully inhibited this induction. However, GW9662 only partially inhibited 15d-PGJ2-induced luciferase activity, suggesting that 15d-PGJ2 may also activate either of the other isoforms. The expressions of PPAR-α and -δ were identified in amnion, choriodecidua and placental membranes and PPAR-δ was significantly increased all tissues with labour. PPAR-α expression was reduced in chorio-decidua, but was significantly higher in placenta with labour. The changes observed with labour suggest that regulation of PPAR expression and function may have a role in the mechanisms that maintain pregnancy or initiate labour. The anti-inflammatory effects of 15d-PGJ2 were also investigated by measuring interleukin (IL)-1β-stimulated prostaglandin and cytokine productions by WISH cells after treatment with 15d-PGJ2 for 3 hours. 15d-PGJ2 exerted differential effects that were dependent upon its concentration. At low nanomolar physiologic concentrations (1-10 nM), 15d-PGJ2 inhibited IL-1β-stimulated PGE2, but not cytokine (IL-6/IL-8) production or cyclooxygenase (COX)-2 expression. This effect was attenuated by GW9662, by transfection with dominant negative PPAR constructs, and was reproduced by rosiglitazone. At micromolar (1-10 µM) concentrations, 15d-PGJ2 inhibited IL-1β-stimulated COX-2, PGE2 and cytokine productions and these effects were not blocked by GW9662 or mimicked by rosiglitazone. GW501516 (PPAR-δ agonist) also inhibited IL-1β-stimulated PGE2 production, but only at high concentrations (1 µM). IL-1β-induced NF-kB DNA binding activity was significantly inhibited by 15d-PGJ2 (10 µM) and GW501516 (1 µM), but increased by rosiglitazone (10 µM). In conclusion, this is the first report of an effect of 15d-PGJ2 at low nanomolar physiologic concentrations and 15d-PGJ2 mediates its actions through PPAR-γ (<0.1 µM) and PPAR-γ-independent(1-10 µM) pathways, the latter through inhibition of NF-kB and/or activation of PPAR-δ. Further studies on the effect of physiologic concentrations of 15d-PGJ2 on primary gestational tissues will provide understanding on the role(s) 15d-PGJ2 plays in fetal membrane remodelling and its involvement in the inflammatory processes associated with labour and parturition. / Whole document restricted, but available by request, use the feedback form to request access.

Accumulating evidence over the past 20 years has indicated that the brain has an endogenous immune system, mediated at the local level predominantly by microglial cells. The primary role of an inflammatory response is to protect the host against a foreign stimulus, remove damaged cells and to initiate repair and regeneration of damaged tissue. However, it has become increasingly evident that this inflammatory response, in addition to its role in host defence and repair, can exert detrimental effects in the central nervous system (CNS). Neuroinflammation, as this response is now commonly known, has been implicated in the pathogenesis of many neurodegenerative diseases. Microglia, the resident macrophage of the brain, play a central role in sustaining this inflammatory response through the release of proinflammatory and potentially cytotoxic mediators. Hypothermia is neuroprotective, and these properties are thought to be mediated, in part, by the suppression of microglial activation. The BV-2 microglial cell line was used to investigate the mechanisms involved in the activation and inhibition of microglia. Lipopolysaccharide (LPS)-induced activation of BV-2 cells led to the up-regulation of inducible nitric oxide synthase (iNOS) and cyclooxygenase (COX)-2, the release of inflammatory mediators such as nitric oxide (NO), prostaglandin (PG)-E2, interleukin (IL)-6 and tumour necrosis factor (TNF)-α, and a change to a phagocytic phenotype. This response was mediated by the initiation of signal transduction pathways that culminated in transcription and translation of the inflammatory genes. Hypothermia (33°C) caused complete suppression of iNOS and NO whilst displaying little or no effect on IL-6 and TNF-α respectively. In contrast, LPS-induced COX-2 expression and PGE2 release was super-induced in response to hypothermia. A co-culture model of neuroinflammation was developed to investigate microglial-neuronal interactions. LPS/interferon (IFN)-γ activated BV-2 cells required direct contact with SK-N-SH neuroblastoma cells to elicit a cytotoxic phenotype. Hypothermia and the selective iNOS inhibitor S-methylisothiourea (S-MT) protected against this BV-2-induced SK-N-SH cell death, strongly implicating NO as the major candidate molecule in microglial-induced neuronal cell death. Thus, microglial activation sustains the chronic inflammatory response in the CNS and in doing so contributes to further neuronal death. Inhibition of the detrimental facets of microglial activation may provide some protection against neurodegeneration.

The aim of this thesis was to study xenobiotic metabolising enzymes (XME) in the human placenta, in particular the uridine diphosphate glucuronosyltransferases' (UGTs) role in the metabolism, transfer and disposition of the drug AZT, and to develop the human placental perfusion model into the first trimester. UGT, β-glucuronidase, Cytochrome p4501A (CYPIA) and CYP reductase were present and active in the human placenta. CYP2E1 protein was expressed but not active. CYPIA, CYP2EI and the UGT2B subfamily were expressed across gestation, but the UGTIA subfamily was only expressed in first trimester placentas. The localisation of XME (the syncytiotrophoblast layer bordering the placental villi) did not change with gestation but enzyme activity and affinity did. Greater activify but lower affinity of UGT and CYPIA were observed in first trimester placentas than at term. In contrast, β-glucuronidase had a high affinity, low activity profile in early gestation but the opposite at term. UGT and CYPIA activities in the first trimester placenta were induced by maternal cigarette smoking and may be synergistically induced by combined alcohol consumption and smoking. A significant correlation with higher UGT activity and earlier gestational age was observed. CYPIA exhibited a significant, negative correlation with maternal age across gestation. Maternal variables had no effect on β-glucuronidase and their effects on the activity of CYP2EI and CYP reductase were not established. AZT caused apoptosis in the placenta and also increased reactive oxygen species and altered XME. The absence of serum enhanced these effects. Alterations in XME expression and activities included a decrease in UGT activity and increases in CYPIA, β-glucuronidase, CYP reductase and glutathione-Ѕ-transferase activity in response to AZT exposure. AZT transport in a perfusion model was bi-directional and reached equilibrium approximately 3 h after addition of AZT into the maternal reservoir. In contrast, transfer of AZT glucuronide (AZT-G) showed significantly greater transport rates out of the fetal compartment resulting in AZT-G concentrations approximately 2-fold higher in the maternal circuit. Transfer of the co-factor UDPGA was significantly greater in the fetal direction and almost complete after 4h of perfusion. The low levels of glucuronidation of AZT catalysed by the human placenta (approximately 2% of a dose) are not due to insufficient transport of the co-factor UDPGA and are unlikely to be significant in terms of maternal whole-body clearance. Therapeutic failure of AZT in protecting the fetus is unlikely to be due to metabolism and clearance performed by the placenta but may be due to placental cytotoxicity. The development of the perfusion model into the first trimester placenta is technically achievable, but was impossible with the tissue available for these studies due to the method used to obtain placentas. / Whole document restricted, but available by request, use the feedback form to request access.

Neurotrophic factors are a family of polypeptides that promote the differentiation, growth and survival of numerous central nervous system neurons during development and adulthood. It has been proposed that alterations in neurotrophic factor protein or receptor expression may be involved in the pathogenesis of human neurodegenerative disorders. Recent research supports the therapeutic use of neurotrophic factors in neurodegenerative disorders. However, while information has been obtained regarding the structure and function of neurotrophic factors and their receptors (trk receptors) in the developing and mature rodent central nervous system, little research has been performed examining the expression and functional role of these factors in the normal and diseased human brain. This thesis investigated the role neurotrophic factors and trk receptors play in the pathogenesis of human neurodegenerative disorders. Using immunohistochemical and in situ hybridisation techniques, the regional distribution and cellular localisation of neurotrophic factors and trk receptors was examined throughout both the adult rat and normal human brain. The expression of individual neurotrophic factors and trk receptors was also examined in human post mortem normal, Alzheimer's and Huntington's disease brain tissue, as well as in an animal model of apoptotic nerve cell death. Individual neurotrophic factors exhibited a specific and heterogeneous regional pattern of distribution throughout the adult human brain. Neurotrophic factor expression was detected in several neuronal populations which exhibit selective vulnerability in various neurodegenerative disorders. Alterations in the expression of neurotrophic factors within specific regions of the human brain may result in neuronal atrophy, possibly via apoptotic mechanisms. A significant reduction in the level of brain-derived neurotrophic factor (BDNF) was observed within the hippocampus and temporal cortex of the Alzheimer's disease brain. A loss of neuroprotection afforded by BDNF may contribute to the progressive atrophy of neurons in Alzheimer’s disease. The high-affinity trk receptors, trkA and trkB (full-length and truncated) were also altered within the Alzheimer's disease brain. TrkA receptor-immunoreactivity was observed in astrocytes in the CA1 region of the Alzheimer's disease hippocampus, some of which were associated with β-amyloid plaques. Truncated trkB receptors were found in high levels in senile plaques while the full-length trkB receptor was expressed in glial-like cells in the Alzheimer's disease hippocampus. The appearance of trkA and trkB receptors in astrocytes and plaques in the Alzheimer's disease brain might be related to β-amyloid deposition and could be implicated in the development of Alzheimer's disease. Alterations in insulin-like growth factor-I (IGF-I) protein expression were also observed within the Alzheimer's disease brain. IGF-I-immunoreactivity was expressed in a subpopulation of reactive astrocytes in the Alzheimer's disease temporal cortex. These observations may indicate that IGF-I is involved in the neuropathology of Alzheimer's disease. The induction of IGF-I in response to neuronal injury may be an attempt to inhibit mechanisms that result in delayed neuronal death. In addition, neurotrophic factor expression was examined in the Huntington's disease brain. Glial cell line-derived neurotrophic factor (GDNF) and transforming growth factor-α (TGF-α) were significantly reduced within both the Huntington's disease globus pallidus and substantia nigra. Reduced GDNF and TGF- α levels within the Huntington's disease brain may produce a loss of local or target-derived neurotrophic support within the basal ganglia and contribute to the preferential degeneration of medium-sized spiny projection neurons within the Huntington's disease striatum. Moderate hypoxic-ischemic (Hl) injury was used as an animal model of apoptotic nerve cell death. In agreement with the observations made in the Alzheimer's disease brain, moderate Hl injury resulted in the loss of BDNF within the rat hippocampus. In contrast, an increase in trkB (truncated) receptor expression was detected within glial cells in the rat brain. Alterations in BDNF and trkB receptor levels may lead to a loss of neuroprotection and the initiation of downstream mechanisms resulting in the induction of apoptotic processes. A cascade of events similar to those observed within the rat Hl model may occur within human neurodegenerative disorders. This study demonstrated that, while the neuropathogenesis of both Alzheimer's and Huntington's disease is complex, alterations in individual neurotrophic factor or trk receptor expression within selectively vulnerable cortical or subcortical regions may play a role in their pathophysiology. Furthermore, these results support the proposal that neurotrophic factors may be considered for the treatment of neurodegenerative disorders by protecting against neuronal cell loss and by increasing the function of surviving neuronal populations. / Whole document restricted, but available by request, use the feedback form to request access.

The aims of this thesis were to examine:(1) the effects of acute and chronic dietary exposure of the pro-carcinogen, IQ on the expression and activity of xenobiotic metabolising enzymes; (2) the ability of foodstuffs chronically (individually and in combination with the dietary pro-carcinogen IQ) to alter the xenobiotic metabolising enzymes; (3) whether tissue-specific effects of the foodstuffs on XME could explain pathological results; and (4) whether any isolated fraction of the foodstuffs acted directly as a toxicant or stimulated cell growth in vitro. For the chronic study, male Fischer 344 rats were fed defined AIN-76A diet ± 10% foodstuff (taro, kumara, pineapple, or coconut), with or without IQ (300ppm for 52 weeks). In the acute study, male Fischer 344 rats were fed defined AIN-76A diet and gavaged with IQ (20 mg.kg-1, 3 days). Pathology was observed and changes in enzyme activities and expression examined. Simple fractionation of the foodstuffs was also performed and their effects on cell viability and enzyme expression were investigated in vitro in freshly isolated rat hepatocytes and immortal human HT29 and P388 cells. In the liver, acute exposure to IQ affected enzymes consistent with complex xenobiotic response element / glucocorticoid response element activation, whereas chronic exposure was consistent with activation of the antioxidant response element. No pattern of enzyme change was obvious in other organs. Inclusion of South Pacific foodstuffs in the diet at 10% had a marked effect on enzyme activity in the liver and lung, although there was no consistent pattern for enzyme changes for a particular enzyme or foodstuff. Enzyme activity in the colon was below the level of detection. There was a complex response when IQ was included with the foodstuffs, which did not resemble the response to either IQ or the foodstuff alone. Fractionation of foodstuffs and in vitro cell-based work did not reveal the presence of compounds in the foodstuffs that explained the complex changes in pathology or XME observed in vivo. The interaction between foodstuffs and IQ in the body was complicated. This may be explained by the presence of many compounds within the foodstuffs that have additive, synergistic, or inhibitory actions on many pathways in the body and underlines the complexity of using whole foodstuffs, rather than a single, purified constituent, in chemoprevention studies.

Alzheimer's disease (AD) is a late onset neurodegenerative disorder, affecting tens of thousands of people in New Zealand, as well as their families and caregivers. AD is characterised by extra cellular β-amyloid (βA) deposition, Tau-containing neurofibrillary tangles (NFTs) and progressive cortical atrophy. Abnormal protein accumulation is also a common feature of other late onset neurodegenerative diseases, including the heritable polyglutamine (polyQ) disorders such as Huntington disease (HD) and the spinocerebellar ataxias (SCAs). One of this family of disorders, SCA17, is caused by an expansion of a polymorphic polyQ repeat in TATA binding protein (TBP), an essential transcription factor. Surprisingly, the wild type TBP repeat length ranges from 25-42, and in Caucasian populations the most common allele is 38, a size large enough to cause HD if within the huntingtin protein. Wild type length TBP accumulates in the disease structures of HD and in at least some of the SCAs. The work described in this thesis investigates the hypothesis that the TATA binding protein (TBP) contributes to AD. This thesis describes the discovery that TBP accumulates in AD brain, localising to NFTs. Also, a proportion of the detectable TBP present is insoluble; a signature of the polyQ diseases. TBP positive structures are shown to be present differentially between patients and its amount and distribution is not directly proportional to that of Tau or β-amyloid positive structures. To investigate if the polyQ repeat length in TBP is associated with AD, alleles were genotyped from two large case/control cohorts. In one cohort, but not the other, older patients carried longer TBP polyQ repeats than controls, and the effect was enhanced when subjects homozygous for the ApoE4 allele were removed from the dataset. Evidence is presented that transcriptional activation by HMGBI, a chromatin associated protein that binds to the polyQ tract within TBP, is altered when co-expressed with TBP carrying 90 glutamines. Taken together this data provides evidence for the hypothesis that the accumulation or misfolding of this polyQ containing protein may be a contributing factor in Alzheimer's disease.