Browsing by Title

Sort by: Order: Results:

Now showing items 1-3 of 3
  • Vellonen, Kati-Sisko (2010)
    Drug discovery and development from its very onset up to market approval is a long process lasting 10-15 years. New research tools are needed to accelerate and rationalize this process. Ocular drug research still relies heavily on animal testing with rabbits and other rodents. Computational methods and cell culture models are promising tools for early pharmacokinetic studies and may partly replace the animals in pharmacokinetic and toxicological studies. Computational methods are initially based on experimental data, but thereafter their application is straightforward and they can be used to reduce, partly replace and refine further experimental studies. Similarly, cell culture models may enable absorption and toxicity testing of drug candidates with continuously growing cells of human origin, and thereby reduce the need for animal experiments. The cornea is the main route of ocular drug absorption after topical administration, and the corneal epithelium is the most important barrier to drug permeation. Membrane transporter proteins play an important role in the general pharmacokinetics and toxicology. However, their role in ocular pharmacokinetics is still poorly understood. Based on literature analysis many ocular drugs seem to be substrates of transporters, but the expression of these proteins in the eye is largely unknown. The goal of this work was to develop and evaluate cellular and computational tools for ocular pharmacokinetics and toxicology, and to characterise the active drug transporters in the corneal epithelium. The expression of monocarboxylate transporters and ATP-binding cassette (ABC) class efflux proteins was studied in the corneal epithelium and human corneal epithelial (HCE) cell model. Human corneal epithelium expressed monocarboxylate transporters 1 and 4 (MCT1 and MCT4), efflux transporters multidrug resistance-associated protein 1 and 5 (MRP1 and MRP5), and breast cancer resistance protein (BCRP). Cultured human corneal epithelial cells over-expressed several ABC class efflux proteins and MCT1 and MCT4. The functionality of efflux and monocarboxylate transport was demonstrated in HCE cells and in the rabbit cornea ex vivo. The MTT test is a widely used cytotoxicity test in cell research. It was demonstrated that substrates and inhibitors of ABC class efflux proteins may interfere with the MTT test, presumably by inhibiting dye efflux from the cells. This may lead to an underestimation of toxicity in the MTT test. Quantitative structure property relationship (QSPR) models are commonly used in early drug discovery to predict ADME properties of novel compounds. Multivariate analysis was used to develop QSPR models for in silico prediction of the corneal permeability. Two factors, the distribution coefficient (logD7.4 /logD8.0) and hydrogen binding potential, were shown to be the parameters that determine the transcorneal permeability of a compound. These models were able to predict intracameral steady state drug concentrations in rabbit eyes. In conclusion, the new in silico QSPR model can make reliable predictions for passive drug permeability in the cornea, while the HCE model seems to over-express some membrane transporters as compared to the normal human corneal epithelium. Even if these investigated methods have some restrictions they are still very useful tools for drug discovery purposes.
  • Shawesh, Amna Mohammed (Helsingin yliopisto, 2015)
    Indomethacin (IND) is a potent non-steroidal anti-inflammatory drug used in the treatment of rheumatoid arthritis, osteoarthritis, acute gout and other disorders. IND is available worldwide mostly in the form of capsules and suppositories, however, these formulations usually create side effects. Consequently, an alternate route of administration to avoid or minimize side effects may be found in the form of semisolid dermatological formulations, now available in few countries. The specific goals of this study were: (I) to determine the solubility of IND using different co-solvents: hexylene glycol (HG), propylene glycol (PG), polyethylene glycol 300 (PEG), butylene glycols (1,2 BG; 1,3 BG and 1,4 BG) and ethanol (ETOH). 1% (w/w) Tween® 80 or polyvinyl pyrrolidone (PVP-25) were used as enhancers; (II) to develop suitable topical gel preparations using 20% (w/w) Pluronic® (PF-127) or 1% (w/w) Carbopol ETD® 2001 (C2001) as gelling agents and HG or PEG 300 as solvents (1% (w/w) Tween® 80 and PVP-25 were added as excipients); (III) To evaluate the effect of composition of prepared gel formulations on the following parameters: appearance, crystallization, pH and rheological behaviour and (IV) to investigate the influence of storage time and storage conditions on the characteristics of the gels. These results indicate that all the solvents tested increased the solubility of IND to varying degrees. Tween® 80 and PVP-25 only slightly enhanced the solubility of IND. 1% (w/w) IND was able to form a structural gel with both PF-127 and C2001. Storing the INDPF-127 gels at 6°C resulted in the precipitation of IND. All gels stored at room temperature exhibited good stability. The gels stored at 45°C developed a dark yellow colour. Gels with C2001 and PF-127/PEG had slightly decreased viscosities with increasing storage time, while the gels with PF-127/HG showed increase in viscosities with time. In conclusion, the water solubility of IND was increased by the addition of co-solvents. 1% (w/w) IND gel can be suitable for using as a gel formulation and it is stable at room temperature. The search for suitable gels for IND topical formulation needs to be continued with more stability studies. Moreover, in-vitro and in-vivo experiments will be necessary for providing data on bioavailability.
  • Zhou, Fang (Helsingin yliopisto, 2012)
    We propose network abstraction as a research area. It is motivated by the growth of networks in many areas of life. Consider, for instance, networks of thousands of genes, millions of people, or billions of web pages. They are too large to be directly analyzed by users. The aim of network abstraction is to summarize a large network as a smaller one. An abstracted network can then help users to see the overall topology of a large network, or to understand the connections of distant nodes. The general network abstraction task is: given a large network, transform it into a smaller one, which contains in some well-specified sense the most relevant information. In this thesis, we analyze this research area and propose methods to solve some instances of the problem. The methods also provide different trade-offs between the graph quality and simplicity, as well as between result quality and efficiency. More specifically, we propose two approaches to abstracting a network. The first one is to simplify a weighted network by removing edges under the constraint that distances between all pairs of nodes are preserved. We first empirically show that a number of edges can be removed from real biological networks without losing any graph connectivity. We next relax the constraint of fully preserving original graph connectivity, extend lossless network simplification to lossy network simplification, and demonstrate that many more edges can be removed with little loss of quality. The second approach we give for network abstraction is to compress a weighted network by grouping nodes and edges. We propose novel methods and experimentally show that real graphs can be compressed efficiently with relatively little error. We next consider graphs with weights also on the nodes, and utilize them as node importances to extend the definition of weighted graph compression. We present new compression operations and demonstrate that the compressed graph can preserve more information related to more important nodes. Furthermore, we propose the idea of using node weights and compression to summarize the metabolisms in a set of organisms, and apply the methodology to better understand the metabolic biodiversity between Archaea and Eubacteria, the two most fundamental branches of life.