From protein structure to function with bioinformatics

It has long been known that amino acids are the building blocks for proteins and govern their folding into specific three-dimensional structures. However, the details of this process are still unknown and represent one of the main problems in structural bioinformatics, which is a highly active research area with the focus on the prediction of three-dimensional structure and its relationship to protein function. The protein structure prediction procedure encompasses several different steps from searches and analyses of sequences and structures, through sequence alignment to the creation of the structural model. Careful evaluation and analysis ultimately results in a hypothetical structure, which can be used to study biological phenomena in, for example, research at the molecular level, biotechnology and especially in drug discovery and development. In this thesis, the structures of five proteins were modeled with templatebased methods, which use proteins with known structures (templates) to model related or structurally similar proteins. The resulting models were an important asset for the interpretation and explanation of biological phenomena, such as amino acids and interaction networks that are essential for the function and/or ligand specificity of the studied proteins. The five proteins represent different case studies with their own challenges like varying template availability, which resulted in a different structure prediction process. This thesis presents the techniques and considerations, which should be taken into account in the modeling procedure to overcome limitations and produce a hypothetical and reliable three-dimensional structure. As each project shows, the reliability is highly dependent on the extensive incorporation of experimental data or known literature and, although experimental verification of in silico results is always desirable to increase the reliability, the presented projects show that also the experimental studies can greatly benefit from structural models. With the help of in silico studies, the experiments can be targeted and precisely designed, thereby saving both money and time. As the programs used in structural bioinformatics are constantly improved and the range of templates increases through structural genomics efforts, the mutual benefits between in silico and experimental studies become even more prominent. Hence, reliable models for protein three-dimensional structures achieved through careful planning and thoughtful executions are, and will continue to be, valuable and indispensable sources for structural information to be combined with functional data.

Online Ultrasound Measurements of Membrane Compaction

There are several filtration applications in the pulp and paper industry where the capacity and cost-effectiveness of processes are of importance. Ultrafiltration is used to clean process water. Ultrafiltration is a membrane process that separates a certain component or compound from a liquid stream. The pressure difference across the membrane sieves macromolecules smaller than 0.001-0.02 μm through the membrane. When optimizing the filtration process capacity, online information about the conditions of the membrane is needed. Fouling and compaction of the membrane both affect the capacity of the filtration process. In fouling a “cake” layer starts to build on the surface of the membrane. This layer blocks the molecules from sieving through the membrane thereby decreasing the yield of the process. In compaction of the membrane the structure is flattened out because of the high pressure applied. The higher pressure increases the capacity but may damage the structure of the membrane permanently. Information about the compaction is needed to effectively operate the filters. The objective of this study was to develop an accurate system for online monitoring of the condition of the membrane using ultrasound reflectometry. Measurements of ultrafiltration membrane compaction were made successfully utilizing ultrasound. The results were confirmed by permeate flux decline, measurements of compaction with a micrometer, mechanical compaction using a hydraulic piston and a scanning electron microscope (SEM). The scientific contribution of this thesis is to introduce a secondary ultrasound transducer to determine the speed of sound in the fluid used. The speed of sound is highly dependent on the temperature and pressure used in the filters. When the exact speed of sound is obtained by the reference transducer, the effect of temperature and pressure is eliminated. This speed is then used to calculate the distances with a higher accuracy. As the accuracy or the resolution of the ultrasound measurement is increased, the method can be applied to a higher amount of applications especially for processes where fouling layers are thinner because of smaller macromolecules. With the help of the transducer, membrane compaction of 13 μm was measured in the pressure of 5 bars. The results were verified with the permeate flux decline, which indicated that compaction had taken place. The measurements of compaction with a micrometer showed compaction of 23–26 μm. The results are in the same range and confirm the compaction. Mechanical compaction measurements were made using a hydraulic piston, and the result was the same 13 μm as obtained by applying the ultrasound time domain reflectometry (UTDR). A scanning electron microscope (SEM) was used to study the structure of the samples before and after the compaction.

Virulence properties of Aggregatibacter actinomycetemcomitans biofilm and characterisation of its putative cytokine exploitation

Biofilms are surface-attached multispecies microbial communities that are embedded by their self-produced extracellular polymeric substances. This lifestyle enhances the survival of the bacteria and plays a major role in many chronic bacterial infections. For instance, periodontitis is initiated by multispecies biofilms. The phases of active periodontal tissue destruction and notably increased levels of proinflammatory mediators, such as the key inflammatory mediator interleukin (IL)-1beta, are typical of the disease. The opportunistic periodontal pathogen Aggregatibacter actinomycetemcomitans is usually abundant at sites of aggressive periodontitis. Despite potent host immune system responses to subgingival invaders, A. actinomycetemcomitans is able to resist clearance attempts. Moreover, some strains of A. actinomycetemcomitans can generate genetic diversity through natural transformation, which may improve the species’ adjustment tothe subgingival environment in the long term. Some biofilm forming species are known to bind and sense human cytokines. As a response to cytokines, bacteria may increase biofilm formation and alter their expression of virulence genes. Specific outer membrane receptors for interferon-γ or IL-1β have been characterised in two Gram-negative pathogens. Because little is known about periodontal pathogens’ ability to sense cytokines, we used A. actinomycetemcomitans as a model organism to investigate how the species responds to IL-1beta. The main aims of this thesis were to explore cytokine binding on single-species A. actinomycetemcomitans biofilms and to determine the effects of cytokines on the biofilm formation and metabolic activity of the species. Additionally, the cytokine’s putative internalisation and interaction with A. actinomycetemcomitans proteins were studied. The possible impact of biofilm IL-1beta sequestering on the proliferation and apoptosis of gingival keratinocyte cells was evaluated in an organotypic mucosa co-culture model. Finally, the role of the extramembranous domain of the outer membrane protein HofQ (emHofQ) in DNA binding linked to DNA uptake in A. actinomycetemcomitans was examined. Our main finding revealed that viable A. actinomycetemcomitans biofilms can bind and take up the IL-1β produced by gingival cells. At the sites of pathogen-host interaction, the proliferation and apoptosis of gingival keratinocytes decreased slightly. Notably, the exposure of biofilms to IL-1beta caused their metabolic activity to drop, which may be linked to the observed interaction of IL-1beta with the conserved intracellular proteins DNA binding protein HU and the trimeric form of ATP synthase subunit beta. A Pasteurellaceaespecific lipoprotein, which had no previously determined function, was characterized as an IL-1beta interacting membrane protein that was expressed in the biofilm cultures of all tested A. actinomycetemcomitans strains. The use of a subcellular localisation tool combined with experimental analyses suggested that the identified lipoprotein, bacterial interleukin receptor I (BilRI), may be associated with the outer membrane with a portion of the protein oriented towards the external milieu. The results of the emHofQ study indicated that emHofQ has both the structural and functional capability to bind DNA. This result implies that emHofQ plays a role in DNA assimilation. The results from the current study also demonstrate that the Gram-negative oral species appears to sense the central proinflammatory mediator IL-1beta.

Targeting oncogenic signaling proteins : new insights using a FRET-based chemical biology approach

Cancer affects more than 20 million people each year and this rate is increasing globally. The Ras/MAPK-pathway is one of the best-studied cancer signaling pathways. Ras proteins are mutated in almost 20% of all human cancers and despite numerous efforts, no effective therapy that specifically targets Ras is available to date. It is now well established that Ras proteins laterally segregate on the plasma membrane into transient nanoscale signaling complexes called nanoclusters. These Ras nanoclusters are essential for the high-fidelity signal transmission. Disruption of nanoclustering leads to reduction in Ras activity and signaling, therefore targeting nanoclusters opens up important new therapeutic possibilities in cancer. This work describes three different studies exploring the idea of membrane protein nanoclusters as novel anti-cancer drug targets. It is focused on the design and implementation of a simple, cell-based Förster Resonance Energy Transfer (FRET)-biosensor screening platform to identify compounds that affect Ras membrane organization and nanoclustering. Chemical libraries from different sources were tested and a number of potential hit molecules were validated on full-length oncogenic proteins using a combination of imaging, biochemical and transformation assays. In the first study, a small chemical library was screened using H-ras derived FRET-biosensors. Surprisingly from this screen, commonly used protein synthesis inhibitors (PSIs) were found to specifically increase H-ras nanoclustering and downstream signalling in a H-ras dependent manner. Using a representative PSI, increase in H-ras activity was shown to induce cancer stem cell (CSC)-enriched mammosphere formation and tumor growth of breast cancer cells. Moreover, PSIs do not increase K-ras nanoclustering, making this screening approach suitable for identifying Ras isoform-specific inhibitors. In the second study, a nanoncluster-directed screen using both H- and K-ras derived FRET biosensors identified CSC inhibitor salinomycin to specifically inhibit K-ras nanocluster organization and downstream signaling. A K-ras nanoclusteringassociated gene signature was established that predicts the drug sensitivity of cancer cells to CSC inhibitors. Interestingly, almost 8% of patient tumor samples in the The Cancer Genome Atlas (TCGA) database had the above gene signature and were associated with a significantly higher mortality. From this mechanistic insight, an additional microbial metabolite screen on H- and K-ras biosensors identified ophiobolin A and conglobatin A to specifically affect K-ras nanoclustering and to act as potential breast CSC inhibitors. In the third study, the Ras FRET-biosensor principle was used to investigate membrane anchorage and nanoclustering of myristoylated proteins such as heterotrimeric G-proteins, Yes- and Src-kinases. Furthermore, Yes-biosensor was validated to be a suitable platform for performing chemical and genetic screens to identify myristoylation inhibitors. The results of this thesis demonstrate the potential of the Ras-derived FRETbiosensor platform to differentiate and identify Ras-isoform specfic inhibitors. The results also highlight that most of the inhibitors identified predominantly perturb Ras subcellular distribution and membrane organization through some novel and yet unknown mechanisms. The results give new insights into the role of Ras nanoclusters as promising new molecular targets in cancer and in stem cells.