Projects available for 2010
- Computational drug discovery of novel antibacterial agents effective against MRSA
- Statistical and experimental analysis of gene expression to uncover the mechanisms underlying unwanted side effects of anti-HIV therapy
- Nutrient - Gene Interactions: Novel insights into nutrient sensitive genetic targets
- Diabetic nephropathy: decomposing and engineering BMP-receptor and BMP7-Noggin/Gremlin interactions using structure-based energy calculations
Staphylococcus aureus (MRSA) is implicated in numerous life-threatening diseases. These include sepsis, septic arthritis, infective endocarditis and biofilm formation on indwelling devices such as coronary stents. Some of the major virulence factors responsible are the clumping factors (ClfA, ClfB) and the fibronectin binding proteins (FnbpA, FnbpB), both members of the microbial surface components recognizing adhesive matrix molecules (MSCRAMM) adhesion family. Both the human platelet integrin αIIbβ3 (vital to thrombus formation) and the MSCRAMMs bind fibrinogen (fg). However, the ClfA-fg and αIIbβ3-fg recognition motifs are sufficiently different to allow the development of novel therapeutics (Ganesh VK., 2008, PLoS Pathog 4(11)).
ClfA binding to the C-terminal region of the fibrinogen γ-chain uses a "Dock, Lock and Latch" mechanism that is clearly illustrated by the crystallographic model of this binding interaction. The blue C-terminal beta-sheet connects into the beta-sheet structure of the C-domain of ClfA, after docking of the fibrinogen ligand (green carbons), thus locking the ligand in place.
The 'locked' binding site forms a defined, solvent excluded site that is suitable for drug discovery. Both the ClfA structure, and homology models of ClfB, FnBPA and FnBPB based on the ClfA structure, will be used as templates for virtual screening studies.
Virtual High Throughput Screening will thus involve the creation of novel databases of combinatorial peptides, both linear and cyclically constrained, containing both natural and non-natural amino acids. Non-natural amino acids will be sourced from databases of commercially available compounds, using SMARTS to define the leaving groups that will be removed (both during SPPS and by the final trifluoroacetic acid cleavage step). The final virtual products of SPPS will then be linked combinatorially using SMILES (Simplified Molecular Input Line Entry System), energy minimised, and searched using docking and pharmacophore screening. Particular care will be taken to ensure that hit compounds do not bind to the human platelet integrin αIIbβ3. The active site will also be searched using databases of clinically tested, safe compounds, and commercially available drug-like molecules.
Ligands predicted in silico to have selective and high-affinity binding to MSCRAMMs would then be tested in Dr. O'Gara's laboratory for binding affinity, inhibition of bacterial-fg binding and human cell cytotoxicity. Iterative cycles of drug design and testing should provide a new therapeutic agent for septic arthritis, infective endocarditis, and improve the longevity of indwelling devices. This is likely to be an increasing important issue, given the rise of methicillin resistant Staphylococcus aureus (MRSA) in Irish hospitals.
Statistical and experimental analysis of gene expression to uncover the mechanisms underlying unwanted side effects of anti-HIV therapySupervisors: Dr Patrick Mallon, Prof. Brendan Murphy
HIV is now a treatable condition but requires life-long therapy with a combination of drugs. As a result, focus has shifted from AIDS illnesses towards long-term treatment toxicities, which now account for significant morbidity and mortality. The HIV Molecular Research Group at UCD focuses on translational research into long-term toxicities of antiretroviral therapy. Our two main research areas are HIV-associated lipodystrophy (arising from antiretroviral-induced adipose tissue toxicity) and HIV-associated cardiovascular disease. We are part of European and International collaborations that examine potential pathogenic mechanisms underlying these toxicities as part of clinical trials in HIV-infected patients.
The aim of this PhD project is to build on previous molecular work in this area to examine genome-wide responses in adipose tissue gene expression in HIV-infected patients treated with antiretroviral therapy as part of randomised controlled clinical trials. The candidate will be expected to perform a series of group gene array experiments in samples derived from these clinical trials, identify patterns of change that can be attributed to exposure to particular drugs or drug classes and relate these molecular changes to changes in clinical parameters collected as part of the clinical trial. In addition, we would like to validate candidate genes identified within these analyses in primary human adipocyte cell models. The project will involve a mixture of laboratory-based molecular research (including extraction and purification of RNA, real-time PCR, microarray and primary cell culture) and bioinformatics relating to interpretation of array results and correlation with clinical parameters. In addition to practical experience and development of statistical analysis skills, the candidate will also gain an in-depth understanding of issues relating to HIV and lipid metabolism.
Genes interact with the environment to modulate the phenotype. Food and nutrition are examples of environmental stressors which can alter the expression of disease-susceptibility genes. This project will build on existing expertise relating to nutrient sensitive genetic targets which are related to greater risk of type 2 diabetes, wherein dietary saturated fatty acids seem to interact with genes associated with greater risk of insulin resistance to augment risk of type 2 diabetes. However we need a greater understanding of the functional nature of these associations. Therefore the first aim of this PhD project is to identify putative functional gene nutrient interactions. Briefly, this will expand the set of 840 SNPs investigated to a larger dataset of candidate SNPs using imputation (using data derived from the International HapMap Project and genome-wide association studies (GWAS)). In this study, we will take our LIPGENE dataset (of 1784 cases and matched controls; patients and up to 30 key phenotypes relevant to insulin resistance), which consists of genotypes at a modest number of genetic markers, compare this to a reference sample derived from the International HapMap Project and map this to existing GWA studies with a view to identifying putative functional nutrient-sensitive functional genotypes.
In phase two of the PhD Programme the focus will be on network analysis. The aim will be to integrate genomics with transcriptomic data sets with a view to creating networks that reflect genetic diversity and differential nutrient sensitivity within the context of metabolic health. Genetic and PBMC transcriptomic profiles will be derived from an on-going study JINGO, which will have genetic and two transcriptomic profiles in response to fat and carbohydrate challenges in 200 subjects. This dataset will be built upon using pre-existing data sets which we will add to within the context of determining transcriptional networks / signatures that reflect genetic background and differential metabolic response to fat and carbohydrate challenges. References:
- Phillips CM, Goumidi L, Bertrais S, Ferguson JF, Field MR, Kelly ED, Peloso GM, Cupples LAC, Shen J, Ordovas JM, McManus R, Hercberg S, Portugal H, Lairon D, Planells R, Roche HM (2009) Complement component 3 polymorphisms interact with polyunsaturated fatty acids to modulate risk of the metabolic syndrome. Am J Clin Nutr 90(6):1665-73. Oct 14. [Epub ahead of print]
- Ferguson JF, Phillips CM, Tierney AC, Pérez-Martínez P, Defoort C, Helal O, Lairon D, Planells R, Shaw DI, Lovegrove JA, Gjelstad IMF, Drevon CA, Blaak EE, Saris WHM, Leszczyńska-Gołąbek I, Kiec-Wilk B, Risérus U, Karlström B, López Miranda J and Roche HM (2010) Gene-nutrient interactions in the metabolic syndrome: SNPs in ADIPOQ and ADIPOR1 interact with plasma saturated fatty acids to modulate insulin resistance. Am J Clin Nutr 2009 Dec 23. [Epub ahead of print]
- JINGO - Joint Irish Nutrigenomics Study http://www.ucd.ie/jingo
Diabetic nephropathy: decomposing and engineering BMP-receptor and BMP7-Noggin/Gremlin interactions using structure-based energy calculationsSupervisors: Prof. Finian Martin, Dr. Jens Erik Nielsen
Bone morphogenetic proteins (BMPs) are a group of extracellular-matrix-associated molecules that belong to the transforming growth factor β (TGFβ) superfamily . The classically defined role for a BMP is in limb development and the induction of cartilage and endochronal bone growth. The mature protein then undergoes dimerisation and glycosylation to form the bioactive moiety. Active BMP dimers then engage specific type I/II tyrosine kinase receptors at the plasma membrane, propagating their signal via Smad1/5/8 phosphorylation.
BMP action is tightly regulated by a family of soluble, extracellular secreted BMP antagonists. Characterised by their ability to block the action of BMPs through direct binding, BMP antagonists, such as the CAN (Cerberus and DAN) family of proteins which includes Gremlin (Grem1) and Cerberus (Cer1), Twisted gastrulation Twsg1), Chordin (Chrd) and Crossveinless 2 (Cv2), and Noggin (Nog) have been identified. The timing and concentration of both BMP and BMP antagonist expression is crucial to normal developmental processes in the levels of either BMP or BMP antagonist can lead to deformities in bone, limb and kidney formation. Unexpectedly in adult tissues, the reactivation of previously quiescent BMP antagonist expression may contribute to the progression of human diseases such as fibrosis and cancer.
BMP antagonists have been implicated in fibrotic disorders of multiple organs including lung, liver and kidney. Fibrotic responses typically reflect activation of fibroblasts and transformation of epithelial cells to a mesenchymal myofibroblast phenotype at a site of injury [epithelial mesenchymal transformation, EMT]. In the kidney unresolved tubular damage and interstitial fibrosis is the final common pathway resulting in endstage renal disease and eventual organ failure irrespective of the initiating causes e.g. ischemic injury, diabetes. Grem1 expression is upregulated in vitro models of diabetic nephropathy. Importantly, Grem1 upregulation is observed in renal biopsy material from patients with diabetic nephropathy where the level of grem 1 expression was shown to correlate with the degree of functional renal impairment. The low or undetectable levels of Grem1 in adult kidney suggest that upregulation in renal disease may occur as part of recapitulation of developmental gene expression contributing to a maladaptive repair process. In this regard it is noteworthy that haploinsufficiency of Grem1 has recently been shown to attenuate several important markers of kidney damage in type 1 diabetic mice. Furthermore, variants in the Grem1 gene have recently been shown to be associated with diabetic nephropathy in a case control study.
The present project aims to analyze the molecular basis of the interaction between BMP7 and its receptors: the BMP type I and II receptors, and its antagonists: Noggin and Gremlin. The project will consist of an extensive experimental phase that includes the construction of multiple mutants of BMP, Noggin and Gremlin that will be assayed for their ability to bind to each other or the receptors. The project will furthermore contain a theoretical phase where the candidate will develop an algorithm for predicting the effect of point mutations on protein-protein interactions and use this algorithm to analyze the experimental results. In the final phase of the project the candidate will design a number of BMP7-based proteins to bind to Noggin/Gremlin but not to the BMP7 receptor thus sequestering BMP7 antagonists in the cell and functioning as a molecular protector against diabetic nephropathy.
The project will consist of the following phases:
- Lab period I (16 months)
- Becoming familiar with general cell biology and molecular biology techniques (3 months)
- Setup of BMP cell-based assays (2 months)
- Feasibility study of biophysical interaction assay (1-2 months)
- Limited mutation scan of Noggin, Gremlin and BMP7 to establish sites of interaction in BMP7-Noggin, BMP7-Gremlin and BMP-receptor complexes. Measurement of response in cell-based assays and in biophysical assays (the latter only if feasible). (9 months)
- Theoretical investigations (11 months)
- Introduction to Nielsen group algorithms (1 month) and structural bioinformatics software in general.
- Construction of automatic computational workflow for predicting the effect of point mutations on protein-protein interactions. This pipeline will be based on PEAT_SA (http://enzyme.ucd.ie/main/index.php/PEAT_SA) - a program developed by the Nielsen group. (4 months)
- Visit to structural biology lab - details to be determined. (2 months)
- Collection of published mutational data on the effect of point mutations on protein-protein interactions. Calibration of pipeline using this data (2 months)
- Benchmarking the performance of the pipeline on experimental data obtained during lab period. Decomposition of BMP-Noggin and other BMP ligand interactions (2 months)
- Design of 3-5 BMP-based proteins that will be protective against diabetic nephropathy by sequestering the Noggin and Gremlin antagonists in cell-based assays.
- Lab period II (8 months)
- Construction of designed proteins (3 months)
- Measurement of effect in cell-based assays (2 months)
- Biophysical characterization (3 months)
- Theoretical investigations II (9 months)
- Analysis of BMP designed proteins and experimental data (2 months)
- Refinement of developed pipeline in light of experimental findings (2 months)
- Industry placement - details to be determined (1 month)
- Application of pipeline to other protein-protein interaction pairs of relevance for human disease (exact targets to be decided) (4 months).
- Thesis write-up (4 months)
Location: Galway, Ireland
Almost a decade after the publication of the sequence of the human genome, the short arms of the acrocentric chromosomes (HSA13, 14, 15, 21 and 22) remain unsequenced. Each of these chromosomal arms contains an array of ribosomal gene repeats termed the nucleolar organiser region (NOR). The precise organization of rDNA repeats within NORs remains poorly understood but recent research using molecular combing techniques suggests that as much 30% of rDNA gene repeats may be inverted or rearranged. Such alternative rDNA repeat configurations are hypothesized to have important implications in cancer and aging. At NUI Galway we have developed protocols for preparing DNA from nucleoli isolated from human cells. This DNA is highly enriched for acrocentric chromosome short arm DNA, including NORs, as determined by back painting onto metaphase chromosomes. Molecular combing of nucleolar DNA confirms that a large proportion of rDNA repeats within NORs are rearranged (see Fig 1). We have used nucleolar DNA as a template in next generation (454) sequencing and have obtained 7.3x105 sequence reads of average length 287bp (i.e. 208 Mb of sequence) with approximately 9.5% of sequence reads mapping to ribosomal RNA genes. The objectives of this project will be to attempt the assembly or partial assembly of these high-throughput sequence reads and to develop a custom annotation pipeline. Assembled or individual sequence reads will then be used to investigate the structure of rDNA repeats as well as the sequences flanking the rDNA gene clusters. Coverage of the acrocentric chromosomes is sufficient (approximately four-fold) to provide a high probability of confirming or refuting the existence of rearranged rDNA gene clusters. Having characterised the structure of rearranged rDNA repeats, a series of PCR primers will be developed to permit the analysis of rearranged rDNA in human disease and aging.
To the extent possible, high-throughput sequence reads will also be used to investigate divergence of distinct rDNA gene clusters and spacer regions as well as sequences proximal and distal to the rDNA gene clusters.
Figure 1. Nucleoli were isolated from HeLa cells and resuspended in low-melting point agarose blocks. Blocks were digested with proteinase K. Blocks were melted and the agarose removed by digestion with agarase. Nucleolar DNA was then combed onto treated coverslips using a molecular combing apparatus supplied by Genomic Vision (Paris). After fixation probes were hybridised using 18S (green) and 28S (red) rDNA probes. Upper and lower panels show canonical (normal) and non-canonical (rearranged) repeats respectively.Further reading
- McStay B, Grummt I (2008) The epigenetic regulation of ribosomal gene expression. Annu Rev Cell Dev Biol. 24:131-57.