Introduction: According to the World Health Organization, dental cavities are the number one chronic disease in children. Saliva coats the teeth all day and can serve many functions to maintain and protect teeth. Saliva has many proteins that can be both detrimental and essential to the preservation of tooth enamel. The purpose of this study is to determine if a correlation exists between the total protein concentration in saliva and the prevalence of cavities in the mouth. We hypothesized that there would be a positive correlation with total salivary protein concentration and the prevalence of cavities in the participant. Methods: Saliva samples were taken from the patient during their comprehensive exam at the University of Central Florida (UCF) Dental Center and were analyzed using the DC assay to determine the protein concentration in the saliva. These results were compared to the number of cavities found in the mouth of each patient to determine if a correlation exists between protein concentration and cavity number. Results: The correlation between the variables was fairly weak, indicating that the data from this study does not support a correlation between salivary protein concentration and cavity number. Discussion: Future research should look at specific salivary proteins, control the time of day of collection, and take into account more variables in order to get a more precise study.

One of the most prevalent diseases acquired in older populations and currently the most common form of dementia, exists in the form of Alzheimer's [1]. Progressing over time, Alzheimer's begins intramolecularly through the buildup of amyloid precursor protein derived Aβ peptides which aggregate into neurotoxic fibrils. The fibrils result in damage to memory and cognitive systems, leading to depression of routine function and eventual death. There is currently no cure nor treatment by which this plaque buildup can be prevented or eliminated, and as such, significant work is being made towards this topic. It has been recently discovered that the human matriptase protein is capable of cleaving both the amyloid precursor protein from which Aβ peptides are made as well as the Aβ1-42 peptide itself, facilitating interest into its potential to reduce fibril formation [2][3]. In this study we set out to determine if the human matriptase serine protease domain can cleave Aβ fibrils or prevent the Aβ fibril formation in vitro. A recombinant matriptase serine protease domain (r-MatPD) was subcloned into a pET-28a-c(+) expressing vector, expressed, and purified via Ni-NTA affinity resin. The purified his-tagged r-MatPD was further auto-activated and incubated with the purified recombinant Aβ1-42 peptides. We observed that r-MatPD can cut polymerized Aβ1-42 into smaller fragments and prevent Aβ1-42 fibril formation. Effectively, this study suggested that the matriptase protease domain can be further investigated for its role in Aβ fibril clearance in vivo with a possibility of developing matriptase therapeutic potentials in treating Alzheimer's patients.

No description available.

No description available.

Uromodulin (UMOD) is a protein made exclusively in the thick ascending limb. Clinical studies have demonstrated that rare missense mutations in UMOD result in autosomal dominant tubulointerstitial kidney diseases manifest by tubulointerstitial fibrosis (TIF), tubular cysts and a rapid progression to renal failure. In addition, several genome wide association studies reported that common single nucleotide polymorphisms in the UMOD gene are associated with an increased risk of chronic kidney disease (CKD) and hypertension. Interestingly, Dahl salt-sensitive (SS) rats exhibit many of the same pathologies observed in these clinical populations with alterations in UMOD. The goal of this study was to assess the qualitative and quantitative aspects of UMOD via western blotting, and the extent of SS hypertension and proteinuria in Dahl SS vs. a consomic rat strain in which chromosome 1 of the salt-resistant Brown-Norway (BN) rat, harboring the UMOD gene, has been introgressed into the Dahl SS background (SS.BN1). We hypothesized that differences in UMOD would be apparent in SS vs. SS.BN1 rats maintained on a low salt-diet and that the extent of SS hypertension and proteinuria would be attenuated in SS.BN1 vs SS rats. Western blot of urinary UMOD was performed in 16 week old SS (n=5), SS.BN1 (n=7) and BN (n=6) rats maintained on a low salt (LS) diet. BP (radiotelemetry) and proteinuria were assessed during LS feeding and during three weeks of high salt (HS) feeding in a different group of 8-10 week old SS (n=9) and SS.BN1 (n=8) rats. For western blotting, urine was normalized based on the protein concentration, and the density of the 85 kDa UMOD band in SS and SS.BN1 samples were normalized to the average density observed in BN rats. The UMOD band was 4.5 fold higher (p In summary, these data demonstrate striking qualitative and quantitative differences in UMOD between SS and SS.BN1 rats. The pattern of UMOD expression in SS rats is consistent with that observed in some patient populations of UMOD associated kidney disease. Finally, the evidence that SS.BN1 rats, harboring the UMOD gene from BN rats, exhibit significant protection against SS hypertension and proteinuria is consistent with the notion that an alteration in UMOD function may, in part, be responsible for such pathologies in SS rats.

This thesis explores the possibility of using patient reported symptom descriptions to predicthow patients will report their symptoms on 3-D body models. The feasibility of using embeddingmodels to map self-reported symptom descriptions to symptom categories is investigated usingdimensionality reduction techniques such as t-SNE and UMAP. The findings highlight how thereexists promising clustering between various correlated groups of symptoms, indicating that theembedding space of patient reported symptom descriptions may be granular enough todistinguish between symptom groups. The thesis also discusses pathways and potentialimplementations utilizing embeddings and other features from patient reported symptoms toimplement predictive medical-AI models. A thorough discussion regarding bias, marginalization,safety, privacy concerns and the need for transparency in a clinical setting is discussed.Potential benefits of a clinical implementation are discussed, such as enhanced patient-doctorcommunication for marginalized groups, such as non-native speakers, and more efficiency ininitial assessment, specifically for emergency care or local clinical centers. The findings suggestfurther research in applying embeddings to patient reported symptoms could show promise indeveloping the medical AI models of tomorrow’s healthcare system.

Variation at regulatory elements, identified through hypersensitivity to digestion by Deoxyribonuclease I (DNase I), is believed to contribute to variation in complex traits, but the extent and consequences of this variation are poorly characterized. To investigate the relationship between sequence variation, and the functional consequences of variation in chromatin accessibility, genome-wide DNase I hypersensitive sites (DHS) of terminally differentiated erythroblasts were studied in eight inbred strains of mice studied (A/J, AKR/J, BALBc/J, C3H/HeJ, C57BL/6J, CBA/J, DBA/2J, and LP/J). These strains were selected because of the availability of their genome sequence and quantitative trait loci (QTL) data. After confirming that next generation sequencing could identify DNase I hypersensitive sites with high sensitivity and specificity, and that differential peaks could be found, an automated peak calling pipeline was developed and optimized. 36,693 DHS peaks were identified covering 9.1 Mb (0.29%) of mouse genome. There was no indication of within strain variation. Between strains reproducible variation was observed at approximately 5% of DNase hypersensitive sites (1,397 DHSs). Variable DHSs were more likely to be enhancers than promoters and less likely to occur at conserved regions of the genome. Only 36% of such variable DHSs contain a sequence variant predictive of site variation and 12% contain at least one variant that disrupts transcription factor binding sites. The majority (86%) of variable DHSs differ in size/shape and the remaining 14% demonstrate discrete variation in single peak or cluster of peaks. Sequence variants within variable DHS are more likely to be associated with complex traits than those in non-variant DHS, and variants associated with complex traits preferentially occur in enhancer-like elements. Changes at a small proportion (7%) of discretely variable DHS are associated with changes in nearby transcriptional activity. Our results show that whilst DNA sequence variation is not the major determinant of variation in open chromatin, where such variants exist they are likely to be causal for complex traits.

Mouse embryonic stem cells (ESCs) are capable of differentiating into any somatic cell type and are known to display fragile immune privilege in vivo and in vitro. The extent to which the depletion of essential amino acids (EAAs) by ESCs contributes to this phenomenon was investigated. ESCs were found to express various enzymes capable of catabolising EAAs within the culture medium. In particular, depletion of threonine, valine and lysine was found to have significant impact on T cell proliferation and differentiation, biasing their polarisation towards a FoxP3⁺ T regulatory (T_reg) phenotype. Supplementing ESC conditioned medium with these three EAAs alone rescued normal T cell proliferation, whereas artificially limiting their availability was sufficient to induce T_reg cell differentiation to a level equivalent to general EAA depletion. The pattern of EAA catabolism by mouse ESC was shared by induced pluripotent stem cells, while mouse melanoma cell lines and human ESCs displayed distinct patterns of EAA depletion. The cytosolic branched chain aminotransferase enzyme, Bcat1, catalyses the first step of branched chain amino acid catabolism (isoleucine, leucine and valine), and is highly expressed by both mouse and human ESCs. The contribution of this enzyme to the establishment of acquired immune privilege by ESC-derived tissues was, therefore, investigated. ESC lines were derived from mice lacking Bcat1 activity and were characterised. Bcat1^−/− ESC lines displayed no difference to their wildtype counterparts (Bcat1^LoxP) in terms of in vitro proliferation and their capacity to form teratomas in vivo. Furthermore, the loss of Bcat1 function had little impact on the inhibition of T cell proliferation in culture, ability to induce T_reg cell commitment or their ability to prevent rejection by T cell receptor transgenic recipients, suggesting the minimal contribution of Bcat1 to the depletion of EAAs by ESCs. In conclusion, EAA depletion by mouse ESC may provide a mechanistic explanation for the previously described immune-suppressive capacity of ESC.

Developmental language impairments are neurodevelopmental disorders in which the acquisition of language, a task which children typically perform with ease, is hindered or fraught with difficulty. This work focuses on specific language impairment (SLI), a common and highly heritable language impairment in which language development is abnormal while other developmental domains are normal. Additionally, a case-study of a child with a broader linguistic and behavioural phenotype is also presented. The work described in this thesis includes both genetic and functional investigations which were aimed at identifying candidate genes for language impairment and provide insight into the genetic mechanisms that underlie language development. I performed a genome-wide association study of SLI which included child genotype effects, maternal genotype effects, parent-of-origin effects, and maternal-foetal interaction effects. This study found significant paternal parent-of-origin effects with the gene NOP9 on chromosome 14, and suggestive maternal parent-of-origin effects with a region on chromosome 5 which had previously been implicated in autism and ADHD. Case-control and quantitative association analyses of HLA genes and SLI identified several risk alleles and protective alleles. A case-control association analysis for related individuals which used an isolated population affected by SLI identified a non-synonymous coding variant in the gene NFXL1 which was significantly more frequent in affected individuals than in unaffected individuals. High-throughput sequencing of the coding regions of NFXL1 and LD blocks surrounding associated variants in ATP2C2, CMIP and CNTNAP2 (as reported in previous studies) identified novel or rare non-synonymous coding variants in NFXL1 and ATP2C2 in SLI families as well as intronic variants in all four genes that were significantly more frequent in SLI probands than in population controls. I describe a functional study of NFXL1 examining its expression in various brain regions, the presence of different splice variants across several tissues, its effect on genes it potentially interacts with, and the subcellular localisation of the protein. Finally, I present the case-study of a child with language impairment who had chromosomal rearrangements which spanned the location of FOXP2. I examine the potential influence the chromosomal rearrangements had on FOXP2 expression and describe a lincRNA gene which was disrupted by the chromosomal inversion. In conclusion, this work identified new candidate genes for language impairment, provided further support for the involvement of previously-identified candidate genes in SLI and contributed to the understanding of the molecular function of a newly-identified candidate gene for SLI.

Brain metastasis is a frequent occurrence in cancer patients and carries a high mortality rate. The incidence of brain metastasis is on the rise, highlighting the need for improved therapeutic intervention. Immune cells have been shown to promote disseminated tumour cells to colonise the lung and liver. Therefore, we aim to determine whether immune cells also facilitate brain metastasis by describing the host immune response to tumour cells attached to the brain vasculature. We developed a model of brain metastasis by using ultrasound guidance to perform intracardiac injection of tumour cells. Using this method, we identified highly and weakly brain metastatic cell lines. To understand how cancer cells develop into brain metastases, we analysed brains harvested 4 h- 14 d after tumour injections. At 4 h after intracardiac injection, only cell lines that developed into brain metastases were found adhered to the brain vasculature in high numbers. A small number of arrested tumour cells clustered with CD45⁺ immune cells. These tumour-CD45 clusters persisted over time whilst the frequency of solitary tumour cells declined. Tumour-associated CD45⁺ immune cells were identified to be Ly6G⁺Gr-1⁺CD11c⁻ myeloid cells. Considerably more tumour-CD45⁺ immune cell clusters were found within the brain vasculature when tumour cells were injected into mice bearing a primary tumour. Increased tumour-CD45⁺ immune cells clusters correlated with an increased number of brain metastases in the same group of mice. We also found a positive association between increased tumour-immune clusters and levels of tumour and host derived G-CSF. To establish a causal relationship between tumour cell-CD45 clusters and metastases, we developed an experimental setup for transcranial imaging. Our results suggest that tumour recruited immune cells may promote tumour cell colonisation of the brain and provides a framework for further investigation.