Diss. (sammanfattning) Stockholm : Karol. inst., 2006. / Härtill 4 uppsatser.

Attention Deficit Hyperactivity Disorder (ADHD) is one of the most prominent neurobehavioral disorders of childhood that is heavily researched and often revised. Deficits in attention/concentration, impulsivity, inhibition, information processing speed, working memory and executive functioning manifest differently according to subtype diagnosis for both children and adults. As a result, previous attempts to construct a unifying theory of ADHD with neural correlates and quantifiable performance discrepancies have resulted in a proliferation of literature reviews citing both significant and insignificant research findings. The purpose of the current study was to construct a homogenous sample of children diagnosed with ADHD and examine purported subtype differences in working memory and information processing speed using the Wechsler Intelligence Scales for Children - Forth Edition (WISC-IV). Sixty participants were selected from archival data from two clinical sites and separated into three groups according to the current DSM-IV-TR ADHD subtype criteria: ADHD Predominately Inattentive type (ADHD-IN), ADHD Predominately Hyperactive/Impulsive type (ADHD-HY), and ADHD Combined type (ADHD-C). Significant differences within groups were revealed on the Coding subtest and Processing Speed Index (PSI) relative to performance on the Perceptual Reasoning Index (PRI). No significant between groups or interaction effects were revealed. While the goal of the current study was aimed at discovering evidence of greater deficits in processing speed by the ADHD-IN group compared to the other two groups, processing speed was reduced for the entire sample providing more evidence of a possible neurological deficit/basis inherent to ADHD. The results of the current study provided minimal evidence of differences on WISC-IV measures of working memory within the current sample. The current study was successful in correcting previous methodological flaws inherent in the relevant literature by constructing a truly homogenous sample of ADHD and provided strong evidence for the necessity of an accurate diagnosis of ADHD subtypes.

Nonverbal Learning Disability (NVLD) is a syndrome characterized by impaired social perception, visual-spatial skills, fine motor coordination, and mathematics abilities. Researchers have found that children with NVLD often have significant symptoms of inattention, and there is evidence that the majority of children with NVLD also meet clinical criteria for Attention Deficit/Hyperactivity Disorder, Predominantly Inattentive Subtype (ADHD:PI) (Brown, 2000; Gross-Tsur & Shalev, 1995; Voeller, 1996). Although significant overlap is observed between NVLD and behavioral symptoms of ADHD, little research has focused on the specific attention problems of children with NVLD. Given the high incidence of co-morbid attention problems with NVLD (Brown, 2000), many researchers have proposed that overlapping neural regions are responsible for the similarity in attention impairments observed in both NVLD and ADHD:PI (Denckla, 2000; Stefanatos, 2001). Other researchers suggest that there are distinct neurological impairments in children with NVLD and both subtypes of ADHD that result in attention problems. Specifically, Rourke (1995) suggested a developmental sequence that results in generally intact auditory attention with impaired attention for visual stimuli in children with NVLD. This study sought to reconcile the discrepancy between conceptualizations of attention problems in children with NVLD. It was hypothesized that children with NVLD would exhibit distinct profiles of strengths and weaknesses on neuropsychological measures of attention compared to children with ADHD, Predominantly Inattentive Subtype (ADHD:PI) and ADHD, Combined Subtype (ADHD:C). Specifically, it was expected that the three diagnostic groups would differ on the neuropsychological measures depending on the attention modality (auditory vs. visual). Extant neuropsychological data from 88 children between the ages of 9 and 15 years of age with diagnoses of NVLD, ADHD:PI, and ADHD:C were analyzed. Neuropsychological measures of processing speed, working memory, vigilance, and inhibition were examined to compare specific domains of attention functioning in the three groups. Evidence from the current study supported the model in which NVLD and the two ADHD subtypes represent a continuum of dysfunction dependant on overlapping neural regions. Moreover, specific attention strengths and weaknesses in children with NVLD compared to children with ADHD:PI, ADHD:C, and normative data were identified in order to inform clinical diagnosis and intervention. / text

The current project examined the ability of a novel series of guandine and amidine-containing nicotine analogs to interact with several native and recombinantlyexpressed mammalian neuronal nicotinic-acetylcholine receptor (nAChR) subtypes. Rational drug design methods and parallel organic synthesis was used to generate a library of guanidine-containing nicotine (NIC) analogs (AH compounds). A smaller series of amidine-containing nicotine analogs (JC compounds) were also synthesized. In total, >150 compounds were examined. Compounds were first assayed for affinity in a high-throughput [3H]epibatidine radioligand-binding screen. Lead compounds were evaluated in subtype-selective binding experiments to probe for affinity at the α4β2* and α7* neuronal nAChRs. Several compounds were identified which possess affinity and selectivity for the α4β2* subtype [AH-132 (Ki=27nm) and JC-3-9 (Ki=11nM)]. Schild analysis of binding suggests a complex one-site binding interaction at the desensitized high-affinity nAChR. Whole-cell functional fluorescence (FLIPR) assays revealed mixed subtype pharmacology. AH-compounds were identified which act as activators and inhibitors at nAChR subtypes, while lead JC-compounds were found which possess full agonist activity at α4β2* and α3β4* subtypes. Compounds were identified as partial agonists, full agonists and inhibitors of multiple nAChR subtypes. Several SAR-based, ligand-receptor pharmacophore models were developed to guide future ligand design. Second-generation lead compounds were identified.

<p>Southern Africa is facing one of the most serious HIV epidemics. This project contributes to the HIVNET, Network for Prevention Trials cohort for vaccine development. HIV’s biology and rapid mutation rate have made vaccine design difficult. We examined HIV-1 subtype C diversity and how it relates to CTL epitope location along viral gag sequences. We found a negative correlation between codon sites under positive selection and epitope regions / suggesting epitope regions are evolutionarily conserved. It is possible that epitopes exist in non-conserved regions, yet fail to be detected due to the reference strain diverging from the circulating viral population. To test if CTL clustering is an artifact of the reference strain, we calculated differences between the gag codons and the reference strain. We found a weak negative correlation, suggesting epitopes in less conserved regions maybe evading detection. Locating conserved and optimal epitopes that can be recognized by CTLs is essential for the design of vaccine reagents.</p>

The genus Blastocystis has recently attracted the attention of scientists, especially parasitologists. Similarly to the related opalines and proteromonads, Blastocystis is anaerobic and lives endobiotically in the intestine of various animals. This organism is also often found in humans, where it is associated with irritable bowel syndrome, though its pathogenic potential remains uncertain. The genus Blastocystis is remarkable for its rich genetic diversity. The taxonomy of Blastocystis is inconsistent and problematic. The strains isolated from homoiothermic vertebrates are divided into 17 subtypes, while strains from poikilotherms are either classified as separate species or are not considered in taxonomic studies at all. The aim of the study was to further examine the genetic diversity of the genus Blastocystis. We determined SSU rDNA sequences of 38 strains isolated from poikilothermic vertebrates and arthropods. The results of our phylogenetic analysis showed that Blastocystis is considerably diverse in these hosts, and we defined 21 new subtypes. The total number of known subtypes of Blastocystis has thus increased to 38. We also examined light-microscopical morphology of some strains. Most of the newly defined subtypes show identical morphology, ST20 (Blastocystis geocheloni) is an exception,...

PhD (Molecular Medicine and Haematology), Faculty of Health Sciences, University of the Witwatersrand / Background: The availability of highly active antiretroviral (ARV) treatment in the South African government sector has reduced the morbidity and mortality associated with HIV-1 infection. However, ARV drug resistance and toxicity are major obstacles to achieving and maintaining virus suppression, but there is no provision for ARV drug resistance testing in the public sector. To date, most studies of ARV drug resistance in HIV-1 reverse transcriptase (RT) and protease (PR), are based on sequence data from HIV-1 subtype B, whereas subtype C is the predominant circulating subtype in South Africa. Moreover, host genetic polymorphisms associated with ARV drug toxicity have not been investigated in South African populations. This study evaluated viral and host genetic factors associated with ARV treatment outcome in 812 ARV drug-naive South African AIDS participants enrolled on the CIPRA-SA study from Johannesburg and Cape Town. Methodology: An affordable in-house genotyping protocol (subtype C specific) was established and validated to monitor the emergence of ARV drug resistance. This assay was used to genotype all CIPRA-SA participants failing the first- and second-line ARV drug regimens. Allellic discrimination assays to identify the G1344A, A6986G, G516T and C3435T SNPs in CYP3A4, 3A5, 2B6 and MDR-1, respectively, associated with ARV metabolism and absorption were performed. Results: The in-house ARV drug resistance assay successfully genotyped 95% of patient samples, including non-C subtypes from 8 African sites. Treatment failure was experienced in 371 participants, mainly due to toxicity (n=134) or virological failure (n=83). Overall, CIPRA-SA participants with a lower CD4+ T-cell count at study onset were more likely to experience viral failure. Genotyping using the in-house assay revealed that 6 participants had ARV drug resistance mutations at study entry. Treatment failure of 58 participants was a result of ARV drug resistance mutations, whereas 19 had no known ARV drug resistance mutations. The most frequent mutations were M184V (67%) and K103N (50%). K65R was present (3%) and one participant harboured TAMs. Longitudinal genotypic analysis showed that NNRTI mutations accumulated at a rate of one per three months left on failing therapy. No PR mutations were detected amongst participants experiencing second-line failure. The four SNPs analysed occured in similar frequencies between a background and the CIPRA-SA cohort. Furthermore, no statistically significant association could be found between these four SNPs and viral failure and/or toxicity. Conclusion: Overall, HIV-1 subtype C-infected individuals receiving ARV therapy develop many of the known subtype B drug resistance mutations. However, the ARV drug resistance patterns in the closely monitored CIPRA-SA cohort were less complex compared to published data from the region, confirming that more frequent viral load monitoring, genotyping, and a virological failure cut-off value of 1000 RNA copies/ml ensure a better prognosis, and preserve future ARV treatment options.

Human respiratory syncytial virus (HRSV) is a negative sense, single strand RNA virus that causes respiratory tract infection with common cold-like symptoms, which can be severe in children, immunocompromised, and the elderly. Even with 60 years of research, the need for vaccine and effective treatment has not been met. In this work, recombinant viruses have been generated which will be valuable in gaining a better understanding of HRSV subtypes, glycoprotein evolution, and the polymerase localization, which would contribute to HRSV vaccine and therapeutics development. The differences in the fitness of A and B antigenic subtypes of HRSV and how it affects the regional circulation pattern is not well understood. To study and compare the two subtypes, it is important to use clinically relevant recombinant viruses and to use animal models that best represent human infection. Using a wild-type virus strain (A11 and B05) from each HRSV subtype, a wild-type like recombinant (r) virus, rHRSVA11, and recombinant viruses expressing fluorescent proteins, rHRSVA11EGFP(5) and rHRSVB05dTom(5), were generated. Characterization of rB05 viruses demonstrated that the differences in the fluorescent protein expressed did not affect virus growth kinetics. To prepare for an experiment in cotton rats, recombinant HRSVs generated were used to infect cotton rat lung cells in vitro. With confirmation of infection of cotton rat lung cells by rHRSV, cotton rat co-infection experiment was planned for the recombinant A11 and B05 viruses and a microneutralization assay was developed for post-infection processing of the in vivo samples. The BA genotype of HRSV B subtype is a strain of HRSV B subtype containing a 60 nucleotide duplication in the glycoprotein (G) gene. HRSV BA genotype was first isolated in 1998 and has quickly become the predominant genotype circulating globally. Although a role of immune evasion by the strains of BA genotype has been suggested to explain this phenomenon, few studies have supported this hypothesis. To compare the HRSV B subtype virus with and without the duplication, rB05 virus lacking the duplication, rHRSVB05EGFP(5)GΔ60b, and containing an epitope tag within the duplication, rHRSVB05EGFP(5)Gmycb, were generated. A serial passage experiment was set up using rHRSVB05EGFP(5) and rHRSVB05EGFP(5)GΔ60b to understand the mutations that accumulate in the G protein gene of each virus. This will be valuable in setting up a similar experiment in the presence of immune pressure to understand the advantage that is conferred to the virus containing the duplication. Expression of Gmyc was confirmed in rHRSVB05EGFP(5)Gmyc infection, which validated that this virus can be used to study the HRSVB05 G protein and modifications in the duplicated region. The HRSV large (L) protein is essential in HRSV transcription and replication, but is difficult to study due to lack of immunologic reagents and challenges with purification. Recombinant viruses expressing reporter and polymerase fusion proteins have been generated and used for studying various other viral polymerases. Expression plasmids for HRSV L protein containing a reporter protein in its variable region 2 have been published. However, the modification resulted in downregulation in the function of the protein and rHRSV expressing modified L protein have not yet been published. In this study, rHRSVB05LVenus was generated to study the effects of modification of HRSV L protein variable region and the localization of HRSV L protein. LVenus protein in rHRSVB05LVenus infected cells was visualized by confocal laser scanning microscopy and the expression levels were examined by immunoblotting. rHRSVB05LVenus was compared to rHRSVB05EGFP(5) with unmodified L protein to show that modification of HRSV L protein had no effect on virus replication. Viruses had equivalent growth kinetics and were equally sensitive to ribavirin, a known HRSV inhibitor. The recombinant viruses generated in this study are valuable tools in answering questions that are difficult to pursue without clinically relevant recombinant viruses. Characterization of the rHRSVs demonstrated that these viruses will have many applications. In this study, viruses were characterized for the basic growth kinetics, expression of proteins of interest, and assay development. With these validated tools, questions such as the cause of the epidemiological pattern observed for HRSV A and B subtypes, the role of host immune response in advantage conferred to HRSV BA genotype, and the effects of inhibitors to formation of HRSV polymerase complex can be addressed. / 2018-10-31T00:00:00Z

Influenza is a contagious respiratory disease that has caused at least four pandemics and countless epidemics since the 20th century, impacted millions of people worldwide and the global economy. To date, the predominant influenza species circulating in humans are influenza A and B. Influenza may cause serious illness in all age groups but individuals such as the newborns and senior population whose immune systems are compromised are at higher risk for severe disease. Interestingly, during the outbreak of pandemic 2009 H1N1 (H1N1pdm), it was found that the elderly had the lowest hospitalization rate and an increased proportion of healthy adults developed severe disease. Furthermore, several clinical studies have demonstrated that most H1N1pdm infected children experienced mild to moderate illness and led to the least mortality. The difference of disease outcome in age groups between different influenza infections may be due to several factors, which include differing pathogenicity between the viruses, differential immune status and composition among the age groups, and pre-existing immunity from previous encounter(s) with a similar virus. Since the human clinical data are often complicated by secondary factors such as co-morbidities, I used the ferret model to address these questions. I first compared the clinical and pathological patterns among the pandemic and seasonal influenza strains and found H1N1pdm caused the most severe illness to healthy ferrets. Importantly, the disease severity did not correlate with viral burden but immunopathology. To study the age effect, I found that H1N1pdm infected young ferrets with mild clinical symptoms developed specialized ectopic lymphoid structures and a distinct cytokine expression profile in the lungs, which were absent in adult ferrets with severe illness. I also examined antigenic change in historical H1N1s and anti-H1 responses to explain the pre-existing immunity of H1N1pdm found in the elderly. However, low similarity was found between historical H1N1s and H1N1pdm. Lastly, I conducted a detailed influenza B comparative study. I observed the pathogenic B strain was capable to cause lower respiratory tract infection and pathology like the influenza A viruses. Overall, this thesis provides novel insights for developing therapeutic and prophylactic strategies against influenza infection.

This dissertation shows that nonnegative matrix factorization (NMF) can be extended to a general and efficient clustering method. Clustering is one of the fundamental tasks in machine learning. It is useful for unsupervised knowledge discovery in a variety of applications such as text mining and genomic analysis. NMF is a dimension reduction method that approximates a nonnegative matrix by the product of two lower rank nonnegative matrices, and has shown great promise as a clustering method when a data set is represented as a nonnegative data matrix. However, challenges in the widespread use of NMF as a clustering method lie in its correctness and efficiency: First, we need to know why and when NMF could detect the true clusters and guarantee to deliver good clustering quality; second, existing algorithms for computing NMF are expensive and often take longer time than other clustering methods. We show that the original NMF can be improved from both aspects in the context of clustering. Our new NMF-based clustering methods can achieve better clustering quality and run orders of magnitude faster than the original NMF and other clustering methods. Like other clustering methods, NMF places an implicit assumption on the cluster structure. Thus, the success of NMF as a clustering method depends on whether the representation of data in a vector space satisfies that assumption. Our approach to extending the original NMF to a general clustering method is to switch from the vector space representation of data points to a graph representation. The new formulation, called Symmetric NMF, takes a pairwise similarity matrix as an input and can be viewed as a graph clustering method. We evaluate this method on document clustering and image segmentation problems and find that it achieves better clustering accuracy. In addition, for the original NMF, it is difficult but important to choose the right number of clusters. We show that the widely-used consensus NMF in genomic analysis for choosing the number of clusters have critical flaws and can produce misleading results. We propose a variation of the prediction strength measure arising from statistical inference to evaluate the stability of clusters and select the right number of clusters. Our measure shows promising performances in artificial simulation experiments. Large-scale applications bring substantial efficiency challenges to existing algorithms for computing NMF. An important example is topic modeling where users want to uncover the major themes in a large text collection. Our strategy of accelerating NMF-based clustering is to design algorithms that better suit the computer architecture as well as exploit the computing power of parallel platforms such as the graphic processing units (GPUs). A key observation is that applying rank-2 NMF that partitions a data set into two clusters in a recursive manner is much faster than applying the original NMF to obtain a flat clustering. We take advantage of a special property of rank-2 NMF and design an algorithm that runs faster than existing algorithms due to continuous memory access. Combined with a criterion to stop the recursion, our hierarchical clustering algorithm runs significantly faster and achieves even better clustering quality than existing methods. Another bottleneck of NMF algorithms, which is also a common bottleneck in many other machine learning applications, is to multiply a large sparse data matrix with a tall-and-skinny dense matrix. We use the GPUs to accelerate this routine for sparse matrices with an irregular sparsity structure. Overall, our algorithm shows significant improvement over popular topic modeling methods such as latent Dirichlet allocation, and runs more than 100 times faster on data sets with millions of documents.