Classifying noisy protein sequence data: a case study of immunoglobulin light chains

SUMMARY: The classification of protein sequences obtained from patients with various immunoglobulin-related conformational diseases may provide insight into structural correlates of pathogenicity. However, clinical data are very sparse and, in the case of antibody-related proteins, the collected sequences have large variability with only a small subset of variations relevant to the protein pathogenicity (function). On this basis, these sequences represent a model system for development of strategies to recognize the small subset of function-determining variations among the much larger number of primary structure diversifications introduced during evolution. Under such conditions, most protein classification algorithms have limited accuracy. To address this problem, we propose a support vector machine (SVM)-based classifier that combines sequence and 3D structural averaging information. Each amino acid in the sequence is represented by a set of six physicochemical properties: hydrophobicity, hydrophilicity, volume, surface area, bulkiness and refractivity. Each position in the sequence is described by the properties of the amino acid at that position and the properties of its neighbors in 3D space or in the sequence. A structure template is selected to determine neighbors in 3D space and a window size is used to determine the neighbors in the sequence. The test data consist of 209 proteins of human antibody immunoglobulin light chains, each represented by aligned sequences of 120 amino acids. The methodology is applied to the classification of protein sequences collected from patients with and without amyloidosis, and indicates that the proposed modified classifiers are more robust to sequence variability than standard SVM classifiers, improving classification error between 5 and 25% and sensitivity between 9 and 17%. The classification results might also suggest possible mechanisms for the propensity of immunoglobulin light chains to amyloid formation.     

Biological actions of extra-renal 25-hydroxyvitamin D-1alpha-hydroxylase and implications for chemoprevention and treatment

The Vitamin D-activating enzyme 25-hydroxyvitamin D-1alpha-hydroxylase (1alpha-hydroxylase) is now known to be expressed in a much wider range of tissues that previously thought, suggesting a role for 1,25-dihydroxyvitamin D(3) (1,25(OH)(2)D(3)), which is more in keeping with a cytokine than a hormone. In this capacity, the function of 1alpha-hydroxylase in tumors is far from clear. Studies from several groups including ours have shown altered expression of 1alpha-hydroxylase in different types of neoplasm including breast, prostate and colon cancers. However, functional analysis of Vitamin D metabolism in cancer is complicated by the heterogenous composition of tumors. Immunohistochemical analysis of breast tumors has shown that 1alpha-hydroxylase is expressed by both epithelial cells and by tumor-infiltrating macrophages, suggesting an immunomodulatory component to 1,25(OH)(2)D(3) production in some types of cancer. The demonstration of 1alpha-hydroxylase activity in tumors and their equivalent normal tissues has implications for both the treatment and prevention of cancers. For example, in tumors chemotherapy options may include the use of non-1alpha-hydroxylated Vitamin D analogs to increase local concentrations of active metabolites without systemic side-effects. The role of 1alpha-hydroxylase in protection against cancer is likely to be more complicated and may involve anti-tumor immune responses.     

Transgenic proteoid roots of white lupin: a vehicle for characterizing and silencing root genes involved in adaptation to P stress

White lupin (Lupinus albus L.) has become an illuminating model for the study of plant adaptation to phosphorus (P) deficiency. It adapts to -P stress with a highly coordinated modification of root development and biochemistry resulting in short, densely clustered secondary roots called proteoid (or cluster) roots. In order to characterize genes involved in proteoid root formation and function in a homologous system, we have developed an Agrobacterium rhizogenes-based transformation system for white lupin roots that allows rapid analysis of reporter genes as well as RNA interference (RNA(i))-based gene silencing. We used this system to characterize a lupin multidrug and toxin efflux (Lupinus albus MULTIDRUG AND TOXIN EFFLUX, LaMATE) gene previously shown to have enhanced expression under -P stress. Here, we show that LaMATE had high expression in proteoid roots not only under -P, but also under -Fe, -N, -Mn and +Al stress. A portion containing the putative LaMATE promoter was fused to GUS and enhanced green fluorescence protein (EGFP) reporter genes, and a translational LaMATE::EGFP fusion was constructed under control of the LaMATE promoter. The LaMATE promoter directed P-dependent GUS and EGFP expression to proteoid roots. Confocal microscopy in white lupin and Arabidopsis point to the plasma membrane as the likely location of the LaMATE protein. LaMATE displayed homology to FRD3 in Arabidopsis, but did not complement an Arabidopsis ferric reductase defective 3 (FRD3) mutant. RNA(i)-based gene silencing was shown to effectively reduce LaMATE expression in transformed white lupin roots. LaMATE RNAi-silenced plants displayed an about 20% reduction in dry weight.     

Synaptic activity regulates interstitial fluid amyloid-beta levels in vivo

Aggregation of the amyloid-beta (Abeta) peptide in the extracellular space of the brain is central to Alzheimer's disease pathogenesis. Abeta aggregation is concentration dependent and brain region specific. Utilizing in vivo microdialysis concurrently with field potential recordings, we demonstrate that Abeta levels in the brain interstitial fluid are dynamically and directly influenced by synaptic activity on a timescale of minutes to hours. Using an acute brain slice model, we show that the rapid effects of synaptic activity on Abeta levels are primarily related to synaptic vesicle exocytosis. These results suggest that synaptic activity may modulate a neurodegenerative disease process, in this case by influencing Abeta metabolism and ultimately region-specific Abeta deposition. The findings also have important implications for treatment development.     

Cortical localization of the Galpha protein GPA-16 requires RIC-8 function during C. elegans asymmetric cell division

Understanding of the mechanisms governing spindle positioning during asymmetric division remains incomplete. During unequal division of one-cell stage C. elegans embryos, the Galpha proteins GOA-1 and GPA-16 act in a partially redundant manner to generate pulling forces along astral microtubules. Previous work focused primarily on GOA-1, whereas the mechanisms by which GPA-16 participates in this process are not well understood. Here, we report that GPA-16 is present predominantly at the cortex of one-cell stage embryos. Using co-immunoprecipitation and surface plasmon resonance binding assays, we find that GPA-16 associates with RIC-8 and GPR-1/2, two proteins known to be required for pulling force generation. Using spindle severing as an assay for pulling forces, we demonstrate that inactivation of the Gbeta protein GPB-1 renders GPA-16 and GOA-1 entirely redundant. This suggests that the two Galpha proteins can activate the same pathway and that their dual presence is normally needed to counter Gbetagamma. Using nucleotide exchange assays, we establish that whereas GPR-1/2 acts as a guanine nucleotide dissociation inhibitor (GDI) for GPA-16, as it does for GOA-1, RIC-8 does not exhibit guanine nucleotide exchange factor (GEF) activity towards GPA-16, in contrast to its effect on GOA-1. We establish in addition that RIC-8 is required for cortical localization of GPA-16, whereas it is not required for that of GOA-1. Our analysis demonstrates that this requirement toward GPA-16 is distinct from the known function of RIC-8 in enabling interaction between Galpha proteins and GPR-1/2, thus providing novel insight into the mechanisms of asymmetric spindle positioning.     

Quantitative genetics of natural variation of behavior in Drosophila melanogaster: the possible role of the social environment on creating persistent patterns of group activity

Using a set of nine effectively isogenic lines collected from nature in 1998, we observed unperturbed behaviors of mixed-sex groups of Drosophila melanogaster. We repeatedly scanned replicated groups of genetically identical individuals, five females and five males, and recorded the behavior of each individual (i.e., walking, feeding, grooming, flying, courting, mating, fighting, or resting). From these behaviors, we made a composite variable of activity for our quantitative genetic analysis. Genotypes differed in activity, explaining 14.41% of the variation in activity; 8.60% of the variation was explained by a significant genotype x sex interaction, which signifies genetic variation for sexual dimorphism in behavior. Phenotypic plasticity explained 11.13% of the variation in activity. Different genotypes and sexes within genotypes had different rank orders of the component behaviors that contribute to activity. We found no effect of common rearing environment. Instead, differences between replicate groups within genotype accounted for 19.47% variation in activity, and activity was significantly repeatable across scans. This emergent group behavior is likely caused by differences between groups of interacting individuals, even though individuals were genetically identical across groups. Thus, emergent group behavior explained almost as much variation in activity as the combined sources of genetic variation (23.01%), and this is an additional level on which selection could operate: individuals and groups. We discuss how differences among groups could change patterns of additive genetic variation available for evolution. Furthermore, because the behavior of an individual is influenced by conspecifics, genotype interactions between individuals could contribute to indirect selection. Finally, if we consider activity as a syndrome governing all component behaviors with strong genetic correlations among behaviors within an individual, then these component behaviors cannot evolve independently. These results suggest that reductionist approaches of molecular behavior genetics may be incomplete and/or misleading when considering similar phenotypes at the population level or when trying to understand how behaviors evolve.