ABSTRACT
Small nuclear ribonucleoproteins (snRNPs) represent key constituents of major and minor spliceosomes. snRNPs contain a common core, composed of seven Sm proteins bound to snRNA, which forms in a step-wise and factor-mediated reaction. The assembly chaperone pICln initially mediates the formation of an otherwise unstable pentameric Sm protein unit. This so-called 6S complex docks subsequently onto the SMN complex, which removes pICln and enables the transfer of pre-assembled Sm proteins onto snRNA. X-ray crystallography and electron microscopy was used to investigate the structural basis of snRNP assembly. The 6S complex structure identifies pICln as an Sm protein mimic, which enables the topological organization of the Sm pentamer in a closed ring. A second structure of 6S bound to the SMN complex components SMN and Gemin2 uncovers a plausible mechanism of pICln elimination and Sm protein activation for snRNA binding. Our studies reveal how assembly factors facilitate formation of RNA-protein complexes in vivo.
Subject(s)
Drosophila Proteins/chemistry , Ion Channels/chemistry , snRNP Core Proteins/chemistry , Amino Acid Sequence , Animals , Crystallography, X-Ray , Drosophila melanogaster , Humans , Hydrogen Bonding , Mice , Microscopy, Electron , Models, Molecular , Protein Multimerization , Protein Structure, Quaternary , Protein Structure, Secondary , Protein Structure, Tertiary , Xenopus Proteins/chemistry , Xenopus laevis , snRNP Core Proteins/ultrastructureABSTRACT
The three-dimensional (3D) structure of the genome is organized non-randomly and plays a role in genomic function via epigenetic mechanisms in the eukaryotic nucleus. Here, we analyzed the spatial positioning of three target regions; the SNRPN, UBE3A, and GABRB3 genes on human chromosome 15q11.2-q12, a representative cluster of imprinted regions, in the interphase nuclei of B lymphoblastoid cell lines, peripheral blood cells, and skin fibroblasts derived from normal individuals to look for evidence of genomic organization and function. The positions of these genes were simultaneously visualized, and all inter-gene distances were calculated for each homologous chromosome in each nucleus after three-color 3D fluorescence in situ hybridization. None of the target genes were arranged linearly in most cells analyzed, and GABRB3 was positioned closer to SNRPN than UBE3A in a high proportion of cells in all cell types. This was in contrast to the genomic map in which GABRB3 was positioned closer to UBE3A than SNRPN. We compared the distances from SNRPN to UBE3A (SU) and from UBE3A to GABRB3 (UG) between alleles in each nucleus, 50 cells per subject. The results revealed that the gene-to-gene distance of one allele was longer than that of the other and that the SU ratio (longer/shorter SU distance between alleles) was larger than the UG ratio (longer/shorter UG distance between alleles). The UG distance was relatively stable between alleles; in contrast, the SU distance of one allele was obviously longer than the distance indicated by the genome size. The results therefore indicate that SNRPN, UBE3A, and GABRB3 have non-linear and non-random curved spatial positioning in the normal nucleus, with differences in the SU distance between alleles possibly representing epigenetic evidence of nuclear organization and gene expression.
Subject(s)
Cell Nucleus/ultrastructure , Chromosomes, Human, Pair 15/ultrastructure , Imaging, Three-Dimensional/methods , In Situ Hybridization, Fluorescence/methods , Receptors, GABA-A/ultrastructure , Ubiquitin-Protein Ligases/ultrastructure , snRNP Core Proteins/ultrastructure , B-Lymphocytes , Chromosome Mapping , Chromosomes, Human, Pair 15/genetics , Female , Fibroblasts , Humans , Male , Microscopy, Confocal , Receptors, GABA-A/genetics , Ubiquitin-Protein Ligases/genetics , snRNP Core Proteins/geneticsABSTRACT
A model based method for the accurate quantification of the 3D structure of fluorescently labelled cellular objects similar in size to the optical resolution limit is presented. This method is applied to both simulated confocal images of chromatin structures and to real confocal data obtained on a Fluorescence in situ Hybridization (FISH) labelled gene domain. The model assumes that the object is composed of a small number of discrete points which are convolved with the microscope point spread function to give the image. Fitting this model to image data results in a method to assess object structure which is accurate, shows a low bias, and does not require user intervention or the potentially subjective setting of a threshold.
