In the eukaryotic cell nucleus, spatial organization and dynamics of the genome is important in the regulation of gene expression. This thesis describes the use of the fission yeast, Schizosaccharomyces pombe, to study chromatin regulation and dynamics. We used nitrogen starvation to induce transcription of genes in fission yeast cells. In induced genes, nucleosomes get evicted in both the promoter and in the open reading frame (ORF). In the genes with the highest expression more nucleosomes get evicted from the ORF than from the promoter. This indicates that large rearrangements of the chromatin are occurring during a drastic gene induction. Many of the genes that become expressed early after nitrogen starvation are located together in clusters. In a cell where nitrogen is present in the surrounding media the gene clusters locate close to the nuclear periphery. When the nitrogen source is removed from the media, the clusters move to a more internal position. Thus rearrangement of chromatin due to gene induction, described in the first study, is accompanied by subnuclear changes of localization. Another type of regulation is the silencing of genes. We have studied a factor necessary for correct repression of genes located in silent chromatin, in S. pombe. The protein, Clr2, is part of the SHREC complex containing a remodeler (Mit1) and a histone deacetylase (Clr3). By bioinformatic analysis of Clr2 and newly sequenced fungi genomes, three motifs were identified. To gather more information about important parts of the Clr2 protein, deletions were made. When removing from about 20 to 100 amino acids in the middle of the protein, silencing of a reporter gene inserted at the mating-type region, inner repeats of centromere 1 and at the central core of centromere 2, failed. This indicates that Clr2 has an important role in establishing silent chromatin.

The heterogeneous nuclear ribonucleoproteins (hnRNPs) are a group of proteins intitially characterised in the late 1980’s by their presence in complexes that form on nascent RNA transcripts. This definition was purely operational, and was based on protein isolation techniques available at that time. Since then, the tendency to refer to and view the hnRNPs as a protein family has become increasingly prevalent, although there has been no systematic sequence- or structure-based study of their evolutionary history. While the hnRNPs share some structural characteristics (modular structure, presence of RNA-binding domains) and functional properties (binding to RNA, involvement in multiple steps of RNA processing), these criteria also apply to other types of RNA binding proteins (RBPs), such as the SR and ELAV families of proteins. Thus, we have adopted a more methodical and rigorous approach to the classification of hnRNPs and other RBPs, through the phylogenetic analysis of their sequences and domains. Besides establishing phylogenetic relationships and simplifying nomenclature, studying the evolutionary divergence of the hnRNPs is important for understanding their functional features. The hnRNP A/B subfamily is comprised of paralogues A1, A2/B1, A3 and A0, which exhibit a high level of similarity at both the sequence and structural level. While they are often treated as functional homologues, they are not functionally identical. Hence, we undertook a detailed comparison of their sequences, and found that the introduction of novel splicing signals or mutation of existing sequence elements has led to changes in alternative splicing patterns between the paralogues, which may affect the regulation of their expression and their RNA binding properties. In addition, we also investigated species-specific alternative splicing of the hnRNPs A/B, which has significant implications for the interpretation of current research, since different research groups tend to use different model organisms in their experiments. Hence, exploration of the sequence divergence of the hnRNPs A/B has provided some clues as to how their functional differences arose, and also highlighted the need to take species-specific splicing into consideration. Alternative splicing can create functional variation not only between paralogues, but also between splice variants. hnRNP A2/B1, which has a well-established role in mRNA trafficking in neuronal cells, has four spliceoforms. In order to study the contribution of each isoform to this process, we investigated isoform-specific variations in intracellular localisation, and expression in different developmental stages and species. We found that in rat, minor isoform A2b was the predominant isoform in the cytoplasm, and may be the key player in mRNA trafficking. These findings demonstrate the importance of considering individual isoforms (including those expressed in low abundance) when studying the function of alternatively spliced proteins, especially when the function is restricted to a particular subcellular compartment. In addition to its cytoplasmic role in mRNA trafficking, hnRNP A2/B1, and the other hnRNPs A/B, have multiple nuclear functions, including packaging of nascent transcripts, nuclear export of mRNA, regulation of alternative splicing and telomere maintenance. These processes take place in discrete regions within the nucleus, and thus we examined the subnuclear distribution of the hnRNPs A/B. We found that hnRNP A1 had a localisation pattern distinct from that of A2/B1 and A3, and that these patterns were spatially and temporally regulated. Hence, the evolutionary divergence of the hnRNPs A/B has affected the localisation, expression and splicing patterns of these proteins, which we have examined at multiple levels, including comparisons across all hnRNPs, within the hnRNP A/B paralogues, and between the hnRNP A2/B1 splice variants. As the hnRNPs A/B are involved in almost every step in RNA processing, this functional diversity has significant implications for transcriptomic complexity. Furthermore, our findings highlight the importance of taking species- and isoform-specific differences into account when investigating protein function. In conclusion, this study of the hnRNPs A/B provides a conceptual framework for exploring the relationships between sequence, structural and functional divergence, which may be applicable to protein families in general.

The heterogeneous nuclear ribonucleoproteins (hnRNPs) are a group of proteins intitially characterised in the late 1980’s by their presence in complexes that form on nascent RNA transcripts. This definition was purely operational, and was based on protein isolation techniques available at that time. Since then, the tendency to refer to and view the hnRNPs as a protein family has become increasingly prevalent, although there has been no systematic sequence- or structure-based study of their evolutionary history. While the hnRNPs share some structural characteristics (modular structure, presence of RNA-binding domains) and functional properties (binding to RNA, involvement in multiple steps of RNA processing), these criteria also apply to other types of RNA binding proteins (RBPs), such as the SR and ELAV families of proteins. Thus, we have adopted a more methodical and rigorous approach to the classification of hnRNPs and other RBPs, through the phylogenetic analysis of their sequences and domains. Besides establishing phylogenetic relationships and simplifying nomenclature, studying the evolutionary divergence of the hnRNPs is important for understanding their functional features. The hnRNP A/B subfamily is comprised of paralogues A1, A2/B1, A3 and A0, which exhibit a high level of similarity at both the sequence and structural level. While they are often treated as functional homologues, they are not functionally identical. Hence, we undertook a detailed comparison of their sequences, and found that the introduction of novel splicing signals or mutation of existing sequence elements has led to changes in alternative splicing patterns between the paralogues, which may affect the regulation of their expression and their RNA binding properties. In addition, we also investigated species-specific alternative splicing of the hnRNPs A/B, which has significant implications for the interpretation of current research, since different research groups tend to use different model organisms in their experiments. Hence, exploration of the sequence divergence of the hnRNPs A/B has provided some clues as to how their functional differences arose, and also highlighted the need to take species-specific splicing into consideration. Alternative splicing can create functional variation not only between paralogues, but also between splice variants. hnRNP A2/B1, which has a well-established role in mRNA trafficking in neuronal cells, has four spliceoforms. In order to study the contribution of each isoform to this process, we investigated isoform-specific variations in intracellular localisation, and expression in different developmental stages and species. We found that in rat, minor isoform A2b was the predominant isoform in the cytoplasm, and may be the key player in mRNA trafficking. These findings demonstrate the importance of considering individual isoforms (including those expressed in low abundance) when studying the function of alternatively spliced proteins, especially when the function is restricted to a particular subcellular compartment. In addition to its cytoplasmic role in mRNA trafficking, hnRNP A2/B1, and the other hnRNPs A/B, have multiple nuclear functions, including packaging of nascent transcripts, nuclear export of mRNA, regulation of alternative splicing and telomere maintenance. These processes take place in discrete regions within the nucleus, and thus we examined the subnuclear distribution of the hnRNPs A/B. We found that hnRNP A1 had a localisation pattern distinct from that of A2/B1 and A3, and that these patterns were spatially and temporally regulated. Hence, the evolutionary divergence of the hnRNPs A/B has affected the localisation, expression and splicing patterns of these proteins, which we have examined at multiple levels, including comparisons across all hnRNPs, within the hnRNP A/B paralogues, and between the hnRNP A2/B1 splice variants. As the hnRNPs A/B are involved in almost every step in RNA processing, this functional diversity has significant implications for transcriptomic complexity. Furthermore, our findings highlight the importance of taking species- and isoform-specific differences into account when investigating protein function. In conclusion, this study of the hnRNPs A/B provides a conceptual framework for exploring the relationships between sequence, structural and functional divergence, which may be applicable to protein families in general.

No description available.

In eukaryotes the genome adopts a non-random spatial organisation, which is important for gene regulation. However, very little is known about the driving forces behind nuclear organisation. In the simple model eukaryote fission yeast, Schizosaccharomyces pombe, it has been known for a long time that transcriptionally repressed heterochromatin localise to the nuclear membrane (NM); the centromeres attaches to spindle pole body (SPB), while the telomeres are positioned at the NM on the opposite side of the nucleus compared to the SPB. Studies presented in this thesis aimed at advancing our knowledge of nuclear organisation in Schizosaccharomyces pombe. We show that the heterochromatic mating-type region localises to the NM in the vicinity of the SPB. This positioning was completely dependent on Clr4, a histone methyl transferase crucial for the formation of heterochromatin. Additional factors important for localisation were also identified: the chromo domain protein Swi6, and the two boundary elements IR-L and IR-R surrounding this locus. We further identify two other chromo domain proteins; Chp1 and Chp2, as crucial factors for correct subnuclear localisation of this region. From these results we suggest that the boundary elements together with chromodomain proteins in balanced dosage and composition cooperate in organising the mating-type chromatin. Gene regulation can affect the subnuclear localisation of genes. Using nitrogen starvation in S. pombe as a model for gene induction we determined the subnuclear localisation of two gene clusters repressed by nitrogen: Chr1 and Tel1. When repressed these loci localise to the NM, and this positioning is dependent on the histone deacetylase Clr3. During induction the gene clusters moved towards the nuclear interior in a transcription dependent manner. The knowledge gained from work presented in this thesis, regarding nuclear organisation in the S. pombe model system, can hopefully aid to a better understanding of human nuclear organisation.