Wilhelm bernhard workshop on the cell nucleus

Istituto di Genetica Molecolare del CNR, Pavia, Italy

Nuclear stress bodies are transient structures elicited by stress treatments such as heat shock or exposure to heavy metals. The assembly of these structures initiates with the recruitment of heat shock factor 1 (HSF1). Thereafter a subset of RNA processing factors, including hnRNP M, hnRNP HAP/SAF-B, ASF/SF2, SRp30c, Sam68, is recruited. Numerous data suggest that these bodies are not sites of transcription but rather depots for RNA molecules synthesized before stress. We have recently shown that the pericentromeric heterochromatic regions of human chromosome 9, 12 and 15 act as recruiting centers on which these structures are assembled. Surprisingly, stress bodies are not stained by DAPI or by antibodies against HP1, a typical marker of heterochromatin domains. In contrast, and unexpectedly, they are recognized by antibodies against the hyper-acetylated form of histone H4. In fact they correspond to the main sites of accumulation of this histone form in stressed cells. This result suggests that the higher order chromatin organization of heterochromatic regions of HSA9, 12 and 15 is drastically altered in response to stress, probably due to the recruitment of some, as yet unknown, histone acetylase by HSF1. In order to dissect the molecular mechanism underlying the formation of these structures we have determined the protein domain that mediates the recruitment of ASF/SF2. The recruitment depends on the second RRM of the protein. Amino acid substitutions that abrogate the recruitment also modify the activity of this factor in alternative splicing. A two-hybrid screening for proteins interacting with ASF/SF2 is currently in progress.

SIGNAL SEQUENCES IN THE MOLECULE OF THE NUCLEOLAR PROTEIN FIBRILLARIN DIRECTING ITS SPECIFIC LOCALIZATION WITHIN THE NUCLEOLUS

^1,3Mukharyamova K.Sh., ²Levitsky S.A., ²Veiko V.P., and ^1,3Zatsepina O.V.

¹Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry RAN, ²Institute of Genetics and Selection of Industrial Microorganisms of The Ministry of Economics, and ³A.N. Belozersky Institute of Physical and Chemical Biology, Moscow State University, Moscow, Russia
Fibrillarin (321 amino acids in human) is a key nucleolar protein that is essential for posttranscriptional modifications of rRNA transcripts, including pre-rRNA processing, pre-rRNA methylation and ribosome assembly. Human fibrillarin has three structural domains. NH₂-terminus (~80 amino acids) contains a glycine- and arginine-rich domain (GAR-domain). A central region (~90 amino acids) resembles an RNA-binding (RB) domain that is also present in many small nuclear snRNP proteins. The COOH-terminus contains a small (~30 amino acids) alpha helix domain and apparently has a methyltransferase activity. The domains are separated by the spacer regions (~50 amino acids each), called 1^st and 2^nd. In the present work we studied implication of these domains in specific localization of fibrillarin in the foci of early rRNA synthesis, which can be visualized by run-on transcription assay with BrUTP as precursor. A plasmid encoding full-length fibrillarin/GFP (green fluorescent protein) was kindly donated by Dr. M.O.J. Olson (UMMC, USA) and used for generation of mutant fibrillarin constructs fused to GFP, including pΔ2α/GFP, pGAR1/GFP, pΔRB/GFP, p2α/GFP, pSRB/GFP. HeLa cells were transiently transfected of these plasmids and distribution of fibrillarin/GFP was recorded by fluorescence microscopy after 15-48 h and co-localized with endogenous fibrillarin. Our data showed that a deletion or mutation in any known fibrillarin domain inhibits protein specific (i.e., within the Br-RNA positive foci) localization within the nucleolus. However, together GAR-domain and the 1^st spacer were sufficient to target fibrillarin to the nucleolus. The results are discussed in terms of relevant literature data.

The research was supported by the Russian Foundation of Basic Researches (grant 03-04-48951).

MOLECULAR ARCHITECTURE OF THE SYNAPTONEMAL COMPLEX: SCP3 PROTEIN DOMAINS THAT ARE ESSENTIAL FOR POLYMERIZATION

Baier, A., Alsheimer, M., and Benavente, R.

Biocenter, University of Würzburg, , Germany
The synaptonemal complex (SC) is a karyoskeletal structure of meiotic cells that is critically involved in synaptic pairing and segregation of homologous chromosomes. The SC has a tripartite structure with a central region and two lateral elements (at which the chromatin of the respective homolog is attached). One major structural component of lateral elements of the mammalian SC is protein SCP3 (Lammers et al., 1994). This protein is composed of three domains: a central alpha-helical domain that has a length of 135 amino acids and is flanked by non-helical N- and C-termini which are 116 and 6 amino acids long, respectively. In a previous study (Yuan et al., 1998) it has been shown that both the N-terminus and the alpha-helix plus the C-terminus are required for SCP3 polymerization. For a more detailed study of the polymerization properties of this protein, we have transfected COS-7 cells with different SCP3 constructs. Here we show that most of the N-terminus is dispensable for proper polymerization of SCP3 molecules. In contrast, polymerization could be impaired if parts of the alpha-helix were deleted. We also identified two short sequences flanking the alpha-helix that are essential for polymerization, namely amino acids 87 to 105 of the N-terminus and the six amino acids of the C-terminus. Deletion of one of these two domains impairs SCP3 polymerization. The relevance of these two domains is emphasized by our analysis of data bases which showed that they are the best conserved SCP3 sequences throughout vertebrate evolution.

References:

Lammers J.H.M. et al.: Mol. Cell. Biol. 14:1137-1146, 1994