Chromatin structure and gene expression

Physical, biochemical, and genetic factors that play roles in the organization and expression of the genome were reported in this symposium. The technologies varied from genetic dissection using the subtleties of mutational analysis of position effect variegation in Drosophila (M. Hearn, Seattle; A. Hedrick, Vancouver), to the microdissection and cloning of lampbrush chromosome domains (N. Angelier, France) and the isolation of the mammalian telomere DNA sequence (B. Moyzis, Los Alamos). In addition, the distribution of the high mobility group proteins (HMG 14 and HMG T) was explored immunologically under changing conditions of gene expression (H. Trepte, Marburg), and the effect of DNA sequence positioning with respect to the local environment was explored by transformation (B. Chatterjee, Philadelphia) and through the genotype as seen in the maize ribosomal genes (E. Jupe, Baton Rouge). These studies collectively deal with a number of critical issues in the expression of the genome. It is evident that the expression is tightly coupled to the structure of the chromatin and that this is under strict genetic control; hence, the study of structure or expression must be integrated to further our understanding of the interplay in this tightly integrated process. The symposium began with the studies of N. Anglier on the microdissection and cloning of newt lampbrush bivalent VII globular loops. From the 2000 isolated loops, a total of 32 recombinant clones were identified among 200 plaques. These ranged from 2 -7 kb, and a detailed analysis was reported on 3 of them. The striking feature of the clones PW-14, -1 11, and -130 was their in situ localization. Although they were derived from a single bivalent and exhibited (in the case of PW 11 1) a single band on Southern analysis, they hybridized to all bivalents that had the globular domains. The basis for this observation remains unknown and is the subject of continued investigation. It seems reasonable to speculate that the globular morphology of the loops is related to their expression, and the conditions of in situ hybridization have not differentiated between the genomic coding sequences and bound ribonucleoproteins or RNA. It remains to be seen if these are transcripts and, if so, if they play a role in development. Hans Trepte took an immunological approach to the study of the spatial organization of HMG 14 and HMG T during the cell cycle and in response to various modulators of gene expression. The results with both antibodies were similar and were consistent with the hypothesis that these proteins are in some way involved in active gene expression. Four separate observations support these conclusions: (i) nuclear localization; (ii) they are present only in transcriptionally active nuclei and disappear as nuclei become transcriptionally quiescent (on interphase but not metaphase chromosomes); (iii) they are present in polytene puffs and loops and appear and disappear in co~junction with these morphologically identifiable structures; (iv) puff inactivation by inhibitors (actinomycin D) or heat shock does not remove the immunoreactivity. There is a clear distribution of these HMG proteins in regions of gene activity, while their function in the processes of gene expression or possibly chromatin structure remain obscure. The investigation of the methylation patterns of the ribosomal gene repeat units in two polymorphic inbred maize lines was camed out by Eldon Jupe. The chromosome that camed the additional EcoRI site (8-kb morph) exhibited markedly enhanced sensitivity to HpaII consistent with a reduced methylation and active gene expression. The methylation pattern within the region was not uniform with under methylation in the intergenic spacer region. From studies with oligonucleotide probes specific for the region that surrounds the EcoRI site, it appears that this polymorphism marks the ribosomal arrays that are preferentially expressed. Hence, although both gene structures have the potential to be expressed in their parent lines, there is preferential transcription in the hybrids, which is consistent with the concept that expression can be under more global controls. The joint presentations of M. Hearn and A. Hedrick on the modulation of position effect variegation (PEV) by dominant mutants in Drosophila was possibly the most intriguing of the talks. Owing to the limited understanding of most geneticists about the subtleties of the position effect phenomenon, the implications of the studies may not have had the impact they deserved. Two systems in which PEV can be explored were used. The alteration of expression of either heterochromatic or euchromatic loci placed in juxtaposition with the opposite type of chromatin organization was quantified for eye phenotypes by measuring total eye pigments. Enhancers of the "light" gene (heterochromatic location) exhibit a reciprocal effect on the expression the euchromatically located loci, while suppressors of euchromatic gene variegation fall into two classes: those that enhance the suppression of light and those that have no effect. Taken as a whole these studies and others from T. Grigliatti's laboratory in Vancouver (poster session 3 1.17) provide compelling evidence that the packaging of the chromatin is under genetic control and that the euchromatic and heterochromatic domains are in a dynamic balance. The introduction of foreign DNA (HSV 7k) into cultured L cells was explored by Bishwanath Chatterjee to study the effect of the local environment on the expression of the 7k gene. In transformants with two intact 7k genes adjacently located, only one appears to be transcribed. The gene pair studied was imbedded in mouse centromeric satellite sequences and contained a stretch of vector DNA. By using methylation sensitive restriction techniques, the expression of each 7k gene was correlated with its degree of methylation. The gene with mouse satellite immediately upstream was hypermethylated and inactive, while the one with several kilobases of vector 5' adjacent to it was hypomethylated and active. The spreading effects of the centromeric satellite sequences on methylation appear to be restricted. The lack of an effect on the second copy of the 7k gene was interpreted as a short-range phenomenon, although it