Dual requirement for yeast hnRNP Nab2p in mRNA poly(A) tail length control and nuclear export.

Recent studies of mRNA export factors have provided additional evidence for a mechanistic link between mRNA 3'-end formation and nuclear export. Here, we identify Nab2p as a nuclear poly(A)-binding protein required for both poly(A) tail length control and nuclear export of mRNA. Loss of NAB2 expression leads to hyperadenylation and nuclear accumulation of poly(A)(+) RNA but, in contrast to mRNA export mutants, these defects can be uncoupled in a nab2 mutant strain. Previous studies have implicated the cytoplasmic poly(A) tail-binding protein Pab1p in poly(A) tail length control during polyadenylation. Although cells are viable in the absence of NAB2 expression when PAB1 is overexpressed, Pab1p fails to resolve the nab2Delta hyperadenylation defect even when Pab1p is tagged with a nuclear localization sequence and targeted to the nucleus. These results indicate that Nab2p is essential for poly(A) tail length control in vivo, and we demonstrate that Nab2p activates polyadenylation, while inhibiting hyperadenylation, in the absence of Pab1p in vitro. We propose that Nab2p provides an important link between the termination of mRNA polyadenylation and nuclear export.

Immunoelectron microscopic localization of estrogen receptor on pre-mRNA containing constituents of rat uterine cell nuclei.

The localization and quantitative changes of estradiol receptor (ER) were studied by means of immunogold-electron microscope methods using a polyclonal antibody directed against an amino acid sequence representing the DNA binding site of ER, a monoclonal antibody against hnRNP core protein, and anti-DNA antibody. The uteri of normal rats in estrus and those of ovariectomized females were used. Ovariectomized rats were studied 21 days after surgery at different times after the injection of normal saline or estradiol-17 beta. The density of labeling was measured in interchromatin space, compact chromatin, nucleolus, cytoplasm, and background of epithelial cells, muscle cells, and fibroblasts. In the three types of cells ER was found mainly on extranucleolar ribonucleoprotein (RNP) fibrils. In epithelial and muscle cells the nucleolus was labeled but compact chromatin was not labeled. In epithelial cells there was a low but significant labeling of the cytoplasm. Fibroblasts exhibited a low labeling of the compact chromatin. Ovariectomy did not change these distributions. The estradiol injection increased labeling in all compartments of epithelial and muscle cells but decreased the labeling of compact chromatin of fibroblasts. These results show: (a) that ER is mainly nuclear but it is also present in the cytoplasm, (b) that ER binds to the nuclear particles containing newly synthesized RNA, and (c) that the binding to RNPs does not block the DNA binding domain of the ER.

The C proteins of heterogeneous nuclear ribonucleoprotein complexes interact with RNA sequences downstream of polyadenylation cleavage sites.

The heterogeneous nuclear ribonucleoprotein C1 and C2 proteins were preferentially cross-linked by treatment with UV light in nuclear extracts to RNAs containing six different polyadenylation signals. The domain required for the interaction was located downstream of the poly(A) cleavage site, since deletion of this segment from several polyadenylation substrate RNAs greatly reduced cross-linking efficiency. In addition, RNAs containing only downstream sequences were efficiently cross-linked to C proteins, while fully processed, polyadenylated RNAs were not. Analysis of mutated variants of the simian virus 40 late polyadenylation signal showed that uridylate-rich sequences located in the region between 30 and 55 nucleotides downstream of the cleavage site were required for efficient cross-linking of C proteins. This downstream domain of the simian virus 40 late poly(A) addition signal has been shown to influence the efficiency of the polyadenylation reaction. However, there was not a strict correlation between cross-linking of C proteins and the efficiency of polyadenylation.

A model of senescence as the piecewise loss of control over transcription.

Given that the transcription of DNA into heterogeneous nuclear RNA (hnRNA) is controlled for each species of hnRNA by an on-off switch at the beginning of each hnRNA template, than control over transcription may be independently lost over each control site. If an incorrect switch setting is conserved during replication and mitosis after the pattern of conservation of DNA-binding molecules and DNA bound histones, then the errors are heritable and can accumulate from one cell generation to the next. This type of control system failure is hypothesized to be the basis of senescence. This theory easily explains the Hayflick limit for cultured cells, the link between ageing and cancer and the reason why haploid animals are much rarer than haploid plants. This theory predicts that a haploid animal will accumulate senescence at over 100 times the normal rate, and therefore that obtaining a haploid animal will be virtually impossible. The implications for the control of senescence and possible strategies for preventing or repairing switch mis-settings are discussed.

The nature and nurture of epithelial-mesenchymal interactions during tooth morphogenesis.

Epithelial-mesenchymal interactions during tooth morphogenesis are inductive and instructive developmental processes as well as permissive and regulatory processes. Data is available to support the early influences of enamel organ epithelium upon a responding mesenchyme in the determination of dental morphogenetic fields (Dryburg, 1967; Miller, 1969). Mesenchymal specificity appears to be operant during tooth shape and form and during the induction of secretary amelogenesis (Kollar, 1972). These heterotypic tissue interactions can be observed in vivo and in vitro. The cellular responses to these interactions appear to be transcriptional, translational and post-translational; as a direct consequence of the interactions, new gene products are synthesized and secreted and/or pre-existing gene products are amplified (Hata and Slavkin, 1978). The mechanism(s) by which epithelial-mesenchymal interactions function may best be learned through critical investigations of differentiation alloantigens, receptors, coupling components within the plasma membrane, translating components by which epigenetic external cues become internal chemical information, and the associations between peripheral and integral proteins within the plasma membrane and intracytoplasmic microfilaments and microtubules.