Constitutive nuclear import and stress-regulated nucleocytoplasmic shuttling of mammalian heat-shock factor 1.

Inducible expression of major cytosolic and nuclear chaperone proteins is mediated by the heat-shock transcription factor HSF1 that is activated by derepressive mechanisms triggered by transient heat stress and sustained proteotoxicity. Despite progress in defining essential aspects of HSF1 regulation, little is known about the cellular dynamics enabling this factor to mediate gene responses to cytosolic stress signals. We report that the inactive, stress-responsive form of HSF1 accumulates in the nucleus due to a relatively potent import signal, which can be recognized by importin-alpha/beta, and simultaneously undergoes continuous nucleocytoplasmic shuttling due to a comparatively weak, nonetheless efficient, export activity not involving the classical exportin-1 pathway. Strikingly, experimental stresses at physiological or elevated temperature reversibly inactivate the export competence of HSF1. Likewise, mutations mimicking stress-induced derepression impair export but not import. These findings are consistent with a dynamic process whereby exported molecules that are derepressed in an inductive cytosolic environment are recollected and pause in the nucleoplasm, replacing progressively the inactive pool. While steady-state nuclear distribution of the bulk of HSF1 ensures a rapid gene response to acute heat stress, our results suggest that the capture in the nucleus of molecules primed for activation in the cytosol may underlie responses to sustained proteotoxicity.

A small-scale survey identifies selective and quantitative nucleo-cytoplasmic shuttling of a subset of CREM transcription factors.

Elucidating dynamic aspects of intracellular localization of proteins is essential to decipher their functional interaction networks. Although transcription factors lacking a detectable cytoplasmic fraction have been generally considered compartmentalized in the nucleus, some were found to shuttle into the cytoplasm, suggesting functional interactions therein. To further investigate how common, specific and quantitative is this traffic, we have employed the heterokaryon assay for a small-scale survey of nuclear factors not previously tested for their nucleo-cytoplasmic motion. We show that a subset of cAMP response element (CRE) binding proteins of the CREM type shuttles within a biologically meaningful time frame, revealing a continuous flow into the cytoplasm that persists during signaling. Their dynamic behavior, not involving the classical Exportin-1 pathway, could be ascribed to C-terminal sequences, containing, in addition to the bZIP domain and the NLS, a nuclear export activity and an inhibitory activity at an adjacent site. Other proteins examined in this study either did not shuttle significantly or, like CREB and distinct CREM isoforms, shuttled with markedly delayed kinetics, denoting considerable selectivity of this traffic. These findings raise the possibility that events associated with bi-directional transport and periodic transit through the cytoplasm may modulate activities of select nuclear transcription factors.

The transcriptional corepressor MTG16a contains a novel nucleolar targeting sequence deranged in t (16; 21)-positive myeloid malignancies.

The MTG (Myeloid Translocation Gene) proteins are a family of novel transcriptional corepressors. We report that MTG16a, a protein isoform encoded by the MTG16 gene deranged by the t (16; 21) in myeloid malignancies, is targeted to the nucleolus. The amino acid sequence necessary for nucleolar localization was mapped to the MTG16a N-terminal region. MTG16a, like MTG8, the nuclear corepressor deranged by the t (8; 21), is capable to interact with specific histone deacetylases (HDACs) suggesting that the protein may mediate silencing of nucleolar gene transcription. In addition, MTG16a is capable to form oligomers with other MTG proteins. As a consequence of the t (16; 21) the AML1 DNA-binding domain replaces the MTG16a N-terminal region. The AML1-MTG16 fusion protein is targeted to the nucleoplasm where it is capable to oligomerize with MTG16a and interact with HDAC1 and HDAC3. The deficiency of HDAC-containing complexes at nucleolar sites and the accumulation of HDAC-containing complexes at AML1-sites may be critical in the pathogenesis of t (16; 21) myeloid malignancies.